Pump Control/System Automation - KSB

K S B K n o w - h o w , V o l u m e 4

Pump Control / System Automation

Pump Control / System Automation

August 2006 Edition

Subject to technical changes.

© Copyright byKSB Aktiengesellschaft

Publisher:KSB Aktiengesellschaft

Segment:Building ServicesD-91253 Pegnitz

All rights to disseminate, including by film, radio, tele-vision, video, photo-mechanical reproduction, audio ordata carriers of any type, printing of extracts, or storageand retrieval in data processing systems only with theauthorization of the producer.

Technical Data Technical Data Technical Data Technical Data

Riotronic S

� Q up to 3.5 m3/h,1.0 l/s

� H up to 6 m

� T + 20 °C to + 110 °C

Riotronic ECO

� Q up to 2.5 m3/h,0.7 l/s

� H up to 5 m

� T + 15 °C to + 110 °C

Riotec

� Q up to 60 m3/h, 17 l/s

� H up to 10 m

� T + 20 °C to + 110 °C

Riotec Z

� Q up to 90 m3/h, 25 l/s

� H up to 10 m

� T + 20 °C to + 110 °C

Etaline PumpDrive

� Q up to 788 m3/h, 219 l/s

� H up to 100 m

� T – 10 °C to + 110 °C

� pd up to 16 bar

Etaline Z PumpDrive

� Q up to 479 m3/h, 133 l/s

� H up to 76 m

� T – 10 °C to + 110 °C

� pd up to 16 bar

hyatronic mb

� 1-8 pumps, modular layout

� Up to two frequencyinverters with full permutation

� For motor powers upto 200 kW (higherpowers upon request)

� Mains voltage3~ 400 V, 50 Hz

� Ambient temperature0 to + 40 °C max

KSB Control Concepts

Pumps with integrated control Pump control systems

Glandless pumps Glanded pumps

Riotronic S /Riotronic ECO

Riotec / Riotec Z

Technical Data

Rio-Eco

� Q up to 60 m3/h, 16,7 l/s

� H up to 13 m

� T – 10 °C to + 110 °C

Rio-Eco Z

� Q up to 108 m3/h, 30 l/s

� H up to 13 m

� T – 10 °C to + 110 °C

Rio-Eco /Rio-Eco Z

Etaline PumpDrive /Etaline Z PumpDrive

hyatronic mb

Contents Page

1 Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.1 Hydraulic principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.1.1 Pump principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Positive-displacement pumps. . . . . . . . . . . . . . . . . . . . . . 2

Centrifugal pumps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.1.2 Flow rate adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Flow rate adjustment by throttling . . . . . . . . . . . . . . . . . 3

Flow rate adjustment using a bypass . . . . . . . . . . . . . . . . 4

Flow rate adjustment by parallel operation of pumps . . . 5

Flow rate adjustment by speed adjustment . . . . . . . . . . . 6

Flow rate adjustment by a combination of parallel operation and variable speed operation. . . . . . . . . . . . . . 7

1.1.3 Characteristic curve conversion at variable pump speed . 8

Calculation of the piping parabola . . . . . . . . . . . . . . . . . 9

Calculation of the controlled-operation parabola . . . . . . 9

Pump selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Calculation of the affinity parabola . . . . . . . . . . . . . . . 10

Determination of pump characteristic curves . . . . . . . . 10

Determination of auxiliary points and intermediate characteristic curves . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Addition of the characteristic curves . . . . . . . . . . . . . . . 12

Determination of pump input power. . . . . . . . . . . . . . . 13

Determination of set value . . . . . . . . . . . . . . . . . . . . . . 14

1.1.4 Economy calculation for infinitely variable speed adjustment systems with frequency inverter . . . . . . . . . 15

Influences due to the design of the system. . . . . . . . . . . 15

Influences as a result of the loading of the systemover time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Influences of the pump . . . . . . . . . . . . . . . . . . . . . . . . . 15

Pump power consumption from the electric grid. . . . . . 16

A comparison of three systems with and without

speed control. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Economy calculation. . . . . . . . . . . . . . . . . . . . . . . . . . . 20

1

1.2 Control principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.2.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.2.2 Further control terminology . . . . . . . . . . . . . . . . . . . . . 21

