Fluid Filtration

Equipment
Methods
Measurements

www.GEKEngineering.com 1
 Polymer Damage
From: muds, pills, frac, carriers
Stable? - for years
location - depends on form polymer was in
dispersed properly - surface to deep in
formation
in pills and mass - right in perfs

www.GEKEngineering.com 2
www.GEKEngineering.com 3
 Particles in the Fluid
Solids from tanks, lines and fluids
Severe problem, but often ignored

www.GEKEngineering.com 4
www.GEKEngineering.com 5
www.GEKEngineering.com 6
www.GEKEngineering.com 7
www.GEKEngineering.com 8
www.GEKEngineering.com 9
www.GEKEngineering.com 10
 Particulate Damage
Unintended particulates - dirty fluids
filter fluids to 5 microns at Beta of 1000
maximum NTU of 30, preferable is 20
clean tanks, lines - how about tubulars?
Particles in fluid loss pills
mixed in proper range for perm encountered?
Will it come off formation? Can it come back
thru pack? Thru screen? What about
removal?
www.GEKEngineering.com 11
www.GEKEngineering.com 12
www.GEKEngineering.com 13
www.GEKEngineering.com 14
www.GEKEngineering.com 15
www.GEKEngineering.com 16
www.GEKEngineering.com 17
www.GEKEngineering.com 18
 Beta Rating
Beta = number of filter rating and larger
size particles in dirty fluid divided by
number of those particles in clean fluid.

Beta = 1000/1 = 1000 or 99.99% clean

www.GEKEngineering.com 19
 Filters - Operational
Filters work better at removing solids as
they collect solids bed filtration is very
effective.
The filtration efficiency (both size and
quantity removed) improve with use.
Pressure drop increases with use and is a
measure of filter life.

www.GEKEngineering.com 20
 Cartridge Filter Pods
Have two sets of cartridge filter pods
operate on one while changing filters in
the other.
Set a pressure drop standard to change
the filter depends on cartridge type and
pod type.
Watch dirty fluid discharge not into the
well.

www.GEKEngineering.com 21
 Pod filter schematic up to 36 filter
cartridges in any pod. Socks also used in
place of filters, but filtration quality may
be affected.

Pressure
drop
Dirty fluid inlet through the
filters

www.GEKEngineering.com 22
Clean fluid
Question when cartridges are removed during a
change out, where does the dirty fluid in the pod
housing and the sediment at the bottom go?
Directly into the clean fluids tank!

www.GEKEngineering.com 23
Q/A item: Equip the pod with a wash-out drain to clean
sediment before changing cartridges.

www.GEKEngineering.com 24
 Filtration
Filtration removes solids to prevent build
up of solids and helps prevent plugging in
the formation.
Two basic filtration systems are employed:
cartridges (nominal or absolute)
filter presses (Plate and Frame or Pressure
Leaf)

www.GEKEngineering.com 25
 Particulate Damage
Unintended particulates - dirty fluids
filter fluids to 5 microns at Beta of 1000
maximum NTU of 30, preferable is 20
clean tanks, lines - how about tubulars?
Particles in fluid loss pills
mixed in proper range for perm encountered?
Will it come off formation? Can it come back
thru pack? Thru screen? What about
removal?

www.GEKEngineering.com 26
 Filtration and Fluids
NTU - a turbidity indicator - can be mislead by
natural color of water.
Nominal filter rating - estimate of the size of
particle removed - dont trust it.
Absolute filter rating - size of the holes in the
filter - will change with bed buildup
Beta rating - a ratio of particles before filtration to
after. A good measure of filter efficiency.

www.GEKEngineering.com 27
 Risks and Issues
Filtration can impact pump rate of completion fluids
Filtration may be unnecessary cost in some wells, but
should be assessed on a well-by-well basis (from
potential damage mechanisms
Filter presses have the advantage of high solids
tolerance and throughput - units are large but cheap to
operate.
Cartridges used for small volumes of relatively clean
material - smaller throughputs and less tolerance to dirty
fluids, generally cartridges are more expensive
Can be health and environmental risks associated with
operating and disposing of filtration medium.

www.GEKEngineering.com 28
 Risks and Issues
NTU (a light transmission test) measurements rely
solely on the cloudiness of the fluids, corrosion
products in return fluids give high values, NTU and
solids content are not directly related. Solids content
will only assess materials collected at the bottom of the
test tube during centrifuging.
Coloured water will give higher NTU values.
Using sample sizes of a few mls, it is difficult to draw
conclusions about a well with an annular volume of 500
bbl, 0.1% vol/vol solids equates to +/- 1 cuft of solids
deposited for tubing contents of 200 bbl.

