FUSE Application Experiment 26770

Dissemination / Demonstrator document

NEW REGULATOR FOR PROPORTIONAL VALVE

The DSP gives a new heart to electric-hydraulic components

AE number: 26770
New Technology: DSP
Product industrial sector: Taps and Valves (Prodcom 2913)
Contact TTN: COREP

FUSE 1 ATOS demonstrator document

AE abstract
ATOS, a 280 employees company with a 40 MEUR annual turnover, is a leading manufacturer of
components & systems in electric-hydraulics, the advanced technology that integrates hydraulics
with electronics.
In many thousands of applications world-wide, ATOS components work today improving the
machines performances and confirming their outstanding reliability: in plastic machinery,
transfer lines, presses and metal forming machines, food processes and saw-mills, steel plants, off-
shore and civil engineering installations, etc.

In this AE, ATOS has developed an intelligent, fully programmable electronic regulator to
control the proportional valves for hydraulic systems used in factory automation and machine
tools.
This involved DSP technology in place of their existing analogue PCB system. The DSP enables
to minimise and solve the disadvantages that are typical of an analogue control (component
ageing, temperature drift, etc.) by means of a fast inner architecture capable of eliminating
computing errors, increasing performances together with a high degree of integration of data and
signal mutual exchange "peripherals". Moreover ATOS introduced for the first time in its product
a dedicated CAN interface in line with the trends of the distributed control and automation

The advantages that we realised are:

- More precise and flexible control with linearization of control curve
- Much more complex algorithms realised with digital filters
- Strongly reduced calibration and trimming costs (-50%)
- Increased reliability and flexibility (programmability)
- Easy connection with process automation components via CAN Bus interface

It is believed that the company's market shares, following the program's realisation, will be
increased both on the home and the international markets (about 10 % and 12 %, respectively).
The economic break-even period for the considered investment, according to our estimation, can
be less than 1 year after the market introduction. The return of investment ROI can be as large as
5 times in 5 years.
The total cost of the AE is 44kEUR and it lasted 6 months, as originally planned, from February
to July 1998.
The success of NINTER experiment induced ATOS to plan a complete technology update aimed
to upgrade all electronic products with the new digital circuitry based on DSP components.

LESSONS LEARNT from the AE

1. Internal Know-how Must be guaranteed that, at the end of the project, the new
technology became a stable and full acquired property of our
company
2. Training and Consultants Necessary to quickly learn how to correctly implement the new
technology and transmit it to the involved persons.
3. Project Work-plan A detailed planning of all phases of the project give a more
precise and reliable results of timing and costs evaluation.

Keywords:
Proportional valve, electric-hydraulic components, pressure/flow regulator, DSP technology,
internal replication, Mathlab for high level simulation, CAN bus, product diversification

Signature:
2 0112 555 1324 1 2913 2 29 I

FUSE 2 ATOS demonstrator document

1. Company name and address

ATOS Spa
Via alla Piana 57
21018 Sesto Calende (VA), Italy
Tel. 0039 0331 922078
Fax 0039 0331 920005
e-mail: scmail@atos.com
http://www.atos.com
Contact person: Paolo Fabbri

2. Company size
The company ATOS counts 280 employees of which 4 are involved in electronics. Turnover in
1997 was 40 MEUR.

ATOS decided to hire one electronic engineer with a good skill in digital electronic circuits to
guarantee a quick and stable acquisition of the new technology.

3. Company business description

ATOS designs, manufactures and sells a wide range of hydraulics components. The products are
delivered world-wide with target market the one of machine and modern plant manufacturers
that use electro-hydraulic components for intensive process automation, particularly concerning
the following fields:

Market segmentation
Company production is organised in four different departments:
1. Electrohydraulic components, pumps, cylinders, on-off and proportional valves, electronic
regulators
2. Customised electrohydraulic blocks and units
3. Electrohydraulic servo-actuators
4. Complete electrohydraulic systems and equipment
Turnover: (31/12/1997):

Products Domestic [MEUR] Foreign [MEUR] TOTAL[MEUR]

Pumps 1.705 1.456 3.161
Cylinders, servo-cylinders 5.358 0.958 6.316
Distribution and on–off 11.514 11.483 22.997
valves
Proportional valves 2.279 1.967 4.246
Complete units 2.715 0.212 2.927
Boards 0.290 0.216 0.506
Spare parts 0.192 0.258 0.450
Components 0.188 0.188 0.376
Total 24.241 16.738 40.979

Atos S.p.A speciality is electrohydraulic technology the one that includes oil, hydraulics and
electronics to enhance the performance of high precision tooling machines.

FUSE 3 ATOS demonstrator document

Oil-hydraulics performs the energy transmission which actuates the machine movements via a
pressurised oil (or equivalent fluid) flow. The main feature of oil-hydraulic controls is the ability
to handle very high power forces in a quite simple and reliable way since the hydraulic circuits are
sealed and lubricated from the oil itself and the size and weight of the component is very small.

The usage of oil-hydraulic and electric-hydraulic systems is wide and we can mention as
significant examples:
• Metallurgy :movement for melting, rollers, flying shears, thickness control,
lamination
• Sheet steel (punching, bending stations, shearing): movement and positioning, cutting
and punching sequence control, bending
• Moving and handling materials: robots, cranes, foundry manipulation
• Food industry: metering needles, level meters, mixers
• Tooling machines and transfer: transfer unit, drilling unit, control axes, hydraulic
copiers
There are many other industrial applications which proves the wide spread demand for this kind
of technology
On the contrary the offer is restricted to a relatively small and well reputed number of
manufacturers with a world wide production and sales network.
The electric-hydraulic techniques add to the outstanding advantages of the oil-hydraulics (high
power, reliability) the unique flexibility offered from electronics, allowing new and more advanced
control solutions to the modern industry automation systems.
The oil-hydraulic component market (including the complete units) is relevant. The estimation
for the European market is 6 Bl $ with more than 40% in Germany and more than 15% in Italy.
USA and Japanese markets have quite similar sizes.
A recent survey gives an increase trend of about 5% per year.
In this favourable scenario Atos offer includes a complete rang of products to most of the
mentioned industry applications of electric-hydraulics.
Atos has been acknowledged by specialised print among the 10 most highly reputed and
experienced companies; the list includes multinational giants like REXROTH (1000 MEUR
turnover), Vickers, Bosch..
Our current technology level is considered very advanced and we sell production licences to
Japan, China and Iran.

