ANSI/HI 9.

8-2018

American National Standard for

Rotodynamic Pumps
for Pump Intake Design

`---
` `
`
````-`-`
```
--`
 ANSI/HI 9.8–2018

American National Standard for

Rotodynamic Pumps
for Pump Intake Design

Sponsor
Hydraulic Institute
www.Pumps.org

Approved January 8, 2018

American National Standards Institute, Inc.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

American Approval of an American National Standard requires verification by ANSI that the requirements for
due process, consensus and other criteria for approval have been met by the standards developer.
National
Consensus is established when, in the judgment of the ANSI Board of Standards Review,
Standard substantial agreement has been reached by directly and materially affected interests.
­
Substantial agreement means much more than a simple majority, but not necessarily unanimity.
Consensus requires that all views and objections be considered, and that a concerted effort be
made toward their resolution.

The use of American National Standards is completely voluntary; their existence does not in
any respect preclude anyone, whether he has approved the standards or not, from manufacturing,
marketing, purchasing, or using products, processes, or procedures not conforming to the
standards.

The American National Standards Institute does not develop standards and will in no circum-
stances give an interpretation of any American National Standard. Moreover, no person shall
have the right or authority to issue an interpretation of an American National Standard in the
name of the ­American National Standards Institute. Requests for interpretations should be
addressed to the secretariat or sponsor whose name appears on the title page of this standard.

CAUTION NOTICE: This American National Standard may be revised or withdrawn at any
time. The procedures of the American National Standards Institute require that action be taken
periodically to reaffirm, revise, or withdraw this standard. Purchasers of American National
Standards may receive current information on all standards by calling or writing the American
National Standards Institute.

Published By

Hydraulic Institute
6 Campus Drive, First Floor North
Parsippany, NJ 07054-4406
www.Pumps.org

Copyright © 2018 Hydraulic Institute

All rights reserved.

No part of this publication may be reproduced in any form,

in an electronic retrieval system or otherwise, without prior
written permission of the publisher.
Recycled
Printed in the United States of America
paper
ISBN 978-1-935762-71-3

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

Contents
Page

Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

9.8 Pump intake design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

9.8.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

9.8.1.1 Purpose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

9.8.1.2 Scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

9.8.1.3 Nomenclature, primary symbols, and units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

9.8.2 Design objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

9.8.3 Intake structures for clear liquids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

--

9.8.3.1 Rectangular intakes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

-

9.8.3.1.1 Approach flow patterns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

9.8.3.1.2 Open versus partitioned structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

9.8.3.1.3 Trash racks and screens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

9.8.3.1.4 Recommendations for dimensioning rectangular intake structures. . . . . . . . . . . . . . . . . . . . . . . . 11

9.8.3.2 Formed suction intakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

9.8.3.2.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

9.8.3.2.2 Recommended dimensions for FSI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

9.8.3.2.3 Application standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

9.8.3.2.4 Alternative FSI designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

9.8.3.3 Circular pump stations (clear liquids). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

9.8.3.3.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

9.8.3.3.2 Recommendations for dimensioning circular pump stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.3.2.1 Nomenclature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.3.2.2 Floor clearance Cf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.3.2.3 Wall clearance Cw. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.3.2.4 Inlet bell clearance Cb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.3.2.5 Sump diameter Ds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.3.2.6 Inlet bell or volute diameter Db. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.3.2.7 Inflow pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

9.8.3.4 Trench-type intakes (clear liquids). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved iii

 9.8.3.4.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

9.8.3.4.2 Objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

9.8.3.4.3 Orientation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

9.8.3.4.4 Approach flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.4.5 Width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.4.6 Intake submergence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.4.7 End wall clearance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.4.8 Floor clearance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.4.9 Centerline spacing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.4.10 Inlet conduit elevation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.5 Tanks – pump suction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

--

9.8.3.5.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

9.8.3.5.2 Objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
-

9.8.3.5.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

9.8.3.5.4 Submergence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

9.8.3.5.5 Application options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

9.8.3.5.6 NPSH considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

9.8.3.5.7 Simultaneous inflow and outflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

9.8.3.5.8 Multiple inlets or outlets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

9.8.3.6 Can vertical turbine pump intakes (clear liquids), including those with submersible motors
(refer to Appendix G)���������������������������������������������������������������������������������������������������������������������26

9.8.3.6.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

9.8.3.6.2 Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

9.8.3.6.3 Design considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

9.8.3.6.4 Open bottom can intakes (Figure 9.8.3.6.4). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

9.8.3.6.5 Closed bottom can (Figure 9.8.3.6.5). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

9.8.3.7 Unconfined intakes (Figure 9.8.3.7). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

9.8.3.7.1 Scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

9.8.3.7.2 Cross-flow velocities and pump location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

9.8.3.7.3 Debris and screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

9.8.3.7.4 Submergence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

iv Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 9.8.4 Intake structures for solids-bearing liquids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

9.8.4.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

9.8.4.1.1 Scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

9.8.4.1.2 Objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

9.8.4.1.3 Principles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

9.8.4.1.4 Vertical transitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

9.8.4.1.5 Horizontal surfaces near inlet. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

9.8.4.1.6 Cleaning procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

9.8.4.1.7 Wet-well volume. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

9.8.4.2 Trench-type wet wells for solids-bearing liquids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

9.8.4.2.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

9.8.4.2.2 Objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

9.8.4.2.3 Approach flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

9.8.4.2.3.1 Inlet transition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

9.8.4.2.3.2 Inlet floor clearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

9.8.4.2.3.3 Inlet splitters and cones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

9.8.4.2.3.4 Anti-rotation baffle and vanes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

9.8.4.2.3.5 Cleaning procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

9.8.4.3 Circular plan wet pit for solids-bearing liquids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

9.8.4.3.1 Wet-pit design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

9.8.4.3.2 Accessories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

9.8.4.3.3 Cleaning procedure and low liquid level. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

9.8.4.3.4 Floor clearance C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

9.8.4.4 Rectangular wet wells for solids-bearing liquids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

9.8.4.4.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

9.8.4.4.2 Objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

9.8.4.4.3 Control of sediments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

9.8.4.4.4 Confined wet-well design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

9.8.4.4.4.1 Suction inlet clearance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

9.8.4.4.4.2 Anti-rotation baffle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

9.8.4.4.4.3 Cleaning procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved v

 9.8.5 Inlet bell design diameter (D). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

9.8.5.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

9.8.5.2 Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

9.8.6 Required submergence for minimizing surface vortices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

9.8.6.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

9.8.6.2 Controlling parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

9.8.6.3 Application considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

9.8.7 Physical model studies of intake structures and pump suction piping. . . . . . . . . . . . . . . . . . . . . . 48

9.8.7.1 Need for a physical model study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

9.8.7.2 Physical model study objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

9.8.7.3 Physical model similitude and scale selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

9.8.7.4 Physical model study scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

9.8.7.5 Instrumentation and measuring techniques. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

9.8.7.6 Test plan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

9.8.7.7 Acceptance criteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

9.8.7.8 Report preparation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

9.8.8 Use of computational fluid dynamics (CFD). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

9.8.8.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

9.8.8.2 Simulation methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

9.8.8.3 Acceptable uses of CFD modeling in pump suction hydraulics. . . . . . . . . . . . . . . . . . . . . . . . . . . 55

--

Appendices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Appendix A Remedial measures for problem intakes (informative). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

-

A.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

A.2 Approach flow patterns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56

A.2.1 Open versus partitioned structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

A.3 Controlling cross-flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

A.4 Expanding concentrated flows. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

A.4.1 Free surface approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

A.4.2 Closed conduit approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

A.5 Pump inlet disturbances. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

vi Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

A.5.1 Free surface vortices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

A.5.2 Subsurface vortices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

A.5.3 Pre-swirl. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

A.5.4 Velocities in pump bell throat. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

A.6 Deviations from standard dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

A.7 Tanks – pump suction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

A.7.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

A.7.2 Vertical tank, simultaneous inflow and outflow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

A.7.3 Horizontal tank, simultaneous inflow and outflow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Appendix B Sump volume (informative). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

B.1 Scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

B.2 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

B.3 Construction of a graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

B.4 Example for a simple controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

B.5 Example for programmable controllers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

Appendix C Intake basin entrance conditions, trench-type wet wells for solids-bearing liquids (informative). . 76

C.1 Scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

C.2 Entrance conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

C.3 Variable-speed pumps in trench-type wet wells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

C.4 Constant-speed pumps in trench-type wet wells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

C.4.1 Auxiliary storage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

C.4.2 Approach pipes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

C.4.3 Transition manhole, sewer to approach pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

C.4.4 Lining. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

C.5 Design examples. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Appendix D Performance enhancements for trench-type wet wells (informative). . . . . . . . . . . . . . . . . . . . . . . 81

D.1 Scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

D.2 Performance of bare trenches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

D.2.1 Normal operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

D.2.2 Cleaning operations for wastewater and stormwater wet wells . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved vii

D.3 Enhancing normal operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

D.3.1 Inlet baffles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

D.3.2 Suction bell vanes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

D.3.3 Floor cones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

D.3.4 Flow splitters in wastewater wet wells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

D.3.5 Fillets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

D.3.6 Maintaining cleaning velocity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

D.3.7 Last pump. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

D.3.8 Ramps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

D.3.9 Choice of enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

D.3.10 Omission of enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Appendix E Aspects of design of rectangular wet wells for solids-bearing liquids (informative). . . . . . . . . . . . 87

E.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

E.2 Design capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

E.3 Design alternatives – general. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

E.4 Front – high-level entry intake structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

E.5 Side – high-level entry intake structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

E.6 Side – low-level entry intake structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

E.7 Cleaning procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

E.8 Sump dimensions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

Appendix F Suction bell design (informative) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

F.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

F.2 Bell outside diameter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

F.3 Ratio of bell outside diameter to throat diameter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

F.4 Suction bell length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

F.5 Bell intake shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

Appendix G Submersible pumps – well motor type (informative). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

G.1 Submersible pumps – well motor type. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Appendix H Modification of existing pumping systems (informative). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

H.1 Scope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

viii Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

` ``` ```` ` ` ` ` ` `
H.2 Purpose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

H.3 Recommendations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

Appendix I Alternate formed suction intake designs (informative). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

I.1 Stork-type formed suction intake. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

I.2 Shoe-box-type formed suction intake. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

Appendix J Rectangular intakes for shallow liquid source (informative). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

J.1 General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

J.2 Entrance conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

J.3 Vertical transition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

J.4 Pump bay dividing walls and details near the entrance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
--

J.5 Pump bay details near the pumps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

Appendix K Influence of pump selection on intake design (informative). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

-

K.1 Influence of pump selection on intake design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

Appendix L Sediment and debris issues at surface water pump station intakes (informative) . . . . . . . . . . . . 107

L.1 Issues with surface water intakes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

L.2 Selection of intake location. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

L.2.1 River intakes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

L.2.2 Fresh water lakes, reservoirs, aqueducts, and canals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

L.2.3 Flow regime at the intake. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

L.3 Sediment and debris control and removal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

L.3.1 Eliminating/minimizing sediment into intakes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

L.3.2 Collection, removal and sediment treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Appendix M References (informative) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

Appendix N Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved ix

 This page initentionally blank

` ``` ```` ` `

` ` ` `
Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved
Foreword

Purpose and aims of the Hydraulic Institute

The purpose and aims of the Hydraulic Institute are to promote the advancement of the pump manufacturing indus-
try and further the interests of the public and to this end, among other things:

a) Develop and publish standards.

b) Address pump systems.

c) Expand knowledge and resources.

d) Educate the marketplace.

e) Advocate for the industry.

Purpose of Standards and Guidelines

a) Hydraulic Institute Standards and Guidelines are adopted in the public interest and are designed to help elimi-
nate misunderstandings between the manufacturer, the purchaser, and/or the user and to assist the purchaser
in selecting and obtaining the proper product for a particular need.

b) Use of Hydraulic Institute Standards and Guidelines is completely voluntary. Existence of Hydraulic Institute
Standards does not in any respect preclude a member from manufacturing or selling products not conforming to
the standards.

Definition of a Standard of the Hydraulic Institute

Quoting from Article XV, Standards, of the By-Laws of the Institute, Section B:

“An Institute Standard defines the product, material, process or procedure with reference to one or more of the
following: nomenclature, composition, construction, dimensions, tolerances, safety, operating characteristics, per-
formance, quality, rating, testing and service for which designed.”

Definition of a Hydraulic Institute Guideline

A Hydraulic Institute Guideline is not normative. The guideline is tutorial in nature, to help the reader better under-
stand the subject matter.

Comments from users

Comments from users of this standard will be appreciated, to help the Hydraulic Institute prepare even more useful
future editions. Questions arising from the content of this standard may be directed to the Technical Director of the
Hydraulic Institute. If appropriate, the inquiry will then be directed to the appropriate technical committee for provi-
sion of a suitable answer.

Revisions

American National Standards of the Hydraulic Institute are subject to constant review, and revisions are under-
taken whenever it is found necessary because of new developments and progress in the art. If no revisions are
made for five years, the standards are reaffirmed using the ANSI canvass procedure.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved xi

Disclaimer

This document was prepared by a committee of the Hydraulic Institute and approved by following ANSI essential
requirements. Neither the Hydraulic Institute, Hydraulic Institute committees, nor any person acting on behalf
of the Hydraulic Institute: 1) makes any warranty, expressed or implied, with respect to the use of any informa-
tion, apparatus, method, or process disclosed in this document or guarantees that such may not infringe privately
owned rights; 2) assumes any liabilities with respect to the use of, or for damages resulting from the use of, any
information, apparatus, method, or process disclosed in this guideline. The Hydraulic Institute is in no way respon-
sible for any consequences to an owner, operator, user, or anyone else resulting from reference to the content of
this document, its application, or use.

This document does not contain a complete statement of all requirements, analyses, and procedures necessary
to ensure safe or appropriate selection, installation, testing, inspection, and operation of any pump or associated
products. Each application, service, and selection is unique with process requirements that shall be determined by
the owner, operator, or its designated representative.

Units of measurement

Metric units of measurement are used, and corresponding US customary units appear in parentheses. Charts,
graphs, and sample calculations are also shown in both metric and US customary units. Because values given in
metric units are not exact equivalents to values given in US customary units, it is important that the selected units
of measure to be applied be stated in reference to this standard. If no such statement is provided, metric units shall
govern.

Consensus

Consensus for this American National Standard was achieved by use of the canvass method. The following orga-
nizations, recognized as having an interest in the standardization of pumps, were contacted prior to the approval of
this revision of the standard. Inclusion in this list does not necessarily imply that the organization concurred with the
submittal of the proposed standard to ANSI.

Alden Research Laboratory, Inc. MWI Pumps

Camtrack, LLC MWH Americas, Inc.
Carollo Engineers, Inc. Patterson Pump Company
Ekwestrel Corp Pentair – Berkeley
ITT Industrial Process Pentair – Fairbanks Nijhuis
Kemet Inc. Sulzer
Las Vegas Valley Water District Taco, Inc.

Committee list

Although this standard was processed and approved for submittal to ANSI by the canvass method, a working com-
mittee met many times to facilitate its development. At the time it was developed, the committee had the following
members:

Chair - Jack Claxton, Patterson Pump Company

Vice Chair – Andrew Johansson, Alden Research Laboratory, Inc.

Committee Member Company

Charlie Allaben HDR, Inc.
Paul Behnke ITT – Industrial Process
Ravindra Birajdar Kirloskar Brothers Ltd.
Michael Cugal Weir Minerals North America
Thomas Demlow Northwest Hydraulic Consultants Inc.
Anton de Fockert Deltares

` ``` ```` ` ` ` ` ` `

xii Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

Committee Member (continued) Company (continued)
Carl Frizzell CDM Smith – Water Services Group
Bob Hawboldt AE – Associated Engineering
Patricia McCarthy Xylem Inc. – Water Solutions
Paul Moulton AECOM
Michael Mueller Flowserve Corporation
David Nailor Hazen and Sawyer, P.C.
John Saenz Simflo Pumps, Inc.
Arnold Sdano Pentair – Fairbanks Nijhuis

`---
Constantino Senon MWH Americas, Inc.

` `
Jerry Szofer Grundfos Water Utility

`
````-`-`
Dave Werth Clemson Engineering Hydraulics, Inc.
Mike Zappone Carollo Engineers, Inc.

```
--`
Alternates Company
Mark Allen CDM Smith – Water Services Group
Lech Bobowski Flowserve Corporation
David Cowan ITT – Industrial Process
Zbigniew Czarnota Xylem Inc. – Water Solutions
Steven Fehniger CDM Smith – Water Services Group
Kara Hurtig Northwest Hydraulic Consultants Inc.
Jason Kerns HDR, Inc.
Randy Kosmicki Weir Minerals North America
Michael Murazzi Patterson Pump Company
Joseph Orlins Alden Research Laboratory, Inc.
Edward Pascua MWH Americas, Inc.
Jan Schyberg Xylem Inc. – Water Solutions
Ernest Sturtz CDM Smith – Water Services Group
George Tey MWH Americas, Inc.
Femke Verhaart Deltares
Ed Wicklein Carollo Engineers, Inc.
Kristel Zaman Xylem Inc. – Water Solutions

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved xiii

Flowchart for use of standard

Note: This flowchart is intended as a guide to the use of this standard and can be used to locate the appropriate
sections in this standard. The chart is not a substitute for the understanding of the complete standard.
--
-

xiv Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

9.8  Pump intake design

9.8.1 Introduction

This standard applies to the design of new intakes as well as the modification of existing designs used with rotody-
namic pumps. It outlines standard intake designs based on certain criteria, beyond which requires a physical model
study to be in compliance with the standard.

In the application of this standard, the pump rated flow shall be used as the design flow for the basis of the intake
design. However, other considerations in the installation design and contemplated modes of operation may require
the pumps to function under higher rates of flow. In such instances, the maximum flow rate shall be used for the
purposes of pump intake design unless specified otherwise. Refer to Appendix K for discussion on this topic.

In applications where piping is used to connect the intake to the pump, the latest version of ANSI/HI 9.6.6 Rotody-
namic Pumps for Pump Piping complements this standard.

9.8.1.1 Purpose

To provide normative criteria for the design and acceptance of new or modified free surface intakes used with roto-
dynamic pumps.

9.8.1.2 Scope

New or existing free surface intakes used with rotodynamic pumps.

Intake structures for clear liquid are given as follows:

• Rectangular intakes

• Formed suction intakes

• Circular intakes

• Trench-type intakes

• Partially filled tanks

• Open bottom and closed bottom can intakes

• Unconfined intakes

Intake structures for solids-bearing liquids are given as follows:

• Trench-type intakes

• Circular intakes

• Rectangular intakes

9.8.1.3  Nomenclature, primary symbols, and units

Preferred terms, units, and symbols to be used throughout this standard are shown in Tables 9.8.1.3.1 and 9.8.1.3.2.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 1

Rotodynamic Pumps for Pump Intake Design — 2018

Table 9.8.1.3.1  Glossary

Term Definition

Active storage Liquid stored between low and high liquid levels in the wet well and in upstream piping.
Air core vortex A vortex strong enough to form an elongated core of air (see type 6,
Figure 9.8.7.5a).
Anti-rotation baffle Device used to inhibit the rotation of fluid at or near the suction.
Approach channel A structure that directs the flow to the pump or wet well.
Approach pipe A pipe laid at a gradient sufficient to cause supercritical flow and used to contain a
portion of the active storage requirement for a constant speed pump.
Axial flow (propeller) pump High flow rate, low head, high specific speed pump.
Backwall A vertical surface behind the inlet to a suction fitting.
Backwall clearance The distance between the backwall and the point of closest approach of the
suction fitting.
Backwall splitter A device formed or fabricated and attached to the backwall that guides the
movement of flow at or near a suction.
Baffles Obstructions arranged to provide a more uniform flow at the approach to a pump
or suction inlet.

`---
Bay A portion of an intake structure configured for the installation of one pump.

` `
Bell The entrance to an axial flow pump or the flared opening leading to pump inlet

`
````-`-`
piping.
Benching A type of fillet used to minimize stagnant zones by creating a sloping transition

```
--`
between vertical and horizontal surfaces. Benching is applied between sump walls
and the sump bottom, or between the backwall and the sump bottom. It is also
referred to as fillets, such as sidewall fillets and backwall fillets.
Cavitation Formation and implosion of liquid vapor bubbles.
Cell A structure intended to confine the liquid approaching the intake to a pump
(see Bay).
CFD See Computational fluid dynamics.
Check valve Piping component used to prevent reverse flow.
Circular well A circular suction chamber in plan.
Computational fluid The systematic application of computing systems and computational solution
dynamics (CFD) techniques to mathematical models formulated to describe and simulate fluid
dynamic phenomena.
Cone See Floor cone.
Critical Depth The liquid depth that has the minimum specific energy for a given open channel
flow, corresponding to a Froude number equal to 1.
Curtain wall A near-vertical plate or wall located in an intake that extends below the normal low
liquid level to suppress vortices.
Double suction impeller An impeller provided with a single suction connection that separates and conveys
the fluid to two suction areas.
Dry-pit suction Suction from a well that conveys fluid to a pump located in a nonwetted
environment.
Dual flow screens Screening that provides two flow paths for liquid, not in-line with the main flow.
Eddy A local rotational flow pattern disturbing regular streamlines (a vortex).

2 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Table 9.8.1.3.1  Glossary (continued)

Fillet A triangular element at the vertex of two surfaces to guide the flow.
Floor clearance The distance between the floor and the suction bell or opening.
Floor cone A conical fixture placed below the suction between the floor and the suction bell.
Forebay The region of an intake before individual partitioning of flow into individual suctions
or intake bays.
Formed suction intake A shaped suction inlet that directs the flow in a particular pattern into the pump
suction.
Free surface flow Open channel or unconfined flow.
Froude number A dimensionless grouping of parameters used in flow analysis and modeling that
indicates the relative influence of inertial compared to gravitational forces (see
Equation 9.8.7.3-1).
Guide vanes Devices used in the suction approach that direct the flow in an optimal manner.
Hydraulic jump A turbulent sudden increase in liquid depth as the flow decelerates from
supercritical to subcritical flow.
Intake The structure or piping system used to conduct fluid to the pump suction.
Intake velocity The average or bulk velocity of the flow in an intake.
Invert The bottom of a conduit.
Mixer A mechanical device that produces an axial propeller jet, often used for
maintaining suspension of solids-bearing liquids in wet wells and tanks.
NPSHR A minimum net positive suction head (NPSH) given by the manufacturer, which is
required for a pump to achieve a specified performance at a specified flow rate,
speed, and pumped liquid.
Ogee ramp or spillway The gradual change in shape/slope in the floor of an intake, shaped like an
elongated letter S.

`---
Physical hydraulic model A reduced-scale replicate of the geometry that controls approach flow patterns,

` `
`
operated according to certain similitude laws for flow, velocity, and time.

````-`-`
Piezometric head The sum of pressure head at a point in a body of liquid plus the elevation of the

```
point. In a pressurized conduit, it is seen as the elevation to which a liquid column

--`
will rise in a vertical tube, if a small hole is drilled in the conduit and connected to
the tube.
For open-channel flow, it is seen as the elevation of the liquid surface.
Pre-swirl Rotation of the flow at the pump suction due to the approach flow patterns.
Pump A device used to convey fluid from a low energy level to a higher one.
Pump column Part of the pump assembly that both connects the pump to the discharge head and
nozzle and conveys fluid into the system.
Pump suction bell A part of the pump that provides an opening to convey flow into the suction eye of
the impeller.
Rectangular wet well Any wet well in which pumps are arranged along a wall opposite the influent
conduit. The shape may be square, rectangular, or trapezoidal.
Reynolds number A dimensionless grouping of parameters used in flow analysis and modeling
that indicates the relative influence of inertial compared to viscous forces (see
Section 9.8.7.3).
Scale The ratio between geometric characteristics of the physical model and prototype.
Scale effect The impact of reduced scale on the applicability of test results to a full-scale
prototype.
Sediment Settleable materials suspended in the flow.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 3

Rotodynamic Pumps for Pump Intake Design — 2018

Septicity A condition in which stagnant domestic sewage turns septic, due to a lack of
oxygen.
Sequent depth The depth of liquid following a hydraulic jump.
Solids Material suspended in the liquid.
Specific energy For open-channel flow, the energy per unit weight of liquid relative to the channel
bottom.
For a channel with small slope and uniform velocity distribution, it is the sum of the
depth and the velocity head of the liquid.
For closed-conduit flow, the energy per unit weight of liquid relative to a vertical
datum, it is the sum of piezometric head and velocity head.
Specific speed An index of pump performance (developed total head) at the pump’s best efficiency
point (BEP) flow rate, with the maximum diameter impeller, and at a given rotative
speed. For more information refer to ANSI/HI 2.1-2.2 Rotodynamic (Vertical)
Pumps for Nomenclature and Definitions.
Soffit Inside top of a conduit.
Strainer A device located at the inlet to either protect the pump or provide flow stability at
the suction.
Submergence The height of liquid level over the suction bell or pipe inlet.
Submersible pump A close-coupled pump and drive unit designed for operation while immersed in the
pumped liquid.
Suction bell diameter Overall OD of the bell-shaped fitting at the suction inlet.
Suction head Pressure available at the pump suction, usually positive if the liquid level is at a
higher elevation than the pump suction.
Suction lift Negative pressure at the pump suction, usually a result of the liquid level being at a
lower elevation than the pump suction.
Sump A pump intake basin or wet-well. See Forebay.
Swirl Rotation of fluid around its mean, axial flow direction.
Swirl angle The angle formed by the axial and tangential (circumferential) components of a
velocity vector (see Equation 9.8.7.5-1).
Swirl meter A device with four flat vanes of zero pitch used to determine the extent of rotation
in otherwise axial flow.
Trench intake An intake design that aligns the pump suctions in-line with but below the inflow. A
type of forebay.
Turning vanes Devices applied to the suction to alter the direction of flow.
Unconfined intake Suction in a free-flow field with no lateral physical boundaries.
Vane See Floor vane.
Volute The pump casing for a centrifugal type of pump, typically spiral or circular in shape.
Vortex A well-defined swirling flow core from either the free surface or from a solid
boundary to the pump inlet (see Figures 9.8.7.5a and b).
Vortex, free surface A vortex that originates at the free surface of a flow field.
Vortex, subsurface A vortex that originates on the floor or sidewalls of an intake.
Vortex suppressor A fixed or floating device used to help minimize surface or subsurface vortices.
Wall clearance Dimensional distance between the suction and the nearest vertical surface.
Wastewater Description of fluid that typically carries suspended waste material from domestic
or industrial sources.

4
` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Table 9.8.1.3.1  Glossary (continued)

Weber number A dimensionless grouping of parameters used in flow analysis and modeling that
indicates the relative influence of inertial compared to surface tension forces (see
Section 9.8.7.3).
Wet-pit suction A suction with the pump fully wetted.
Wet well A pump intake basin or sump having a confined liquid volume with a free liquid
surface designed to hold liquid in temporary storage to even out variations between
inflow and outflow. See Forebay.