www.GEKEngineering.com 29
 Determination of Well
Cleanliness
The determination of how clean the well is usually
based on the cleanliness of fluids returning from the
wellbore. The most common measures are normal
turbidity units (NTU) and solids content, neither
relate to what is left in the well.
Junk baskets, gauge rings and the SPS
WellPatroller do give some positive indication of
solids removal.
Other indicators of a clean well are torque and drag
(related to the friction coefficient of fluid coating the
casing walls) and cleanliness of the clean up string
when pulled.

www.GEKEngineering.com 30
 Determination of Well
Cleanliness
The determination of how clean the well is usually based on the
cleanliness of fluids returning from the wellbore. The most
common measures are normal turbidity units (NTU) and solids
content, neither relate to what is left in the well.
Junk baskets, gauge rings and the SPS WellPatroller do give
some positive indication of solids removal.
Other indicators of a clean well are torque and drag (related to
the friction coefficient of fluid coating the casing walls) and
cleanliness of the clean up string when pulled.

www.GEKEngineering.com 31
 Learnings
1. Cartridge units are not usually appropriate for dirty or
viscosified fluids
2. DE press materials require good HSE control
3. Tendency is to over-specify filtration requirements.
4. Dont filter oil with a DE press.
5. Filtration less required when underbalance
perforating(?)
6. Kill pills are usually not filtered
7. Absolute cartridge filters are 2-5 times cost of nominal

www.GEKEngineering.com 32
 Best Practices
For general applications, coarse filtration to 80 microns
is all that should be considered when fluids do not
penetrate the formation
When a fluid penetrates the formation, filtration is more
likely to be required. Filtration should be tailored to the
pore throat size of the formation. A simple guide to
setting a specification is 14% (1/7th) of the average pore
throat diameter
Filtration below 2 microns is usually impractical
Always use a guard filter downstream of a DE press
For a DE press filtration, rate is approximately 1 bpm per
100 sqft for sea water and 0.5 bpm per 100 sqft for a
www.GEKEngineering.com 33
brine
 Best Practices, Pt 2
A solids content of about 0.05% determined by an electric centrifuge
is acceptable for most operations. For gravel packing this should be
reduced to 0.02%.
To determine solids content the standard tubes for sand content for
the centrifuge are not usually appropriate, centrifuge tubes with
more accurate calibrations should be obtained.
NTU values should be used to track clean up with regular samples
taken for analysis of solids content at a later date.
A target value of 50 NTU above surface pits should be used.
For critical wells e.g. high angle wells/where milling has
occurred/previous problems with debris, a circulating junk basket
(e.g., the SPS WellPatroller is recommended).

www.GEKEngineering.com 34
 Best Practices, Pt 3
Where the string is rotated during clean up the increase
in torque can be used as an indicator, the coefficient of
friction in sea water/brine is more than twice that of
OBM.
Visual inspection of the clean up string, if it is mud free
and water wet mud displacement has been successful, if
the string is mud coated run a gauge ring/junk basket or
SPS WellPatroller.
For critical application (e.g. gravel pack) use particle
size analysers on location, laser particle size systems
are recommended.

www.GEKEngineering.com 35
 Measurement Learnings
To deal with rust color (or precipitation)
interference with NTU readings, add a
small amount of HCl after the first reading
and re-measure.
If the junk basket is full, rerun.

www.GEKEngineering.com 36
 Filtration Considerations
Very clean fluids have high leakoff rates
Fluids with properly blended fluid loss
control materials added after filtration have
a better chance of cleanup than with the
initial particles in the fluid. Particles must
be sized to stop at the face of the
formation.
All gelled fluids should be sheared and
filtered even the liquid polymer fluids.
www.GEKEngineering.com 37
 Sizing of Filters
Methods to determine size of pore throats:
Kozeny
Coberly
Scanning Electron Microscope

Comments from Ken Troupe of Baker

www.GEKEngineering.com 38
 Pore Throat Size- Kozeny
d= ( D50 )/ (3(1- )) 0.128 (k/)1/2
where:
d = avg pore throat diam (m)
D50 = avg formation grain diameter (m)
= porosity (fraction)
k = absolute permeability (md)

www.GEKEngineering.com 39
 Pore Throat Size - Coberly

d = D50 / 6.5

where:
d = average pore throat diameter (m)
D50 = avg formation grain diameter (m)

www.GEKEngineering.com 40
 SEM Pore Throats
In addition to other methods, pore size
openings can be physically measured by
looking at a formation sample with a
scanning electron microscope.
Avg pore size determined statistically.