Production: over 650.000 electric-hydraulic components per year

Main manufacturing plants:

• Sesto Calende factory - Italy
20.000 m_ surface under roof
Pumps, valves, electronics and electric-hydraulic systems
• Modena factory - Italy
4.000 m_ surface under roof
Hydraulic cylinders and servocylinders
• Trieste factory - Italy
2.000 m_ surface under roof
Modular valves

4. Company markets and competitive position at the start of the AE

ATOS is a company specialised in electric-hydraulics. The state-of-the-art in this field is
represented by the production of high-performance electric-hydraulic proportional valves
complete with position transducer and controlled by analogue electronic regulators integrated into
the valve.

The current ATOS market concerning high-performance proportional valves amounts to 4.5
MEUR (in comparison with an overall 40 MEUR turnover) with an increasingly growing share
due to the increased product reliability obtained in the last five years.

FUSE 4 ATOS demonstrator document

The profitability for this class of products is the best in the whole range and amounts to 25 %
(operative/invoiced return) against an overall average value of 15 %.
In the following charts we reported the competition shares in the proportional valves market:

Home Market

OTHERS
10% ATOS
30%
Company B
30%
Company A
30%

International Market

ATOS
OTHERS
3%
27%
Company A
40%

Company B
30%

Currently the production of proportional valves is addressing for the 54% the national market
and for the 46% the international one. At the start of the AE the total quantities of proportional
valves sold and forecasted are showed in the following chart:

Proportional valves

20000

15000

10000

5000

0
1995 1996 1997 1998 1999

As you can see the proportional valve market is today increasing. This chart doesn't show how
the market is quickly evolving and demanding new features that cannot be added to our current
product as it is the case of the field bus interface requested by the process control and automation
industry.
Indeed our reputation in the market at the start of the AE was quite good since we succeeded in
improving the reliability of our valves in the last years while keeping the product price lower
than our competitors, as such, there is some room for cost increases. Therefore we were
confident that if we could succeed in meeting the customers request for added facilities we could
further increase our market share.

Our current proportional valve production is segmented into the following main sectors
FUSE 5 ATOS demonstrator document
Thus the NINTER project fully matched the ATOS business strategy turned to
• strengthening its position in a fast evolving market
• studying innovative solutions to maintain a technologically updated product line

5. Product to be improved and its industrial sectors

The product to be improved is the proportional control valve applied to the hydraulic pump and,
in particular, its electronic regulator.
Electric-hydraulic controls (proportional valve unit + electronic regulator + oleodynamic
actuator) add the benefits of electronic regulations to the natural capability of hydraulic to easily
control high power. They are usually used to adjust the volume and/or the pressure of the oil that
actuates the operation of several kinds of tooling machines, under the supervision of an axis
controller so to achieve high reliability and increased flexibility.
The proportional valve controller is the result of continued developments and improvements.
Throughout a period of twenty years it has turned from a simple acceleration and deceleration
control device into a sophisticated component fitted with a position transducer and integrating
complex analogue processing.
The rugged electronics of the proportional valve receives its commands directly from the NC
axis control card (supervising the operation of the machine electromechanical axis) and,
moreover, adapts the input signal to the characteristics of the hydraulic actuator, such as
differential commands, non-linear gains, high gains by small movements, "plug and play"
connection without any adjustment etc...
The technology used in the realisation of this electronics is based on conventional analogue
signal-processing circuits with operational amplifiers, and no digital (complex logic or
microprocessor)-operated electronics is presently being used on any electronic products
developed by ATOS, or by most competitors in this field.
Electronic regulators, realised with digital circuits, are actually already available from some of our
competitors but they have two important drawbacks which prevent a massive introduction into
the market.
- Their cost/performance ratio is still too high
- Their level of integration into the target valve is too small
The new, high-performance integrated hydraulic pump we developed, is made up of four parts
assembled together (figure 1):

A) the electronic pressure controller: performs both flow and pressure closed loop
control
B) the integrated electronic: feeds the proportional valve with current signal suitable to
align valve regulation to the reference signal provided by the axis control unit and/or the
integrated transducers .
The two parts A+B represents the electronic system which is the subject of the innovation and
togheter performs a three axis control
C) the proportional valve is the electromechanical part which controls the hydraulic
features (pressure, flow rate, etc.) and consequently the pump movement.
D) The pump is the hydraulic actuator moved by the valve and controlled by the regulated
pressure controller and the associated electronic

FUSE 6 ATOS demonstrator document

fig.1 – Product to be improved – 3 axis control

FUSE 7 ATOS demonstrator document

Product functional description (fig.2)

The core of electric-hydraulic

controls is the proportional valve.
The electronic driver regulates a
proper electrical current supplied to
the valve's solenoid according to the
reference signal (normally ±1OVDC).
The solenoid converts the electrical
current into a mechanical force acting
the spool/poppet against a return
spring: increasing of the current
produces a corresponding increase in
output force and consequent
compression of return spring thus
movement of the spool or poppet.
In pilot operated executions the
proportional pilot regulates flow and
pressure acting on the spool/poppet
of main operated stage.

There are 3 types of proportional valves to perform different control systems according to the
kind of requested regulation (position, velocity, pressure or flow), the number of transducers can
vary accordingly (1 2 or 3).
Due to the wide range of requests from our customers we assemble more than 70 different kind of
proportional valves control subsystems, with different values of passive components (resistors
and capacitors); this also means several calibration points (from 10 to 20 in average).
Calibration requires from 10 to 20 minutes per unit, which is a very long testing time.
Currently our electronic control system is composed from three different boards, a power supply
board, a driver PWM board and an analogue control board.
Two separate resistive trimmers, are used for adjusting the zero level and the full-scale range.

The next picture shows the electronics as included in the current product. It should be noticed
that here only the power driver part is mounted on board of the pump, while the analogue control
processor is located onto a separate Euro card PCB.

FUSE 8 ATOS demonstrator document

The technical/economic reasons that pushed ATOS toward the innovation of the product are
summarised below:

⇒ Difficulty in realising a flexible production of the integrated regulator with analogue

electronics, due the great variety of the required components for the settings of every valves
foreseen.
⇒ Very high number of analogue electronic components causing problems of thermal drifts,
offsets, noise, ageing (analogue processing part is composed of 220 electronic components).
⇒ Test and calibration difficulties and cost.
⇒ Hardware stiffness makes it hard to add new features to the device, in contrast to industrial
automation market requests (for the 3 axis version is now necessary to add an additional
electronic card for the pressure control see module A fig.1).
⇒ Regulator size is too large to allow housing in the pump body (requires a separate PCB)

6. Description of the technical product improvements

Besides addressing the limitations listed above, the introduced improvements are aimed to
enhance the proportional valve controllability in a distributed control environment.