1.2.3 Control terminology based upon the example of pump control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.2.4 Controlled quantities for closed hydraulic circuits . . . . 22

Differential pressure dependent control. . . . . . . . . . . . . 22

Differential temperature dependent control. . . . . . . . . . 24

Return temperature dependent control . . . . . . . . . . . . . 25

Supply temperature dependent control . . . . . . . . . . . . . 26

1.2.5 Controlled quantities for open circuits . . . . . . . . . . . . . 27

Pressure dependent control . . . . . . . . . . . . . . . . . . . . . . 27

Level dependent control . . . . . . . . . . . . . . . . . . . . . . . . 28

Flow rate dependent control . . . . . . . . . . . . . . . . . . . . . 29

1.2.6 Compensation of additional interference factors . . . . . . 30

Compensation by the selection of the correct measurement location . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Compensation by means of additional measured variable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

1.3 Principles of integral drive. . . . . . . . . . . . . . . . . . . . . . . 33

1.3.1 “Intelligent” integrated drives for pumps . . . . . . . . . . . 33

1.3.2 Advantages of integration . . . . . . . . . . . . . . . . . . . . . . . 33

1.3.3 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

1.3.4 Pump-specific functions . . . . . . . . . . . . . . . . . . . . . . . . 34

1.3.5 Economy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

1.4 Principles of communication technology . . . . . . . . . . . . 36

2 System automation technology and planning notes . . . . 38

2.1 General electrical notes . . . . . . . . . . . . . . . . . . . . . . . . . 38

Power supply system types . . . . . . . . . . . . . . . . . . . . . . 39

Earth leakage circuit breakers . . . . . . . . . . . . . . . . . . . . 40

Power system dependent protective measures . . . . . . . . 40

Ambient temperature . . . . . . . . . . . . . . . . . . . . . . . . . . 40

Starting method for squirrel-cage motors . . . . . . . . . . . 40

2

Contents

2.2 Control functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Controlled quantity. . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Set value / Set value switching . . . . . . . . . . . . . . . . . . . . 42

Optimization of the controlled-operation curve . . . . . . 42

Monitoring the pumps and the hydraulic system in theautomatic operating mode . . . . . . . . . . . . . . . . . . . . . . 43

Measuring equipment . . . . . . . . . . . . . . . . . . . . . . . . . . 44

Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Commissioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

3 Project planning example . . . . . . . . . . . . . . . . . . . . . . . 47

3.1 System description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

3.2 Calculation of the piping characteristic curve . . . . . . . . 47

3.3 Further steps in accordance with the “Project Planning Sequence Plan” . . . . . . . . . . . . . . . . . 48

4 Reasons for pump automation and control . . . . . . . . . . 53

4.1 Operational reliability. . . . . . . . . . . . . . . . . . . . . . . . . . 53

4.2 Improving operating behaviour. . . . . . . . . . . . . . . . . . . 53

4.3 Increasing product quality. . . . . . . . . . . . . . . . . . . . . . . 54

4.4 Reducing operating cost / life-cycle cost . . . . . . . . . . . . 54

4.5 Improving system information . . . . . . . . . . . . . . . . . . . 54

5 An overview of automation concepts . . . . . . . . . . . . . . 55

5.1 Parallel connection of identical pumps with one frequency inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

5.2 Parallel connection of identical pumps with twofrequency inverters. . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

5.3 Parallel connection on non-identical pumps . . . . . . . . . 58

5.4 Further electric configuration concepts from theKSB product range . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Overview of project planning sequence . . . . . . . . . . . . . 60

3

Contents

1Principles

1.1Hydraulic Principles

1.1.1Pump Principles

The task of a pump is to gener-ate pressure and flow in a liquid.Various pump configurationshave been developed to achievethis task. The most importantdesigns are the positive-displace-ment pump and the centrifugalpump.

Positive-displacement pumps

These pumps are primarily usedin cases where low flow ratesare required in combinationwith a large pump head. Theirworking principle is based uponthe periodic change in volume ofcavities that are separated fromthe suction and discharge pipeby separating elements.