www.GEKEngineering.com 41
 SEM Rules of Thumb
If Dp < d, bridging occurs
If d/7 <Dp < d/3, shallow invasion (worst case)
If Dp < d/7, deep invasion (desirable??)

where:
Dp = diameter of particle (m)
d = average pore throat diameter (m)

www.GEKEngineering.com 42
 Particles in the Fluid
Solids from tanks, lines and fluids
Severe problem, but often ignored

www.GEKEngineering.com 43
 Filter Example for a Specific
Formation
k = 2000 md and = 22%:

d = 0.128 (2000/0.22)1/2 = 12.2 m

Dp = 1.7 m

Minimum filtration requirement = 2 m

www.GEKEngineering.com 44
 Example
k = 200 md and = 22%:
d = 0.128 (200/0.22)1/2 = 3.9 m
Dp = 3.9 / 7 = 0.55 m
Dp = 0.55 m

Filter = ? = low as possible, probably a 1 to 2

www.GEKEngineering.com 45
 Beta Rating
Beta = number of filter rating and larger
size particles in dirty fluid divided by
number of those particles in clean fluid.

Beta = 1000/1 = 1000 or 99.99% clean

www.GEKEngineering.com 46
 Saucier Method (SPE 4030)
Impairment versus Sand Size Ratio
If (D50/dp50) < 6, surface bridging occurs
If 6 < (D50/dp50) < 6, shallow invasion
If (D50/dp50) > 18, deep invasion occurs
where:
D50 = mean formation grain diameter
dp50 = mean particle diameter

www.GEKEngineering.com 47
 Saucier
Based on Saucier, the largest particle size
in our completion fluid should be less than
D50/18,
This correlation indicates a much less
stringent filtration requirement than
Kozeny's or Coberly's method.

www.GEKEngineering.com 48
 Saucier Formation Example
D50 = 105 micron (control w/ 20/40 Mesh gravel)
Dp = 105/18 = 5.8 m

Minimum filtration requirement = 6 m

www.GEKEngineering.com 49
 Saucier Formation Example
D50 = 60 micron(control w/ 40/60 mesh
gravel
Dp = 60/18 = 3.3 m

Minimum filtration requirement = 3 m

www.GEKEngineering.com 50
 Polymer Damage
From: muds, pills, frac, carriers
Stable? - for years
location - depends on form polymer was in
dispersed properly - surface to deep in
formation
in pills and mass - right in perfs

www.GEKEngineering.com 51
 Tank/Pit Cleaning
Use pit washing tools to speed up pit
cleaning, reduce pit entry and waste
volumes.
Draft plan including pumping schedules
and pit requirements, discuss at toolbox
talk with all relevant parties
Check pit isolation valves to reduce risk of
U-tube contamination

www.GEKEngineering.com 52
 Areas to Clean:
1. Tanks used for clean fluid and pumping
2. Blender or cement unit and piping to rig floor
3. Kill line
4. Choke and rig floor manifolds
5. Pump bleed over line to pits
6. Shaker area (header box) and sample point
7. Stand pipe and back pressure manifold
8. Degasser tank and unit
9. Reverse lines
10.Flowline where accessible
11.Mixing system
12.Trip Tank
13.Drain lines and drip pans

www.GEKEngineering.com 53
 Performance Best Practices
A brine returns with an NTU < 30 above
value of fluid in pits is excellent
performance
Clean returns after circulating < 150% of
hole volume is good performance
Interface volumes between pills of <20 bbl
is excellent performance

www.GEKEngineering.com 54
 Risks and Issues
Filter presses use beds of diatomaceous earth (DE) or
perlite,
they have the advantage of high solids tolerance and throughput,
units are large but cheap to operate.
Cartridges should be used where small low volumes of relatively
clean material have to be processed. They have smaller
throughputs than and less tolerance to dirty fluids. Cartridges are
typically more expensive
There may be HSE risks associated with operating and
disposing of filtration medium (dust, chemical residue,
etc.).