In fact the trend of automation market underlines the following evolution

• The supervision in a complex automated plant is quickly moving from centralised control
toward distributed control, or rather toward the use of automation devices with integrated
intelligence.
• These new intelligent devices, besides their locally performed machine control task, should
feature functions of command/data communication with the distributed supervision system;
This is usually realised via a standardised field-bus.

No solution of this kind was available for digital control that could be integrated into a hydraulic
system.
Therefore the Atos main objective is to produce a new family of electric-hydraulic components
able , not only to manage the valve regulation, but also to behave as an intelligent subsystem able
to inteface a standard automation field-bus.

The NINTER project is the fundamental step to reach this ambitious goal.
FUSE 9 ATOS demonstrator document
NINTER foresees in the realisation of o series of prototypes embedding a DSP, A/D and D/A
converters, additional digital circuitry into a mixed TH-SMD PCB, with an optional added SMD
CAN Bus board

6.1) Description of the improved product

HARDWARE SYSTEM
The modular approach to hardware design involves the separation of the regulating modules and
the interfacing and power supply modules.

The whole system is divided into modules according to the following structure:
Ä DSP processing unit
Ä Power supply
Ä Analogue/Digital conversion unit
Ä Inductive transducer
Ä Communication unit
Ä proportional valve

BLOCK A - DIGITAL CONTROLLER

D.S.P. processing unit: the reduced size constraint asks for using an 80 pin PQFP processor
version with the minimum number of external “ports” needed by the system under consideration.
The following characteristics are specially required:

Ä no. 2 serial ports

ª a port directly controls the D to A converter
ª the second port carries out the communication to the external SMD module .
Ä Parallel bus for controlling the program and data storing devices
Ä Timer for timing the A to D conversio
Ä A non-volatile memory was chosen for containing information concerning control
coefficients, characteristic valve data, operating trajectories, etc.
The external BOOT-FLASH-EEPROM type 29F010 is large enough to allow the boot from 8
different control programs and the storing of the associated parameters.
As a further very important advantage, the
BLOCK B - DC/DC converter
Power supply: starting from 24 V d.c., three inner voltages are generated: + 5 V and ± 15 V by
using a switching mode conversion system.

BLOCK C Analogue/Digital conversion unit:

Ä the A to D converter converts 4 analogue signals in 32 µs with 12 bit precision of these:
ª one is used as the analogue reference (setpoint);
ª two are used for reading the position of the transducers inside the valve
(feedback signals):
Both converters are selected to be fast enough (8 µs per channel) comply with the system
requested passband.

The V/I converter function is eliminated implementing a PWM converter in the DSP to drive
directly the power drivers. The power driver current is also monitored in a dedicated feedback
loop.

BLOCK D - COMMUNICATION INTERFACE

FUSE 10 ATOS demonstrator document

A serial communication interface (on a separate optional board), provides a link with an external
fieldbus (CAN Bus) in case the operational set points and the control parameters are supplied by
the distributed supervision system.

SOFTWARE SYSTEM
From the software point of view, all processes defined in the system can be developed under the
control of an "ad hoc" operation language provided for the DSP.
The development followed a rigorous “top-down” approach and in this way the production of a
software documentation following all development stages was easily obtained.
CONTROL ALGORITHMS
The project included also the development of optimised control algorithms in order to optimise
the valve-transducer system regarding:
- the speed
- the steady state position accuracy
- the dynamic range of certain control parameters
The following control types have been analysed:
Ø Loop regulation with a PID control type with many non linear functions necessary
for the good control of valve spool position
Ø . Non linear functions fig.4
Ø Derivative filters
Ø Integral filters
Ø Asymmetrical functions

FUSE 11 ATOS demonstrator document

6.2) Summary of the realised product improvements.

The new integrated controller allows to avoid the need for assembling different board releases
while eliminating most of the requested calibrations, thanks to the integrated digital control that
manages via software the setting of the required operating conditions. The NINTER project and
his success will lead to the upgrade of the whole range of the ATOS proportional valves, bringing
a series of important technological advantages.

♦ Flexibility of production with much reduced manual work requested for the calibration
♦ Test cost reduction
♦ Easy configurability of the product
♦ Increased integration of the electronic components
♦ Size reduction since the software reduces the requested hardware; total components number is
reduced to 170
♦ Easy to start up and use
♦ Elimination of the V/I driver board since a PWM is implemented via software on the DSP
♦ Possibility to comply with the new demand of distributed intelligent devices from the modern
industrial automation scenario, that requires easy remote and flexible programmability.
♦ Easy connection to an industrial fieldbus (CAN interface)
♦ New added features like an extended number of closed control loops and the creation of
customised movement profiles.
♦ Increased reliability due to the elimination of about 35% of the analogue components
♦ Simpler maintenance thanks to auto-diagnostics of the valves status .

7. Choices and rationale for the selected technologies, tools and

methodologies
Once we had understood the advantages of a digital control approach with respect to an analogue
one,
we had to choose the most suitable technology for the implementation

The issues to be considered are:

FUSE 12 ATOS demonstrator document

♦ The control system should be interfaced with analogue signals
♦ It should strongly reduce the usage of analogue components
♦ It should not require long and difficult calibration and testing procedures
♦ It should guarantee a reasonable manufacturing and assembling cost
♦ It should exhibit high reliability and easy maintenance
♦ It should feature the maximum of programmability and flexibility

A conventional PCB (similar to the one already used in the current product) approach was
discarded because of the size reduction and programmability requests: In fact even with smaller
SMD components and more compact PAL components we would have obtained a 30% larger
circuit then required. Moreover the programmability would have been very limited and the testing
and calibration complexity almost unchanged.
The usage of a more complex hardware programmable device, FPGA or CPLD, was also
considered, (its price at this level of complexity would not have been so different from the DSP
one) but we considered it less convenient at system level because A/D and D/A converters are not
embedded in this kind of devices.
Another possibility would have been to develop a mixed signal ASIC, but the cost of the masks
(10 kEUR at least) and the medium volume request (initially some thousands) discouraged this
choice.
The only left possibility was to use a software programmable device, and this meant to decide
between Microcontroller and DSP.
When dealing with real time processing systems, we need to compare the different alternatives in
view of:
♦ processing throughput
♦ easiness of programming
♦ complexity of the algorithms to be run
♦ cost