Table 9.8.1.3.2  Units and Symbols

US Customary Conversion
Symbol Definition Metric Unit Abbreviation Abbreviation
Unit Factor1
Distance from the
pump inlet centerline
A Meter m Feet ft 0.3048
to the intake structure
entrance
Ae Empty area Square meter m2 Square feet ft2 0.0929
At Total area Square meter m 2
Square feet ft2
0.0929
--

Length of constricted
a bay section near the Meter m Feet ft 0.3048
-

pump inlet
Distance from the
B backwall to the pump Meter m Feet ft 0.3048
inlet bell centerline
Distance between the
C Meter m Feet ft 0.3048
inlet bell and floor
Inlet bell or volute
Cb clearance for circular Millimeter mm Inch in 25.4
pump stations
Floor clearance on
Cf Millimeter mm Inch in 25.4
circular pump stations
Wall clearance on
Cw Millimeter mm Inch in 25.4
circular pump stations
Inlet bell diameter
or inlet bell design
diameter (may also
D refer to pipe inside Meter m Feet ft 0.3048
diameter if a pipe is
used instead of a bell
inlet)
Tank outlet fitting inside
D Meter m Feet ft 0.3048
diameter
Vertical can riser inside
D1 Meter m Feet ft 0.3048
diameter
D1 Can inside diameter Meter m Feet ft 0.3048
Inlet bell or volute
Db Millimeter mm Inch in 25.4
diameter

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 5

Rotodynamic Pumps for Pump Intake Design — 2018

Diameter of circle with

area equivalent to
De Meter m Feet ft 0.3048
rectangular area at FSI
entrance
Well motor cooling
DM Meter m Feet ft 0.3048
shroud diameter
Inside diameter of
Dp Millimeter mm Inch in 25.4
approach pipe
Ds Sump diameter Millimeter mm Inch in 25.4
Diameter at outlet of
d Meter m Feet ft 0.3048
formed suction intake
Diameter of the pipe at
d Meter m Feet ft 0.3048
the swirl meter
EGL Energy grade line Meter m Feet ft 0.3048
Froude number
F Dimensionless N/A Dimensionless N/A N/A
(general)
Froude number
FD (calculated at diameter Dimensionless N/A Dimensionless N/A N/A
D)
Froude number ratio,
Fr Dimensionless N/A Dimensionless N/A N/A
Fm/Fp
Froude number of
Fm Dimensionless N/A Dimensionless N/A N/A
physical model
Froude number of
Fp Dimensionless N/A Dimensionless N/A N/A
prototype
G Geometry Dimensionless N/A Dimensionless N/A N/A
Meters per Feet per square
g Acceleration of gravity m/s2 ft/s2 0.3048
square second second
H Minimum liquid depth Meter m Feet ft 0.3048
Hf Height of FSI Meter m Feet ft 0.3048
Minimum height of
h constricted bay section Meter m Feet ft 0.3048
near the pump
A characteristic length
L (usually bell diameter or Meter m Feet ft 0.3048
submergence)
Geometric scale of
Lr Dimensionless N/A Dimensionless N/A N/A
physical model
Characteristic length
Lv of a cubic cage-type Millimeter mm Inch in 25.4
vortex suppressor
NΓ Circulation number Dimensionless N/A Dimensionless N/A N/A
Revolutions/second of Revolutions Revolutions per
n rev/s rev/s 1
the swirl meter per second second
n Manning’s number Dimensionless N/A Dimensionless N/A N/A
Liters per Gallons per
Q Flow L/s gpm 0.06309
second minute

` ``` ```` ` ` ` ` ` `

6 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Table 9.8.1.3.2  Units and Symbols (continued)

Liters per Cubic feet per

Q Inflow into sump L/s ft3/s 28.3168
second second
Flow at physical model Liters per Gallons per
Qm L/s gpm 0.06309
scale second minute
Liters per Gallons per
Qp Flow at prototype scale L/s gpm 0.06309
second minute
Flow scale ratio,
Qr physical model/ Dimensionless N/A Dimensionless N/A N/A
prototype
R Reynolds number Dimensionless N/A Dimensionless N/A N/A
r Radius of curvature Meters m Feet ft 0.3048
Radius of tangential
r Meters m Feet ft 0.3048
velocity component
Minimum submergence
S Meter m Feet ft 0.3048
depth
Total cycle time in
T Seconds s Seconds s 1
seconds
Time at physical model
Tm Seconds s Seconds s 1
scale
Tp Time at prototype scale Seconds s Seconds s 1
Time scale ratio,
Tr physical model/ Dimensionless N/A Dimensionless N/A N/A
prototype
Average axial velocity
Meters per Feet per
u (such as in the suction m/s ft/s 0.3048
second second
bell)
Average axial velocity Meters per Feet per
u m/s ft/s 0.3048
at the swirl meter second second
Meters per Feet per
V Velocity m/s ft/s 0.3048
second second
Meters per Feet per
Vol Active sump volume m/s ft/s 0.3048
second second
Meters per Feet per
Vc Cross-flow velocity m/s ft/s 0.3048
second second
Velocity at physical Meters per Feet per
Vm m/s ft/s 0.3048
model scale second second
Velocity at prototype Meters per Feet per
Vp m/s ft/s 0.3048
scale second second
Velocity scale ratio,
Vr physical model/ Dimensionless N/A Dimensionless N/A N/A
prototype
Meters per Feet per
Vt Tangential velocity m/s ft/s 0.3048
second second
Meters per Feet per
Vx Pump bay velocity m/s ft/s 0.3048
second second
VT Vortex type Dimensionless N/A Dimensionless N/A N/A

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 7

Rotodynamic Pumps for Pump Intake Design — 2018

We Weber number Dimensionless N/A Dimensionless N/A N/A

Pump bay entrance
W Meter m Feet ft 0.3048
width
W Width of FSI Meter m Feet ft 0.3048
Constricted bay width
w Meter m Feet ft 0.3048
near the pump
X Pump bay length Meter m Feet ft 0.3048
Distance from pump
Y inlet bell centerline to Meter m Feet ft 0.3048
traveling screen
y Depth Meter m Feet ft 0.3048
Distance from pump
Z1 inlet bell centerline to Meter m Feet ft 0.3048
diverging walls
Distance from pump
Z2 inlet bell centerline to Meter m Feet ft 0.3048
sloping floor
a Angle of floor slope Degree ° Degree ° 1
Angle of wall
b Degree ° Degree ° 1
divergence
Angle of sidewall of
ε Degree ° Degree ° 1
trench
f A function Dimensionless N/A Dimensionless N/A N/A
Kilograms per Pound-mass
ρ Liquid density kg/m3 lbm/ft3 16.0185
cubic meter per cubic foot
Square meter Square foot per
Γ Circulation of the flow m2 /s ft2/s 0.0929
per second second
Kinematic viscosity of Square meter Square foot per
v m2/s ft2/s 0.0929
the liquid per second second
θ Swirl angle Degree ° Degree ° 1
Surface tension of Newton per
σ N/m Pound per inch lb/in 0.1130
liquid/air interface meter
Angle of divergence
f from constricted area to Degree ° Degree ° 1
bay walls

Notes.

1. Conversion factor × US customary units = metric units

9.8.2  Design objectives

Specific hydraulic phenomena have been identified that can adversely affect the performance of pumps. Phenom-
ena that must not be present to an excessive degree are:

• Submerged vortices

• Free surface vortices

` ``` ```` ` ` ` ` ` `

8 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

• Pre-swirl magnitude and fluctuation with time

• Nonuniform distribution of velocity in space and time at the impeller eye

• Entrained air or gas bubbles

In general, the flow of liquid into any pump should be uniform, steady, and free from swirl and entrained air. Lack
of uniformity can cause the pump to operate away from the optimum design condition, and at a lower hydraulic
efficiency. Unsteady flow causes the load on the impeller to fluctuate, which can lead to noise, vibration, bearing
problems, and fatigue failures of pump shafts.

Swirl in the pump intake can cause a significant change in the operating conditions for a pump, and can produce
changes in the flow capacity, power requirements, and efficiency. It can also result in local vortex-type pressure
reductions that extend into the pump. This and any other air or vapor ingestion can cause reductions in pump flow
and fluctuations of impeller load that result in cavitation, noise and vibration, which may lead to physical damage.

The negative impact of each of these phenomena on pump performance depends on pump specific speed and
size, as well as other design features of the pump that are specific to a given pump manufacturer. In general, larger
pumps and axial flow pumps (high specific speed) are more sensitive to adverse flow phenomena than smaller
pumps or radial flow pumps (low specific speed). A comprehensive assessment of which pump types may be
expected to withstand a given level of adverse phenomena with no ill effects has not been performed.

The intake structure should be designed to allow the pumps to achieve their optimum hydraulic performance for
all operating conditions. A good design ensures that the adverse flow phenomena described above are within the
limits outlined in Section 9.8.7.7.

In designing an intake structure, the following points must be considered:

• Flow from the forebay should be directed toward the pump inlets in such a way that the flow reaches the inlets
with a minimum of swirl.

• To prevent the formation of air-entraining surface vortices in the sump, the walls must be designed to avoid
stagnation regions in the flow. A properly placed wall close to the inlet can reduce the tendency toward localized
swirl and vortex formation. The liquid depth also must be great enough to suppress surface vortices.

• Although excessive turbulence or large eddies should be avoided, some turbulence does help to prevent the
--

formation and growth of vortices.

• Station inflow may approach the wet well at a relatively high elevation. In such cases, the liquid may fall a signif-
-

icant distance as it enters the sump. Such a drop can also occur whenever the pumps have lowered the liquid
level in the sump to the point at which all pumps are about to be switched off. Therefore, the path between the
sump entrance and the pump inlets must be sufficiently long for the air bubbles to rise to the surface and escape
before reaching the pumps. The energy of the falling liquid should be dissipated sufficiently so that excessively
high and irregular velocities do not occur within the sump. This can be accomplished with properly designed
and placed baffle walls.

• The sump should be small and as simple as feasible to minimize construction costs. However, the required
sump volume may be specified for other reasons, such as to provide adequate storage capacity to prevent
pumps from excessive start cycles, process retention time or other factors.

Additional criteria for solids-bearing liquids are covered in Section 9.8.4.

If an intake is designed to a geometry other than that presented in this standard, and this design is shown by proto-
type testing or a physical model study performed in accordance with Section 9.8.7 to meet the acceptance criteria
in Section 9.8.7.7, then this alternative design shall be deemed to comply with this standard.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 9

Rotodynamic Pumps for Pump Intake Design — 2018

Notes:

1) For intake designs where piping is used to connect the intake to the pump, the piping shall comply with ANSI/
HI 9.6.6 Rotodynamic Pumps for Pump Piping, but model study of piping is covered herein.

2) For intake designs for pumps used for pumping and/or transporting mixtures of solids and liquids or so called
“slurries,” refer to ANSI/HI 12.1-12.6 Rotodynamic (Centrifugal) Slurry Pumps and the pump manufacturer for
guidance.

9.8.3  Intake structures for clear liquids

9.8.3.1  Rectangular intakes

This section is applicable to intake designs for both wet-pit and dry-pit pumps, where an intake structure with a free
liquid surface exists. Pipeline intakes (no free liquid surface) for dry-pit pumps are covered under ANSI/HI 9.6.6
Rotodynamic Pumps for Pump Piping.

9.8.3.1.1  Approach flow patterns

The characteristics of the flow approaching an intake structure are among the most critical considerations for the
designer. When determining direction and distribution of flow at the entrance to a pump intake structure, the follow-
ing must be considered:

• The orientation of the structure relative to the body of supply liquid

• Whether the structure is recessed from, flush with, or protrudes beyond the boundaries of the body of supply
liquid

• Strength of currents in the body of supply liquid perpendicular to the direction of approach to the pumps

• The number of pumps required and their anticipated operating combinations

The ideal conditions – and the assumptions on which the geometry and dimensions recommended for rectangular
intake structures are based – are that the structure draws flow so that there are no cross-flows in the vicinity of the
intake structure that create asymmetric flow patterns approaching any of the pumps, and the structure is oriented
so that the supply boundary is symmetrical with respect to the centerline of the structure. As a general guide,
cross-flow velocities are significant if they exceed 50% of the pump bay entrance velocity. Section 9.8.7 provides
recommendations for analyzing departures from this ideal condition based on a physical hydraulic model study.

9.8.3.1.2  Open versus partitioned structures

If multiple pumps are installed in a single intake structure, then dividing walls placed between the pumps result in
more favorable flow approaching the pump than found in open sumps. Adverse flow patterns can frequently occur
if dividing walls are not used. For pumps with design flows greater than 315 L/s (5000 gpm), dividing walls between
pumps are required.

9.8.3.1.3  Trash racks and screens

Partially clogged trash racks or screens can create severely skewed flow patterns. If the application is such that
screens or trash racks are susceptible to clogging, they must be inspected and cleaned as frequently as necessary
to prevent adverse effects on flow patterns.

Any screen-support structure that disrupts flow, such as dual-flow traveling screens, otherwise known as
­double-entry, single-exit screens or single-entry, double-exit screens can create high-velocity jets and severe insta-
bility near the pumps. A physical hydraulic model study must be performed in every such case. The screen exit

` ``` ```` ` ` ` ` ` `

10 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

should be placed a minimum distance of six bell diameters, 6D, (see Section 9.8.5 for definition of D) from the
pumps. However, this distance should be used only as a general guideline for initial layouts of structures, with final
design developed with the aid of a physical model study.

The recommendations in this standard should be followed even if suction bell strainers are used.

9.8.3.1.4  Recommendations for dimensioning rectangular intake structures

The basic design requirements for satisfactory hydraulic performance of rectangular intake structures include the
following:

• Adequate depth of flow to limit velocities in the pump bays and reduce the potential for formulation of ­surface
vortices

• Adequate pump bay width, in conjunction with the depth, to limit the maximum pump approach velocities to 0.5
m/s (1.5 ft/s), but narrow and long enough to channel flow uniformly toward the pumps

The minimum submergence S required to prevent strong air core vortices is based in part on a dimensionless flow
parameter, the Froude number, defined as:

V  (Eq. 9.8.3.1.4-1)
FD =
( gD )0.5

Where:

FD = Froude number at D (dimensionless)

V = Velocity at suction inlet = Flow/Area, based on D
D = Outside diameter of bell or inside diameter of pipe inlet (see Figure 9.8.3.5.5)
g = gravitational acceleration

Consistent units must be used for V, D, and g so that FD is dimensionless. The minimum submergence S shall be
calculated from (Hecker, G.E., 1987)

S = D(1 + 2.3FD) (Eq. 9.8.3.1.4-2)

where the units of S are those used for D. Section 9.8.6 provides further information on the background and devel-
opment of this relationship.

It is appropriate to specify sump dimensions in multiples of pump bell diameters D (see Section 9.8.5). Basing
dimensions on D ensures geometric similarity of hydraulic boundaries and dynamic similarity of flow patterns.
There is some variation in bell velocity among pump types and manufacturers. However, variations in bell inlet
velocity are of secondary importance to maintaining acceleration of the flow and converging streamlines into the
pump bell.

The basic recommended layout for rectangular sumps, dimensioned in units of pump bell diameter D, is shown in
Figure 9.8.3.1.4a. This figure applies to any number of adjacent pumps. The dimension variables and their recom-
mended values are defined in Table 9.8.3.1.4a.

Through-flow traveling screens usually do not clog to the point where flow disturbances occur. Therefore, they may
be located such that Y is 4D or more in dimension. For non-self-cleaning trash racks or stationary screens, the
dimension Y shall be increased to a minimum of 5D, with the presupposition of reasonable cleaning practices by
the user to remove trapped material that if left untended may generate large nonuniformities in the pump approach
flow.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 11

Rotodynamic Pumps for Pump Intake Design — 2018

The effectiveness of the recommended pump bay dimensions depends on the characteristics of the flow approach-
ing the structure, and on the geometry of hydraulic boundaries in the immediate vicinity of the structure. Section
9.8.3.1.1 provides a discussion of the requirements for satisfactory approach flow.

Negative values of b (the angle of wall divergence) might require flow distribution or straightening devices, and
should be developed with the aid of a physical hydraulic model study.

Occasionally, it is necessary to increase the bay width W to greater than 2D to prevent velocities at the entrance to
the pump bays from exceeding 0.5 m/s (1.5 ft/s). Greater bay widths may also result because of the arrangement
of mechanical equipment. In these cases, the bay width in the immediate vicinity of the pumps must be decreased
to 2D. The dimension of the filler required to achieve the reduction in bay width is as shown in Figure 9.8.3.1.4b.

For pumps with design flows of 315 L/s (5000 gpm) or less, no partition walls between pumps are required, and the
minimum pump spacing shall be 2D.

Some pump station applications, such as cooling water circulating pumps withdrawing from cooling tower basins,
can have relatively shallow depth at the entrance to the intake structure. Designers should ensure that the gravi-
ty-driven inflow is not restricted by the entrance condition to the extent that pumping requirements are not met. A
consequence could be dewatering of the intake while the pumps are operating.

Liquid depth at the entrance to the structure will be adequate if the following condition is met:
0.667
 Q  (Eq. 9.8.3.1.4-3)
H1 ≥ C  
 W1 

Where:

Q = total flow at W1, in L/s (ft3/s)

H1 = liquid depth at the entrance to the intake structure, in m (ft)
W1 = width at the entrance to the intake structure, in m (ft)
C = 0.01 if flow is in L/s and lengths are in m
C = 0.7 if flow is in ft3/s and lengths are in ft

` ``` ```` ` ` ` ` ` `

12 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

--
-

Figure 9.8.3.1.4a  Rectangular intake structure layout

Table 9.8.3.1.4b provides a sequence of steps to follow in determining the general layout and internal geometry of
a rectangular intake structure.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 13

Rotodynamic Pumps for Pump Intake Design — 2018

`---
` `
`
````-`-`
```
--`

Figure 9.8.3.1.4b  Filler wall details for proper bay width

14 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Table 9.8.3.1.4a  Dimensions for Figures 9.8.3.1.4a and b

Dimension
Description Recommended Value
Variable
A Distance from the pump inlet bell centerline to A = 5D minimum, assuming no significant cross-
the intake structure entrance. flowa at the entrance to the intake structure and
assuming through flow screens are used.

a Length of constricted bay section near the a = 2.5D minimum

pump inlet.
B Distance from the backwall to the pump inlet B = 0.75D
bell centerline.
C Distance between the inlet bell and floor. C = 0.3D to 0.5D
D Inlet bell design outside diameter. See Section 9.8.5 and Figure 9.8.3.5.5.
--

H Minimum liquid depth. H=S+C

H1 Liquid depth at the entrance to the intake Refer to Eq. 9.8.3.1.4-3.
-

structure.
H2 Liquid depth just before the basin floor Refer to Eq. 9.8.3.1.4-3, substituting H2 for H1,
intermediate step boundary. Refer to Figure J.1 substituting the boundary length of the
intermediate step for W1, and using the total
station flow as Q.
h Minimum height of constricted bay section near h = (Greater of H or 2.5D)
the pump inlet bell.
S Minimum pump inlet bell submergence. S = D(1.0 + 2.3 FD). (See Section 9.8.6 for
detailed discussion on determining minimum
submergence.)
W Pump bay entrance width. W = 2D minimum
W1 Width at the entrance to the intake structure. Refer to Figure 9.8.3.1.4a.
w Constricted bay width near the pump inlet bell. w = 2D
X Pump inlet bay length. X = 5D minimum, assuming no significant cross-
flow at the entrance to the intake structure and
assuming through flow screens are used.
Y Distance from pump inlet bell centerline to the Y = 4D minimum for self-cleaning screens.
downstream face of the optional screen. Y = 5D for manual cleaning screens. Dual-flow
screens require a physical model study.
Z1 Distance from pump inlet bell centerline to Z1 = 5D minimum, assuming no significant cross-
diverging walls. flow at the entrance to the intake structure.

Z2 Distance from inlet bell centerline to sloping Z2 = 5D minimum

floor.
a Angle of floor slope. a = –10 to +10 degrees
b Angle of wall convergence. b = 0 to +10 degrees. (Negative values of b,
if used, might require flow distribution devices
developed through a physical model study.)
f Angle of convergence from constricted area to f = 10 degrees maximum
bay walls.
a
Cross-flow is considered significant when VC > 0.5 VX average (see Figure 9.8.3.1.4a).

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 15

 Rotodynamic Pumps for Pump Intake Design — 2018

Table 9.8.3.1.4b  Design sequence, rectangular intake structures

Design Step Description

1 Consider the flow patterns and boundary geometry of the body of liquid from which the pump
station is to receive flow. Compare with the approach flow condition described in Section
9.8.3.1.1 and determine from Section 9.8.7.1 if a physical hydraulic model study is required.
2 Determine the number and size of pumps required to satisfy the range of operation likely to be
encountered.
3 Identify pump inlet bell diameter. If final bell diameter is not available, refer to Section 9.8.5 to
determine the inlet bell design diameter.
4 Determine the bell-floor clearance, see Figure 9.8.3.1.4a and b. A good preliminary design
number is 0.5D.
5 Determine the required bell submergence, using the relationship in Section 9.8.6.
Note: If a submergence greater than recommended herein is needed to provide the required
NPSH for the pump, that greater submergence governs and must be used.
6 Determine the minimum allowable liquid depth in the intake structure from the sum of the floor
clearance and the required bell submergence.
7 Check bottom elevation near the entrance to the structure and determine if it is necessary
to slope the floor upstream of the bay entrance. If the resulting depth at the entrance to the
intake structure is shallow, then check to ensure that gravity-driven flow is not restricted by the
--

entrance condition.
8 Check the pump bay velocity for the maximum single-pump flow and minimum liquid depth with
-

the bay width set to 2D. If bay velocity exceeds 0.5 m/s (1.5 ft/s), then increase the bay width or
consider decreasing the floor level to reduce to a maximum flow velocity of 0.5 m/s (1.5 ft/s).
9 If it is necessary to increase the pump bay width to greater than 2D, then decrease bay width in
the vicinity of the pumps according to Figure 9.8.3.1.4b.
10 Compare cross-flow velocity (at maximum system flow) to average pump bay velocity. If cross-
flow value exceeds 50% of the bay velocity, a physical hydraulic model study is necessary.
11 Determine the length of the structure and dividing walls, giving consideration to minimum
allowable distances to a sloping floor, screening equipment, and length of dividing walls. If dual
flow traveling screens or drum screens are to be used, a physical hydraulic model study is
required (see Section 9.8.7.1).
12 If the final selected pump bell diameter and inlet velocity is within the range given in Section
9.8.4, then the sump dimensions (developed based on the inlet bell design diameter) need not
be changed and will comply with these standards.

9.8.3.2  Formed suction intakes

9.8.3.2.1 General

This portion of the standard applies to formed suction intakes (FSI). The standard uses an FSI adapted from the
“TYPE 10” design developed by the US Army Corps of Engineers (ETL No. 1110-2-327). The FSI may eliminate the
need for the design of sumps with approach channels and appurtenances to provide satisfactory flow to a pump.
The recommended FSI design is relatively insensitive to the direction of approach flow and skewed velocity distri-
bution at its entrance. In applying the FSI design, consideration should be given to the head loss in the FSI that will
affect to some extent the system curve calculations, and the NPSH available to the pump impeller, typically located
near the FSI exit.

16 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.2.2  Recommended dimensions for FSI

The recommended FSI design dimensions are indicated in Figure 9.8.3.2.2. The wall shown in Figure 9.8.3.2.2
above the FSI opening reduces the tendency for flow separation at the FSI entrance plane when the FSIs are
installed in individual bays. The wall is not necessary for unrestricted approach flow conditions, but may be bene-
ficial for reducing flow separation at the entrance plan. To minimize flow separation at the FSI intake, the indicated
radii at the vertical and sidewalls are recommended. FSI geometry is based on a maximum velocity of 6.1 m/s
(20 ft/s) at d. Consult pump manufacturer for assistance in determining d. For velocities between 6.1–7.6 m/s
(20–25 ft/s), consult pump manufacturer or perform a physical model study. For additional considerations for
­physical model study requirements, refer to Section 9.8.7.1.
--

Figure 9.8.3.2.2  Formed suction intake

-

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 17

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.2.3  Application standards

Minimum submergence (see Section 9.8.6) is calculated as follows:

S
= 1.0 + 2.3FD  (Eq. 9.8.3.2.3-1)
De

Where:

The distance from the minimum recommended liquid level to the centerline of the FSI
S =
­opening in the elevation view
De = The diameter of a circle having an area equivalent to the rectangular FSI opening,

()
0.5

De =  WHf 
4
 π 
V (used in FD) = The average velocity through the FSI opening

9.8.3.2.4  Alternative FSI designs

Alternative FSI designs are shown in Appendix I, for reference. At the time of publication of this standard, sufficient
data pertaining to these designs were not available to an extent that would allow this material to be considered
Hydraulic Institute standard designs. Alternative FSI designs can be utilized if they comply with criteria given in
Section 9.8.7.7.

9.8.3.3  Circular pump stations (clear liquids)

9.8.3.3.1 General

A circular design is suitable for many types and sizes of pump stations (see Figures 9.8.3.3.1a through f). It can be
used with most types of pumps and for most types of liquids. A circular design may offer a more compact layout that
often results in reduced construction costs.

The circular geometry results in a smaller circumference, and hence minimizes excavation and construction mate-
rials for a given sump volume. The circular geometry lends itself to the use of the caisson construction technique.
The availability of prefabricated circular construction elements has made this design the most popular for smaller
pump stations. Fully equipped prefabricated pump stations often have a circular design for the above reasons.

The recommended designs of circular stations are categorized in two groups: duplex and triplex. Stations with four
or more pumps are not addressed in this standard because of complex flow patterns; such designs require a phys-
ical model study. Circular pump sumps for flows exceeding 315 L/s (5000 gpm) per pump require a physical model
study. Circular pump sumps per Figures 9.8.3.3.1c and 9.8.3.3.1f with station flows exceeding 315 L/s (5000 gpm)
require a physical model study.

The designs shown in this section are based on one pump being an installed spare.

` ``` ```` ` ` ` ` ` `

18 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Figure 9.8.3.3.1a Figure 9.8.3.3.1b Figure 9.8.3.3.1c

Wet-pit duplex sump Wet-pit duplex sump Dry-pit/wet-pit duplex
with pumps offset with pumps on centerline sump

Figure 9.8.3.3.1d Figure 9.8.3.3.1e Figure 9.8.3.3.1f

Wet-pit triplex sump, Wet-pit triplex sump, Dry-pit/wet-pit triplex
pumps in line compact sump

` ``` ```` ` ` ` ` ` `
Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 19
Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.3.2  Recommendations for dimensioning circular pump stations

9.8.3.3.2.1 Nomenclature

Cf = Floor clearance
Cw = Wall clearance
Cb = Inlet bell or volute clearance (as applicable)
Ds = Sump diameter
Db = Inlet bell or volute diameter (as applicable)
Submergence, the vertical distance from minimum sump liquid level to pump inlet, usually pump
S =
inlet bell (see Section 9.8.6 for details)

9.8.3.3.2.2  Floor clearance Cf

The floor clearance should not be greater than necessary because excessive floor clearance increases the occur-
rence of stagnant zones as well as the sump depth at a given submergence. The conditions that determine the
minimum floor clearance (Cf) are the risk of increasing inlet head loss and flow separation at the bell. Submerged
vortices are also sensitive to floor clearance. Recommended floor clearance is between 0.3 Db and 0.5 Db.

9.8.3.3.2.3  Wall clearance Cw

The minimum clearance between an inlet bell or a pump volute and a sump wall is 0.25 Db or at least 100 mm
(4 in).

9.8.3.3.2.4  Inlet bell clearance Cb

The minimum clearance between adjacent inlet bells or volutes (as applicable) is 0.25 Db or at least 100 mm (4 in).

9.8.3.3.2.5  Sump diameter Ds

Minimum sump diameter shall be as indicated for each type of pump sump as shown in Figures 9.8.3.3.1a through
9.8.3.3.1f.

9.8.3.3.2.6  Inlet bell or volute diameter Db

This parameter is given by the proposed pump type and model.

For submersible and other pumps with a volute in the wet pit, use the volute diameter. Refer to manufacturer for the
volute diameter information.

For pumps without a volute in the wet pit, use the inlet bell diameter.