www.GEKEngineering.com 55
 Learnings
Single stage cartridges not appropriate for very
dirty or some viscosified fluids
DE presses present HSE issues.
Tendency is to over specify filtration specs.
Do not filter oil with DE.
If underbalanced perforation is planned, is
filtration required?
Filter kill pills? (solids range more important).
Absolute cartridges are 2 to 5 x Nominal cost

www.GEKEngineering.com 56
 Best Practices
For general applications, coarse filtration to 80
microns is all that should be considered when
fluids do not penetrate the formation
When a fluid penetrates the formation filtration is
more likely to be required, it should be tailored to
the pore throat size of the formation.
A simple guide to setting a specification is 14% (1/7th)
of the average pore throat diameter

www.GEKEngineering.com 57
 Best Practices
Filtration below 2 microns is usually
impractical in well operations
Always use a guard filter downstream of a
DE press
For a DE press filtration rate is
approximately 1 bpm per 100 sqft for sea
water and 0.5 bpm per 100 sqft for a brine

www.GEKEngineering.com 58
 How Clean?
The determination of how clean the well is usually based
on the cleanliness of fluids returning from the wellbore.
The most common measures are normal turbidity units
(NTU) and solids content, neither relate to what is left in
the well.
Also:
junk baskets, gauge rings and recovery systems give some
indication of solids removal.
torque and drag (related to the friction coefficient of fluid coating
the casing walls) and
cleanliness of the clean up string when pulled.

www.GEKEngineering.com 59
 Filter Level
Minimum filtration criteria
particles passing through the filter need to be
small enough to freely flow through the
formation without plugging up the critical near
wellbore area.
filtering should achieve a maximum particle
size in the completion fluid which is less than
the pore diameter divided by 7.

www.GEKEngineering.com 60
 Risks and Issues
Measurement systems:
NTU: (a light transmission test) measurements rely solely on the
cloudiness of the fluids. NTU and solids content are not directly
related. NTU is affected by color of water, stains, and surfactants.
Common limits are 20 to 30 NTU.
Centrifuging (grindouts/shakeouts) Solids content will only assess
materials collected at the bottom of the test tube during
centrifuging. These tests use the sample sizes of a few mls and it
is difficult to draw conclusions about a well with an annular
volume of 500 bbl, 0.1% vol/vol solids equates to +/- 1 cuft of
solids deposited for tubing contents of 200 bbl.

www.GEKEngineering.com 61
 Learning
To deal with the problem of free rust (oxidation
of free iron in water) impacting turbidity
readings, a little acid (HCl) should be added
after the initial reading.
Oxidized fluids (at surface) will precipitate
disolved ferrous iron.
If the junk basket is pulled full it should be re-
run.

www.GEKEngineering.com 62
 Best Practices
A solids content of >0.05% determined by an
electric centrifuge is acceptable for most
operations. For gravel packing this should be
reduced to 0.02%
To determine solids content the standard tubes
for sand content for the centrifuge are not
usually appropriate, centrifuge tubes with more
accurate calibrations should be obtained

www.GEKEngineering.com 63
 Best Practices
NTU can be used to track clean up by regular
sampling and later analysis of solids content.
NTU from the well should be no more than 50
above the source.
For critical wells e.g. high angle wells/where
milling has occurred/previous problems with
debris, baskets and debris recovery systems are
needed.

www.GEKEngineering.com 64
 Best Practices
Where the string is rotated during clean up the
increase in torque can be used as an indicator,
the coefficient of friction in sea water/brine is
more than twice that of OBM.
Visual inspection of the clean up string, if it is
mud free and water wet mud displacement has
been successful, if the string is mud coated run
a gauge ring/junk basket
For critical application (e.g. gravel pack) particle
size analysers on location may be useful.

www.GEKEngineering.com 65
 Completion Fluid - Checklist
Is the formation liquid sensitive to liquid relative permeability effects?
Compatability with formation? (clay and minerals)
Compatability with formation fluid? (emulsion, sludge, foam, froth)
Tanks and surface equipment clean?(pumps, lines, hoses, blenders)
Are polymers breakable? How is breaker added?
Polymers, hydrated, sheared and filtered? Filter level? Beta rating?
Minimize the pipe dope?
Corrosion reactions understood?
Erosion potential understood?

www.GEKEngineering.com 66
 Unwanted Foams
Foam is an emulsion where gas is the internal
phase (mist is an emulsion with gas as external
phase).
Foaming conditions
some oils (ppm conc. of C6-C9organic acids and
alcohols)
diesel - (particularly bad and varies from lot to lot)
some acid additives (emulsifiers, salts, inhibitors)
XCD polymers (and others)
Look for and destroy the stabilizer to break the foam.