Several digital control technologies based upon microprocessors, microcontrollers, digital signal
processors (DSP) have been analysed.
The microprocessor-based systems offer a control system improvement by eliminating all the
drawbacks, because most functions are carried out digitally.
Usually the modification or updating of a microprocessor based system involves a simple
modification of the software, whereas the hardware needs no changes.
In spite of that, even the best microprocessors/microcontrollers have limitations, since they are
too slow when carrying out complex calculations, such as those needed for real time axis controls,
with the ensuing danger of overload and information loss or the introduction of unwanted delays
in the control loop with consequent phase displacement among the controlled signals. Moreover,
the implementation of adaptive controls on microprocessors often turns out to be problematic.
Finally, these components appear to be inefficient in complex applications because they dispose
of a single bus for controls and data and because the arithmetical logic unit (ALU) carries out
multiplication by repeated summations through the use of several costly machine cycles.
Digital signal processors are microcontrollers realised with a particular inner architecture, with
two separate buses for programs and data (Harvard architecture), an arithmetical logical unit
capable of performing multiplication in a single machine cycle. These features appear to be of
particular importance in movement control applications because control algorithms are mainly
represented by multiplication and summation instructions.
While a 5 to 20 µs time for 16-bit number multiplication is required by general purpose
processors (Von Newman architecture), a DSP device requires only 60 to 150 ns, i.e. it is about
100 times faster and performs the operation with 32 bit precision . This processing speed makes
it possible to use high sampling frequencies thus enabling the handling of fast analogue to digital
signal conversion as well as preliminary processing (e.g., changing a position into a speed) before
delivering data to the controller. Another DSP advantage is the presence of barrel registers
FUSE 13 ATOS demonstrator document
enabling to shift numbers in a single instruction cycle with minimum truncation error while the
fixed point architecture grants the full control of transducer signals and saturation operation.
Digital signal processors offer new advanced programming features with a "signal processing"-
oriented instruction set: for instance, there is a single command for multiplication, additions,
two-number shifts and the 24 bit instruction allows the reduction of the whole number of needed
code thus reserving more space for implementing complex algorithms in the processor memory.
Also, the implementation of complex mathematical functions, such as the introduction of digital
filters to eliminates mechanical resonances or unwanted low/high frequency noise while making
much simpler the realisation of signal spectrum analysis.
Summing up, the choice of a DSP instead of microcontroller-based systems enables the solution
of signal control problems and the realisation of advanced control algorithms such as:

Ä closed loop control;

Ä in-line diagnostics;
Ä trajectory control.

A DSP of the company Analogue Device (family ADSP-2101) has been selected for its high
quality (fixed point architecture with extended reliability performance), extended operating
temperature range, noise immunity, fast multiply operations (24 bit instructions and little case
dimensions (PQFP-80 pins)
The NINTER project has been developed by realising a first prototype in double Eurocard format
to easy connect with an ICE (In Circuit Emulator) and test all electronic signals (this activity was
charged to the project) On the other side, we produced 10 more prototypes the first level test in
the final engineered size to be integrated into our first innovated product an electric-hydraulic
pump on the valve to properly evaluate the reliability in full operative conditions.
Mathlab and custom developed software were used for assessing the control algorithms.
Non linear functions were described and simulated using the object oriented G graphic language of
Labview and a special serial interface was developed allow the easy in circuit programming of the
system parameters
As far as fabrication technology is concerned it should be noticed that the developed electronic
hardware should be housed into a small metallic case on the proportional valve and therefore size
and extended operating temperature range becomes the key point; SMT assembly is in this case
the most natural choice for the required compactness of the device.

Fig.6 the NINTER prototypes

The tests have been undertaken in two different steps (not all charged to the AE):
♦ a first level test on 10 prototypes integrated on the valve, including severe reliability checks:

FUSE 14 ATOS demonstrator document

 ⇒ temperature cycles in temperature test chamber : 144 cycles each one from -25 to +95°C
in 2,5 hours.
⇒ mechanical stress and shock : 625.000 cycles at 1 Hz step command, vibrations 50Hz /4,5
g
EMC compliance to EN500081-2, EN500082-2, for the industrial environment. Tests have
been performed by C.R.F. (Centro Ricerche Fiat) Torino-IT.
During this phase also the software developed during the whole design process has been debugged
and carefully checked in real operating conditions.
♦ Extended reliability tests were carried out on a special low volume production of the final
product (typically 100 pcs).
It should remarked that only the first test step was charged to the project.

8. Expertise and experience in microelectronics of the company and the

staff allocated to the project
Company experience prior the AE
ATOS is a ISO9001 certified company for design and production of electric-hydraulic
components.
The company's experience of over 25 years in the field of industrial electronics associated tothe
control and regulation of electric-hydraulic valves with proportional control features. Design and
production of actual electronics exploited analogue circuits with pulse-width modulation (PWM)
power section, "traditional" PCB construction technologies and discrete components.
The experience relating to the use of surface mount technology existed within the company
before the start of the experiment though we have never used fine pitch QFPs before.
Before the AE the existing know-how enabled the design of simple analogue control circuits using
operational amplifiers for the realisation of proportional, derivative, integrative control actions.
Now we have two engineers with experience in digital controls and signal processing techniques.

Skills of the company’s personnel

ATOS personnel involved in the NINTER project:
• Dr. Ing. P. Fabbri, Director of the Electronic Division since 1990 and with the company
for 15 years , has a consolidated experience in designing all ATOS actual electronic
production , including management and resources planning. He was involved for all the
NINTER project as project leader, for work organisation, tests validation, reports and
dissemination documents.
• Dr. Ing. C Caccia , Hardware and Software designer, with large experience in digital
software and hardware design with the company from January to September 1998. He was
involved mainly in developing and debugging the software of the project..
• Mr. P. Faccin Research and development technician, with the company for 11 years,
involved in NINTER project for prototypes, hardware tests, software validation.
• Mr. C. Brusa Research and development technician, with the company for 10 years,
involved in NINTER project for prototypes, hardware tests reliability test, Quality control

9. Workplan and rationale

Risk perception
The innovation step is not very critical since the risk in using an established technology is low.
The real risk was the need for the company to learn a new design methodology (digital instead of
analogue, software oriented instead of hardware oriented). Moreover the major concern is on
software consistency and safety issues and this is why the final debug of the software has been
performed in real operating conditions and in a fully stressed environment.