9.8.3.3.2.7  Inflow pipe

The inflow pipe shall not be placed at an elevation higher than that shown in the figures. This placement minimizes
air entrainment for liquid cascading down into the sump from an elevated inflow pipe. It is important to position the
inflow pipe(s) radially to the pumps, as shown in the figures, to minimize rotational flow patterns. For the last five
--

pipe diameters before entering the sump, the inflow pipe(s) shall be straight and have no valves or fittings.
-

Note: High inlet pipe velocity can cause excessive turbulence in this type of wet well. An appropriate inlet pipe
velocity has not been determined for this configuration. The designer is cautioned to rely on prior successful expe-
rience with a similar wet-well configuration and similar flows.

20 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.4  Trench-type intakes (clear liquids)

This section establishes criteria for design of trench-type wet wells using both formed suction and bell-type pump
inlets for clear liquid applications.

9.8.3.4.1 General

Trench-type wet wells differ from rectangular intake structures (see Section 9.8.3.1) by the geometry used to form
a transition between the dimensions of the influent conduit or channel and the wet well itself. As illustrated in
­Figure 9.8.3.4.1, an abrupt transition is used to create a confined trench for the location of the pump inlets.

Most applications of the trench-type design have been with the incoming flow directed along the wet well’s long axis
(coaxial). Physical model studies shall be conducted for any installation with individual pump capacities exceeding
1260 L/s (20,000 gpm) or stations with capacities greater than 3150 L/s (50,000 gpm).

Figure 9.8.3.4.1  Trench-type wet well

9.8.3.4.2 Objectives

The purpose of the trench-type wet well is to shield the pump intakes from the influence of the concentrated inflow.
The shielding is accomplished by locating the inlets well below the invert of the influent channel or conduit.

9.8.3.4.3 Orientation

Align the long axis of the wet well with the centerline of the upstream conduit or channel. Offset centerlines are not
recommended.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 21

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.4.4  Approach flow

The velocity in the approach channel or conduit, upstream from the wet well, shall be no greater than:

• 1.2 m/s (4.0 ft/s) with the axis of the channel or conduit coaxial with the axis of the wet well for vertical pumps
or submersible pumps

• 0.9 m/s (3.0 ft/s) with the axis of the channel or conduit coaxial with the axis of the wet well for pumps with suc-
tion piping extending from the dry well into the wet well

The influent pipe upstream from the trench shall be straight and free of fittings or devices that will disrupt the flow
from uniformity entering the trench for a distance equal to at least eight times the influent pipe diameter.

9.8.3.4.5 Width

The recommended width of the bottom of the trench for trench-type wet well is twice the diameter of the pump
intake bell. The width of the sump above the trench must be expanded to produce an average limiting velocity in the
trapezoidal area above the trench of 0.3 m/s (1.0 ft/s). See Figure 9.8.3.4.1.

9.8.3.4.6  Intake submergence

See submergence section (Section 9.8.6).

9.8.3.4.7  End wall clearance

Clearance between the centerline of the most upstream intake bell and the upstream end wall of the trench shall
be 2.5D or greater. Clearance between the centerline of the downstream intake bell and the downstream end wall
of the trench should be 0.75D.

`---
` `
9.8.3.4.8  Floor clearance

`
````-`-`
Clearance between the floor of the trench and the rim of the inlet bell shall be 0.3D to 0.5D. Floor cones or splitters

```
are recommended under each of the pump inlet bells. See Section 9.8.4.2.3.2 for solids-bearing liquids.

--`
9.8.3.4.9  Centerline spacing

Centerline spacing of adjacent intake bells shall be no less than 2.5D.

9.8.3.4.10  Inlet conduit elevation

The elevation of the incoming conduit shall be adjusted so that a cascade is avoided at the minimum liquid level.

9.8.3.5   Tanks – pump suction

9.8.3.5.1 General

This section applies to partly filled tanks, pressurized or nonpressurized, handling non-solids-bearing liquids where
the outflow occurs with or without simultaneous inflow. The following design features are considered:

Tank geometry
Vertical cylindrical
Horizontal cylindrical
Rectangular

22 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Outlet orientation and location

Vertical, downwards
Horizontal, side
Horizontal, bottom
Vertical, upwards

Outlet configuration
Flush with tank interior surface
Protruding through tank interior surface

Outlet fitting
Straight
Cone
Bell
Note: In suction tank applications where piping is used to connect the intake to the pump, the piping shall comply
with ANSI/HI 9.6.6 Rotodynamic Pumps for Pump Piping.

9.8.3.5.2 Objectives

The purpose of this section is to recommend features of tank connections to minimize air or gas entrainment
during the pumping process. It is assumed that the pump is far enough downstream of the tank outlet, such that the
requirements of ANSI/HI 9.6.6 govern.

9.8.3.5.3 Discussion

Due to the formation of vortices inside the tank, air or gas entrainment can occur in pump suction tanks, even when
the tank outlet is totally submerged. Severe cases of air entrainment can cause erratic or noisy pump operation or
reduction in pump performance. A pump is affected by entrained air that can collect, and in severe cases, block the
impeller eye and cause loss of prime.
The extent of air entrainment, caused by vortex formation in a suction tank, depends on the vortex strength, sub-
mergence of the tank outlet, and the fluid velocity in the tank outlet. Vortices may occur in tanks under vacuum or
pressure, whether or not the level is varying or steady due to inflow.

9.8.3.5.4 Submergence

See Figure 9.8.3.5.4, examples 1 through 4. The recommended minimum submergence S of the outlet fitting below
the free surface of the liquid within the tank to prevent air core vortices, given tank outlet diameter D, may be
obtained from the relationship

S
= 1.0 + 2.3FD  (Eq. 9.8.3.5.4-1)
D
`---
` `

Where:
`
````-`-`

FD = Froude number = V
```

( gD)0.5
--`

D = outlet fitting diameter (see Figure 9.8.3.5.5)

V = outlet fitting velocity
g = acceleration of gravity
For further discussion of submergence, see Section 9.8.6.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 23

Rotodynamic Pumps for Pump Intake Design — 2018

Figure 9.8.3.5.4  Datum for calculation of submergence

9.8.3.5.5  Application options

Whereas Figure 9.8.3.5.4, examples 1 through 4, show how the calculated submergence value is to be applied,
Figure 9.8.3.5.5, examples 5 through 8, show where values of V and D are obtained for the three types of outlet fit-
ting designs: straight, cone-shaped, and bell-shaped. If the desired minimum submergence is less than that calcu-
lated by the above relationship, the outlet size, and therefore fluid velocity, may be adjusted to reduce the required
minimum submergence. It may be desirable to use a bell-shaped or cone-shaped fitting to reduce the head loss
in the fitting. In such cases, shown in Figure 9.8.3.5.5, examples 5 through 8, the largest diameter of the fitting is
used in the above equations to calculate velocity, V. If the inlet is fitted with a suction umbrella, the diameter of the
inlet, not the outside diameter of the umbrella, should be used as shown in Figure 9.8.3.5.5. Owing to the uncertain
approach conditions typically encountered in a closed tank or vessel, outlet vortex breakers as illustrated in Appen-
dix A, Figure A.13, should be considered.

24
` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Figure 9.8.3.5.5  Definitions of V and D for calculation of submergence

9.8.3.5.6  NPSH considerations

All the head losses incurred from the free liquid surface to the pump inlet must be considered when calculating the
NPSH available for the pump. If a submergence greater than the required NPSH for the pump is needed to provide
the recommended minimum submergence of a pump bell or pipe intake to reduce the probability that strong free
surface air core vortices will occur, then greater submergence governs must be used.

9.8.3.5.7  Simultaneous inflow and outflow

In general, tanks should not have the inlet pipe close to the tank outlet when inflow and outflow occur simultane-
ously. Suitable baffling or other flow distribution devices may be required to isolate the outlet or reduce the inlet
effects on flow patterns. Special attention should also be given to the design to avoid air entrainment with a non-
submerged inlet pipe.

Guidelines provided in Appendix A.7 may be helpful.

9.8.3.5.8  Multiple inlets or outlets

The design of tanks with multiple inlets and/or outlets should be such that unsatisfactory flow interaction does not
occur. Baffling or other flow distribution devices may be required to eliminate such effects.

Guidelines provided in Appendix A.7 may be helpful.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 25

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.6  Can vertical turbine pump intakes (clear liquids), including those with submersible motors
(refer to Appendix G)

9.8.3.6.1 General

A can pump is a pump that has a barrel around the pumping unit.

The purpose of this section is to establish criteria for the design of clear liquid intakes for open bottom and closed
bottom can vertical turbine pumps as well as for submersible (well motor driven) vertical turbine pumps. It is neces-
sary to avoid designs to simply fit into a piping arrangement without considering flow patterns to the can inlet or in
the barrel itself. For submersible turbine pumps, the cooling of the immersed motor must also be considered.

The intake design information provided is for vertical turbine type pumps with specific speed less than 100 (5000 US
customary units). Higher specific speed vertical mixed flow and propeller pumps may perform in a barrel; however,
they are more sensitive to hydraulic suction design. Refer to the pump manufacturer for specific can intake designs
for these pumps.

9.8.3.6.2 Objective

The following provides guidelines to avoid unfavorable flow conditions for both open bottom and closed bottom
vertical turbine can pump intakes.

9.8.3.6.3  Design considerations

It is necessary to design the can intake such that the first-stage impeller suction bell inflow velocity profile is uni-
form. An asymmetrical velocity profile may result in hydraulic disturbances, such as swirling, submerged vortices,
and cavitation that may result in performance degradation and accelerated pump wear.

It is recommended that the vertical pump be allowed to hang freely suspended and without restraining attachments
to its vertical pump can (riser). However, if it is necessary to install restraining attachments between the pump and
barrel, such as for seismic compliance, binding of the pump must be avoided.

The pump manufacturer should be consulted regarding the design of any component that affects the pump hydrau-
lic intake performance. These include the suction barrel, 90-degree turning vane elbow, and vortex suppressor.

9.8.3.6.4  Open bottom can intakes (Figure 9.8.3.6.4)

The minimum liquid level is considered a minimum operational level. When the pump is started, the liquid level will
reduce momentarily until the pump flow velocity is achieved. The intake piping must be large enough to limit draw-
down below the recommended minimum suction level to a period of less than 3 seconds during start-up.

Open bottom can intakes with flows greater than 315 L/s (5000 gpm) per pump require a physical model study.

Example 1 This pump intake configuration is particularly effective when liquid elevations (pump submergence) is
limited. Flows through a horizontal suction header with velocities up to 2.4 m/s (8.0 ft/s) can be effectively directed
into a vertical turbine pump by use of a 90-degree vaned elbow.

The 90-degree turning vane inlet diameter shall be sized to limit the inflow velocity to 1.5 m/s (5.0 ft/s). Attachment
of a 90-degree vaned elbow to the horizontal header is recommended to provide hydraulic thrust restraint. Caution
is necessary when using this intake configuration in liquids containing trash or crustaceans that attach to the turn-
ing vanes.

Example 2 The vortex suppressor and pump are an integral assembly that can be removed for repair, cleaning,
and inspection. A vortex suppressor is necessary to break up abnormal flow patterns ahead of the pump suction
bell. For vertical turbine pumps with rated flows less than 315 L/s (5000 gpm) the maximum horizontal header

26
` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

velocity is 1.8 m/s (6.0 ft/s) and the maximum riser velocity is 1.5 m/s (5.0 ft/s). The installation must allow the pump
to hang centered in the vertical riser pipe.

Example 3 When the vertical riser is located at the end of a suction header, a 90-degree vaned elbow must be
used to direct flow into the pump’s suction. This intake configuration is effective when liquid elevation (pump sub-
mergence) is limited. The 90-degree turning vane inlet diameter shall be sized to limit the inflow velocity to 1.5 m/s
(5.0 ft/s).

Example 4 A 90-degree long radius elbow may be used at the end of a suction header to direct flow into the pump
suction when velocities are less than 0.6 m/s (2.0 ft/s). Installing vanes in the elbow (although difficult) promotes a
uniform velocity flow profile. Velocities up to 1.5 m/s (5.0 ft/s) are acceptable when the elbow is fully vaned.

A flexible joint between the pump suction and the elbow is recommended to isolate the pump from piping loads.
Because this is a dry-pit application, the joints throughout the pump should be sealed against leakage by the use
of O-rings, gaskets, etc.

Figure 9.8.3.6.4  Open bottom can intakes

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 27

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.6.5  Closed bottom can (Figure 9.8.3.6.5)

The most typical can pump configurations are closed bottom. See Figure 9.8.3.6.5 for design recommendations
with various inlet pipe positions relative to the bell. For flows greater than 189 L/s (3000 gpm) refer to the manufac-
turer for applicability of flow straightening or vortex suppression devices and associated details. Closed bottom can
intakes for pump flows exceeding 440 L/s (7000 gpm) require a physical model study.

The can installation tolerances have a significant impact on the flow patterns and interaction with the pump. The
pump should be centered within the can with both the can and pump as plumb and level as possible to assure
concentricity along the submerged length of the pump. This becomes increasingly difficult as the pump and can
lengths increase, and is difficult to correct after the can installation is complete. Prior to pouring concrete and grout
during the can installation, the can should be secured in such a way to counteract the expected buoyancy effect
from the pour that will cause the can to float out of plumb and out of level. The following can installation tolerances
shall be achieved and confirmed prior to installing the pump.

1. For a can with an integral flange for mounting the pump, the mounting flange surface shall be level in all direc-
tions to within 0.4 mm/m (0.005 in/ft) or better if specified by the pump manufacturer.

2. For a can with a mounting flange separate from the barrel, the barrel element of the can shall be plumb to
ensure that it is centered at the suction bell elevation to within 3% of the suction bell diameter and separate
mounting flange shall be level in all directions to within 0.4 mm/m (0.005 in/ft) or better if specified by the pump
manufacturer.

Because of the limited volume provided by a can-type intake, surging of the liquid level within the can may be a
problem when operating with a partially filled can.

The intake piping must be large enough to limit drawdown below the recommended minimum liquid level to a
period of less than 3 seconds during start-up.

Under steady state operation, the hydraulic grade line shall allow for a minimum liquid level of 1.0D above the
crown of the inlet pipe. See Figure 9.8.3.6.5.

As shown in Figure 9.8.3.6.5, for nominal velocities up to 1.2 m/s (4 ft/s) the minimum offset height between the
inlet piping and suction bell shall be 2.0D1 and have a minimum straight length of five pipe diameters. For inlet pip-
ing with nominal velocities between 1.2 m/s (4 ft/s) and 1.8 m/s (6 ft/s) the minimum offset height between the inlet
piping and suction bell shall be 4.0D1 and shall have a minimum straight length of two pipe diameters.

28 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

` ``` ```` ` ` ` ` ` `
 Rotodynamic Pumps for Pump Intake Design — 2018
--
-

Figure 9.8.3.6.5  Closed bottom can

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 29

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.3.7  Unconfined intakes (Figure 9.8.3.7)

9.8.3.7.1 Scope

Unconfined intakes involve pumps installed on platforms or other structures where the intake lacks guide walls,
walls of a sump, or other flow-guiding structures. Typical installations include intakes on rivers, canals or channels,
lakes, and pumps located on platforms for seawater systems.

The unconfined intake design information provided is for pumps having flow rates no greater than 315 L/s (5000 gpm).

9.8.3.7.2  Cross-flow velocities and pump location

Pumps with unconfined intakes are often located where a unidirectional cross-flow occurs, or on platforms where
tidal variations may cause highly complex current conditions around the pump inlet bell. The minimum recom-
mended distance from an obstruction to the pump suction in the direction of any current that could cause wake
effects is five times the maximum cross-sectional dimension of the obstruction. If an obstruction to the flow is down-

`---
stream of the pump, the minimum recommended distance is 0.75D.

` `
`
````-`-`
Cross-flow velocities shall be less than 25% of the bell velocity, but the designer may have little control over this
variable. Installations with higher cross-flow velocities require special flow correction devices, which are beyond this

```
design standard (see Appendix A for reference information). For higher cross-flow velocities, supplemental lateral

--`
support of the pump may be required.

If debris or bottom sediments are not a concern, the inlet bell shall be located 0.3D to 0.5D above the bottom to
minimize submerged vortices. For applications where suspension of bottom debris may be a problem, a 5D mini-
mum clearance is suggested.

For installations on platforms along the seashore, suspension of sand during storms is unavoidable due to wave
action. In some cases, a bed of armor stone around the intake has proved useful in minimizing suspension of sed-
iments. The design of such armor layers should be performed with the assistance of an engineer with experience
in sediment transport and design of riprap protection, as the proper design of armor stone protection requires spe-
cialized techniques.

30 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Figure 9.8.3.7  Unconfined intakes

9.8.3.7.3  Debris and screens

Debris is of particular concern for unconfined intakes. Light debris loading may be accommodated by screens
attached to the pump bell. Special design considerations are required to accommodate heavy debris loading.

Large floating debris and ice that could damage the pump is also of concern. A barrier may be required to protect
the pump. These barriers should not introduce wake disturbances into the pump.

9.8.3.7.4 Submergence

S
= 1.0 + 2.3FD  (Eq. 9.8.3.7.4-1)
D

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

` ``` ```` ` ` ` ` ` `
31
Rotodynamic Pumps for Pump Intake Design — 2018

Where:

FD = Froude number = V
( gD)0.5
D = outside diameter of suction bell
V = average suction bell velocity

For further discussion of submergence, see Section 9.8.6.

9.8.4  Intake structures for solids-bearing liquids

9.8.4.1 General

Wet wells for solids-bearing liquids require special considerations to allow for the removal of floating and set-
tling solids. These considerations include wet-well geometry and provisions for cleaning of the structure to remove
material that would otherwise be trapped and result in undesirable conditions such as odor and poor approach flow
to the pumps. See Section 9.8.4.1.2 for a complete list.

Some dimensions given in this section may be impacted by actual pump geometry, or by specific product opera-
tional limitations. Where appropriate, these dimensions are indicated by the note “consult with manufacturer”.

9.8.4.1.1 Scope

This section applies specifically to installations where the pumped liquid contains solids that may float or settle in
the wet well. Fluids such as wastewater, industrial discharges, storm or canal drainage, combined wastewater, and
some raw water supplies are included in this category.

9.8.4.1.2 Objectives

The objective of this part of the standard is to introduce special design features recommended for wet wells used
in solids-bearing liquid applications. These features are intended to eliminate or minimize accumulations of solids,
thereby reducing maintenance. Organic solids accumulations not removed may become septic, causing odors,
increasing corrosion, and releasing hazardous gases.

9.8.4.1.3 Principles

The design of a solids-bearing wet well must both provide for proper approach flow to the pumps as described in
Section 9.8.2, and prevent the accumulation of sediments and surface scum in the sump. The main principle is to
minimize horizontal surfaces in the wet well anywhere but directly within the influence of the pump inlets, thereby
directing all solids to a location where they may be removed by the pumping equipment. Vertical or steeply sloped
sides shall be provided for the transition from upstream conduits or channels to pump inlets.

The following points shall be considered in addition to those found in Section 9.8.2: `---
` `

• Flow of liquid from the sump entrance should be directed toward the pump inlets in such a way that the flow
`
````-`-`

reaches the inlets with a minimum of swirl.

```

• Although excessive turbulence or large eddies should be avoided, some turbulence could help to prevent the
--`

formation and growth of vortices.

• Sediment, which could become septic, must not accumulate within the sump. Stagnant regions or regions of
such low velocity where sedimentation might occur shall be avoided. A sloping floor and fillets (or benching)
­often helps to reduce sedimentation. For large variations in flow, part of the sump can be dedicated to low
­inflows with a lower floor level and a smaller pump.

32 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

• Surface scum, floating sludge, and debris can accumulate in any relatively calm region of the liquid surface;
and this material must be pumped away. The liquid level should be intermittently lowered as much as possible
to increase both velocity and turbulence near the sump bottom; however, air should not be drawn into the pump.
The occasional increases in velocity will also assist in removing the accumulation of sediment on the floor.

• The sump should be as small and as simple as feasible; however, a minimum required sump volume may be
specified for other reasons, such as to provide for a minimum retention time or to ensure that only a certain
number of pump starts per hour occur.

9.8.4.1.4  Vertical transitions

Transitions between levels in wet wells for solids-bearing liquids shall be at steep angles (60 degrees minimum
for concrete, 45 degrees minimum for smooth-surfaced materials such as plastic and coated concrete; all angles
relative to horizontal) to prevent solids accumulations and promote movement of the material to a location within
the influence of the currents entering the pump intakes. Horizontal surfaces should be eliminated where possible
except near the pump inlet. See Figure 9.8.4.1.4.

Figure 9.8.4.1.4  Open trench-type wet well

33
` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.4.1.5  Horizontal surfaces near inlet

The horizontal surface immediately in front (for formed suction inlets) or below (for bell inlets) should be limited to
a small, confined space directly in front of or below the inlet itself. To make cleaning more effective, the walls and
floor forming the space must be confined so that currents can sweep floating and settled solids to the pump inlet.
See Figure 9.8.4.4.4.

9.8.4.1.6  Cleaning procedures

Removal of solids from wet wells, designed in accordance with these principles, can be achieved by operating the
pumps selectively to lower the level in the wet well until the pumps lose prime. Both settled and floating solids are
removed by the pumping equipment and discharged to the force main (or discharge conduit). This cleaning pro-
cedure momentarily subjects the pumps to vibration, dry running, and other severe conditions. Consult the pump
manufacturer before selecting the pumping equipment. A system analysis should be performed to verify that air
and solids entering the system during the cleaning procedure can be removed, as applicable. The frequency of
cleaning cycles depends on local conditions, and therefore should be determined by experience at the site.

Alternatively, liquid jets or mixers positioned to create horizontal and vertical currents can be used intermittently or
continuously to maintain suspension and direct floating and settled solids toward the pump intakes. The solids are
swept into the pump intake for removal. Caution should be exercised when using jets or mixers to avoid inducing
continuous currents near pump inlets that could result in damage to the pumping equipment.

9.8.4.1.7  Wet-well volume

--

Wet wells for constant-speed pumps should be constructed to minimize size for economy and to facilitate cleaning.
-

One approach is to provide storage for pump regulation in the upstream conduit or channel, as well as in the wet
well itself. Refer to Appendix B for guidance on sump volume for constant-speed pumps and Appendix C for stor-
age in the upstream conduit.

Wet wells for variable-speed pumping stations designed to match outflow with inflow need not be designed for
storage, but rather only to accommodate the inlets and the geometry required for velocity limitations and cleaning.

CAUTION: While no storage volume is required for continuous operation, the designer should provide ade-
quate volume to avoid inappropriately short pump cycle intervals between pump starts, which could dam-
age pump motors.

9.8.4.2  Trench-type wet wells for solids-bearing liquids

9.8.4.2.1 General

The purpose of this section is to establish criteria for design of trench-type wet wells for solids-bearing liquids such
as stormwater, wastewater, and canal-type pumping stations. Physical model studies shall be conducted for any
installation with individual pump capacities exceeding 1260 L/s (20,000 gpm) or stations with capacities greater
than 3155 L/s (50,000 gpm).

9.8.4.2.2 Objectives

Trench-type wet wells are designed to provide for cleaning with the periodic operation of the pumping equipment
using a special procedure. This standard provides guidance on the geometry necessary to induce scouring veloc-
ities during the cleaning procedure. Experience has shown that trench-type wet wells with an ogee transition
between the entrance conduit and the trench floor provides optimum geometry for efficient cleaning operations.

Refer to Sections 9.8.4.2.3 to 9.8.4.2.3.5 and Figure 9.8.4.1.4 for recommendations for trench-type wet wells.
Trench-type wet wells can be used with both constant-speed and variable-speed pumping equipment.

34 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

There is no difference between wet wells for variable as compared with constant-speed pumps, but there is a dif-
ference between inlet conduits for the two kinds of pumping stations. With variable-speed pumps, there is no need
for storage if pump discharge equals wet well inflow. Consequently, the liquid level in the wet well can be made to
match the liquid level in the upstream conduit.

When constant-speed pumps are used, the liquid level must fluctuate – rising when pumps are off and falling when
they are running. There must be sufficient active storage to prevent excessive frequency of motor starts. As trench-
type wet wells are inherently small and not easily adapted to contain large volumes of active storage, it is desirable
to dedicate a portion of the upstream conduit to storage. The dedicated portion is called an approach pipe. It is usu-
ally 75–150 mm (3–6 in) larger than the conduit upstream of the dedicated portion, and it is laid at a compromise
gradient of 2% (although other gradients could be used.) At low liquid level, the velocity in the approach pipe is
supercritical, thus leaving a large part of the cross section empty for storage as the liquid level rises. The design of
approach pipes is not a part of these standards, but the essentials of design are given in Appendix C.

9.8.4.2.3  Approach flow

The velocity in the approach channel or conduit, upstream from the wet well, shall be no greater than:

• 1.2 m/s (4.0 ft/s) with the axis of the channel or conduit coaxial with the axis of the wet well for vertical pumps

`---
or submersible pumps

` `
`
````-`-`
• 0.9 m/s (3.0 ft/s) with the axis of the channel or conduit coaxial with the axis of the wet well for pumps with suc-
tion piping extending from the dry well into the wet well

```
--`
The influent pipe upstream from the trench shall be straight and free of fittings or devices that will disrupt the flow
uniformity entering the trench for a distance equal to at least eight times the influent pipe diameter.

9.8.4.2.3.1  Inlet transition

The ogee spillway transition at the inlet to the wet-well trench is designed to convert potential energy in the influent
liquid to kinetic energy during the wet-well cleaning cycle. The curvature at the top of the spillway should follow the
trajectory of a free, horizontal jet issuing from under the sluice gate and discharging approximately 75% of the flow
rate of the last pump. The radius of the curvature, r, shall be at least 2.3 times the pressure head upstream of the
sluice gate during cleaning. The radius of curvature at the bottom of the ogee needs to be large enough only for a
smooth transition to horizontal flow: 0.5 to 1.0r is sufficient.

To produce smooth flow down the ogee ramp and avoid standing waves, the discharge under the sluice gate should
be uniform in depth across the 2D width of the trench. Either (1) a short transition from a circular to a rectangular
section, as shown in Figure 9.8.4.1.4 or (2) a short rectangular recess in front of the sluice gate is recommended.

9.8.4.2.3.2  Inlet floor clearance

All bell-type pump inlets, except that farthest from the wet-well inlet, shall be located 0.5D above the floor of the
wet-well trench. The inlet for the last pump (farthest from the wet-well inlet) shall be located 0.25D above the floor
of the trench. See Figure 9.8.4.1.4.

For pumps that may be sensitive to loss of prime (due to entrainment of air from surface vortices), the last pump
inlet can be lowered by 0.25D provided the floor near the intake is lowered by the same amount. All other dimen-
sions and velocities for this arrangement shall comply with those given in Figure 9.8.4.1.4.

For submersible wastewater pumps, an inlet extension and nozzle, sized for a peak entrance velocity not to exceed
the velocity of 1.7 m/s (5.5 ft/s) and fitted to the pump inlet, is necessary to meet the dimensional requirements for
development of the trench.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 35

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.4.2.3.3  Inlet splitters and cones

Floor-mounted flow splitters aligned with the axis of the trench are recommended. They must be centered under
the suction bells for all but the pump inlet farthest from the wet-well entrance. A floor cone should be installed under
the pump inlet farthest from the wet-well inlet conduit or pipe as shown in Figure 9.8.4.1.4.

9.8.4.2.3.4  Anti-rotation baffle and vanes

An anti-rotation baffle positioned on the wall at the last pump inlet, shown in Figure 9.8.4.1.4, is needed to ensure
satisfactory performance during the cleaning cycle. The anti-rotation baffle should protrude toward the pump as far
as practicable. Vanes in line with this baffle are needed on either side of the floor cone, with a height as high as
practicable yet compatible with the 0.25D bell clearance at that pump inlet.