www.GEKEngineering.com 67
 Filtration and Cleaning
Brine and ? (tanks, lines, pumps, etc.)
Pickle the tubulars?
NTU or particle count as a measure? How clean
is clean? = Avg pore throat x 0.2?
Beta rating and micron rating important
Tank arrangement when filtering (from dirty tank
to clean, not in a loop)
Filter type
DE or Cartridge (no resin coated cartridges)

www.GEKEngineering.com 68
 Filtration Ratings
NTU - a turbidity indicator - can be mislead by natural
color of water. An NTU of 20 to 30 is generally clean.
Nominal filter rating - estimate of the size of particle
removed - dont trust it. Filtration efficiency improves with
bed build-up
Absolute filter rating - size of the holes in the filter.
Filtration efficiency improves with bed build-up
Beta rating - a ratio of particles before filtration to after.
A good measure of filter efficiency.
Suggestion use a 5 to 10 micron rating with a beta
rating of 1000 for most clear brine applications.

www.GEKEngineering.com 69
 Beta Rating
Beta = number of filter rating and larger
size particles in dirty fluid divided by
number of those particles in clean fluid.

Beta = 1000/1 = 1000 or 99.99% clean

www.GEKEngineering.com 70
 Equipment Risks and Issues
Equipment failure may result in additional junk or a stuck clean up
string, (not a common problem.
There have been instances with casing scrapers and brushes where
the clean out tool has introduced junk to the hole when these are not
of a single piece construction or have retaining mechanisms for
blocks.
Circulating subs (particularly hydraulically operated) carry some risk
associated with failing to close thus losing the ability to circulate to
the bottom of the well.
Some of the equipment is only available from niche / specialist
suppliers which carries risks around QA/QC and tool availability.
There is concern with running well clean up tools with minimum by
pass area pushing large pieces of junk ahead and into CIV/FIV tools

www.GEKEngineering.com 71
 Equipment Learnings
Conventional casing scrapers with spring actuated
blocks have failed leaving junk in the hole or even a
stuck tool.
Conventional scrapers are not built for extended
rotation or drilling. Scrapers work in reciprocation.
Failure rate of multifunction ball opening circulating
subs has been high in some regions.

www.GEKEngineering.com 72
 Equipment Learnings
Weight actuated tools rely on maintaining weight set
down on the tool to maintain the circulation path (all
circulation is at one point). These tools should be
run with the clutch option allowing drillpipe to be
rotated independently of the pipe inside the liner.
(Rotation assists cleanout)
Recent experience with the SPS WellPatroller
highlighted it is very effective at removing gunk and
solids debris, this tool provides a level of assurance
not available with any of the other tools (including
conventional junk subs).

www.GEKEngineering.com 73
 Best Practices and Design
Casing scrapers should create maximum contact with ID of casing,
be a one piece design with full drill pipe strength and no weak
internal connections.
Casing scrapers should enable rotation at rates up to 50 rpm, these
should be of a design so that blocks are either retained in the body
of the scraper. (UWG/Global) or are an integral design (SPS
Razorback).
The preferred circulating sub should be of the type utilising weight
set down to function, with a clutch mechanism. If a reliable hydraulic
tool is developed this may become preferable as it allows
reciprocation with circulation.
For clean out strings in high angle cased well a WellPatroller should
be run (or if concern about debris or junk in the well).
Casing brushes are not considered necessary if an effective scraper
is selected. If a brush is used it should be redressed after each
application.
www.GEKEngineering.com 74
 Surface Learnings
Use pit washing tools to speed up pit cleaning,
reduce pit entry and waste volumes.
Plan pumping schedules and pit requirements,
Check pit isolation valves to reduce risk of U-
tube contamination
Focus clean up work from flow line/shakers
through to the pits.

www.GEKEngineering.com 75
 Surface Learnings Check
Areas
1. All pits to be used for clean fluid and pumping
2. Cement unit to rig floor
3. Kill line
4. Choke and rig floor manifolds
5. Pump bleed over line to pits
6. Shaker area (header box) and sample point
7. Stand pipe and back pressure manifold
8. Degasser tank and unit
9. Reverse lines
10. Flowline where accessible
11. Mixing system
12. Trip Tank
13. Drain lines and drip pans
www.GEKEngineering.com 76