FUSE 15 ATOS demonstrator document

 With the assistance of our TTN, a work-plan was actually prepared at the beginning of the
project, and for each phase a time-schedule was discussed, forecasted and planned.
The Management decided that the prototype in Eurocard format was the most suitable to
evaluate all functional performances, including the possibility to add new functions and circuitry
for the new product, but this assembly could not allow to assess the reliability of the NINTER in
real working conditions. Therefore a second prototype on a much smaller SMD board has been
made to fit into the existing housing on the target valve.
Thus the 10 prototypes, forecasted for the first level test, have been realised with a completely
different assembly in order to be as much similar as possible to the final product and were tested,
once integrated into the valve, in very hard test conditions to ensure the expected results (see
7.2).

• Workplan
The development of the application experiment was divided in 7 phases:

WP1. MANAGEMENT
Roles
• ATOS: Organises the project management, with associated resources, supervises the
different phases of the Experiment and deals with TTN for reporting, monitoring and
dissemination.

Tasks : • to organise manage the different phases of the experiment (Responsibility: ATOS)
• to coordinates the internal personnel with the external resources (Responsibility:
ATOS)
• to verify the respect of timing and the preparation of the deliverables
(Responsibility: ATOS)
Effort : 13 person days
Involved person and • ATOS 1 person
resources :
deliverables: presentation material and experiment report

WP2. TRAINING
Roles
• ATOS: receives on job training on the different topics of Digital control design and the
different issues of the DSP technology
• Digital Instruments: supplies the requested training to ATOS staff
• Baselectron and Sear: provide training on PCB and layout design

tasks : • to enable the designers to define the DSP hardware configuration for
application for industrial digital controls/actuators through on job training
(Responsibility: DIGITAL INSTRUMENTS)
effort : 20 person days
Involved person and resources : • ATOS 4 persons
• - Digital Instruments
• Baselectron and Sear for PCB with SMT technology

Note about TRAINING : It was intended that training would consist mainly of on-job activities, therefore after a short initial period
of traditional training, it was merged and integrated into the specification and design phases.

WP3.FUNCTIONAL SPECIFICATIONS
Roles:
• ATOS: defines the specifications for the whole system
• Digital Instruments: co-operates to identify the constraints with a software implementation

tasks : • Definition of HW/SW architecture (Responsibility: ATOS)

effort : 14 person days
involved person and • ATOS 3 persons - Digital Instruments supervision, training
resources:
deliverables: Detailed functional and electrical specifications

FUSE 16 ATOS demonstrator document

 WP4. HW/SW Design
Roles
• ATOS: carries out the design of the hardware and participates in the definition and
detailed development of the software
• Digital Instruments: defines the control algorithms carries out the complete design of the
software and the detailed development and debug of the firmware code

• Development of software specifications for system and subsystems;

tasks : • Mastering
• Definition of engineering; mechanical and connective aspects of the device;
• on job training
effort : 25 person days
involved person and • ATOS 3 persons
resources: • - Digital Instruments supervision, training
success : first level prototype executive drawings
deliverables: Detailed electrical and logic schematics, firmware code and test-bench

WP5. PROTOTYPE
Roles:
• Sear and Baselectron: master , process and assembly of the PCB EUROCARD prototype
• ATOS: provide s assistance and advice for the EMC and reliability issues

tasks : • PCB prototype in eurocard format (Responsibility: Baselectron)

• to mount a SMT circuit prototype (Responsibility: Sear)
effort : 14 person Days
involved person and Baselectron and Sear: PCB layout, routing and board prototype
resources:
success : prototype (correctly mounted) and correctly working
deliverables: final test report

WP6. REVISION OF THE PROJECT

Roles
• ATOS: carries on the testing of the prototype, identifies the flaws and suggests the
requested modifications
• Digital Instruments: implements the changes under ATOS supervision

tasks : • analysis of the problems emerging from the first prototype experimentation
effort : 12 person days
involved person and • ATOS 4 persons – Digital Instruments supervision, training
resources:
success : success of the whole experiment and plan for the new products design
deliverables: report on final performance

WP7 Dissemination
Roles
• ATOS: to prepare all the material for the dissemination (including the final report of the
Application Experiment) and to organize dedicated dissemination events

tasks : • to prepare documentation and material for dissemination (demonstrator and Flier)
(Responsibility: ATOS)
• to disseminate the results of AE through documentation, presentations, attendance
at exhibitions and seminars. (Responsibility: ATOS)
effort : 15person days
involved person and • ATOS 1 person
resources :
success : dissemination material completed and available
deliverables: presentation dissemination material (demo and flier)

Thanks to the very good design modularity and experience of the subcontractor, the project was
conducted without changing the original plan as workpackages/tasks. The only changes with respect the
original proposal are due at the lower number of working day from the FU since an experienced SW
engineer was engaged with higher salary than the one expected in the original plan. This has been
necessary for the followings reasons:
FUSE 17 ATOS demonstrator document
• Complex SW algorithms
• Know-how in house for future developments

Application experiment timing diagram

Month 1 Month 2 Month 3 Month 4 Month 5 Month 6

WP1
T11
WP2
.
T21
T22
WP3
T31
WP4
T41
T42
T43
T44
T45
WP5
T51
2.
T52
T53
WP6
T61
T62
WP7
T71
T72
EFFORT & COSTS

Task Company’s Company’s Subcontractors’

effort effort costs (kEUR)
(person/days) (person/days)
Planned Actual
Management 24 13
Training 18 20 2,5
Specifications 17 14 2
Design 32 25 6
Prototype 12 14 4,3
Validation and test 15 12 0,5
Dissemination 20 15
Total 138 113 15,3

NINTER Workplan (FU person/days)

35
30
25
20
15
10
5
0
Management

Dissemination
Design

Validation and
Training

Prototype
Specifications

Planned
18
test

FUSE ATOS demonstrator document

Actual
Technical problems encountered during the development.

The engineering work has not yet come to a definite stop, since we are confident to be able to further
reduce the number of components and the size of the boards
• A difficulty was found in implementing digital filters on DSP for the derivative and integral closed
loop PID control of valve position similar to the analogue one: the solution was found with the
strong cooperation with Digital Instruments using MatLab for the description and simulation on PC
and them implement and testing the code on DSP-
• The introduction of non linear functions in the main control program was not as easy as expected. A
typical non linear curve had to be simulated using LabView object program and an associated look-up
table was created. Under of Digital Instrument assistance a special software was developed to create a
non linear curve easy to modify that was then translated into DSP code.
• Due to the large amount of parameters requested to define the control of the valve, the use of In-
circuit-emulator was considered too slow, therefore we decided, with our consultant, to realise a
dedicated serial interface able to communicate on the fly, in real time the parameter changes. The
device perform also a easy to use software running and PC able to upgrade programs and change
parameters.