9.8.4.2.3.5  Cleaning procedure

Trench-type wet wells for solids-bearing liquids can be quickly cleaned by choosing a time when the inflow is about
half of the capacity of the last pump. If that pump, operating at full speed, takes more than about a minute to lower
the liquid level to the middle of the trench, two pumps can be activated. The liquid flowing down the ramp reaches
supercritical velocity and forms a hydraulic jump that, taking all solids with it, moves to the last pump. The Froude
number before the jump at the last pump should be no less than 3.5. Informative material regarding trench-type,
wet-well calculations is located in Appendix C. If the inflow is insufficient for cleaning, enough liquid to complete
the cleaning cycle can be stored in the upstream conduits by stopping all pumps for a short time. If the inflow is too
high, two pumps can be operated to produce enough turbulence to clean the trench. The hydraulic jump should
move from the toe of the ramp to the last pump in no more than 30 seconds, because operation at low intake sub-
mergence is severe service for the pump. As the hydraulic jump passes under each pump intake, the pump loses
prime and should be stopped.

Note: Pumps must be reprimed prior to the next start.

This cleaning procedure, a push-button operation that theoretically can be completed in less than three or four min-
utes, rids the wet well of all sludge, grit, and scum; however, grease accumulations on walls between normal high
and low liquid levels must be manually hosed off from time to time. An epoxy coating or PVC lining is substantially
better than concrete for ease and speed of the washing process. Varying the high liquid level a few inches changes
the rim of grease formed at high liquid level into a band and somewhat prolongs the intervals between hosings.
A liquid flow rate of 1.6 L/s (25 gal/min) at a nozzle Pitot pressure of 620 kPa (90 psi) is adequate. The wet well
should be designed for convenience and ease in washing the walls.

9.8.4.3  Circular plan wet pit for solids-bearing liquids

9.8.4.3.1  Wet-pit design

The design of the wet pit should adhere to the general recommendations given in Section 9.8.3.3. As stated in that
--

section, circular pump sumps for flows exceeding 315 L/s (5000 gpm) per pump require a physical model study.
Additionally, the bottom of the wet pit shall have sloped surfaces around the inlet bells or pumps, as shown in
Figures 9.8.4.3.1a, b, and c. The designs shown in this section are based on single pump operation, i.e., one duty
-

pump and one installed spare.

9.8.4.3.2 Accessories

The use of pump and sump accessories that cause collection or entrapment of solids should be limited to a practi-
cal minimum.

36 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

9.8.4.3.3  Cleaning procedure and low liquid level

The frequency of cleaning cycles depends on local conditions, and therefore should be determined by experience
at the site. Removal of settled solids is effected each time a pump is activated, but to aid in the removal of floating
solids, the wet well may need to be pumped down below the level of minimum submergence, to a level equal to a
submergence of 0.5D to 1.0D. Such a submergence level is lower than that recommended in Section 9.8.6. In this
case, the liquid surface area is at a minimum and the pump intake submergence is low enough to create a strong
surface vortex (numbers 4 through 6 in Figure 9.8.7.5a–i). This level is called the low liquid level. Pumping under
these severe conditions will cause noise, vibration, and high loads on the impeller and hence should be limited to a
period as recommended by the pump manufacturer. The pumps should be stopped as soon as they lose prime, or
as soon as the sump is free of floating debris.

9.8.4.3.4  Floor clearance C

The recommended floor clearance C is between 0.3D and 0.5D. In certain cases this clearance may need to be
larger, depending on the size of expected solids in the liquid relative to the calculated value of C.

Figure 9.8.4.3.1a  Circular wet pit with sloping walls and minimized horizontal floor area (dry-pit pumps)
`---
` `
`
````-`-`
```
--`

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 37

Rotodynamic Pumps for Pump Intake Design — 2018

--
-

Figure 9.8.4.3.1b  Circular wet pit with sloping walls and minimized horizontal floor area
(submersible pumps shown for illustration)

38 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

`---
` `
`
````-`-`
```
--`

Figure 9.8.4.3.1c  Circular wet pit with sloping walls and minimized horizontal floor area
(wet-pit pumps shown for illustration)

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 39

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.4.4  Rectangular wet wells for solids-bearing liquids

9.8.4.4.1 General

The geometry of rectangular wet wells is not particularly suited for use with solids-bearing liquids, but with special
provisions for frequent cleaning, such wet wells may be acceptable. Physical model studies shall be conducted
for any installation with individual pump capacities exceeding 2520 L/s (40,000 gpm) or stations with capacities
greater than 6310 L/s (100,000 gpm).

9.8.4.4.2 Objectives

The objective of this section is to describe several means for minimizing or eliminating accumulations of solids
before they interfere with the operation of the pumps or before they become septic and generate excessive odors
that must be treated.

9.8.4.4.3  Control of sediments

Several means for controlling the accumulation of sediments are possible, such as:

• Designing the wet well to provide currents swift enough (e.g., 1.0 m/s [3.0 ft/s] or more) to carry settleable solids
to the pump intakes.

• Violent mixing to suspend sediments while the mixture is being removed by the main pumps. These methods
include:

– Use of submerged mixers.

– Connecting the pump discharge or force main to a valve and into the wet well. About half of the pump dis-
charge is allowed to recirculate back into the wet-well.

• Dewatering the wet well and sweeping solids to the pumps with a high-pressure hose.

• Vacuuming both floating and settled solids out of the wet well, usually by an external pump and hose.

• Dewatering one side of the wet well (if possible) and removing the solids.

9.8.4.4.4  Confined wet-well design

In this arrangement each suction inlet bell is located in a confined pocket to isolate the pump from any flow distur-
bances that might be generated by adjacent pumps, to restrict the area in which solids can settle, and to maintain
higher velocities at the suction inlet to minimize the amount of solids settling out of the flow.

See Figure 9.8.4.4.4 for the arrangement of a confined wet well.

` ``` ```` ` ` ` ` ` `

40 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Figure 9.8.4.4.4  Confined wet-well design

9.8.4.4.4.1  Suction inlet clearance

All suction inlets shall be located D above the floor of the wet well unless otherwise recommended by the man-
4
ufacturer. The sidewalls of the individual cell should be 1.5D to 2.0D in dimension. The depth of the individual cell
must be a minimum of 2.0D. A cone shall be installed under each suction inlet.

9.8.4.4.4.2  Anti-rotation baffle

Anti-rotation baffles are required for individual flows in excess of 189 L/s (3000 gpm).

9.8.4.4.4.3  Cleaning procedure

--

Removal of settled solids from wet wells, designed in accordance with Figure 9.8.4.4.4, can be achieved by oper-
ating the pumps one at a time at full speed for about two minutes. Typically, only one pump should be operated at a
-

time to avoid excessive drawdown of the liquid level in the sump.

The majority of floating solids are removed from the sump by operating the pumps one at a time at full speed while
restricting the flow into the wet well to 60% to 80% of the flow rate of the pump at full speed. Adjusting the sluice
gate is the normal method of flow restriction. As the liquid level in the wet well falls, swift currents will suspend most
of the floating debris, causing them to be pumped from the trench. The pump will eventually lose prime and must
be stopped immediately.

Both settled and floating solids are removed by the pumping equipment and discharged to the force main (or
discharge conduit). This cleaning procedure momentarily subjects the pumps to vibration, dry running, and other
severe conditions. The frequency of cleaning cycles depends on local conditions, and therefore should be deter-
mined by experience at the site. Normally, the cleaning operation will take less than five minutes to perform and the
duration between cleaning cycles would typically be one to two weeks.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 41

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.5  Inlet bell design diameter (D)

9.8.5.1 General

The purpose of this section is to provide criteria for the selection of a pump or suction pipe inlet bell design diame-
ter, whether or not the actual pump ultimately to be used has been selected. The primary use of the inlet bell design
diameter is for designing the sump. The actual pump bell diameter shall fall within the range dictated by the allowed
range of inlet bell velocities.

9.8.5.2 Objective

Designing a sump to achieve favorable inflow to the pump or suction pipe bell requires control of various sump
dimensions relative to a pump suction bell or inlet pipe bell diameter. For example, the clearance from the bell to
the sump floor and sidewalls and the distance to various upstream intake features is controlled in these standards
by expressing such distances in multiples of the pump suction bell or inlet pipe bell diameter. Such standardization
of conditions leading to, and around, the inlet bell reduces the probability that strong submerged vortices or exces-
sive pre-swirl will occur. Also, the required minimum submergence to prevent strong free surface vortices is related
to the inlet bell diameter (see Section 9.8.6).

However, only the use of bell sizes within the guidelines provided in this section will produce sump dimensions that
comply with these standards. Use of bell diameters outside the range recommended herein will also comply with
these standards if a physical model study is conducted in accordance with Section 9.8.7 to confirm acceptable
inflow conditions to the pump as required by Section 9.8.7.7.

It is recommended that the inlet bell diameter be chosen based on achieving the bell inlet velocity that experience
indicates provides acceptable inflow to the pump. The bell inlet velocity1 is defined as the flow through the bell
(i.e., the pump flow) divided by the area of the bell, using the diameter of the bell (see Figure 9.8.3.5.5). Information
on acceptable average bell inlet diameters and velocities are provided in Tables 9.8.5.2a and b, and Figure 9.8.5.2,
based on a survey of inlet bell diameters used by pump vendors and industry experience. The solid line represents
the average pump bell diameter from the survey, corresponding to a bell inlet velocity of 1.7 m/s (5.5 ft/s).

Industry experience indicates that the recommended inlet bell velocity V may vary as follows:

`---
a) For flows less than 315 L/s (5000 gpm), the inlet bell velocity shall be ≤2.7 m/s (9.0 ft/s).

` `
`
b) For flows equal to or greater than 315 L/s (5000 gpm), but less than 1260 L/s (20,000 gpm), the velocity shall

````-`-`
be ≤2.4 m/s (8.0 ft/s).

```
c) For flows equal to or greater than 1260 L/s (20,000 gpm), the velocity shall be ≤2.1 m/s (7.0 ft/s).

--`
These permissible ranges in inlet bell velocity are given in Tables 9.8.5.2a and b and are shown on Figure 9.8.5.2
(by the dashed lines) in terms of the recommended bell diameter range for a given flow per pump or inlet. Although
the survey indicated that pumps with bells outside this range may be proposed, experience indicates that inlet bell
velocities higher than the recommended range can cause hydraulic problems and shall be verified with a physical
model study. Inlet bell velocities lower than the recommended bell velocity of 1.7 m/s (5.5 ft/s) are acceptable and
may be necessary to accommodate mechanical components. Also at sites where the depth of the sump will be lim-
ited, lower inlet bell velocities can be used to reduce submergence requirements. It should be noted that the use of
bell velocity values below the recommended bell velocity would produce larger pump bells and, therefore, sumps.

For sump design prior to pump selection, the recommended inlet bell diameter shown by the solid line on Figure
9.8.5.2 shall be used. This recommended bell diameter is based on an inlet velocity of 1.7 m/s (5.5 ft/s).

1 For submersible pumps, see 9.8.1.3 Nomenclature.

42 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Table 9.8.5.2a  Acceptable velocity ranges for inlet bell diameter D (metric units)

Recommended Inlet Bell Design

Pump Flow Range Q, L/s Acceptable Velocity Range, m/s
Velocity, m/s
<315 V = 1.7 V ≤ 2.7
≥315
V = 1.7 V ≤ 2.4
<1260
≥1260 V = 1.7 V ≤ 2.1

( )
0.5
Q
Note: See Figure 9.8.5.2 for corresponding inlet diameters calculated according to D = 785V  (Eq. 9.8.5.2-1)

where D = outside bell diameter, m.

Table 9.8.5.2b  Acceptable velocity ranges for inlet bell diameter D (US customary units)

Recommended Inlet Bell Design

Pump Flow Range Q, gpm Acceptable Velocity Range, ft/s
Velocity, ft/s
<5000 V = 5.5 V≤9
≥5000
V = 5.5 V≤8
<20,000
≥20,000 V = 5.5 V≤7

( )
0.5
0.409Q
Note: See Figure 9.8.5.2 for corresponding inlet diameters calculated according to D = V  (Eq. 9.8.5.2-2)

where D = outside bell diameter, in.

43
` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

Rotodynamic Pumps for Pump Intake Design — 2018

Figure 9.8.5.2  Inlet bell design diameter

This process will allow the sump design to proceed. When the pump is specified and selected, the outside diameter
of its bell (without added horizontal rings or “umbrellas,” sometimes used as vortex suppressors) shall fall within the
acceptable range to produce an inlet velocity within the limits indicated in Tables 9.8.5.2a and b. An inlet bell diam-
eter within this range will produce a sump geometry that complies with these standards on minimum submergence
and sump dimensions (see Figure 9.8.5.2).

9.8.6  Required submergence for minimizing surface vortices

9.8.6.1 General

This section concerns the recommended minimum submergence of a pump bell or pipe intake to reduce the prob-
ability that strong free surface air core vortices will occur. Submerged vortices are not believed to be related to
submergence and are not considered in this section.

If a submergence greater than recommended herein is needed to provide the required NPSH for the pump, that
greater submergence would govern and must be used.

Approach-flow skewness and the resulting circulation have a controlling influence on free surface vortices in spite
of adequate submergence. Due to the inability to predict and quantify approach-flow characteristics for each par-
ticular case without resorting to physical model studies, and the lack of available correlation between such char-
acteristics and vortex strength, the recommended minimum submergence given herein is for a reasonably uniform
approach flow to the pump suction bell or pipe inlet. Highly nonuniform (skewed) approach flows will require the

` ``` ```` ` ` ` ` ` `

44 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

application of vortex suppression devices (not part of this standard) such as those offered for information in Appen-
dix A. Such devices can be more practical in suppressing vortices than increased submergence.

Even for constant flows, vortices are not steady in position or strength, usually forming and dissipating sporadically.
This is due to the random nature by which eddies merge to form coherent circulation around a filament and by
which turbulence becomes sufficient in intensity to disrupt the flow pattern. For these reasons, the strength of vor-
tices versus time shall be observed to obtain an average and a maximum vortex type for given conditions, and this
process is enhanced by defining a measure of vortex strength, as illustrated in Figure 9.8.7.5a.

9.8.6.2  Controlling parameters

By use of dimensional analysis, it may be shown that a given vortex type, VT, is a function of various dimensionless
parameters.

(
VT = f FD , NΓ ,
S
D
,G  ) (Eq. 9.8.6.2-1)

Where:

VT = vortex type (strength and persistence)

f = a function

FD = V
Froude No. =
( gD)0.5
NΓ = Circulation No., ΓD/Q, of approach flow
S = Submergence
D = Outside diameter of bell or inside diameter of pipe inlet
G = Geometry
Γ = Circulation (2prVt for concentric flow about a point with a tangential velocity Vt at radius r)

V = (
Velocity at inlet =
4Q
π D2 )
g = Gravitation acceleration
Q = Flow

For a given geometry and approach flow pattern, the vortex strength would only vary with the remaining parame-
ters, that is

(
VT = f FD ,
S
D )  (Eq. 9.8.6.2-2)

This formula indicates that a plot of S/D versus FD would contain a family of curves, each representing different val-
ues of vortex strength, VT (refer to Figure 9.8.7.5a). Selection of one vortex strength of concern, such as a vortex
without air entrainment, would yield a unique relationship between S/D and FD, which corresponds to that vortex, all
for a given geometry and approach flow pattern (circulation).

For typical intake geometry and relatively uniform approach flow (i.e., low values of the circulation parameter), data
and experience suggests that the following recommended relationship between submergence and the Froude num-
ber reduces the likelihood of strong air core vortices (Hecker, G.E., 1987). While the following equation provides a
good balance between intake depth and strong vortex potential, further reduction of vortex activity to acceptable
levels using mitigation devices may be required.
--
-

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 45

Rotodynamic Pumps for Pump Intake Design — 2018

S
= 1.0 + 2.3FD  (Eq. 9.8.6.2-3)
D
Where:

S = Submergence above a horizontally oriented inlet plane (vertical inlet pipe) or above the centerline
of a vertically oriented inlet plane (horizontal inlet pipe)
D = Diameter of inlet opening (equivalent diameter for noncircular openings, giving the same area as a
­circular opening)

FD = Froude No. = V
( gD)0.5
V = Velocity at inlet face = Flow/Area

This equation indicates that a minimum of one diameter of submergence must be provided, even at negligible inlet
flows or velocities, and that the relative submergence, S/D, increases from that value as the inlet velocity increases.
This is reasonable, since the inlet velocity (flow) provides the energy to cause a potentially greater vortex strength
if the relative submergence were not increased.

The relative submergence would only be constant if the Froude number for various inlets were constant. Informa-
tion collected by the Hydraulic Institute (not included herein) shows that the average inlet Froude number for bells
of typical pump applications is not constant, and that a range of Froude numbers would be possible at a given
design flow. Even the restricted range of inlet bell diameters (and velocities) at a given flow recommended in Sec-
tion 9.8.5 allows some variation in the Froude number. Thus, Equation 9.8.6.2-3 is recommended, rather than a
fixed relative submergence.

9.8.6.3  Application considerations

For a given flow, Q, an inlet diameter may be selected in accordance with Section 9.8.5. Rearranging Equation
9.8.5.2-2 the recommended minimum submergence for that diameter D would be given by

Metric units (S is in meters for g = 9.81 m/s2, Q in L/s, and D in meters):

S=D+
Q
( )
D1.5  (Eq. 9.8.6.3-1)
1069
US customary units (S is in inches for g = 32.2 ft/s2, Q in gpm, and D in inches):

0.574Q
S=D+
D1.5

The above illustrates that the actual submergence depends on the selection of D for a given flow. As D increases,
the first term causes an increase in submergence, whereas the second term causes a decrease. These opposing
trends imply a minimum value of S at some D for a given flow, and differentiating S with respect to D, allows deter-
mining that value. However, for the range of recommended bell diameters in Section 9.8.5, the change of S with D
for a given flow is minimal, and D for pump bells should be selected based on other considerations.

For the inlet bell design diameter recommended in Section 9.8.5, the required minimum submergence for reducing
the severity of free surface vortices is shown on Figures 9.8.6.3. These figures also show the recommended min-
imum submergence for the limits of the bell diameter that comply with these standards, see Figures 9.8.5.2 and
Tables 9.8.5.2a and b. Due to the small change in submergence, no change in submergence from that calculated
with the recommended bell diameter is needed, as long as the final selected bell diameter is within the limits that
comply with these standards.

` ``` ```` ` ` ` ` ` `

46 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

`---
` `
`
````-`-`
```
--`

Note: See Figure 9.8.5.2 for corresponding inlet diameters.

Figure 9.8.6.3  Minimum submergence to minimize free surface vortices

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 47

Rotodynamic Pumps for Pump Intake Design — 2018

9.8.7  Physical model studies of intake structures and pump suction piping

9.8.7.1  Need for a physical model study

A properly conducted physical model study is a reliable method to identify unacceptable flow patterns at the pump
suction for given sump or suction piping design and to derive acceptable intake sump or piping designs. Consider-
ing the cost for a physical model study, an evaluation is needed to determine if one is required. A physical hydraulic
model study shall be conducted for pump intakes with one or more of the following features:

• A suction intake arrangement with elevation relative to water level that does not provide the minimum submer-
gence requirement of this standard, irrespective of pump manufacturer’s stated submergence values.

• The intake design is not a standard intake design presented in this standard or the geometry (such as bay width,
bell clearances, sidewall angles, bottom slopes, distance from obstructions, the bell diameter, submergence, or
piping changes, etc.) deviates from this standard.

• There is no prior physical model study for the intake design considered in terms of physical features and flow
rates.
--

• Nonuniform or nonsymmetric approach flow to the pump sump exists (e.g., intake from a significant cross-flow,
use of dual flow or drum screens; use of elbows, bends, or multiple screens just upstream of a trench-type
-

wetwell; or a short-radius pipe bend near the pump suction, etc.).

• Proper pump operation of a critical service or application as defined by the customer (such as a safety-related
system).

• Pump repair, remediation of a poor design, and the impacts of inadequate performance or pump failure all to-
gether would cost more than 10 times the cost of a physical model study.

• Circular stations with four or more pumps.

• For trench type wet wells (clear or solids-bearing liquids) the pumps have flows greater than 1260 L/s (20,000 gpm)
per pump or the total station flow with all pumps running would be greater than 3155 L/s (50,000 gpm).

• Circular pump sumps (clear or solids-bearing liquids) with flows exceeding 315 L/s (5000 gpm) per pump re-
quire a physical model study (see Sections 9.8.3.3 and 9.8.4.3). Circular pump sumps (clear liquids) per Figures
9.8.3.3.1c and 9.8.3.3.1f with station flows exceeding 315 L/s (5000 gpm) require a physical model study.

• The pumps of an open bottom barrel or riser arrangement with flows greater than 315 L/s (5000 gpm) per pump
(see Section 9.8.3.6).

• The pump of a closed bottom can intake has flows greater than 440 L/s (7000 gpm) (see Section 9.8.3.6).

• The pumps have flows greater than 2520 L/s (40,000 gpm) per pump or the total station flow with all pumps
running would be greater than 6310 L/s (100,000 gpm).

When evaluating the impacts of inadequate performance or pump failures, the probability of failure may be con-
sidered, such as by comparing the proposed intake design to other intakes of essentially identical design and
approach flow that operate successfully. The physical model study shall be conducted by a hydraulic laboratory
using personnel that have experience in modeling pump intakes.

9.8.7.2  Physical model study objectives

Adverse hydraulic conditions that can affect pump performance include free and subsurface vortices, swirl approach-
ing the pump impeller, flow separation at the pump bell, and a nonuniform axial velocity distribution at the suction.

48 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

Free surface vortices are detrimental when their core is strong enough to cause a (localized) low pressure at the
impeller and because a vortex core implies a rotating rather than a radial flow pattern. Subsurface vortices also
have low core pressures and originate closer to the impeller. Strong vortex cores may induce fluctuating forces on
the impeller and cavitation. Subsurface vortices with a dry-pit suction inlet are not of concern if the vortex core and
the associated swirling flow dissipate well before reaching the pump suction flange.

Pre-swirl in the flow entering the pump exists if a tangential component of velocity is present in addition to the axial
component. Swirl alters the inlet velocity vector at the impeller vanes, resulting in undesired changes in pump per-
formance characteristics, including potential vibration.

A reasonably uniform axial velocity distribution in the suction flow (approaching the impeller) is assumed in the
pump design, and nonuniformity of the axial velocity may cause uneven loading of the impeller and bearings.

`---
A properly conducted physical model study can be used to derive remedial measures, if necessary, to alleviate

` `
these undesirable flow conditions due to the approach upstream from the pump impeller. The typical hydraulic

`
````-`-`
model study is not intended to investigate flow patterns induced by the pump itself or the flow patterns within the
pump. The objective of a model study is to ensure that the final sump or piping design generates favorable flow

```
conditions at the inlet to the pump.

--`
9.8.7.3  Physical model similitude and scale selection

Physical models involving a free surface are operated using Froude similarity because the flow process is con-
trolled by gravity and inertial forces. The Froude number, representing the ratio of inertial to gravitational forces, can
be defined for pump intakes as:

u
F =  (Eq. 9.8.7.3-1)
( gL)0.5

Where:

u = average axial velocity (such as in the suction bell)

g = gravitational acceleration
L = a characteristic length (usually bell diameter or submergence)

The choice of parameter used for velocity and length is not critical, but the same parameter must be used in the
model and prototype when determining the Froude number.

For similarity of flow patterns, the Froude number shall be equal in model and prototype:

Fm
Fr = = 1 (Eq. 9.8.7.3-2)
Fp

where m, p, and r denote model, prototype, and the ratio between model and prototype parameters, respectively.

In physically modeling a pump intake to study the potential formation of vortices, it is important to select a reason-
ably large geometric scale to minimize viscous and surface tension scale effects, and to reproduce the flow pattern
in the vicinity of the intake. Also, the model shall be large enough to allow visual observations of flow patterns,
accurate measurements of swirl and velocity distribution, and sufficient dimensional control. Realizing that larger
models, though more accurate and reliable, are more expensive, a balancing of these factors is used in selecting a
model scale. However, the scale selection based on vortex similitude considerations, discussed below, is a require-
ment to avoid scale effects and unreliable test results.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 49

Rotodynamic Pumps for Pump Intake Design — 2018

It can be shown by the principles of dimensional analysis that such flow conditions at an intake are governed by the
following dimensionless parameters:

uD u D uD u2D
, , , , and
Γ ( gD)0.5 S v σ
 ρ

Where:

u = average axial velocity (e.g., at the bell entrance)

Γ = circulation of the flow
D = diameter (of the bell entrance, or equivalent diameter for noncircular openings, giving the same
area as a circular opening)
S = submergence (at the bell entrance)
v = kinematic viscosity of the liquid
g = acceleration due to gravity
σ = surface tension of liquid/air interface
ρ = liquid density

The influence of viscous effects is defined by the parameter uD = R , the Reynolds number, and surface tension
u2D ν
effects are indicated by  σ  , the Weber number. Based on the available literature, the influence of viscous forces
 ρ
and surface tension on vortexing may be negligible if the values of R and We in the model fall above 3 × 104 and
120, respectively (Daggett, L., and Keulegan, G.H., 1974; Jain, A.K. et al., 1978).

With negligible viscous and surface tension effects, dynamic similarity is obtained by equating the parameters uD ,
Γ
u D
0.5 , and in the model and prototype. An undistorted geometrically scaled Froude model satisfies this condition,
( gD) S
provided the approach flow pattern in the vicinity of the sump, which governs the circulation, Γ, is properly simulated.

Based on the above similitude considerations and including a safety factor of 2 to ensure minimum scale effects,
the model geometric scale shall be chosen so that the model bell entrance Reynolds number and Weber number
at the pump rated flow are above 6 × 104 and 240, respectively, for the test conditions based on Froude similitude.
No specific geometric scale ratio is recommended, but the resulting dimensionless numbers must meet these min-
imum values. For practicality in observing flow patterns and obtaining accurate measurements, the model scale
shall yield a bay width of at least 300 mm (12 in), a minimum liquid depth of at least 150 mm (6 in), and a pump
throat or suction diameter of at least 80 mm (3 in) in the model. If auxiliary service water pumps are not the main
focus of the study and pump flows are a small fraction of the main pumps to be studied, scaling requirements do
not apply to the auxiliary service water pumps. If auxiliary service water pumps are the main focus, minimum scal-
ing criteria shall apply. In instances where service water flows may influence flow patterns, flow withdrawal shall be
included.

In a model of geometric scale Lr, with the model operated based on Froude scaling, the velocity, flow, and time
scales are, respectively:

Vm
Vr = = Lr0.5  (Eq. 9.8.7.3-3)
Vp

Qm
Qr = = Lr2Vr = Lr2.5  (Eq. 9.8.7.3-4)
Qp

Tm L
Tr = = r = Lr0.5  (Eq. 9.8.7.3-5)
Tp Vr

50
` ``` ```` ` ` ` ` ` `
Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved
 Rotodynamic Pumps for Pump Intake Design — 2018

Models of closed-conduit piping systems leading to a pump suction are not operated based on Froude similitude,
but need to have a sufficiently high pipe Reynolds number, R = uD , such that flow patterns are correctly scaled.
v
Based on available data on the variation of loss coefficients and swirl with Reynolds number, a minimum value of
1 × 105 is recommended for the Reynolds number at the pump suction.

9.8.7.4  Physical model study scope

Selection of the model boundary is extremely important for proper simulation of flow patterns at the pump. As the
approach flow nonuniformities contribute significantly to the circulation causing pre-swirl and vortices, a sufficient
area of the approach geometry or length of piping has to be modeled, including any channel or piping transitions,
bends, bottom slope changes, control gates, expansions, and any significant cross-flow past the intake.

All pertinent sump structures or piping features affecting the flow, such as screens and blockage due to their struc-
tural features, trash racks, dividing walls, columns, curtain walls, flow distributors, and piping transitions must be
modeled. In modeling screens, the screen head loss in the model shall be the prototype screen head loss times
the model scale ratio. This may require modifications to the geometric scaling of the screens. The head loss coef-
ficient is a function of the screen Reynolds number, the porosity (percent open area), and the screen (wire) geom-
etry. Scaling of the prototype screen wire diameter and mesh size to the selected model geometric scale may be
impractical and improper due to the resulting low model Reynolds number. In some cases, a model could use the
same screen as the prototype to allow equal loss coefficients. Scaling of trash racks bars may also be impractical
and lead to insufficient model bar Reynolds number. Fewer bars producing the same porosity and the same flow
guidance effect (bar width to bar depth aspect ratio) may be more appropriate.

The inside geometry of the bell (and hub, if modeled) up to the bell throat (section of maximum velocity) shall
be scaled. Consideration should be given to modeling the hub if the hub occupies 10% or more of the area. In
such cases, the hub shall extend downstream beyond the throat to prevent flow separation in the annular velocity
measuring plane. Supports for the hub shall be round rods placed so as to not dissipate any pre-swirl generated
by the approach flow or influence the velocity data. Additionally, any vanes in the bell shall not be modeled. The
bell should be modeled of clear plastic or smooth fiberglass, the former being preferred for flow visualization. The
outside shape of the bell may be approximated except in the case of multistage pumps, in which case the external
shape may affect flow patterns approaching the inlet bell. The impeller is not included in physical models, as the
objective is to evaluate the effect of the intake design on flow patterns approaching the impeller. A straight pipe
equal to the throat diameter or pump suction diameter shall extend at least five diameters downstream from the
throat or pump suction.

For free surface intakes, the model shall be deep enough to cover the range of scaled submergence.

9.8.7.5  Instrumentation and measuring techniques

Unless agreed upon circumstances indicate otherwise, the following measurements shall be made. The extent of
the measurements is summarized in Section 9.8.7.6, Test plan, below.

Flow: The outflow from each simulated pump shall be measured with flowmeters. If an orifice or venturi meter con-
forming to American Society of Mechanical Engineers (ASME) standards is used, the meter need not be calibrated.
The accuracy of the flow measurement shall be within ±2% of the actual flow rate.

Liquid level: Liquid surface elevations shall be measured using any type of liquid level indicator accurate to at least
3 mm (0.01 ft) in the model.

Free surface vortices: To evaluate the strength of vortices at pump intakes systematically, the vortex strength scale
varying from a surface swirl or dimple to an air core vortex, shown in Figure 9.8.7.5a, shall be used. Vortex types
are identified in the model by visual observations with the help of dye and artificial debris, and identification of a
coherent dye core extending to the pump bell or pump suction flange is important. Vortices are usually unsteady in
strength and intermittent in occurrence. Hence, an indication of the persistence of varying vortex strengths (types)

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 51

Rotodynamic Pumps for Pump Intake Design — 2018

shall be obtained through observations made at short intervals in the model (e.g., every 15 seconds) for at least
10 minutes, so that a vortex type versus frequency evaluation can be made and accurate average and maxi-
mum vortex types may be determined. Such detailed vortex observations are needed only if coherent dye core
(or ­stronger) vortices exist for any test. Photographic or video documentation of vortices is recommended.

Subsurface vortices: Subsurface vortices usually originate at the sump floor and walls, and may be visible only
when dye is injected near the vortex core. The classification of subsurface vortices, given in Figure 9.8.7.5a, shall
be used. The possible existence of subsurface vortices must be explored by dye injection at all locations on the wall
and floor around the suction bell where a vortex may form, and documentation of persistence shall be made, as for
free surface vortices.

`---
` `
`
````-`-`
```
--`
Figure 9.8.7.5a  Classification of free surface and subsurface vortices

Swirl in the suction pipe: The intensity of flow rotation shall be measured using a swirl meter, see Figure 9.8.7.5b.
The swirl meter shall consist of a straight-vaned propeller with four vanes mounted on a shaft with low-friction bear-
ings. The tip-to-tip vane diameter is 75% of the pipe diameter and the vane length (in the flow direction) is equal to
0.6 pipe diameters. The location of the swirl meter should be about four suction pipe diameters downstream from
the bell or pump suction flange to allow for convenient installation of velocity traverse instrumentation. The revolu-
tions per unit time of the swirl meter are used to calculate a swirl angle, θ, which is indicative of the intensity of flow
rotation. The swirl meter shall have sufficient support to prevent vibration during testing. The location of the support
and area occupied should be such that it does not influence the swirl measurement.

θ = arctan ( π udn )  (Eq. 9.8.7.5-1)

Where:

u = average axial velocity at the swirl meter

d = diameter of the pipe at the swirl meter
n = revolutions/second of the swirl meter

52 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

`---
` `
`
````-`-`
Figure 9.8.7.5b  Typical swirl meter

```
--`
Flow swirl is often unsteady, both in direction of rotation and speed of rotation. Therefore, swirl meter readings shall
be obtained continuously; for example, readings during consecutive intervals of 30 seconds, covering a period of at
least 10 minutes in the model. Swirl meter rotation direction shall also be noted for each short duration. The max-
imum short duration swirl angle and an average swirl angle shall be calculated from the swirl meter rotations (see
Section 9.8.7.7 Acceptance criteria below). When averaging the swirl meter reading over a timed interval, absolute
values should be used, irrespective of rotation direction. Swirl at a dry-pit suction inlet is not of concern if the swirl
dissipates before reaching the pump suction flange.

Velocity profiles: Cross-sectional velocity profiles of the approach flow may be obtained using a suitable device at
a sufficient number of measuring points to define any practical skewness in the approach flow. The cross-section
location shall be selected to be representative of the approaching flow prior to being influenced by the pump, such
as at a distance of two intake widths upstream from the pump centerline. Such measurements are in themselves
not critical or required, but allow a better understanding of how the approach flow may be contributing to other flow
irregularities and what type of remedial devices may be effective.

Velocity measurements to assess the axial distribution and time varying fluctuations shall be obtained at a min-
imum for the final design at the pump and operating condition indicating the highest swirl and/or vortex activity.
Velocity traverses along at least two perpendicular axes at the throat of the model suction bell or the plane of the
pump suction in a piping system shall be obtained for the final design. A Pitot-static tube or other suitable instru-
ment capable of determining the axial velocity component with a repeatability of ±2% or better shall be used, with
minimal damping. Examples of unnecessary damping effects that may be eliminated are excessive length of instru-
mentation tubing, extremely soft-sided tubing, or tubing that is too small in diameter. Electronic damping associated
with instrumentation shall be minimized.

9.8.7.6  Test plan

Tests should be performed for scenarios that characterize the full range of potential operations, though the poten-
tial for model scale effects should be considered if scenarios involving individual pump discharges below the rated
capacity are evaluated. Unusual or infrequent operating combinations do not need to be included in the test plan.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 53

 Rotodynamic Pumps for Pump Intake Design — 2018

Even though vortices are probably most severe at maximum flows and minimum submergence, there are instances
where stronger vortices may occur at higher liquid levels and lower flows, perhaps due to less turbulence.

Vortex observations and swirl measurements shall be made for all tests. Axial velocity measurements at the bell
throat or suction inlet for each pump in the model are recommended at least for the one test indicating the maxi-
mum swirl angle for the final design. Still-photographic documentation of typical tests showing vortexing or other
flow problems shall be made.

The initial design shall be tested first to identify any hydraulic problems. If any objectionable flow problems are
indicated, modifications to the intake or piping shall be made to obtain satisfactory hydraulic performance. Modifi-
cations may be derived using one or two selected test conditions indicating the most objectionable performance.

Practical aspects of installing the modifications should be considered. The performance of the final modified design
shall be documented for scenarios that characterize the full range of potential operation. If any of the tests show
unfavorable flow conditions, further revisions to the remedial devices shall be made. It is recommended that repre-
sentative tests of the final design be witnessed by the user, the pump manufacturer, and the station designer.

9.8.7.7  Acceptance criteria

The acceptance criteria for the model test of the final design shall be the following:

• Free surface and subsurface vortices entering the pump must be less severe than vortices with coherent (dye)
cores (free surface vortices of Type 3 and subsurface vortices of Type 2 in Figure 9.8.7.5a). Dye core vorti-
ces may be acceptable only if they occur for less than 10% of the time or only for infrequent pump operating
conditions.

• Swirl angles, both the short-term (30-second model) maximum and the long-term (10-minute model) average
indicated by the swirl meter rotation, must be less than 5 degrees. Maximum short-term (30-second model) swirl
angles up to 7 degrees may be acceptable, only if they occur no more than 10% of the time or for infrequent
pump operating conditions. The swirl meter rotation should be reasonably steady, with no abrupt changes in
direction when rotating near the maximum allowable rate (angle).

• Time-averaged velocities at points in the throat of the bell or at the pump suction in a piping system shall be
within 10% of the cross-sectional area average velocity. Time-varying fluctuations at a point shall produce a
standard deviation of less than 10% of the time averaged signal.

• For the special case of pumps with double suction impellers, the distribution of flow at the pump suction flange
shall provide equal flows to each side of the pump within 3% of the total pump flow.

9.8.7.8  Report preparation

--

The final report of the model study shall include an illustration of the model design, model description, scaling and
similitude criteria, instrumentation, test procedure, results (data tabulated and plotted), recommended modifica-
-

tions, and conclusions. The report shall contain photographs of the model showing the initial and final designs,
drawings of any recommended modifications, and photographs of relevant flow conditions identified with dye or
other tracers. A brief video recording of typical flow problems observed during the tests is recommended.

54 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Rotodynamic Pumps for Pump Intake Design — 2018

9.8.8  Use of computational fluid dynamics (CFD)2

9.8.8.1 General

Computational Fluid Dynamics (CFD) is an analysis method used in fluid mechanics that deals with numerical
solutions of the general flow equations for mass, momentum, and heat transfer. Its origin can be traced to the
1930s, but its rapid development followed the advancements in computing power combined with the needs of the
aerospace industry in the 1950s and 1960s. Today, CFD is used within many sciences and industries that involve
fluid flow and transport phenomena, including the pump industry.

9.8.8.2  Simulation methods

CFD encompasses a range of methods from very simple meshes and algorithms that can be run on an ordi-
nary personal computer (PC) to the most advanced methods that require high performance computing. Most gen-
eral-purpose, commercial CFD codes use the Reynolds Averaged Navier–Stokes equations – an approach that
offers a reasonable combination of accuracy and computational efforts that is suitable for many industrial applica-
tions. For a given code, there are options for establishing boundary conditions, meshing of the flow domain, and
turbulence model selection. These factors can influence computational speed and accuracy.

When simulating pump stations (including approach geometry, sump, and pump suction piping) steady-state CFD
models are often selected due to their computational efficiency. Steady-state models are time averaged, and thus
cannot predict fluctuating phenomena or short-term extreme values such as transient vortex activity and fluctua-
tions in swirl or point velocities. Advanced techniques to simulate time-dependent phenomena are available; how-
ever, there are practical limitations in predicting time-dependent and highly curvilinear flow patterns, such as swirls
and vortices.

CFD modeling poses many limitations that require skilled and experienced modelers to properly select methods
and parameter settings to produce results that correctly represent the behavior of a full-scale prototype, even when
using an otherwise well-proven CFD code. Validation of simulations against experimental data is absolutely nec-
essary for a given class of problems. To aid CFD modelers and to eliminate most common pitfalls, some industries
have developed best practice guidelines suitable for their specific field.

9.8.8.3  Acceptable uses of CFD modeling in pump suction hydraulics

As of the writing of this standard, there is a lack of generally available correlations of CFD simulations to experimen-
tal results for the complex flow patterns near pumps, and there have been no best practice guidelines established for
CFD modeling of pump intake and pump suction piping. Until there is satisfactory evidence and appropriate accep-
tance criteria developed and verified, CFD modeling cannot be used to demonstrate compliance with this standard.

CFD may be useful in determining the general approach flow to a sump and pump suction piping. In particular,
the CFD simulation may practically cover a much larger area upstream from the pump than would be possible or
practical for a physical model of the scale required to satisfy the minimum Reynolds number given in ANSI/HI 9.8.
Therefore, CFD simulations may be used to determine the extent of the physical model and the velocity distribu-
tion needed at the physical model boundary. Useful applications of CFD would include determining whether or not
physically modeling a single pump bay or single suction pipe would be adequate. CFD simulations may also be
used to compare designs, to aid in the initial selection of a design (or design development option) for testing using
a physical model, and to better define the range of variables to be tested.

The advances in CFD indicate that further uses of these computational simulations may be possible in the future
and the Hydraulic Institute may consider additional applications of CFD in future revisions of this standard.

2 The Intake Design and Pump Piping committees met to discuss the application of CFD modeling with regards to the approach
flow to a pump. Experts nationally and internationally (academic, HI partners, and nonmembers) were invited to discuss the
latest in modeling of intake structures, including current state of the art simulation, postprocessing methods, and case studies.
The outcome of the meeting was used by the committee to revise this section (June 27, 2016, Indianapolis, IN).

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 55

Appendix A – Remedial measures for problem intakes (informative) — 2018

Appendices

These appendices are not part of this standard, but are presented to help the user in considering factors beyond
the standard sump design.

Refer to Section 9.8.2 of the standard which allows for an intake designed to a geometry other than that presented
in the standard, such as those contained in these appendices, to be deemed to comply with the standard if the
intake is tested by prototype testing or a physical model study performed in accordance with Section 9.8.7, and the
test results comply with the acceptance criteria in Section 9.8.7.7.

Requirements for a physical model study are given in Section 9.8.7.

Appendix A

Remedial measures for problem intakes (informative)

Information in this appendix is not part of this standard, but is presented to help the user in considering factors
beyond the standard sump design.

Refer to Section 9.8.2 of the standard, which allows for an intake designed to a geometry other than that presented
in the standard, such as those contained in these appendices, to be deemed to comply with the standard if the
intake is tested by prototype testing or a physical model study performed in accordance with Section 9.8.7, and the
test results comply with the acceptance criteria in Section 9.8.7.7.

Requirements for a physical model study are given in Section 9.8.7.

A.1 Introduction

The material presented in Appendix A is provided for the convenience of the intake design engineer in correcting
unfavorable hydraulic conditions of existing intakes. None of the remedial measures described herein are part of
the standard intake design recommendations provided in Section 9.8. A portion of the material in Appendix A trans-
mits general experience and knowledge gained over many years of improving the hydraulics of intake structures,
and such educational material may not include the specific recommendations appropriate for a standard. Correc-
tions described herein have been effective in the past, but may or may not result in a significant improvement in
performance characteristics for a given set of site-specific conditions. Other remedial fixes not provided herein
may also be effective, and a physical model test is needed to verify whether or not a given remedial design feature
results in acceptable flow conditions. This is particularly true because adding a remedial feature to solve one flow
problem may have detrimental effects on other flow phenomena of concern.

Appendix A concentrates on rectangular intakes for clear liquids, but the basic principles can be applied to other
types of intakes. The material is organized by the general type of hydraulic problem in an upstream to downstream
direction, because proper upstream flow conditions minimize downstream remedial changes.

A.2  Approach flow patterns

The characteristics of the flow approaching an intake structure is one of the foremost considerations for the
designer. Unfortunately, local ambient flow patterns are often difficult and expensive to characterize. Even if known,

` ``` ```` ` ` ` ` ` `

56 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix A – Remedial measures for problem intakes (informative) — 2018

conditions are often unique, and frequently complex, so it is difficult to predict the effects of a given set of flow con-
ditions upstream from an intake structure on flow patterns in the immediate vicinity of a pump suction.

When determining direction and distribution of flow at the entrance to a pump intake structure, the following must
be considered:

• The orientation of the structure relative to the body of supply liquid

• Whether the structure is recessed from, flush with, or protrudes beyond the boundaries of the body of supply
liquid

• Strength of currents in the body of supply liquid perpendicular to the direction of approach to the pumps

• The number of pumps required and their anticipated operating combinations

Velocity profiles entering pump bays can be skewed, regardless of whether or not crosscurrents are present. Sev-
eral typical approach flow conditions are shown in Figure A.1 for rectangular intake structures withdrawing flow
from both moving bodies of liquid and stationary reservoirs. Figure A.2 shows several typical approach flow condi-
tions for different combinations of pumps operating in a single intake structure.

The ideal conditions, and the assumptions on which the geometry and dimensions recommended for rectangular
intake structures in this section are based, are that the structure draws flow so that there are negligible ambient
currents (cross-flows) in the vicinity of the intake structure that create asymmetrical flow patterns approaching any
of the pumps, and the structure is oriented so that the boundary is symmetrical with respect to the centerline of
the structure. As a general guide, cross-flow velocities are significant if they exceed 50% of the pump bay entrance
velocity. Recommendations (based on a physical model study) for analyzing departures from the ideal condition are
given in Section 9.8.7.

A.2.1  Open versus partitioned structures

If multiple pumps are installed in a single intake structure, dividing walls placed between the pumps result in more
favorable flow conditions than found in open sumps. Open sumps, with no dividing walls, have been used with
varying levels of success, but adverse flow patterns can frequently occur if dividing walls are not used. The trench-
type intake structure, described in Sections 9.8.3.4 and 9.8.4.2, is a type of open sump that is an exception. Open
sumps are particularly susceptible to cross-currents and nonuniform approach flow patterns. Even if approach flow
at the entrance to the structure is uniform, open sumps result in nonuniform flow patterns approaching some of the
pumps when operating pumps are arranged asymmetrically with respect to the centerline of the intake structure.
This situation can occur when various combinations of pumps are operating or if the intake structure is designed to
accommodate additional pumps at some future date. Figure A.3 is an example of flow approaching the pumps in a
partitioned structure and an open sump, both operating at partial flow rate.

The example facilities contain four units with two of the four operating. In both structures, flow is withdrawn from a res-
ervoir with no velocity component perpendicular to the longitudinal centerline of the intake structures. In the partitioned
structure, flow enters the bay of pump 1 fairly uniformly. It enters the bay containing pump 2 nonuniformly, with a sepa-
ration area near the right sidewall. However, the length of the bay relative to its width channels the flow and allows it to
become more uniform as it approaches the pump. In Figure A.3, example ii, the dashed line at the wing walls shows a
rounded entrance configuration that minimizes flow separation near the entrance to the outer pump bays.

In open sumps (Figure A.3, example i), flow may enter the structure uniformly with respect to the centerline of the
structure. However, since the location of the two operating pumps is not symmetrical with respect to the centerline
of the structure, flow separates from the wall of the structure and approaches pump 2 with a tangential velocity
component, greatly increasing the probability of unacceptable levels of pre-swirl.

If all four pumps in the open sump were to operate simultaneously, approach flow would be reasonably uniform, but
other adverse phenomena could be present. For example, when two adjacent pumps are operating simultaneously,
submerged vortices frequently form, connecting both pumps.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 57

Appendix A – Remedial measures for problem intakes (informative) — 2018

A.3  Controlling cross-flow

If cross-flow is present (i.e., if the pump station is withdrawing flow from the bank of a canal or stream), trash racks
with elongated bars can provide some assistance in distribution flow as it enters the pump bay, but if the flow pro-
file is skewed when it enters the trash rack, it will be skewed as it exits. To be effective in guiding flow, trash racks
must be placed flush with the upstream edges of the pump bay dividing walls. In this example the trash rack must
be vertical or match the incline of the entrance. Both trash racks and dividing walls must be in line with the stream
bank contour. Trash racks recessed from the entrance to pump bays, and through-flow traveling screens have a
negligible flow-straightening effect (see Figure A.4).

Partially clogged trash racks or screens can create severely skewed flow profiles. If the application is such that
screens or trash racks are susceptible to clogging, they must be inspected and cleaned as frequently as necessary
to prevent adverse effects on flow patterns.

Figure A.1  Examples of approach flow conditions at intake structures and the resulting
effect on velocity, all pumps operating
--

58 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

-
 Appendix A – Remedial measures for problem intakes (informative) — 2018

--
-

Figure A.2  Examples of pump approach flow patterns for various combinations of operating pumps

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 59

Appendix A – Remedial measures for problem intakes (informative) — 2018

`---
` `
`
````-`-`
```
--`
Figure A.3  Comparison of flow patterns in open and partitioned sumps

60 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix A – Remedial measures for problem intakes (informative) — 2018

`---
` `
`
Figure A.4  Effect of trash rack design and location on velocity distribution entering pump bay

````-`-`
```
--`

Figure A.5  Flow-guiding devices at entrance to individual pump bays

Two other flow-straightening devices for minimizing cross-flow effects at bay entrances are shown in Figure A.5.
One or two large guide piers or plates per bay help turn the flow. Although distinct flow separation eddies occur at
each pier, the eddies are smaller than the single flow separation (eddy) that would occur along one bay wall. Alter-
natively, a number of smaller columns or structural members may be placed at the bay entrance, and these are
effective in both turning and creating more uniform velocity by inducing a head loss across the column array.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 61

Appendix A – Remedial measures for problem intakes (informative) — 2018

A.4  Expanding concentrated flows

Two methods for correcting flow disturbances generated by expansion of a concentrated flow are described below.

A.4.1  Free surface approach

In some installations, site conditions dictate that the approach flow channel or conduit, although in line with the
sump axis, is much smaller than the sump width. To avoid concentrated flow and large eddies, the sidewalls
approaching the pump bays must gradually diverge, and flow baffles of varying geometry or guide vanes may be
used to spread the flow at a divergence angle greater than otherwise possible. Figure A.6 shows possible correc-
tive measures.

The flow leaving a dual entry flow screen may require baffling to break up and laterally distribute the concentrated
flow prior to reaching the pump, and one possible arrangement is shown in Figure A.7.

If measures are not taken to mitigate the effects these screens have on flow patterns (see Figure A.8), then the jet
exiting the center of these screens will attach to one wall or the other, and will result in highly nonuniform flow for
an indefinite distance down the channel. The nonuniform flow creates excessive swirl at the pump. The screen exit
must be placed a minimum distance of six bell diameters, 6D, (see Section 9.8.3.1.3) from the pumps. However,
this distance is only a general guideline for initial layouts of structures, with final design to be developed with the aid
of a physical model study.

A.4.2  Closed conduit approach

Flow may be provided to rectangular intake structures through a conduit. When multiple pumps are installed per-
pendicularly to the influent conduit, the flow pattern improves and approach velocities decrease if the sump walls
diverge gradually from the point of influent toward the pump bays. Maintaining a small angle divergence of each
wall from the influent conduit minimizes the difficulty in spreading the flow uniformly. A series of flow distribution
baffles may be installed to dissipate the energy of the entering flow and force a diverging and more uniform flow
pattern approaching the pumps. A typical approach flow pattern in a wet well with a conduit approach, with and
without diverging sidewalls and flow distribution baffles, is shown in Figure A.6.

If a conduit approach is required and there is no room for gradually diverging sidewalls, velocities in the conduit
entering the sump may need to be limited, such as by adding expansion pieces to the downstream end of the con-
duit. In addition to the features described above, a baffle may be needed near the influent point of the conduit(s)
to dissipate the energy from the entering jet and spread the flow toward the pump bays. Increasing the number of
inflow lines together with a flow distributor across the sump and/or each bay may provide an adequate distribution
to the pump bays (see Figure A.9).

The trench-type wet well described in Section 9.8.3.4 is an alternate arrangement, where the pumps are positioned
in line with the approach pipe.

`---
` `
`
````-`-`
```
--`

62 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix A – Remedial measures for problem intakes (informative) — 2018
--

Figure A.6  Concentrated influent configuration, with and without flow distribution devices
-

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 63

Appendix A – Remedial measures for problem intakes (informative) — 2018

--
-

Figure A.7  Baffling to improve flow pattern downstream from dual flow screen.
Note: Physical model study required.

Figure A.8  Typical flow pattern through a dual flow screen

64 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix A – Remedial measures for problem intakes (informative) — 2018

Figure A.9  Improvements to approach flow without diverging sump walls

A.5  Pump inlet disturbances

A.5.1  Free surface vortices

In the Pump Intake Design Standard (Sections 9.8.3.1.4, 9.8.3.2.3 and 9.8.7.2), a formula for submergence is pro-

`---
vided that reduces the likelihood of strong air core vortices. While the equation provides a good balance between

` `
the pump intake depth and strong vortex potential, applying this submergence in a design does not automatically

`
````-`-`
mean that the design will remain free from unacceptable free surface vortices.

```
Surface vortices may be reduced with increasing depth of submergence of the pump bells. However, there are also

--`
situations where increasing depth has negligible effects or even increases surface vortex formation due to stagnant
regions and therefore unstable liquid. Surface vortices are also highly dependent on approach flow patterns and
the stability of these patterns, as well as on the inlet Froude number. This complicates the establishment of a min-
imum depth of submergence as a definitive measure against vortices. To achieve a higher degree of certainty that
objectionable surface vortices do not form, modifications can be made to the intake structures to allow operation at
practical depths of submergence.

Many manufacturers offer a corrective option for a suction bell by adding a suction umbrella. Suction umbrellas are
formed pieces, usually horizontally oriented flat rings or plates attached to the pump bell or supported by radial
vanes to the sump floor. They minimize free surface vortices, prevent entrainment of air, and reduce the minimum
submergence and disturbances at the pump inlet. The umbrella effectively reduces the velocities at the periphery of
the umbrella compared to the bell. Suction umbrella diameters can be up to the full width W of the pump inlet bay.

The most effective use of suction umbrellas is for pumps in drainage or other pump-down services where the mini-
mum liquid depth occurs at the end of the pumping cycle. For more information, see Dicmas, J.L., Vertical Turbine,
Mixed Flow, and Propeller Pumps, McGraw-Hill Book Co.

Curtain walls, such as shown in Figures A.10 and A.12, create a horizontal shear plane that is perpendicular to the
vertical axis of rotation of surface vortices, and prevent the vortices from continuing into the inlet. Vertical curtain
walls have been used with success and are easier to construct than sloping curtain walls. However, the abrupt
changes in flow direction caused by vertical walls can create surface vortices in the upstream corners of those walls.
If the curtain walls are placed at about 45 degrees from the vertical, then all flow near the surface is deflected down-
wards and surface vortices are minimized. Curtain walls also assist in laterally spreading poorly distributed flow.

Horizontal gratings may also be used to suppress free surface vortices when pumping clear liquids. Standard floor
grating 38 mm (1.5 in) deep or greater, or a specially constructed “egg-crate” type grating may be effective. At the

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 65

 Appendix A – Remedial measures for problem intakes (informative) — 2018

low liquid level, the top of the grating should be submerged about 150 mm (6 in). As a temporary measure, floating
rafts of various types may be used to suppress surface vortices.

Knauss (1987) recommends avoiding designs with a submergence lower than S/D = 1.0 + 2.0*Fr. Below this lower
limit, air entraining free surface vortices (Type 5 and 6) are likely to occur and implementation of remedial design
changes and achieving success may be more difficult. Additional information regarding pump submergence can
also be found in the above reference.

It is recommended that the effectiveness of any such modifications or lower than recommended submergences be val-
idated in a physical scale model because a poorly designed modification may worsen the approach flow to the pumps.

A.5.2  Subsurface vortices

The geometry of boundaries in the immediate vicinity of the pump bells is one of the more critical aspects of suc-
cessful intake structure design. It is in this area that the most complicated flow patterns exist and flow must make
the most changes in direction, while maintaining a constant acceleration into the pump bells to prevent local flow
separation, turbulence, and submerged vortex formation. Pump bell clearance from the floor and walls is an inte-
gral part of the design. A sampling of various devices to address subsurface vortices are shown in Figure A.11.