10. Subcontractor information

The design assistance subcontractor for this AE had to demonstrate an established experience in
developing embedded circuits for industrial use; particularly it had to prove an expertise in dealing with
the design with high reliability DSPs
As a matter of fact Digital Instruments had such experience and since we have had a long time fruitful
relationship with this Design-house, also comparing other received offers, we decided to stay with it
• Training and Design Assistance:
Design assistance and on job training has been offered by Digital Instruments. The company has been
represented by Ing. Genova, who assisted ATOS through all the phases of the project. At the beginning a
training was offered from the subcontractor mainly on Digital Signal Processing theory and DSP design
and development tool. This was followed by a continuous on job assistance during the whole project
development.
- Digital Instruments has confirmed the established role as ATOS main consultant and the deep
involvement in the NINTER project has extended its knowledge of the ATOS proportional
valves control needs.
- No special IPR agreement was considered necessary beyond the permanent non disclosure
agreement already signed by Digital Instruments

• Prototype:
Baselectron has been operating in electronics for many years, in particular in design, production and
testing of electronic boards. As PCB manufacturer it is especially reputed for fast prototyping and has a
very high quality product suitable for any professional need.
Sear is a society operating in electronics for many years, in particular in design and mastering PCB.

At the end of AE ATOS developed its own internal specification and criteria for accepting externally
mounted electronic boards, so as to comply with the company total quality assurance policy..

11. Barriers perceived by the company in the first use of the AE technology
ATOS decided to take this important innovation step with the idea to keep his leadership position.
Some fears and doubt somehow delayed the practical realisation.

FUSE 19 ATOS demonstrator document

1. Why to move from a well established implementation technology which also showed excellent
reliability performance?
ATOS infact is widely aknowleged for the performance and quality of its products, also thanks to its
internal “industrial electronics” division established know how in analogue design techniques of signal
and power electronics. Viceversa in the sector of the digitally programmed devices, prior to the AE,
ATOS did not have any proven technical background and this might cause mistrust from its old
customers.
2. Moreover we were not loosing money but instead making good profits with our current analogue
solution and our competitors technology level was the same as ours. .

On the other side the advantages of the innovation were really appealing:
• to be the first to be able to house the whole electronics into the valve
• to get all control flexibility and complex performance that is allowed by DSP techniques
• to acquire brand new (to the company) know-how that could be used further on for new valves
architecture and also for a better process control of the final product.

From the point of view of the knowledge ATOS had the problem to make a more precise
estimate the impact of the microelectronic technology on the product and to identify what the were the
advantages versus the currently marketed valve.
ATOS was also worried of the amount of knowledge increase that should have been acquired by the
company technical staff to jump from our state of the analogue design expertise to the capability of
managing a complete digital control system based on DSP technology. Requested time and effort were
not minor concerns.

The perceived cultural barriers due to the culture of our market sector and in particular to the
distrust in inserting electronics into mechanical components, lead us to foresee an initial slowness to
absorb radical innovations in the way electric-hydraulic systems could be operated.
Moreover we considered what were the main obstacles inside the company that could be met in the
introduction of new technology like, for instance, the attitude of our personnel (not only the technical
one, but for example the quality assurance, the production staff and the purchase and storage offices)
that would have to develop new managerial and operational skills to deal with the new technology.

Finally, the financial barriers that were faced were due to the fear of investing in the
development of a new design, testing and production capability without the clear perception of the
market response for the new high-tech valve.

12. Steps taken to overcome the barriers and arrive at an improved product
Participation to a Fuse public information day and the further discussion with the TTN have persuaded
us that the time has come to move on.
We understood that DSP could provide a much more flexible and high performance solution for the
electronics on the valve and that a good training and knowledge transfer plan could prevent us from
facing undesired surprises
During the feasibility study and the preparation of the FUSE proposal, ATOS has analysed difficulties
and the obstacles to the realisation of the project. These have also been addressed through the help of
external consultants and proper market analysis.
In particular, the barriers of knowledge have been overcome with the help of the TTN who has a lot of
experience in microelectronic technology management.
After several discussions where we examined with the TTN:
• the functional aspects of the product
• the engineering of the device
• the productive aspects
• the reliability requests
we were persuaded that the intervention of a suitable subcontractor was needed who had particular

FUSE 20 ATOS demonstrator document

expertise with digital integrated controllers (particularly DSPs and microcontrollers).
Particular care was paid to keep the reliability performance, to assure easy maintenance and
reconfigurability for future products development.
• The cultural barriers were broken:
Ø with the analysis of market and competition, in accordance to the product and the trend of
the automation sector
Ø with a forecast of the possible sales increase due to the use of digital technology
Ø by verifying the productivity (lower manufacturing and testing cost in the medium-long
term) for the new product
Ø with the analysis of cost and operational income of the project
Ø with a glance to future developments with the acquired technology connected to further
enhancement of the same product or development of brad new one
• Concerning the technological barriers we identified the following solutions:
Ø training and education of the key personnel
Ø select consultants with large experience in using the digital approach in the design of
complex control systems
• The economic risk can be minimised:
Ø with a detailed plan of the foreseen investments during the development of the project and
forecast of the costs of transfer of new technology into the phase of production
Ø with a careful selection of the electronic digitally controlled device (DSP) which should take
in due account performance (speed, bandwidth, temperature range and drift, noise) as well as
reliability figures like the mean time between failures (mtbf)
Ø These constraints were carefully considered during the analysis of the DSPs available on the
market under of the assistance of our subcontractors which explained also us the main
differences between the architectures of the microcontroller and the DSP.

13. Knowledge and experience acquired

The development of the AE has allowed the acquisition of a specific know-how in the field of the
microelectronic technologies, and particularly in the use of the DSPs.
With the help of the TTN we have developed a method for managing multi-partners complex
projects (as it is necessary when using advanced electronic technologies with external
subcontractors) in such a way that the different consultants are given the proper role for each
phase of the activity.
That involved a lot of effort from the project leader to assure that all interactions and
dependencies were assured along the development and, at same time, know-how transfer could be
guaranteed from the selected subcontractor to ATOS engineers.
A particularly delicate activity, was the definition of the specifications of the new control system
and in this phase the project management capability together with the subcontractor assistance,
showed its real advantages allowing to freeze the specifications before the starting of the design
process. This was something which had never achieved before with other projects and consultants.
The most important capability acquired with the project is the new DSP technology which is
considered strategic for our short and medium time product.
We can now plan new products with DSPs on board, define their specification, plan with accuracy
development time and requested resources, design the hardware for the digital boards and develop
and test the firmware. This was obtained thanks to the assistance of our subcontractor but was also
due for an important part to the involvement of ATOS technical staff in the design and to its
continuous auto-training attitude.

As a conclusion we can state that our main objectives have been reached and overcome as it is
demonstrated from the list of the new perspectives opened by the AE results:
1. development of new dynamic mathematical models of the proportional valves and the hydraulic
actuators, aimed to the project of the controller.
2. development of a software environment for the complete simulation of electro hydraulic systems,
based on the modern systems control design.
3. new approach to the project managing and planning of elettrohydraulic components and their
controls, completely computer assisted.
4. use of new technologies developed with the AE (digital electronics with DSP and microcontrollers,
field-bus communication, engineering and miniaturisation of the new integrated digital electronic
module) for the project of digital circuit switch exploit new methods for fabrication, production and
FUSE 21 ATOS demonstrator document
 quality assurance.
5. realisation of a digital electronic module, for the development of a new family of elettrohydraulic
components, able to develop either the valve regulation function or the hydraulic axis control.
6. realisation of elettrohydraulic systems of increasing complexity (servo actuators with digital
electronics, complete elettrohydraulic systems) managing complex functions, made with digital
technology and directly connectable to field-bus.

14. Lessons learned

During the development of the AE ATOS had to withstand different difficulties to come to the
realisation of a product entirely based on the new type of technology.
Our experience allows us to give some advice to companies that want to undertake a similar innovation
step
♦ The planning of the activity should be detailed enough to foresee the interaction between the
project participants (FU and subcontractors) and to monitor the dependencies in the different
phases of the work-plan.
♦ During the phase of the work-plan where the specifications are defined it is important to think not
only of the product under development, but also of the similar products that might be developed
using the same technology in such a way to be sure to introduce the maximum of flexibility and
modularity into the prototype implementation. (Never think to a single customer desires)
♦ If reliability is an important issue, as in our case, it is important to plan in advance the extra tests
(the one that can be necessary to stress the new technology and were not carried out with the old
one)
♦ If your market is process automation and control, like ours, think in advance of the communication
facilities which could cope with the state of the art technologies in this field. In this AE we had to
introduce a new micro onboard to implement the CAN field-bus communication protocol
♦ Use since the beginning an in circuit programmable device because this eases the programming and
testing procedures allowing to change operational parameters in a simple and fast way
♦ Spread the technical information as much as you can throughout the company so that your time to
market is not affected by technical staff turnover

15. Resulting product, its industrialisation and internal replication

The AE has fully convinced ATOS of the technical and commercial validity of the new microelectronic
technology. As a proof, we started a first engineering process even the during the development of the
prototype to exploit, beyond the improved functional characteristics, also the real possibility of quickly
marketing the product and the opportunity to apply such innovation to the whole range of ATOS
products.
This is the reason why ATOS decided to begin introducing the newly developed technology into an
existing product: (i.e. an hydraulic pump) by equipping it with a proportional valve version with digital
control electronic (DSP) on-board.
• In first quarter of 1999 the hydraulic pump PVPC-SLEDR-*/PQ. is available as first hydraulic
product equipped with new integrated electronic regulator: This choice is motivated since in this
particular pump the price/performance relationship seems very advantageous to the customer.
The commercial application of the new digital electronics on the proportional valve still requires a
further development to optimise the production yield so to bring the costs of production to be
comparable to the actual clean cost of the replaced analogue valve.

We can summarise the further steps to bring the developed product to the market.

Industrialisation roadmap (1998 -

2000)
1. Further engineering of the valve to reduce the available space for the electronic card and
improve device reliability (further reduction of the number of discrete components)
2. A closed loop monitor of the solenoid current should be added before starting the mass
production
3. Testing procedures for the full range of new generation digital proportional valves and servo-
actuators should be upgraded.
4. Optimisation of production costs (automation of assembly and test procedures)
5. Full exploitation of field-bus communication opportunities by providing further user
FUSE 22 ATOS demonstrator document
 configurability (for example the standard Can-Bus CANOPEN protocol)
6. Certification of the software for industrial automation
7. Development of standard software for the users (ISO1131)
Obviously in this phase the main problem is to keep the acquired know-how inside the company

The continuity of ATOS internal know-how is guaranteed:

Ø From the permanent relationship with our external consultants,
Ø from the hiring of n°2 software engineers with experience in industrial automation,
Ø further training and update on the new technology
ATOS foresees the beginning of the production of valves with digital control to start in 2Q 2000 and
the application of DSP technology to other electric-hydraulic devices beginning from 2001.
For the industrialisation of the new product, it is planned that production lines are revised and new
equipment is installed for tests and trials, to an overall cost within the 10 and 20 kEUR per year
Costs for launching the new product (marketing and advertising expenses) can be estimated at about 5 to
10 kEUR per year.
An effort of 60 person/days per year was spent till now with about 10 kEUR subcontracting fee

• Budget of investments in new digital integrated electronics after NINTER project

Project NINTER Industrialisation Total

(first half 1998) roadmap Investme
(1998 -2000) nt
Duration 6 months 30 months 3 years
Estimated cost KEUR 42 (actual) 40 per year 142

16. Economic impact and improvement in competitive position

Cost evaluation
The Atos analysis of the new product cost was addressing mainly on the electronic regulator part of the
proportional valve, because the mechanical cost remains almost the same.
At the preliminary evaluation of the new electronic cost cannot be compared with the current analogue
product because digital integrated electronics is very different thanks to the various additional features as
described in section 6.2 .
The main NINTER additional functionalities are :
I. - is fully compatible with automation distributed control systems
II. - is programmable
III.- field bus compatible
Note that the new product is much more complex and involves the global valve costs.
The NINTER types are here catalogued in base of their function complexity (Number of solenoids,
Number of closed loops, in relation to the type of the valve to be controlled) and the diagram is
separated between analogue (old product) and digital (new product) integrated electronic control .
Costs are “normalised” on the cost of actual analogue control type 1 (=100)

FUSE 23 ATOS demonstrator document

 200 187
analogic
180 169
digital
160
140 120 124 123
111
Driver control types
120 110
100
100 TYPE 4 = 3 position loop + 1 solenoid (AE)
80
(*) FUTURE INTERNAL REPLICATION
60
TYPE 1 = 1 position loop + 1 solenoid
40 TYPE 2 = 1 position loop + 2 solenoid
20 TYPE 3 = 2 position loop + 1 solenoid
0
TYPE 4 (AE) TYPE 1 (*) TYPE 2 (*) TYPE 3 (*)

In fact the new electronic control for the hydraulic pump(TYPE 4) realised in digital version has a clean
cost 10% lower than analogue technology, with much superior performances.
After the industrialisation phase, for the final product we foresee a 20% decrease of production cost
(that means the cost of the electronic driver Type 4 would be 149 instead of 169). The future internal
replication will include the other customisations of the valve. As it is clearly shown we preferred to start
the improvements with the most complex control, where the advantages of the introduction of the new
technology would be more evident and the foreseen cost differential was more significant even before
the industrialisation phase.