These and other measures may be used individually or in combination to reduce the probability of flow separation
and submerged vortices.

A.5.3 Pre-swirl

Whether or not pre-swirl exists to an objectionable extent is governed primarily by the approach flow distribution. A
sufficiently laterally skewed approach flow causes rotation around the pump bell, in spite of the local features. Such
rotation causes flow to circulate over the center splitter (Figure A.11, example viii) and potentially produces a sub-
merged vortex emanating from the flow separation at the center splitter. A cone on the floor would not cause such
a submerged vortex problem, but the cone would also not help to control residual pre-swirl.

The most effective way of reducing pre-swirl is to establish a relatively uniform approach flow within each pump bay by
using the baffling schemes discussed in Sections A.2 to A.4 above. Final reductions in swirl may be achieved near the
pump bell by installing a vertical splitter along the backwall, in line and directly behind the pump column, by providing
a horizontal (sloping) floor splitter under the bell as shown in Figure A.11 and perhaps by using a submerged curtain
wall (shown in Figures A.10 and A.12) across the bay width, close to the upstream side of the pump. This wall, if a
few bell diameters high off the floor, has the effect of turning all the flow downward, similar to that in a circular “can”
arrangement, and the basic change in flow pattern may reduce pre-swirl and other undesirable hydraulic phenomena.

Figure A.10  Elevation view of a curtain wall for minimizing surface vortices

` ``` ```` ` ` ` ` ` `

66 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix A – Remedial measures for problem intakes (informative) — 2018

Figure A.11  Methods to reduce subsurface vortices (examples i–ix)

A.5.4  Velocities in pump bell throat

A relatively uniform velocity distribution occurs at the pump bell throat if the flow enters the bell essentially radially,
without pre-swirl or local flow disturbances such as vortices or eddies caused by local flow separation. Therefore,
all of the above-described flow control devices, starting with providing a uniform approach flow and including local
--

antivortex measures near the bell, may be needed to achieve the desired uniformity of velocities.

Alternatively, a properly shaped formed suction intake (FSI) may be provided, as discussed in Section 9.8.3.2.
-

Model tests have shown that the FSI provides the desired uniformity of velocity at the bell throat for reasonable flow
patterns approaching the FSI.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 67

 Appendix A – Remedial measures for problem intakes (informative) — 2018

A.6  Deviations from standard dimensions

When a rectangular intake does not conform to the standard dimensions of Figure 9.8.3.1.4a, it may be helpful
to understand some of the observed effects of such deviations. Designs not built in accordance with the standard
dimensions may or may not operate satisfactorily depending on a variety of factors. Designers may have rea-
sons for selecting nonstandard intake dimensions, in particular, when there exists a successful experience record
with the nonstandard configuration, or certain specific physical constraints or performance objectives are to be
satisfied.

Figure A.12  Combination of remedial corrections: i, v, vi, vii, and viii from
Figure A.11, plus curtain wall from Figure A.10

Depending on the degree of deviation from standard dimensions, adverse flow patterns can result, which may be
difficult to correct. Deviations from the standard intake dimensions are in general not recommended, but when
used, the design should be evaluated by a physical model study.

Narrower bay widths W are possible, particularly when the suction bell diameter (D) is sized generously, and the
resultant pump bay velocity Vx does not exceed the recommended maximum. An advantage of a narrower bay
width is the increased level of turbulence, which can help suppress the formation of surface vortices. On the other
hand, the increased velocities in the vicinity of the pump intake can intensify subsurface vortices and velocity dis-
tortions around the suction bell. Bay widths with values W as low as 1.5D have been made to work satisfactorily.

` ``` ```` ` ` ` ` ` `

68 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix A – Remedial measures for problem intakes (informative) — 2018

The clearance between the bell and sump floor C is typically near 0.5D. Occasionally there may exist a need for
a design outside of the recommended range 0.3D ≤ C ≥ 0.5D. Values of C > 0.5D are possible, but would likely
require one or more methods to reduce subsurface vortices (Figure A.11). When flow velocities surrounding the bell
are relatively uniform, or when a generously sized bell diameter with low peripheral velocity is used, a clearance
C < 0.3D may be possible, for instance, when designing for pumping down to low liquid levels. Clearance values of
C < 0.25D may result in a decelerating velocity with attendant flow instabilities under the bell.

The recommended backwall clearance B is 0.75D. Pumps fitted with generously sized bells or umbrella skirts may
justify a reduced backwall clearance simply based on the clearance geometry and flow velocities surrounding the
impeller eye. The potential benefit of minimizing dimension B is to lessen the tendency of surface vortices to form
behind the pump column, but this may come at the expense of a persistent subsurface vortex under the bell from
the backwall. A reduced B dimension may also make it difficult to achieve an acceptable velocity distribution at the
impeller eye. Axial flow and higher specific speed mixed flow pumps may suffer measurable head loss as the value
of clearance B is reduced. Values of B greater than 0.75D increase the tendency for surface vortexing and create
stagnant areas behind the bell where subsurface vortices may develop.

A.7  Tanks – pump suction

A.7.1 General

For applications involving tanks located on the suction side of pumps, undesirable hydraulic conditions, such as
pre-swirl and vortices, may be created at the tank outlet. Factors that influence the approach flow to the outlet can
include geometric arrangement, inflow-outflow configuration, and the presence of flow obstructions.

The main problem is usually entrainment of air (or other tank gases) due to free surface vortices or aeration of the
tank contents from free fall of the tank inflow, as this air may collect in the piping (causing air binding) or cause
degradation of pump performance.

Preventing the formation of free surface vortices at tank outlets to pumps allows the tank to be drawn to lower lev-
els than would otherwise be possible. This benefit requires the use of various antivortex devices at the tank outlet.
Some common types of such devices are shown in Figure A.13.

A.7.2  Vertical tank, simultaneous inflow and outflow

Inflows to a tank can cause significant disruption of the outflow due to aeration caused by a free-falling inflow. The
best way to suppress this is with either a submerged inlet pipe or a J-pipe; see Figure A.14. The submerged inlet
pipe should extend down to within four diameters of the tank bottom. A J-pipe can be a good alternate if the tank is
tall. The end of the J-pipe should be within one pipe diameter of the tank wall. If the inflow is corrosive or contains
solids, an impingement plate may be needed. Another alternative is to put the inflow nozzle low on the side of the
tank, so the inflow is always submerged.

To suppress vortex formation in order to decrease the minimum submergence to less than that calculated in Sec-
tion 9.8.3.5.4, use a horizontal baffle as shown in Figure A.13, example 1, or a cross as shown in Figure A.13,
example 2 or 3.

Flat-bottom tanks typically have side outlet nozzles. To decrease the minimum submergence below that calculated in
Section 9.8.3.5.4, use a horizontal baffle as shown in Figure A.13, example 4. The baffle radius should be at least equal
to S, as calculated in Section 9.8.3.5.4 and shown in Figure A.14. The baffle should be 2D above the nozzle centerline.

A.7.3  Horizontal tank, simultaneous inflow and outflow

Inflows to a horizontal tank can cause significant disruption of the outflow due to aeration caused by a free-falling
inflow. The two design features that suppress outflow disruption are having the inflow and outflow nozzles at oppo-
site ends of the tank and using a submerged inlet pipe; see Figure A.14. The submerged inlet pipe should extend
down to within four pipe diameters of the tank bottom.

69
` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

Appendix A – Remedial measures for problem intakes (informative) — 2018

To suppress vortex formation in order to decrease the minimum submergence to less than that calculated in Sec-
tion 9.8.3.5.4, use a horizontal baffle as shown in Figure A.13, example 1, or a cross as shown in Figure A.13,
examples 2 or 3.

As an alternative, a cage-type vortex suppressor may be used, as illustrated in Figure A.13, example 5. The cubic
cage may be made of standard 38-mm (1.5-in) deep (or deeper) floor grating (or its equivalent). The length, width,
and height of the cubic cage, each with a characteristic length termed Lv should be about three inlet pipe diame-
ters, and the top of the cage should be submerged about 150 mm (6 in) below the minimum liquid level. Noncubic
cage shapes are also effective if the upper (horizontal) grating is at least three inlet pipe diameters on each side
and is also submerged 150 mm (6 in) below the minimum liquid level. A single horizontal grating meeting these
guidelines may also be effective. Tests on such cage-type vortex suppressors have demonstrated their capability to
reduce air entrainment to nearly zero even under adverse approach flow conditions (Padmanabhan, 1982). How-
ever, it may be noted that the minimum submergence from the tank floor is dictated by the vertical cage dimension
plus the needed 150-mm (6-in) submergence above the top of the cage.

Figure A.13  Tank antivortex devices

` ``` ```` ` ` ` ` ` `

70 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix A – Remedial measures for problem intakes (informative) — 2018

Figure A.14  Tank inflow and outflow configurations

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 71

Appendix B – Sump volume (informative) — 2018

Appendix B

Sump volume (informative)

This appendix is not part of this standard, but is presented to help the user in considering factors beyond the stan-
dard sump design.

B.1 Scope

This section on pump sump volumes pertains to constant-speed pumps. For adjustable-speed pumping, sump
volume may not need to be considered (assuming adequate pump controls) except for a requirement that the
sump volume must be large enough to keep currents sufficiently low and/or if the pump is operated in a start-stop
arrangement to minimize cycle time.

B.2 General

Pump sumps are required in most pumping systems (1) to distribute inflow to the various pumps and (2) to act as a
buffer volume capable of absorbing inflow fluctuations. For constant-speed pumps that operate in an on/off mode,
the sump alternately fills and empties at a cyclic rate slow enough to prevent motors and motor starters from over-
heating. Refer to the pump (not the motor) manufacturer for the allowable cyclic rate.

The active sump volume required for a single constant-speed pump depends on the allowable starts per hour (or
the allowable time per cycle). The total time for a cycle is the time to fill plus the time to empty the basin between
low (LLL) and high liquid levels (HLL).

V V VP  (Eq. B.2-1)
T = + =
Q P − Q Q(P − Q)

Where:

P = pump discharge capacity, in L/s (ft3/s), fixed

Q = inflow to sump, in L/s (ft3/s), variable
T = total cycle time, in s, given
V = active sump volume, in L (ft3), to be determined

The minimum cycle time occurs when the rate of change of volume (as the basin fills and empties) with respect to
inflow is the greatest. Rearranging the equation,

Q2  
V = T Q − (Eq. B.2-2)
`---

 P
` `
`

Differentiating,
````-`-`

( )
```

dV 2Q
= T 1− = 0 (Eq. B.2-3)
--`

dQ P
From which,

P (Eq. B.2-4)
Q=
2

72 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix B – Sump volume (informative) — 2018

Thus, the inflow rate to be used to determine the sump volume is half the capacity of the pump. Substituting P/2 for
in the first expression yields:

P (Eq. B.2-5)
V =T
4
When two or more pumps are to be operated simultaneously, the equations are still valid if P is taken as the incre-
ment of pumpage when another pump is added. Formulas for volume with multiple pump operation including pump
alternation and reduced capacity due to friction are complicated, especially if the pumps are of different sizes.
Graphical solutions are much simpler, offer insights not possible with formulas, provide a definitive analysis of cycle
times and resting times, give start and stop levels, allow the use of pumps with differing capacities, and permit any
sort of pump alternation. Several pumps can be alternated to increase cycle time, and the graphical construction
clearly shows both cycle times and resting times.

B.3  Construction of a graph

Choose a linear scale of volume for the left y-axis. On the right y-axis, plot the corresponding liquid levels; that
scale is linear for wet wells with vertical sides. For basins with sloping sides (such as trench type), the scale is
nonlinear as shown in Figure B.1 plot time on the x-axis. With simple pump controllers, separate the start levels
for successive pumps by at least 150 mm (6 in) because liquid-level sensing devices need that much differential
to operate reliably. Separate the stop levels similarly. Provide a time delay of 5 to 10 seconds on starting to avoid
premature operation caused by turbulence.

B.4  Example for a simple controller

Three pumps of equal size are set in a wet well with sloping sides and auxiliary storage, so although the volume
scale is linear, the liquid-level scale is not. The discharge rate for a single pump is P1, so the minimum cycle time
occurs when the inflow rate, Q, equals half the outflow rate or 0.5P1. Consequently, the slopes of the flow rates
to be plotted are ± 0.5P1. When two pumps operate together the total discharge is P1+2 (which is less than 2P1
because of increased friction losses), and the critical flow rate to be plotted is half the incremental increase of flow
when a second pump is brought on line.

Figure B.1  Graphical analysis for liquid-level controllers

`---
` `
`
````-`-`
```
--`

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 73

Appendix B – Sump volume (informative) — 2018

Assumptions:

Pumps: three, two duty and one standby. Standby pump assumed to be dismantled for repair.

Station Qmax = 400 L/s (6350 gpm), so capacity per pump is PQmax = 200 L/s (3170 gpm).

For single pump operation, P1 = P2 = 230 L/s (3650 gpm).

Pumps to be alternated.

The pump manufacturer recommends 10 starts per hour or less (360 seconds per start) for the particular equip-
ment considered. Note: consult manufacturer for recommended allowable starts.

Liquid-level fluctuation should not exceed 1.2 m (4 ft).

Solution:
TP
V = 4 = 360 seconds × 230 (3650)/4 = 20,700 L (5475 gal) = 20.7 m3 (5475 gal).

But pumps are to alternate, so the volume required is one-half V or 10.35 m3 (2735 gal).
1
Critical flow rate = P 2 = 230 3650 = 115 L/s (1825 gpm).
2
See Figure B.1. Start with low liquid level, LLL, and plot slope of critical liquid inflow at +115 L/s (1825 gpm) until
volume reaches 10.35 m3 (2735 gpm).

Pump 1 is now activated, so plot a line downward at a slope of −230 (−3650) + 115 (1825) = −115 L/s (−1825 gpm)
as the pump empties the basin. The pump is turned off at LLL. The basin refills at +115 L/s (+1825 gpm).

The pumps are alternated by turning on pump 2 at 10.35 m3 and stopping it at the LLL. Basin again refills. Each
pump cycle is 360 s.

When a second (or follow) pump is needed, the total pumpage is 400 m3/s, but the incremental increase in pump-
age is 400 (6350) − 230 (3650) = 170 L/s (2700 gpm).

The critical additional inflow rate is 170 (2700)/2 = 85 L/s (1350 gpm), so the total critical inflow rate is 230 (3650) +
85 (1350) = 315 L/s (5000 gpm).

Start the follow pump, P2, at high liquid level, HLL, which is 10.35 m3 plus 150 mm (6 in) increase in liquid-level
elevation.

Both pumps P1 and P2 reduce the volume at 315 (5000) − 400 (6350) = −85 L/s (−1350 gpm).

Pump 1 is turned off at 150 mm (6 in) above LLL, and the basin refills to the HLL at 315 (5000) − 230 (3650) = 85

L/s (1350 gpm) while pump 2 continues to run.

Pump 1 is turned on at HLL. When the liquid surface again reaches LLL + 150 mm (6 in), pump 2 is turned off and
rests until the basin is refilled. By alternating pumps, either is restarted in 515 seconds.

More pumps and pumps of different sizes can be analyzed in the same manner.

74
` ``` ```` ` ` ` ` ` `
Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved
 Appendix B – Sump volume (informative) — 2018

Other solutions:

Alternators may fail (install a backup) or workers might override the alternator, so designers must decide whether
or not to rely on pump alternation.

If pumps are not alternated, the cyclic rate is doubled and so is the size of the pump intake basin.

B.5  Example for programmable controllers

Smart controllers detect whether one or two (or more) pumps are needed, so any pump (whether lead or follow)
can use the entire volume of the wet well between LLL and HLL. For a station with three pumps, 150 mm (6 in) of
excavation can be saved; with four pumps, the saving is 300 mm (1 ft). The solution, given by Figure B.2, gives a
pump cycle of 490 s (7.3 cycles/h) during two-pump operation.

Figure B.2  Graphical analysis for a “smart” controller

`---
` `
`
````-`-`
```
--`

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 75

Appendix C – Intake basin entrance conditions, trench-type wet wells for solids-bearing liquids (informative) — 2018

Appendix C

Intake basin entrance conditions, trench-type wet wells

for solids-bearing liquids (informative)

This appendix is not part of this standard, but is presented to help the user in considering factors beyond the stan-
dard sump design.

C.1 Scope

This section pertains to entrance conditions and to volume requirements for trench-type wet wells with vari-
able-speed or constant-speed pumps.

C.2  Entrance conditions

There should be a minimum of 8 diameters of straight, uniform, symmetric, and level inlet conduit, and level (or nearly
level) inlet pipe leading into the pump intake basin. The pipe should lie in a vertical plane through the pump intakes
and well above them as shown in Figures 9.8.3.3.1a through f, 9.8.3.4.1, and 9.8.4.2.2. The intake velocity should be
only great enough to keep solids moving, preferably between 0.6 m/s (2.0 ft/s) and 1.2 m/s (4.0 ft/s). In wet wells for
intakes consisting of a bend and flare (leading to dry-pit pumps), inlet velocity should not exceed 0.9 m/s (3 ft/s).

C.3  Variable-speed pumps in trench-type wet wells

Little or no storage volume is required for variable-speed pumping because the discharge of the pumps can typi-
cally be regulated to match the inflow. In cases where inflow is less than the minimum pumping capacity, adequate
storage volume may be required to minimize cycle time.

The liquid level in the wet well should be maintained to match the depth in the upstream conduit. Even a slight drop
generates bubbles and currents that sweep the bubbles to the intake of the first pump.

C.4  Constant-speed pumps in trench-type wet wells

Some types of wet wells, notably trench, confined, and hopper-bottom circular types, are inherently small and may
contain as little as half the active storage volume needed to keep the frequency of motor starts within the manu-
facturer’s recommendations. The active storage volume is obtained by allowing the liquid level to fluctuate, typically
--

about 1.2 m (4.0 ft).

Allowing a free fall or cascade from the inlet into the pool below should be avoided if possible. Even a short drop
-

entrains air bubbles and drives them deep into the pool where they may be drawn into the pumps and thus reduce
pump flow rate, head, and efficiency as well as causing damage to the pumps. If the liquid is domestic wastewater,
the turbulence sweeps malodorous and corrosive gases into the atmosphere.

C.4.1  Auxiliary storage

Active storage need not be confined to the wet well. Some storage may be allocated to an auxiliary storage vessel.
Auxiliary storage should:

• Eliminate any vestige of free fall at all times

• Supply the deficit of the required active storage capacity in an appurtenant structure at low cost

76 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix C – Intake basin entrance conditions, trench-type wet wells for solids-bearing liquids (informative) — 2018

• Discharge liquid horizontally into the wet well pool without turbulence and at low velocities, preferably between
0.6 m/s (2.0 ft/s) and 0.9 m/s (3.0 ft/s)

• Operate automatically with no operator attention

• For wastewater or stormwater, be self-cleaning in every cycle

C.4.2  Approach pipes

One facility that meets the objectives of auxiliary storage is an approach pipe, which is an enlarged pipe laid at a
hydraulically steep gradient between the wet well and an upstream manhole. At a gradient of 2% and a length of
60 m (200 ft), the approach pipe would fill and empty with a 1.2-m (4.0-ft) difference between low liquid level (LLL)
and high liquid level (HLL). See Figure C.1. Low exit velocities can be obtained by setting the LLL at an appropriate
elevation above the invert of the approach pipe so that the turbulence from the hydraulic jump occurs in the pipe
and not in the sump.

Figure C.1  Schematic diagram of approach pipe

Beginning with a full approach pipe, the liquid level lowers as a pump is activated and liquid flows over a sloping
invert in the upstream manhole to the supercritical velocity given in Table C.1 or C.2. On encountering pooled liquid,
as in Figure C.1, the supercritical velocity results in a hydraulic jump with a sequent depth that must not be allowed
to reach the soffit of the pipe because entrapped air might result in a violent eruption that could be destructive. In
the tables, the sequent depth is limited to 60% of the pipe diameter, and that leaves a free liquid surface 20 pipe
diameters long — more than enough for bubbles to rise to the surface and escape up the pipe.

The Froude number (see Section 9.8.3.5.4) for the jump is less than 2.5, so there is little bubble formation and
off-gassing. Note from Tables C.1 and C.2 that the useful active storage cross section of the approach pipe varies
from 73 to 82% of the total pipe cross-sectional area.
`---
` `
`

Tables C.1 and C.2 were originally developed by Wheeler (1995) and modified by Cahoon and Sanks (2002). These
````-`-`

data for approach pipes at 2% gradient are based on Manning’s equation corrected by Escritt (1984) for an n of
0.010. Uncorrected, n varies as much as 25% with depth. Escritt found that by adding half the width of the liquid
```
--`

surface to the perimeter for calculating the hydraulic radius, n was constant within a few percent. A value of 0.0125
for n in the uncorrected Manning equation is roughly comparable to an n of 0.010 in the equation with the Escritt
modification.

These tabular data can be altered for other gradients, n values, and sequent depths by using the program,
Approach, freely available from the Internet at www.Pumps.org/IntakeDesign.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 77

Appendix C – Intake basin entrance conditions, trench-type wet wells for solids-bearing liquids (informative) — 2018

Table C.1  Maximum allowable flow rates in approach pipes (metric units)
Slope = 2%, Manning’s n = 0.010, Escritt’s assumption is used, sequent depth is limited to 60% of pipe diameter.

True Pipe Before Jump After Jump

Dp, mm Area, m2 Q, L/s y ,% v, m/s Ae Froude No. y ,%

,%
Dp Af Dp

254 0.051 20 31 1.4 73 1.91 60

304 0.073 30 31 1.6 74 1.97 60

381 0.114 50 30 1.8 75 2.04 60

457 0.164 80 29 2.0 76 2.10 60

533 0.224 110 28 2.2 77 2.15 60

610 0.292 160 28 2.4 78 2.20 60

686 0.370 210 27 2.6 78 2.24 60

762 0.456 270 27 2.7 79 2.27 60

838 0.552 330 26 2.9 79 2.31 60

914 0.657 410 26 3.0 79 2.34 60

1067 0.894 590 25 3.3 80 2.40 60

1219 1.17 810 25 3.6 80 2.45 60

1372 1.48 1070 24 3.8 81 2.49 60

1524 1.82 1380 24 4.1 81 2.53 60

1676 2.21 1730 24 4.3 82 2.57 60

1829 2.63 2130 24 4.5 82 2.61 60

Escritt’s assumption: add half the width of free liquid surface to wetted perimeter in hydraulic radius.
Omitting Escritt’s assumption ≈ changing n from 0.010 to 0.0125.
Dp is inside diameter of pipe.
y is depth of flow.
Ae is inside area of pipe above liquid surface.
Af is inside area of pipe.
For n = 0.009, multiply Q by 92%.
For n = 0.011, multiply Q by 108%.
For n = 0.012, multiply Q by 115%.
For n = 0.013, multiply Q by 122%.

` ``` ```` ` ` ` ` ` `

78 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix C – Intake basin entrance conditions, trench-type wet wells for solids-bearing liquids (informative) — 2018

Table C.2  Maximum allowable flow rates in approach pipes (US customary units)
Slope = 2%, Manning’s n = 0.010, Escritt’s assumption is used, sequent depth is limited to 60% of pipe diameter.

True Pipe Before Jump After Jump

Q,
Dp, in Area, ft2 y ,% v, ft/s Ae Froude No. y ,%
Mgal/d ,%
Dp Af Dp

10 0.545 0.4 31 4.7 73 1.91 60

12 0.748 0.7 31 5.2 74 1.97 60

15 1.23 1.2 30 6.0 75 2.04 60

18 1.77 1.8 29 6.6 76 2.10 60

21 2.41 2.6 28 7.2 77 2.15 60

24 3.14 3.6 28 7.8 78 2.20 60

27 3.98 4.7 27 8.4 78 2.24 60

30 4.91 6.1 27 8.9 79 2.27 60

33 5.94 7.6 26 9.4 79 2.31 60

36 7.07 9.3 26 9.9 79 2.34 60

42 9.62 13.5 25 10.8 80 2.40 60

48 12.6 18.5 25 11.7 80 2.45 60

54 15.9 24.5 24 12.5 81 2.49 60

60 19.6 31.5 24 13.3 81 2.53 60

--

66 23.8 39.5 24 14.0 82 2.57 60

-

72 28.3 48.6 24 14.8 82 2.61 60

Escritt’s assumption: add half the width of free liquid surface to wetted perimeter in hydraulic radius.
Omitting Escritt’s assumption ≈ changing n from 0.010 to 0.0125.
Dp is inside diameter of pipe.
y is depth of flow.
Ae is inside area of pipe above liquid surface.
Af is inside area of pipe.
For n = 0.009, multiply Q by 92%.
For n = 0.011, multiply Q by 108%.
For n = 0.012, multiply Q by 115%.
For n = 0.013, multiply Q by 122%.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 79

Appendix C – Intake basin entrance conditions, trench-type wet wells for solids-bearing liquids (informative) — 2018

C.4.3  Transition manhole, sewer to approach pipe

The transition in the manhole between the upstream conduit and the approach pipe is designed to accelerate the
liquid to the velocities shown in Tables C.1 and C.2, using a sloping transition between the invert of the upstream
conduit or sewer on one side and the invert of the approach pipe on the other side. The drop (and hence the slope)
of the transition invert can be found by the application of Bernoulli’s equation.

In the sewer, the energy grade line (EGL) lies above the liquid surface by the velocity head, V For a sewer flowing
2

2g
full at maximum design flow rate, the EGL is likely to be somewhat above the soffit. In the approach pipe, the EGL
is 60% of the Dp above the invert plus the velocity head, and the sum is usually about 75% Dp above the invert.

Locate the approach pipe so that its EGL is below the EGL of the sewer by an amount equal to the expected head
loss due to turbulence and friction. As data on head losses are sparse, be conservative and increase the invert
drop somewhat to ensure supercritical flow. A small increase in velocity has an even smaller effect on the sequent
depth. For example, velocities 20% greater than the values shown in Tables C.1 and C.2 increase the sequent
depth from 60 to only 67% Dp. Such an increase reduces safety but, nevertheless, may be tolerable.

C.4.4 Lining

The approach pipe is subject to corrosion caused by sulfuric acid forming above low liquid line by bacteria acting
on sulfur compounds. As with the wet well, all surfaces above low liquid level should either be lined with an impervi-
ous material (e.g., plastic) immune to corrosion or the pipe itself should be plastic.

C.5  Design examples

Examples of wet well designs for

• Variable-speed pumps

• Constant-speed pumps

• Approach pipes

• Transition manholes are given by Jones et al. (2005).

Tables C.1 and C.2 can be modified by changing value of flows, pipeline gradients, or roughness using the spread-
sheet tools available on the website www.Pumps.org/IntakeDesign.

` ``` ```` ` ` ` ` ` `

80 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix D – Performance enhancements for trench-type wet wells (informative) — 2018

Appendix D

Performance enhancements for trench-type wet wells (informative)

Information in this appendix is not part of this standard, but is presented to help the user in considering factors
beyond the standard sump design.

Refer to Section 9.8.2 of the standard, which allows for an intake designed to a geometry other than that presented
in the standard, such as those contained in these appendices, to be deemed to comply with the standard if the
intake is tested by prototype testing or a physical model study performed in accordance with Section 9.8.7, and the
test results comply with the acceptance criteria in Section 9.8.7.7.

Requirements for a physical model study are given in Section 9.8.7.

D.1 Scope

This appendix describes improvements (devices or solutions) that reduce swirling and vortices.

D.2  Performance of bare trenches

D.2.1  Normal operation

During normal operation, the incoming liquid jet travels to the end wall with moderate abatement, dives to the floor,
returns upstream along the floor to the ramp, and moves upward to join with the incoming jet, thus setting up a cir-
culation pattern. The floor current is confined by the trench but nevertheless wanders somewhat so that it may be
relatively strong at one side and weak at the other side. The differential current at the suction bell creates swirling
that, in middle pump intakes, exceeds the standard of acceptance in Section 9.8.7.7. Swirling changes the angle of
attack on the impeller blades, which results in loss of head, capacity, and efficiency, or may substantially increase
motor loads.