Competitive improvements
It is expected that, in the next years, the trend for the valve sales in the world will be a positive one.
The market is in very rapid evolution: customers became more and more sophisticated and demand
extra features which cannot be delivered by the current analogue discrete technology.
The introduction of these new innovative products will enable market share increase both on the home
and on the international markets with reference of most immediate competitors.
The objective market share of ATOS for the proportional valves in next 5 years is the following:
The table below shows the sales expectations for the old analogue valve and the new one embedding the
DSP:

Home Market International Market

ATOS
15%
ATOS
40%

Others
60%

Others
85%

DATA CONCERNING THE SALES OF NEW DIGITAL ELECTRONIC

(only the regulators integrated on the valve)
DATA CONCERNING THE SALES OF NEW DIGITAL ELECTRONIC
(only the regulators integrated on the valve)
YEAR 1999 2000 2001 2002 2003

FUSE 24 ATOS demonstrator document

Total quantities grow Both analogue and digital % 100% 129% 167% 200% 243%

Analogue integrated reg if the investment would 100% 110% 95% 76% 68%
not have been done. [%]

New product quantities[%] of total 0% 17% 26% 40% 100%

Profit margin if the investment would not have 296.963 325.368 281.469 224.917 202.451
been done (Analogue integrated reg only)

Total Profit margins per q.ty/year (new product + 296.963 382.178 497.090 593.925 723.040
old analogue ) EUR

Difference Cumulated Profit EUR (Total Profit - 0 56.810 272.431 641.439 1.162.028
Profit without investment)

Cumulated Investment in digital electronic EUR 42.000 82.000 122.000 162.000 202.000

ROI [%] 575%

For each integrated electronic is evaluated: Profit margin = sale price – estimated production cost
Return of investment is : ROI =(Cumulated Profit-Investment)/ Investment

As you can see the improvement to the market situation of ATOS comes mainly in the market
segments of high-performance proportional valves

From an analysis of the data included in the table it is clear that development costs for the application
experiment which is the subject of this project, together with the related industrialisation costs and the
launching of the product on the market, will be fully paid back already in the first financial year (1999).
The economic break-even period (8 months) for the considered investment, therefore, appears to be
extremely reasonable. The introduction of the new DSP technology will be as profitable as more than 5
times the sustained investment in five years.
The yearly equivalent interest in 5 years investment period of is 100%-(1.162/202)1/5 = 42%

This really means a very good margin for our company and definitely proves that innovation
advantages paid far beyond the money and the effort spent.

17. Target audience for dissemination throughout Europe

The best practice that can be extracted from this AE regard mainly:
Ø The very accurate software architecture and partitioning
Ø The extended usage of DSP development tools (both hardware and software)
Ø The In Circuit programmability of all important system parameters
Ø The thorough rules followed in developing a very small PCB fitted in the valve itself
Ø The introduction of a field bus interface according to the most recent trends in the mechanic
machinery and process automation.
Ø The thorough project planning and the constant monitor of discrepancies with respect to the
forecasted milestones and deadlines

From the company profile point of view, all the companies matching with the ATOS profile should be
interested in acquiring experience from this AE.

FUSE 25 ATOS demonstrator document

 Starting technology: PCB and analogue components
Applications: Valve and tips
Development Traditional, before the AE
Methodology:
Barriers: To change mind from analogue to fully digital control
systems
Company Size: Medium companies with less than 5 employees involved in
electronic
Company turnover: About 40MEUR

Then the target audience for dissemination for this AE is certainly the one of the
machinery and electronic companies with poor experience with electronics which may
improve their own machinery and system by using digital control techniques:

• Machine tools 2940

• Machine for metallurgy 2951
• Machine for food and beverage 2953
• Machine for textile 2954
• Pumps and compressors 2912
• Valve and tips 2911

• Industrial process control 3340

• Precision measurement 3320
instruments

The project demonstrates the advantages in substituting DSP technology to old analogue
technology in industrial control. The following advantages, due to the digital control are the
reasons for a wide target audience:
• Reusability of the same hardware for different applications,
• Ease of modification of the project during development,
• High flexibility in system reprogramming,
• High quality design,
• Fit for complex algorithm performing,
• Hardware options reduction,

But the companies will also learn:

• How to choose the subcontractors
• How to organise the workplan
• How to minimise the economic and the development risks
• How to choose the right technology

The company has already being involved in the TTN dissemination activities and it can be
highlighted its participation at the BIAS fair – Milan November ’98- during the BIAS
Conference the Company presented his AE. The audience was composed by a wide number
of Italian enterprises.

ATOS planned commercial information about New Electronic Integrated Regulators and the related
electric-hydraulic components with realted papers presented on the specialised press and many other
media.

Hereafter a description of few documents printed:

planne Description Document Languages

d
1998 Commercial info on new components F98-1/E I,E,D,F
Feb. 98 Technical article on new hydraulic pumps Magazine “PROGETTARE” I
FUSE 26 ATOS demonstrator document
Mar. 98 Technical article on new hydraulic pumps Magazine “COMPONENTI INDUSTRIALI”Mar98 I
Nov. 98 Presentation at BIAS fair Milan- It FUSE style presentation I
Dec. 98 Technical article on new hydraulic pumps Magazine “HYDRAULIC & PNEUMATIC”-USA E
June 99 Technical article on new hydraulic pumps Magazine “O+P” D
July 99 Technical article on new hydraulic pumps Magazine “BLECH” D
June 99 Technical article on new hydraulic pumps Magazine “MACHINENMARKT” D
June 99 Technical article on new hydraulic pumps Magazine “KEM” D
Nov. 99 Technical article on new hydraulic pumps Magazine “FLUID” D
Dec. 99 Technical article on new hydraulic pumps Magazine “DER KONSTRUKTEUR” D
July 99 Technical article on new hydraulic pumps Magazine “L’USINE NOUVELLE” F

FUSE 27 ATOS demonstrator document