Flow separation creates strong floor vortices under the suction bell and moderately strong subsurface sidewall
vortices opposite the suction bell and somewhat below the rim of the bell. The vortices become smaller but more
intense within the suction throat and may cause cavitation, excessive noise, vibration, and maintenance.

Surface vortices can occasionally form when intake submergence is low, so the submergence should be above that
required by Equation 9.8.3.1.4-2 at all normal flow rates. During pump-down, vortices become strong enough to
suck scum, bubbles, or even a solid core of air into the pump. See Figure 9.8.7.5a.

D.2.2  Cleaning operations for wastewater and stormwater wet wells

Cleaning is accomplished by either of the procedures detailed in Section 9.8.4.1.6. Whatever the procedure, the
--

flow down the ramp attains very high velocity, and a hydraulic jump proceeds down the trench to the last pump as
shown in Figure D.1. The hydraulic jump suspends all solids, which are then swept into the last pump. Friction rap-
-

idly reduces the high velocity and the Froude number. It is important that the Froude number be at least 3 at the last
pump and that the last suction bell be submerged at least D below the sequent depth as shown in Figure 9.8.4.1.4,
2
because a large, air-core vortex usually forms beside the suction bell to prevent lowering the liquid surface much
below Froude number 3.0. The Froude number is defined by Equation 9.8.7.3-1 wherein L is the depth of flow.)

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 81

Appendix D – Performance enhancements for trench-type wet wells (informative) — 2018

Figure D.1  Open trench-type wet-well hydraulic jump

D.3  Enhancing normal operation

D.3.1  Inlet baffles

An inlet or target baffle as in Figure D.2 is advantageous in changing the inlet liquid jet to a gentle downstream
flow above the trench, reducing the upstream current along the floor, and reducing swirling in pump intakes. Both
horizontal and vertical baffles are reasonably effective, but the latter is less critical in size and placement and less
prone to catching stringy material. The inlet baffle in Figure D.2 cut the average swirl in half in one physical model
study with no other enhancements, although the results were still somewhat short of meeting the requirements of
Section 9.8.7.7. The baffle was 1.67D wide and submerged to 1.0D below the ramp apex. The location of the target
baffle can be scaled from the figure with adequate accuracy. Inlet baffles seem to improve pump performance, but
the extent of improvement and the optimum dimensions and location can be determined only by a physical model
study.

Figure D.2  Open trench–type wet-well with inlet baffle

D.3.2  Suction bell vanes

Four (or more) suction bell vanes as shown in Figure D.3 can effectively control swirling if they are large enough.
There must always be a passage sufficient to pass a 75-mm (3-in) sphere for small pumping stations and larger
spheres for large stations. Cast iron suction bells with vanes are unlikely to be available, but for the few needed in a
pumping station, it may be cost-effective for any shop with programmable plasma cutters to fabricate bells of 316L
or 347 stainless steel with welded vanes.

` ``` ```` ` ` ` ` ` `

82 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix D – Performance enhancements for trench-type wet wells (informative) — 2018

Figure D.3  Suction bell vanes

D.3.3  Floor cones

For clear liquids, floor cones, as shown in Figure D.4, can minimize floor vortices. When equipped with vanes, they
can also reduce swirling. They are excellent and relatively inexpensive for all pumps in clear liquid pumping sta-
tions. For very nonuniform approach flow to the wet well, four vanes are not enough to produce acceptable swirling,
but six or eight vanes can be very effective.

Figure D.4  Floor cone with vanes for clear liquids

For solids-bearing liquids, floor cones cannot be used under upstream pumps because the high-velocity flow during
cleaning would be completely disrupted. However, a cone under the last pump is desirable. Two vanes, as shown
in Figure D.5, are all that is necessary. Make the clearance between the suction bell and the vanes at least 75 mm
(3 in). The rear vane can extend to the anti-rotation baffle. If there is a flow splitter under the upstream pumps, then
the splitter should be terminated after the next to last pump bell.

Figure D.5  Floor cone with vanes for solids-bearing liquids

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 83

Appendix D – Performance enhancements for trench-type wet wells (informative) — 2018

D.3.4  Flow splitters in wastewater wet wells

Flow splitters can be made of stainless steel plate 6 to 10 mm (1/4 to 3/8 in) thick. They can be anchored by weld-
ing them to stainless-steel embedments (Figure D.6) set at suitable intervals. Flow splitters can also be set in a
shallow depression and anchored with straps welded to the bottom of the flow splitter. The straps are fastened in
the trench with anchor bolts set in epoxy (or other two-component systems) in drilled holes. Bolt heads and nuts
are then covered with grout. Embedments are preferred.

`---
` `
`
````-`-`
```
--`
Figure D.6  Flow splitter in wet well

In some wet wells, the flow splitters have been terminated at the foot of the ramp. However, regardless of the geom-
etry of the terminating nose, the high-velocity liquid striking the nose bursts into spray and loses much of its energy,
so it is far better to extend the flow splitter to the top of the ramp where the liquid velocity is much lower and there is
sufficient length down the ramp for recovery of small disturbances.

On ramp curves with radius, r, the radius of curvature, R, of the plate is

r
R=  (Eq. D.3.4-1)
sin α
where α is the angle between the concrete floor and the plate. At the top of the ramp, the nose can be made short
and blunt if the velocity is not much above critical (critical velocity corresponds to a Froude number of 1.0). If the
velocity significantly exceeds critical, the nose should be gradually tapered to zero at the top of the ramp. As ramps
and floors are not likely to be accurately formed, it is wise to make a light wood model of the splitter, place it on the
concrete, and scribe offset lines to the concrete surface so that any irregularity can be precisely met.

A tapered nose can be made by cutting the plate at its intersection with the ramp on a radius of

r  (Eq. D.3.4-2)
R=
sin α cos β

where β is the angle between the ramp centerline and the intersection of plate and ramp. The outside radius can
r+h r
vary from to or even less depending on the wanted refinements in the nose shape (where h is the
sin α sin α
height of the splitter). The height of the nose varies from zero at the top of the ramp to h where it joins the rest of the

84 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix D – Performance enhancements for trench-type wet wells (informative) — 2018

flow splitter. That junction requires the insertion of a small triangular filler plate. These refinements can be devel-
oped in the wooden model.

D.3.5 Fillets

Fillets with a 45 degree (preferred) to 55 degree slope are effective in eliminating sidewall vortices. The top of the
fillets must be no lower than D below the suction bell. Fillets can be made larger if desired to reduce channel width,
8
increase flow depth, and decrease friction loss during pump-down for cleaning.

Fillets are best made of shotcrete (Gunite) sprayed, screeded, and troweled smooth. They should be reinforced and
anchored into the corner. As with flow splitters, fillets should be extended to the top of the ramp to produce a good
flow pattern down the ramp.

D.3.6  Maintaining cleaning velocity

The high velocity produced by the ramp quickly dissipates due to friction. See Appendix M for informative material
regarding calculations for trench-type wet wells. It quickly solves the velocities and the hydraulic and energy pro-
files down the ramp and along the trench, calculates the Froude numbers, and plots sequent depths at intervals.
(This program cannot be used if a flow splitter is terminated at the foot of the ramp because of the unknown loss of
energy when liquid strikes the splitter nose and bursts into spray.)

Velocity along the floor can be maximized by (1) specifying concrete to be smoothly troweled, (2) by lining the bot-
tom of the trench with plastic or other smooth coatings, (3) by confining the flow with large fillets and a flow splitter
with maximum (45 degree) side slopes, and, as a last resort, (4) by sloping the bottom of the trench beginning at
the point where the velocity is as low as can be tolerated.

D.3.7  Last pump

The one device universally required during cleaning is the anti-rotation baffle between the end pump and the back-
wall shown in Figures 9.8.4.2.2 and D.5. Without it, liquid circulates between the pump and the wall so that the
current on one side of the pump actually goes upstream and keeps the hydraulic jump far upstream. It may be nec-
essary to weld a part of the baffle to the pump suction nozzle to limit sufficiently the size of any opening. The vanes
attached to the cone are not only desirable, they may be necessary to prevent the circulation of liquid beneath the
suction bell.

During pump-down, the pump is subjected to severe service due to excessively low submergence and cavitation.
Select robust pumps. Clean the wet well at the smallest practical flow rate so as to dewater the basin and complete
--

cleaning as quickly as possible.

-

D.3.8 Ramps

Concrete for ramps can be cast in stair steps with dowels placed to anchor a reinforced blanket of shotcrete.
Screed the shotcrete to templates temporarily bolted to the sides of the trench and trowel the surface smooth.

D.3.9  Choice of enhancements

Judgment as to which enhancements to use should be based primarily on their effectiveness in improving perfor-
mance. Note that their cost as a percentage of the cost of the pumping station is insignificant whereas the effect on
performance and reducing maintenance may be very significant. Life-cycle costs are likely to favor these enhance-
ments and the improved performance.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 85

Appendix D – Performance enhancements for trench-type wet wells (informative) — 2018

D.3.10  Omission of enhancements

In wet wells for station capacities less than about 0.44 m3/s (10 mgd), it may be more cost-effective to omit enhance-
ments such as flow splitters and fillets in favor of using impeller materials that are cavitation-resistant. See ANSI/
HI 9.1-9.5 Pumps – General Guidelines for Types, Definitions, Application, Sound Measurement and Decontamina-
tion. Although flow splitters and fillets can be placed in trenches that are only 0.9 m (3.0 ft) wide, their cost is at a
premium because of the crowded work space. The premium is less in trenches 1.1 m (3.5 ft) wide and disappears
for trenches 1.2 m (4 ft) wide.

Without suction bell vanes, the average swirl in a middle pump operating alone may exceed 5 degrees (the allow-
able limit) about half the time. If another pump is also running, the swirl may be excessive most of the time. Vanes
are always desirable, and a target baffle may be added for certainty.

` ``` ```` ` ` ` ` ` `

86 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix E – Aspects of design of rectangular wet wells for solids-bearing liquids (informative) — 2018

Appendix E

Aspects of design of rectangular wet wells

for solids-bearing liquids (informative)

Information in this appendix is not part of this standard, but is presented to help the user in considering factors
beyond the standard sump design.

Refer to Section 9.8.2 of the standard, which allows for an intake designed to a geometry other than that presented
in the standard, such as those contained in these appendices, to be deemed to comply with the standard if the
intake is tested by prototype testing or a physical model study performed in accordance with Section 9.8.7, and the
test results comply with the acceptance criteria in Section 9.8.7.7.

Requirements for a physical model study are given in Section 9.8.7.

E.1 Introduction

The material presented in this appendix is provided for the convenience of the intake design engineer who faces

`---
the challenge of optimizing sump geometry for new or existing rectangular wet wells with solids-bearing liquids.

` `
Many retrofit installations require the upgrade/replacement of mechanical equipment and an increased station

`
capacity, yet intake hydraulics are many times neglected. This can cause a variety of problems ranging from a noisy

````-`-`
installation to adverse hydraulic conditions at the pump inlet, which can reduce equipment life.

```
--`
Further, this material transmits general experience and knowledge gained over many years of improving the hydrau-
lics of intake structures, and such educational material may not include the specific recommendations appropriate
for a standard. Features described herein have been effective in the past, but may or may not be able to be incor-
porated in an existing wet well or for a given set of site-specific conditions. Other remedial features not provided
herein may also be effective, and a physical model study is needed to verify whether or not a given remedial design
feature results in acceptable flow conditions. This is particularly true because adding a remedial feature to solve
one flow problem may have detrimental effects on other flow phenomena of concern.

Rectangular wet wells pose special challenges as described in Section 9.8.4.4, but are quite often used because
of the physical size requirements of installations with multiple pumps, ease of construction, and reuse of existing
structures. In such cases, incorporating special provisions to ensure proper inflow to the pumps and to minimize
dead zones where solids can settle and accumulate is essential for optimal station performance.

Two important design requirements are preventing significant quantities of air from reaching the impeller, and dis-
posal of settled and floating solids. The recommendations in this appendix can be used as they are, or with appro-
priate variations to meet the requirements of most installations.

E.2  Design capacity

A sump designed in accordance with the recommendations in this appendix is smaller than a conventional sump.
Consequently, there may be less buffer volume to accommodate transient variations of the flow rate. Also there is
no extra retention volume to store the inflow in excess of the total pump capacity (the pipe volumes are usually
much larger than any pump station volume). All critical aspects of operation should be considered in a proper
design.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 87

Appendix E – Aspects of design of rectangular wet wells for solids-bearing liquids (informative) — 2018

E.3  Design alternatives – general

A well-designed baffle wall minimizes air entrainment due to falling liquid. The flow from the inlet pipe strikes the
partition wall then flows down into the inlet chamber through the slot in the floor of the baffle. The slot distributes
the flow evenly toward all the pump inlets. The partition wall is high enough to ensure that the flow does not surge
over it. Although the flow in the inlet chamber is highly turbulent, various materials can collect there. In such cases,
side overflow weirs or side gaps may be used to carry away debris and thus prevent its accumulation. (The top of
the partition wall, or parts of it, should be below the highest start level of any of the pumps to allow transport of the
floating material into the pump chamber.) Equipping the sump with fillets, baffles, and/or benching is often benefi-
cial depending on the number of pumps and their size. To avoid pre-swirl in the pump chamber, the inlet pipe must
have a straight length of five pipe diameters upstream from the sump.

E.4  Front – high-level entry intake structure

The central front high-level entry is the sump design shown in Figures E.1 and E.2. In this configuration, the flow
does not have to make a horizontal turn, which might induce mass rotation in the sump. The exact sump design
varies with the number of pumps and pump size.

E.5  Side – high-level entry intake structure

If the piping system and the sump location do not allow for a front entry inlet, then a side entry inlet with a baffle
wall modified with ports can be used. This configuration is shown in Figures E.3 and E.4. In this design, the baffle
wall redirects the incoming flow and distributes the flow evenly toward the pumps through the ports.

E.6  Side – low-level entry intake structure

In this arrangement, shown in Figures E.5 and E.6 with a straight baffle wall, either the sump or the sewer is
below the normal liquid level in the sump, or an open channel supplies the sump. In the absence of falling flow in
the entrance, no intense entrainment of air takes place. Consequently, the inlet chamber can be greatly simplified
because its only task is to distribute the flow evenly to the pumps.

E.7  Cleaning procedures

Removal of solids from wet wells, designed in accordance with these principles, can be adequate to prevent the
continued buildup of solids by operating the pumps selectively to lower the level in the wet well until just before the
pumps lose prime. Both settled and floating solids are largely removed by the pumping equipment and discharged
into the force main (or discharge conduit). Scum is removed by the surface vortices that form beside a pump when
submersion of the intake is greatly reduced.

Vortices are small and removal of floating solids can be substantially improved by supplemental features, such as
induced turbulence or liquid sprays that quickly move surface scum to the active pump. This cleaning procedure
momentarily subjects the pumps to vibration, dry running, and other severe conditions. Consult the manufacturer
when selecting the pumping equipment. The frequency of the cleaning cycle depends on local conditions, and
therefore should be determined by experience at the site.

As with all wastewater wet wells, grease accumulates on the walls and must occasionally be removed. It is easily
washed off surfaces lined with plastic by means of a hose or liquid lance using about 1.5 L/s (25 gpm) of liquid at a
nozzle Pitot pressure of 600 kPa (90 psi). Concrete surfaces are more difficult to clean.

E.8  Sump dimensions

Refer to Figure E.7 for recommended sump dimensions. Note: Submersible pumps are shown in the figures but the
designs and dimensions also apply to applications using dry-pit pumps.

Note: Minimum required wet-well levels are as given in Figures 9.8.4.3.1a and b.

` ``` ```` ` ` ` ` ` `

88 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix E – Aspects of design of rectangular wet wells for solids-bearing liquids (informative) — 2018

Figure E.1  Front – high-level entry

Figure E.2  Schematic, front – high-level entry

--

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 89

-
Appendix E – Aspects of design of rectangular wet wells for solids-bearing liquids (informative) — 2018

Figure E.3  Side – high-level entry

Figure E.4  Schematic, side – high-level entry

90 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

` ``` ```` ` ` ` ` ` `
 Appendix E – Aspects of design of rectangular wet wells for solids-bearing liquids (informative) — 2018

Figure E.5  Side – low-level entry

--
-

Figure E.6  Schematic, side – low-level entry

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 91

Appendix E – Aspects of design of rectangular wet wells for solids-bearing liquids (informative) — 2018

Figure E.7  Recommended sump dimensions

` ``` ```` ` ` ` ` ` `

92 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix F – Suction bell design (informative) — 2018

Appendix F

Suction bell design (informative)

The design information in this appendix is provided for general information and is not intended to modify or super-
cede standard or proprietary designs provided by pump manufacturers.

F.1 Introduction

The function of a flared suction bell intake is to direct the flow of liquid into an intake pipe or into the throat of the

`---
first-stage suction impeller of a pump. It is a flared converging bell-shaped section designed to guide the liquid uni-

` `
`
formly into the intake pipe or pump impeller with a minimum of hydraulic losses. Refer to Figure F.1.

````-`-`
F.2  Bell outside diameter

```
--`
The flared-bell outside diameter D can be determined per Section 9.8.5 based on a recommended average inlet
velocity of 1.7 m/s (5.5 ft/s), Table 9.8.5.2a and b.

F.3  Ratio of bell outside diameter to throat diameter

The bell diameter D typically falls within 1.7 to 2.3 times the impeller eye throat diameter d or, in axial flow pumps,
the propeller outside diameter, or intake pipe throat diameter. Smaller or greater ratios of D/d are possible, espe-
cially when it is a flared intake for a conduit upstream of the pump suction impeller.

Bell diameter ratios less than 1.7 are possible when the approach flow is sufficiently symmetrical about the axis of
the suction impeller.

F.4  Suction bell length

Suction bell length (L) is selected based on hydraulic, mechanical, and economic considerations. A bell with a
shaft tail bearing has to accommodate the required bearing hub. A cone attached to the floor or suspended from
the pump eliminates a submerged vortex directly under the bell. The hydraulic loss of a well-formed suction bell is
typically within a range of 0.04 to 0.09 times the velocity head corresponding to the throat diameter.

F.5  Bell intake shape

The bell intake shape is designed to ensure that the flow velocity change is gradual throughout the intake cross
section. The contour is typically designed by one of the following curves that form an arc of a curve:

a) Circle
b) Ellipse
c) Parabola
d) Compound based on two radii
e) Lemniscates of Bernoulli
f) Compound curves based on the above, or other methods

After the dimensions D and d are established, the dimension L can be calculated or selected per the specific curve
shape geometry dimensions.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 93

Appendix F – Suction bell design (informative) — 2018

A pump bell designed with a bearing hub and support vanes (ribs) may have a flow area blockage caused by these
elements if appropriate design adjustments are not made. The bell shape may need to be adjusted to compensate
for blockage elements and ensure a smooth rate of velocity change and minimize energy losses.

`---
` `
`
````-`-`
```
--`
Figure F.1  Bell intake shapes

94 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix G – Submersible pumps – well motor type (informative) — 2018

Appendix G

Submersible pumps – well motor type (informative)

This appendix is not part of this standard, but is presented to help the user in considering factors beyond the stan-
dard sump design.

G.1  Submersible pumps – well motor type

Design criteria are provided for both wet-pit type and closed bottom can below-grade suction intakes (Figure G.1).
Proper placement of this type of submersible pump in a well is beyond the scope of this standard.

Figure G.1  Submersible vertical turbine pump

A submersible well-type motor normally requires a minimum flow of liquid around the immersed motor to provide
for adequate motor cooling. For many applications a shroud is required to ensure proper cooling flow around the
motor. Sizing of the cooling shroud for internal flow velocities must be deferred to the pump manufacturer. The top
of the shroud must include a cover to restrict downward flow of liquid to the pump inlet, while allowing for venting of
air from the shroud.

The intake piping must be large enough to limit drawdown below the recommended minimum liquid level to a
period of less than three seconds during start-up.

The first-stage impeller is located above both the strainer and motor. A pump suction inlet is located below the first-
stage impeller. The confined flow pathway provided by the motor cooling shroud is very desirable in developing a
`---

uniform flow to the first-stage impeller. Therefore, placement of the wet-pit-type submersible per Section 9.8.3.1 is
` `
`

only necessary for flow rates above 315 L/s (5000 gpm).
````-`-`
```
--`

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 95

Appendix H – Modification of existing pumping systems (informative) — 2018

Appendix H

Modification of existing pumping systems (informative)

This appendix is not part of this standard, but is presented to help the user in considering factors involved in the
modification of existing pumping systems.

H.1 Scope

This appendix applies to the modification of existing pumping systems. Typical modifications involve system reha-
bilitation, system upgrades, and system expansions.

H.2 Purpose

The purpose of this appendix is to raise awareness within the scope of the intake design standard for the need to
reconfirm the design of any pumping system when any component of the original design has (or will be) changed.

H.3 Recommendations

Design of even the most basic pump system involves consideration of the site, hydraulic performance, civil and
mechanical design, transient analysis, types of equipment, construction issues, power considerations, and control
logic/instrumentation – all in the light of current standards and accepted practices.

To obtain satisfactory results it is necessary that these aspects all be reconfirmed when any characteristic of the
original design is modified. Failure to do so risks serious operational problems in pumping systems. These prob-
lems usually prove difficult and very costly to resolve.

A common modification to an existing pumping system is to increase the flow rate through the system, which may
be inadequate to accommodate the additional flow satisfactorily.

In such cases the modified design should be reconfirmed as if it were a new design. A physical model study is
­particularly useful in avoiding problems. Refer to Section 9.8.7.

` ``` ```` ` ` ` ` ` `

96 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix I – Alternate formed suction intake designs (informative) — 2018

Appendix I

Alternate formed suction intake designs (informative)

Information in this appendix is not part of this standard, but is presented to help the user in considering factors
beyond the standard sump design.

Consult pump manufacturer when using these types of suction intakes.

Refer to Section 9.8.2 of the standard, which allows for an intake designed to a geometry other than that presented
in the standard, such as those contained in these appendices, to be deemed to comply with the standard if the
intake is tested by prototype testing or a physical model study performed in accordance with Section 9.8.7, and the
test results comply with the acceptance criteria in Section 9.8.7.7.

Requirements for a physical model study are given in Section 9.8.7.

I.1  Stork-type formed suction intake

The principal converging element of a stork-type formed suction intake is shown in Figures I.1a, b, and c. The
dimensions are given as ratios of the suction bell throat (1.0d). This design originated from Dutch polder pumps.
The simple geometry of this design is appropriate for concrete construction. The opening width (4.25d as shown)
--

may vary between 3.25 and 5.40d. The inlet geometry is adapted to this design by filleted or tapered walls extended
to meet the approach channel or intake bay. Though relatively insensitive to upstream conditions, entrance cross-
-

flow velocities exceeding 3.5 m/s (12 ft/s) will adversely affect intake performance and are to be avoided. The flow
splitter located below the suction bell is tapered to a final width of 0.04D to minimize the wake. Depending on FSI
size, it may be most practical to form the upper portion of the splitter using steel plate.

I.2  Shoe-box-type formed suction intake

The fundamental design elements of a shoe-box formed suction intake are shown in Figures I.2a, b, and c. The
primary advantages to this configuration are the ability to use the inlet as a remedial device in existing pump sumps
and that it does not require removal of the pump or pump bell. The dimensions are given as ratios of the bell diame-
ter d. The overall width of the intake is equal to 2d and the height at the bell is set at 0.5d, allowing it to be used in a
typical existing intake. The design has been developed during a series of physical model studies and details of the
development were presented in a comprehensive paper by Werth and Cheek (2004).

The inlet geometry consists of flat sides with the exception of a simple curved backwall. Fillets and splitters are
included within the inlet to prevent submerged vortex activity and flow-straightening vanes are included at the
entrance for use in applications with cross-flow. A flared entrance is used to reduce the intake velocity and reduce
submergence requirements. Tests show the inlet to be highly effective with cross-flow velocities of up to 0.75 m/s
(2.5 ft/s) and minimum submergence requirements as presented in this standard. Higher cross-flow velocities and
lower submergence levels may be acceptable but should be verified with a physical model study. The intake can be
easily fabricated out of steel and can be installed without removal of an existing pump or bell.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 97

Appendix I – Alternate formed suction intake designs (informative) — 2018

Figure I.1a  Stork-type FSI, plan view

Figure I.1b  Stork-type FSI, elevation view

Figure I.1c  Stork-type FSI, perspective view

98 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

` ``` ```` ` ` ` ` ` `
 Appendix I – Alternate formed suction intake designs (informative) — 2018

Figure I.2a  Shoe-box-type FSI, plan view

Figure I.2b  Shoe-box-type FSI, Section B

Figure I.2c  Shoe-box-type FSI, Section A

99
` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

Appendix J – Rectangular intakes for shallow liquid source (informative) — 2018

Appendix J

Rectangular intakes for shallow liquid source (informative)

Information in this appendix is not part of this standard, but is presented to help the user in considering factors
beyond the standard sump design. Details of this appendix are intended to work in conjunction with a physical
model study.

Refer to Section 9.8.2 of the standard, which allows for an intake designed to a geometry other than that presented
in the standard, such as those contained in these appendices, to be deemed to comply with the standard if the
intake is tested by prototype testing or a physical model study performed in accordance with Section 9.8.7, and the
test results comply with the acceptance criteria in Section 9.8.7.7.

Requirements for a physical model study are given in Section 9.8.7.

J.1 General

When the liquid source is shallow relative to the required pump submergence, such as is frequently found with
cooling tower basins, the alternative configuration described in this appendix may be used in place of the rectangu-
lar geometry recommended in Section 9.8.3.1. This alternative configuration is recommended when:

a) The minimum depth of liquid in the source basin is less than one half of the minimum required depth of liquid in
the sump.3
b) The source basin floor is generally level within the vicinity of the sump.
c) The source basin walls adjacent to the sump are vertical.
d) The sump is oriented so that its longitudinal axis is perpendicular to the adjacent source basin walls.
e) The sump contains no more than three pumps.

The intent of this appendix is to provide recommendations for anticipating and preventing the occurrence of inlet
control4 at the entrance to the sump, and to provide a geometric alternative to the long structure that would result
from applying the 10-degree maximum floor slope recommended in Section 9.8.3.1.

J.2  Entrance conditions

When the liquid depth of the source is shallower than required to satisfy the pump, NPSHR, or to control surface
vortices, a vertical transition is necessary at the entrance to the sump to effect the depth increase. Although it is
not strictly necessary to adhere to the maximum allowable floor slope, α, of 10 degrees (Figure 9.8.3.1.4a) for the
intake described in this section, particular attention must be given to the entrance condition to provide flow that is
stable and well-distributed.

To ensure that flow does not pass through critical depth at the entrance to the sump and is reasonably stable,
the Froude number upstream from the entrance to the sump must not exceed 0.3. To satisfy this Froude number
requirement, the depth of liquid upstream from the entrance to the sump must be considered with the sump width
and flow, so that:

3 In this case, the minimum depth of liquid in the sump is the greater of the depth required to satisfy pump NPSHR, or the depth to
minimize surface vortices as determined by Equation 9.8.6.2-3.
4 Inlet control is a term used to describe the restriction on gravity-driven flow at an abrupt transition or entrance to a channel or
culvert, which is imposed when the flow approaches critical depth and velocity.

` ``` ```` ` ` ` ` ` `

100 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix J – Rectangular intakes for shallow liquid source (informative) — 2018

0.667
 Q  (Eq. 9.8.3.1.4-3)
H1 ≥ C  
 W1 

Where:

Q = total flow at W1, in L/s (ft3/s)

H1 = liquid depth at the entrance to the intake structure, in m (ft) (See Figure J.1)
W1 = width at the entrance to the intake structure, in m (ft)
C = 0.01 if flow is in L/s and lengths are in m
C = 0.7 if flow is in ft3/s and lengths are in ft

Increasing the liquid depth immediately upstream from the sump entrance is the most effective means of reducing

`---
the Froude number. The increased depth may be achieved by raising the minimum liquid surface elevation in the

` `
`
source liquid basin, reducing the floor elevation in the source liquid basin, or by incorporating an intermediate step

````-`-`
into the source basin floor design immediately upstream from the entrance to the sump. The plan dimensions of the
intermediate step must be such that L1 and L2 have the minimum dimensions shown in Figure J.1, and the length of

```
--`
boundary between the basin floor and the intermediate step must be such that the Froude number along the entire
length also does not exceed 0.3. That is, calculate the minimum depth above the source basin floor by substituting
H2 for H1 and the boundary length of the intermediate step for W in Equation 9.8.3.1.4-3. In this case, Q in Equation
9.8.3.1.4-3 is the total station flow.

Designers should note that, depending on actual operating conditions, the maximum Froude number at the
entrance to a pump bay may occur when a single pump operates at its runout flow and the liquid surface elevation
in the basin is at its minimum.

J.3  Vertical transition

The transition between the floor of the source liquid basin (or intermediate step) and the floor of the pump sump
may be either sloped or abrupt. Sloped floors may improve the stability of flow through the structure, but typically do
so by increasing overall length and facility cost. If a sloping transition is used, the distance from the toe of the slope
to the pumps should be held to the recommendation for Z2 in Figure 9.8.3.1.4a and Table 9.8.3.1.4a.

J.4  Pump bay dividing walls and details near the entrance

Each pump must be separated from the others with a dividing wall. Each dividing wall must begin at the entrance
flush with the inside surface of the source liquid basin walls and extend completely to the sump’s backwall. Smaller
pumps such as screen wash liquid, service liquid, or auxiliary pumps may be placed upstream from the curtain wall
provided their combined capacity does not exceed 15% of the main pump flow.

Flow-guiding vanes must be used at the entrance to the intake structure to prevent flow from contracting or
becoming poorly distributed. These vanes must be arranged so that they are flush with both the inside face of the
source liquid basin wall and the upstream end of the pump bay dividing walls. The size, shape, and quantity of the
flow-guiding vanes should be determined during the physical model study.

J.5  Pump bay details near the pumps

Geometric details near the pumps must include corner fillets at the junctions between the side- and backwalls
and floor. Center splitters must be centered beneath the pumps to prevent the formation of floor vortices and swirl.
Refer to Figure J.2.

A curtain wall is placed upstream from the pumps to assist in directing surface currents toward the pumps and to
reduce potential for the formation of surface vortices.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 101

Appendix J – Rectangular intakes for shallow liquid source (informative) — 2018

Figure J.1  Configuration for rectangular intakes withdrawing from shallow liquid source,
maximum three pumps (refer to Figure J.2)

` ``` ```` ` ` ` ` ` `

102 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix J – Rectangular intakes for shallow liquid source (informative) — 2018

Figure J.2  Pump bay details near the pump bells for rectangular intakes with a shallow liquid source

`---
` `
`
````-`-`
```
--`

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 103

Appendix K – Influence of pump selection on intake design (informative) — 2018

Appendix K

Influence of pump selection on intake design (informative)

This appendix is not part of this standard, but is presented to help the user in considering the influence of pump
selection on intake design.

K.1  Influence of pump selection on intake design

The pump selection and the system curve characteristics have an influence on the intake design, as will be illus-
trated in this appendix. The selection of a pump “design point” may not represent the normal operating point and
hence, not the best flow to use for the intake design for each pump. An example is when multiple pumps operate in
parallel and there is a significant friction loss increase in the total system.

Figure K.1 represents three pumps in parallel to achieve the design flow of a sewer lift station. Because the station
must meet this maximum wet weather flow, the designer often selects that flow and head per pump as the “design
point.” This may be fine for acceptance of the pump but may not be the proper design flow for the intake design. In
the case of a sewer lift station, for most of the time, only one or two pumps may be needed.

As can be seen from Figure K.1, the flow per pump increases as the number of pumps operated decreases. With
two pumps on, the flow per pump increases by nearly 27% compared to three pumps on. With one pump, the flow
would be 55% higher if the pump could operate successfully at that runout condition. Thus, if the design point is

`---
used for setting the intake design, the intake design may be inadequate and result in unfavorable inlet conditions.

` `
`
````-`-`
For this example, the operation of just one pump probably results in flow well beyond the acceptable or allowable
operating range of the pump. Additionally, the intake based on the original design point will exacerbate the unac-

```
ceptable operation of one pump.

--`
Figure K.1 includes another significant assumption that the system curve is a fixed quantity. In practice this is rarely,
if ever, true. Instead, the friction factor of the system is usually not known with precision and even if it is, friction will
usually change over time as the pipe ages or accumulates corrosion or other deposit layers on the pipe. On top of
this, the static head of a system can also vary when either the intake wet well level changes or the level changes at
the end of the pipe. This case of varying friction as well as varying static head is illustrated in Figure K.2.

As can be seen in Figure K.2, the potential variation in flow per pump becomes even greater than the simplified
system curve case of Figure K.1. Now the potential flow per pump is even greater and this influences the perfor-
mance of the intake to the pump.

One approach that designers often take is to include variable-speed drives to attempt to lower the flow per pump
and keep the pump within its allowable operating range when operating on one or two pumps.

Figure K.3 illustrates the application of a variable-speed drive to one pump operating. Each line on Figure K.3 rep-
resents a 10% reduction in speed from the 100% speed. For the design point to fall within the system curve range,
the speed has to be reduced to about 60% of speed. On clear fluids this may be an acceptable turndown but, for
sewage, there could be a problem with the velocity through the pump being low enough that the pump becomes
plugged with debris.

The pump selection engineer may want to consider selecting two different-sized pumps to cover the range of flow
encountered so that performance is optimized for normal operations and can still meet the required maximum flow
design.

104 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix K – Influence of pump selection on intake design (informative) — 2018

Figure K.1  Simplified system curve

--
-

Figure K.2  Multiple system curves

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 105

Appendix K – Influence of pump selection on intake design (informative) — 2018

Figure K.3  Multiple system curves, variable speed

`---
` `
`
````-`-`
```
--`

106 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix L – Sediment and debris issues at surface water pump station intakes (informative) — 2018

Appendix L

Sediment and debris issues at surface water pump

station intakes (informative)

The design information provided in this appendix is for general information and is presented to help the user in con-
sidering factors in the design pump station intakes on rivers.

L.1  Issues with surface water intakes

Raw water supply for municipal, agricultural or industrial use could come from rivers, lakes, reservoirs, aqueducts
or canals. Because these bodies of waters could contain different types and sizes of solid materials that could
damage or shorten the life of the pumping equipment, these should be removed by using screens and sometimes
followed by pretreatment (siltation basins) process, prior to entering the pump. Debris can range from leaves, tree
branches, rocks, sand, grit, vegetation, algae, fish, marine life and growth, dead animals, snow deposits, ice, or
materials such as plastics, cans, etc. The amount of solids and debris found in intakes is dependent on storm
events and season. Sediment is an everyday issue and changes seasonally and with rain and storm events in the
watershed. Sediment transported with the flow stream can range from fine sediment and organics that are in sus-
pension that may pass through the intake and settle in the downstream equipment in a plant to heavier sediment
that can accumulate at the entrance to the intake, within the intake itself, and in the pumps and discharge piping.

The design objective of a good intake is to provide uniform flow into the pumps and to provide velocities as recom-
mended in ANSI/HI 9.8. Accumulation of sediment and debris in the pumping station intake typically will not settle
uniformly. As a result, the uniform flow in the intake is disturbed and nonuniform flow patterns and velocity distribu-
tion may result.

The handling and accumulation of sediment is critical to long term operation of pumping equipment, piping and
valves. Sediment can cause the following problems:
--

• Reduction in effective process tank volumes, such as in water treatment plants

-

• Internal water passages of pumps, piping and valve erosion

• Pump and pipe plugging

• Inoperable valves

• Clogging of bar screens, strainers, heat exchangers, etc.

• Wear of pump casings and impellers

• Wear on traveling screens and rotating components

• Wear and plugging of process equipment

L.2  Selection of intake location

L.2.1  River intakes

Rivers are the most active landform feature on the planet. Rivers are constantly changing. Changes in flows (e.g.,
storm frequency and duration, overland flows, input locations and stormwater outfalls), sediment supply, adjacent

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 107

 Appendix L – Sediment and debris issues at surface water pump station intakes (informative) — 2018

land use, navigation, etc., will cause the river to alter its geometry. Flow withdrawals upstream and/or downstream
will cause the river to react by increasing/decreasing its sinuosity, width, and sediment transport rates. These trans-
formations lead to changes in bed surface elevations and slopes, possible increases in bank instability and erosion,
and increases/decreases in channel water surface elevations. River channel cutoffs upstream or downstream, will
likely cause lateral and vertical instabilities. Upstream of dams and reservoirs, the river channel will aggrade for
a significant distance upstream, as sediment is stored in the channel area upstream of the dam. Downstream of
dams, the river will degrade and widen due to a diminished supply of sediment. The magnitude of any change is
unique to each river and river reach. It can take decades for the river to reach a quasi-equilibrium.

The location of a river intake should be selected in a stable reach of the river where vertical and lateral movement
is minimal. This requires a thorough assessment of the river geomorphology including geologic and recent river
history, river usage (including past, present, and future), potential for debris deposition at the site, sediment trans-
port (including bed form height and transport rate), flooding rates, and navigation. With this information the lateral
and vertical stability of the reach should then be predicted into the future for the lifespan of the project. In navigable
rivers, care is required to ensure that the intake does not interfere with navigation and that any dredging to maintain
a navigation channel does not interfere with the intake. This can lead to a secondary in-depth investigation. For
example, if river meander cutoffs have occurred in the recent past on a channel, a thorough and quantitative inves-
tigation of the reach is required to predict local, vertical, and lateral river stability at the intake.

At every potential intake location, sediment transport rates should be determined using the method most appro-
priate for the size of sediment in the channel to identify the minimum elevation to withdraw flow without entraining
suspended sediment, dunes, or other bed forms. At the same time, the intake should be sufficiently below the water
surface that air entrainment and surface vortices do not enter the intake, debris loading and ice impacts are mini-
mized, and navigation is not impacted.

Intakes should not be located on the inside of a river bend. This is an area of sediment deposition and is typically
a shallow point in the river cross-section. The outside of a bend can be acceptable as long as the bank is found to
be stable and damage from debris is minimal. In some cases, the section of the river reach just upstream of a river
bend can provide a reasonable balance between channel stability and available water depth.

A very experienced river engineer and fluvial geomorphologist with sediment transport expertise is required to
identify the best location for the intake structure and to develop a design that protects the intake and does not
adversely impact other factors.

L.2.2  Fresh water lakes, reservoirs, aqueducts and canals

Pump station intakes along the banks of lakes, reservoirs, aqueducts, and canals could be located based on the
quality of water, suction hydraulic conditions, natural mixing of raw water, stable ground, reliability of availability of
water, constructability, and construction cost. Most of the lakes and reservoirs have better water quality as com-
pared to river water and solids or debris found on those water are limited to leaves, tumble weeds, twigs, and algae,
which could be removed by the use of mechanical means. In most regions of North America, fish screens (fine
mesh) are required to prevent entry of fish and minimize mortality rate of fish species. Additionally, the invasive
Zebra/quagga mussel (a species or subspecies of freshwater mussel, an aquatic bivalve mollusk in the family) are
found in some fresh water lakes and reservoirs.

Quagga mussels were first found in Arizona in Lake Mead in January 2007. They originally came from Eurasia and
became established in the Great Lakes in the 1980s. Since being discovered, these prolific invaders have spread
rapidly. A single adult quagga mussel can produce up to one million larvae in a single year. They colonize rapidly
on hard surfaces and can ruin boat motors and clog water intake structures, such as pipes and screens, thereby
impacting pumping capabilities for power and water treatment plants. Invasive mussels such as quaggas and the
closely related zebra mussels have cost industries and businesses in the Midwest hundreds of millions of dollars in
maintenance and damage repair.

Zebra/quagga mussels can be controlled by different techniques, such as lake bottom aeration technology, chemi-
cal injection, as well as component material selection.

` ``` ```` ` ` ` ` ` `

108 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix L – Sediment and debris issues at surface water pump station intakes (informative) — 2018

L.2.3  Flow regime at the intake

Intakes should be located so that there is sufficient flow velocity in the channel to maintain transport through the
reach. Constructing an intake in a back water region, side channel, embayment, etc., where low velocities occur
may provide sufficient approach flow rates to the intake structure. However, sediment may deposit in these regions
impacting the operation of the intake, including loss of capacity and increased maintenance. Periodic dredging of
an approach channel to an intake can reduce or alleviate this problem.

The intake should be located in a river reach with subcritical flow, where flow rates are sufficient to transport the
amount of sediment supplied to the reach and that sediment move primarily close to the bed (as bedload) and/or in
bedforms (e.g., dunes). In supercritical flow, the river flow velocities are high, making it a challenge to redirect the
flow to the intake, and sediment transport rates will increase with a majority of the sediment more likely to be trans-
ported throughout the water column as suspended sediment. Debris can become more of an issue in fast moving
water due to collisions and potential damage to the intake.

To determine the optimum location for an intake structure requires a very thorough understanding of river engineer-
--

ing and sediment transport. Since each intake site is unique, local knowledge of the river or similar rivers is highly
recommended. At every intake site maintenance will be required for the life of the project, however, careful investi-
gation and site selection will minimize these costs.
-

L.3  Sediment and debris control and removal

L.3.1  Eliminating/minimizing sediment into intakes

Intake structures cannot keep suspended sediment from entering the intake. Intakes can, however, be designed to
exclude debris and minimize entry of the sediment that travels as bed load. The amount of sediment entering an
intake is dependent on many factors, including the intake location, the layout of the intake, the vertical location of
the withdrawal points, and the withdrawal flow velocities. Assuming the selected location is in a stable river reach,
an intake can be oriented and designed so that sediment transports past the intake. Care is required in orienting
the intake so that scour around the base of the structure is minimized. The bottom elevation of the withdrawal loca-
tion should be a reasonable distance above the highest bed form that may pass in front of the structure to inhibit an
influx of sediment traveling as bed load.

Physical modeling with a mobile bed is an excellent tool for optimizing the design and position of an intake, to
ensure sediment passes the structure, to assess the impact of bed forms, and to assess scour potential.

L.3.2  Collection, removal, and sediment treatment

Passive measures as described above are preferred to managing sediment and debris by periodic cleaning and
dredging to remove sediment near the intake, within the intake, by installing a system to continuously manage the
sediment.

For sediment management in pump intake bays using mechanical methods, the system needs to collect, transport,
and classify or separate the sediment for disposal. Consideration should be given to designing the pump intake
structure with multiple bays so that there is the ability to isolate a bay for cleaning of sediment and debris without
compromising the capacity of the pumping system.

Debris can be controlled by screens either mechanically operated or passive. In either case fish and aquatic animal
protection need to be considered, and these regulations vary by location. The advantage of passive screening systems
is that there is no disposal of debris as these types of screens do not collect and remove debris like mechanically
cleaned screens.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 109

Appendix M – References (informative) — 2018

Appendix M

References (informative)

Section 9.8.1.3

Chaudhry, M.H., Open-Channel Flow, Prentice Hall, Englewood Cliffs, NJ, 1993.

Chow, V.T., Open-Channel Hydraulics, McGraw-Hill, New York, NY, 1959.

Section 9.8.3

Dicmas, John L., Vertical Turbine, Mixed Flow and Propeller Pumps, McGraw-Hill Book Company, 1987.

U.S. Army Corps of Engineers (ETL No. 1110-2-327).

Section 9.8.4

www.Pumps.org/IntakeDesign

Section 9.8.6

Hecker, G.E., Chapter 8, Conclusions, “Swirling Flow Problems at Intakes,” IAHR Hydraulic Structures Design
­Manual 1, J. Knauss, Coordinator-editor, A.A. Balkema Publishers, Rotterdam, 1987. Knauss Chapter 1.

Section 9.8.7

Anwar, H.O., Weller, J.A., and Amphlett, M.B., “Similarity of Free Vortex at Horizontal Intake,” Journal of Hydraulic
Research, IAHR, Vol. 16, No. 2, 1978, p. 95.

Daggett, L., and Keulegan, G.H., “Similitude in Free-Surface Vortex Formations,” ASCE Journal of the Hydraulics
Division, Vol. 100, HY11, November, 1974, p. 1565.
--

Hecker, G.E., “Model-Prototype Comparison of Free Surface Vortices,” ASCE Journal of the Hydraulics Division, Vol.
-

107, No. HY10, October, 1981, p. 1243.

Jain, A.K., Raju, K.G.R., and Garde, R.J., “Vortex Formation at Vertical Pipe Intakes,” ASCE Journal of Hydraulics
Division, Vol. 104, No. HY10, October 1978, p. 1429.

Knauss, J., Coordinator-Editor, “Swirling Flow Problems at Intakes,” IAHR Hydraulic Structures Design Manual 1,
A.A. Balkema Publishers, Rotterdam, 1987.

Padmanabhan, M., and Hecker, G.E., “Scale Effects in Pump Sump Models,” ASCE Journal of Hydraulic Engineering,
Vol. 110, No. 11, November, 1984, p. 1540.

Appendix A

Moore, A., The Use of Vanes to Improve Wide Angle Diffuser Performance, RR 1377, BHRA, 1976.

Padamanabhan, M., Evaluation of Vortex Suppressors, Hydraulic Performance of Single Outlet Sumps, and Sensitiv-
ity of Miscellaneous Sump Parameters, Alden Research Laboratory Report No. 49A-82/M398F, September 1982.

110 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix M – References (informative) — 2018

Appendix C

Cahoon, J.E., and Sanks, R.L., “Hydraulic Profiles in Trench-type Wet Wells for Wastewater Pumping,” Presented at
WEFTEC 2002, Chicago, Ill (October 2, 2002).

Escritt, L.B., Sewerage and Sewage Treatment. International Practice, edited and revised by W.D. Haworth, John
Wiley, New York, 1984.

Jones, G.M., Sanks, R.L., Tchobanoglous, G., and Bosserman, B.E., Pumping Station Design, 3rd Ed., Elsevier,
Burlington, MA, 2005.

Appendix D

www.Pumps.org/IntakeDesign

Appendix I

Werth, D.E., and Cheek, D., “Development of Design Guidelines for Alternative Formed Suction Inlets,” ICE/IAHR
Journal of Water Management, 157(3), 151–158, 2004.

` ``` ```` ` ` ` ` ` `

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 111

Appendix N – Index — 2018

Appendix N

Index

Note: f. indicates figure; t. indicates table. debris, 107–108

control and removal, 109
American Society of Mechanical Engineers (ASME) design of rectangular wet wells, 87
standards, 51 cleaning procedures, 88
approach pipes, 77 design alternatives, 88
maximum allowable flow rates, 78t., 79t. design capacity, 87
schematic diagram of, 77f. high-level entry intake structure, 88, 89f., 90f.
aqueducts, 108 low-level entry intake structure, 88, 91f.
auxiliary storage, 76–77 sump dimensions, 88, 92f.

bare trenches, performance of, 81, 82f. energy grade line (EGL), 80
bell inlet velocity, 42 Escritt modification, 77, 78t., 79t.
bell intake shape, 93, 94f.
fillets, 85
--

canals, 108 floor clearance Cf, 20

can vertical turbine pump intakes, 26 floor cones, 83, 83f.
closed bottom can, 28, 29f. flow of liquid, 9, 51
-

design considerations, 26 flow regime at intake, 109

open bottom can intakes, 26, 27f. flow splitters, in wastewater wet wells, 84–85
circular plan wet pit formed suction intake (FSI), 16, 67, 97
accessories, 36 alternative designs, 18
cleaning procedure and low liquid level, 37 application standards, 18
floor clearance, 37–39 recommended dimensions for, 17, 17f.
minimized horizontal floor area and, 37f.–39f. shoe-box-type formed suction intake, 97, 99f.
wet-pit design, 36 stork-type formed suction intake, 97, 98f.
circular pump stations (clear liquids), 18 free surface approach, 62, 63f.–65f.
designs of, 18, 19f. free surface vortices, 49, 51–52, 52f., 65–66, 66f., 68f.
recommendations for dimensioning, 20 fresh water lakes, 108
clear liquid, intake structures for, 1 Froude number, 46, 49, 65, 77, 81, 85
can vertical turbine pump intakes, 26–29, 27f., 29f.
circular pump stations, 18–20, 19f. horizontal gratings, 65–66
formed suction intakes, 16–18, 17f. horizontal tank, 69–70, 70f., 71f.
rectangular intakes, 10–16, 13f.–14f., 15t.–16t. hydraulic phenomena, 8–9
tanks, pump suction, 22–25, 24f., 25f.
trench-type intakes, 21–22, 21f. inflow pipe, circular pump stations, 20
unconfined intakes, 30–32, 31f. inlet bell clearance Cb, 20
closed bottom can, 28, 29f. inlet bell design diameter (D), 42, 43f.–44f.
closed conduit approach, 62, 65f. acceptable velocity ranges for, 43t.
computational fluid dynamics (CFD) inlet bell velocity, 42
in pump suction hydraulics, 55 inlet bell or volute diameter Db, 20
simulation methods, 55 inlet bell velocity, 42
concentrated flows, expanding, 62 inlet or target baffle, 82, 82f.
closed conduit approach, 62, 65f. intake structure
free surface approach, 62, 63f.–65f. for clear liquid see clear liquid, intake structures for
through dual flow screen, 64f. designing, 9–10
constant-speed pumps, in trench-type wet wells, 76 flow regime at, 109
cross-flow, controlling, 58, 61f.
curtain walls, 65 liquid level, 51

112 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 Appendix N – Index — 2018

minimum submergence open versus partitioned structures, 10, 57, 60f.

formed suction intake, 18 recommendations for dimensioning, 11–12
rectangular intake structures, 11 for shallow liquid source, 100–101, 102f.–103f.
structure layout, 13f., 15t.
ogee spillway transition, 35 trash racks and screens, 10–11
open bottom can intakes, 26, 27f. rectangular wet wells
open sumps, 57 anti-rotation baffles, 41
open trench-type wet-well cleaning procedure, 41
hydraulic jump, 81, 82f. confined wet-well design, 40, 41f.
inlet baffle, 82, 82f. control of sediments, 40
design of, 87–88, 89f.–92f.
performance enhancements, for trench-type wet geometry of, 40
wells, 81 suction inlet clearance, 41
bare trenches performance, 81, 82f. reservoirs, 108
enhancements choice, 85 Reynolds number, 50–51
fillets, 85 river intakes, 107–108
floor cones, 83, 83f.
flow splitters, 84–85, 84f. sediments, 108
inlet baffles, 82 control and removal, 109
last pump, 85 handling and accumulation of, 107
maintaining cleaning velocity, 85 shallow liquid source, rectangular intakes for, 100
omission of enhancements, 86 configuration for, 102f.
ramps, 85 entrance conditions, 100–101
suction bell vanes, 82, 83f. pump bay details near the pumps, 101, 103f.
physical hydraulic model study, 48 pump bay dividing walls, 101
physical model study, 48 vertical transition, 101
acceptance criteria, 54 shoe-box-type formed suction intake, 97, 99f.
instrumentation and measuring techniques, 51–53, solids-bearing liquids
52f., 53f. additional criteria for, 9
--

need for, 48 circular plan wet pit for, 36–37, 37f.–39f.

objectives, 48–49 cleaning procedures, 34
report preparation, 54 design principles, 32–33
-

scope, 51 horizontal surfaces near inlet, 34

similitude and scale selection, 49–51 intake structures for, 1, 32
test plan, 53–54 rectangular wet wells for, 40–41, 41f.
pre-swirl, 49, 66 trench-type wet wells for, 34–36
pump bell throat, velocities in, 67 vertical transitions, 33, 33f.
pumping systems, modification of existing, 96 standard dimensions, deviations from, 68–69
pump inlet disturbances, 65 stork-type formed suction intake, 97, 98f.
free surface vortices, 65–66, 66f., 68f. strong vortex cores, 49
pre-swirl, 66 submergence for minimizing surface vortices,
subsurface vortices, 66 44–46, 47f.
velocities in pump bell throat, 67 submersible pumps, 95, 95f.
pump intake design, 1 subsurface vortices, 49, 52–53, 52f., 66, 67f.
design objectives, 8–10 suction bell design, 93, 94f.
preferred terms, 2t.–5t. suction bell length, 93
units and symbols, 5t.–8t. suction bell vanes, 82, 83f.
pump selection, on intake design, 104, 105f.–106f. suction umbrellas, 65
pump suction hydraulics, computational fluid dynamics sump diameter Ds, 20
in, 55 sump volume, 9
active, 72
ramps, 85 assumptions, 74
rectangular intakes, 10–16, 56 graph construction, 73
approach flow patterns, 10, 56–57, 58f.–59f. inflow rate, 73
design sequence, 16t. programmable controllers, 75, 75f.
filler wall details, 14f., 15t. simple controller, 73–75, 73f.

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved 113

Appendix N – Index — 2018

surface vortices, 65, 81 auxiliary storage, 76–77

application considerations, 46, 47f. cleaning procedure, 35
approach-flow skewness, 44–45 constant-speed pumps in, 76
controlling parameters, 45–46 design examples, 80
submergence for minimizing, 44 inlet floor clearance, 35
surface water intakes, issues with, 107 inlet splitters and cones, 36
swirl inlet transition, 35
in pump intake, 9 intake basin entrance conditions, 76–80, 77f.,
in suction pipe, 52–53 78t., 79t.
swirl meter, 52, 53f. lining, 80
objectives, 34–35
tanks, pump suction performance enhancements for, 81–86, 82f.–84f.
air or gas entrainment, 23 transition manhole, 80
antivortex devices, 70f. variable-speed pumps in, 76
design features, 22–23 typical swirl meter, 53f.
geometry, 22
horizontal, 69–70, 70f., 71f. unconfined intakes, 31f.
inflow and outflow configurations, 71f. cross-flow velocities and pump location, 30
multiple inlets or outlets, 25 debris and screens, 31
NPSH considerations, 25 submergence, 31–32
simultaneous inflow and outflow, 25
submergence, 23, 24f. variable-speed pumps, in trench-type wet wells, 76
V and D for submergence calculation, 24, 25f. velocity profiles, 53, 57
vertical, 69, 70f., 71f. vertical tank, 69, 70f., 71f.
trench-type intakes, 21–22 viscous effects, 50
approach flow, 22 vortex suppression devices, 45
centerline spacing, 22
end wall clearance, 22 wall clearance Cw, 20
floor clearance, 22 Weber number, 50
inlet conduit elevation, 22 wet wells for solids-bearing liquids
orientation, 21 cleaning procedures, 34
wet wells, 21, 21f. design principles, 32–33
width, 22 horizontal surfaces near inlet, 34
trench-type wet wells vertical transitions, 33, 33f.
anti-rotation baffle and vanes, 36 wet-well volume, 34

`---
approach flow, 35

` `
`
approach pipes, 77–79, 77f., 78t.–79t. zebra/quagga mussel, 108

````-`-`
```
--`

114 Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

 `---
` `
`
````-`-`
```
--`

B123

Hydraulic Institute Standards, Copyright © 1997-2018, All Rights Reserved

--` ``` ````-`-` ` ` ` `---