Typical examples are:

• Reciprocating piston pump

• Positive-displacement pump

• Diaphragm pump

• Gear pump

• Screw pump

• Vane pump

• Hose pump, etc.

Their main common featuresare:

The flow rate varies with the ro-tational or stroke speed. Thehead, on the other hand, is inde-pendent of this. Due to this be-haviour, positive-displacement

pumps must be protectedagainst impermissibly high pres-sures. A change in the flow rateis only possible as a result of achange in the rotational orstroke speed, or due to add-itional devices (bypass). Thepump characteristic curve showsthe relationship between flowrate and head (pump pressure)at a constant speed.

If the speed changes there is aproportional change in the flowrate.

4

1 Principles

H [%]

Q [%]Pump flow rate

Pum

p h

ead

120

100

80

60

40

20

0

nmin

0 20 40 60 80 100 120

nmax

Fig. 1 Typical characteristic curves of a positive-displacementpump at various speeds

H [%]

Q [%]Pump flow rate

Pum

p h

ead

120

100

80

60

40

20

00 20 40 60 80 100 120

Qmax

Pump duty limit m

ax

Pum

p d

uty

lim

it m

in

Fig. 2 Typical working range of a centrifugal pump with pumpcharacteristic curves for different pump speeds

5

1Principles

Centrifugal pumps:

Centrifugal pumps are used formost technical applications. Thisis due in particular to the fol-lowing properties:

• Robust construction

• Simple design

• Cost-effective manufacture

• “Good natured” operating behaviour

• Good adjustability

The working principle of thecentrifugal pump is based uponenergy transfer by flow diver-sion as well as an additionalcentrifugal force effect in radialimpellers. In contrast to posi-tive-displacement pumps themaximum pump pressure is limit-ed by the operating principle.Special devices to protect againstoverpressure are seldom neces-sary. Based upon the assumptionthat the drive speed is constant,different flow rates can simplybe achieved by means of a throt-tling valve. The permissibleworking range is shown in thepump characteristic curve.

1.1.2Flow Rate Adjustment

Flow Rate Adjustment byThrottling

The purpose of increasing thesystem resistances – the fitting ofa restriction (throttling) – is tomake the resulting system char-acteristic curve steeper. At a con-stant pump speed the operatingpoint on the pump characteristiccurve is moved to a lower flowrate. The pump thus generates ahigher pressure (head) than isnecessary for the system. Theexcess head thus created isbroken down in the restrictingfitting to create a pressure drop.

6

1 Principles

H [%]

P [%]W

Q [%]

Q [%]

Pump characteristic curve

System characteristic curve(part load)

Throttled operation

Excess pump head

Required pump head

Power saving

System characteristiccurve (full load)

160

140

120

100

80

60

40

20

0

120

100

80

60

40

20

0

20

20

40

40

60

60

80

80

100

100

120

B1

PW1

PW2

B2

120

Fig. 3 Throttling configuration

Fig. 4 Pump and power characteristic curves

+ Lower control cost

+ Advantageous at mainly fullload operation

+ Suitable for applications withshort operating periods

+ Well suited for flat pump char-acteristic curves

– Pump pressure too high, par-ticularly where the pump char-acteristic curve is steep

– Poor pump efficiency in partload operation

– Low power saving in part loadoperation

– Unfavourable control behav-iour when the excess head ishigh

– Throttle valve necessary

– Mechanical load on the throt-tle valve

– Danger of flow noise at highlevels of throttling (e.g. inthermostat valves).

Evaluation

Flow Rate Adjustment Using a Bypass

The bypass line is arranged inparallel to the pump. The pumpflow is thus divided into the use-ful flow, which flows into thesystem, and the bypass flow,which is directly or indirectly re-turned to the inlet pressure sideof the pump (see Fig. 5). Chang-ing the bypass flow rate or thebypass line characteristic curveby means of a control valve thusallows the useful flow to be var-ied. The pump itself runs at al-most the same operating point,i.e. at the system’s design point,in full load operation.

7

1

H [%]

P [%]Q [%]

Q [%]

Pump characteristic curve System characeteristic curve(part load)

Excesspump head

Required pump head

Useful flow rate Bypass flow rate

System characteristic curve(full load)

160

140

120

100

80

60

40

20

0

120

100

80

60

40

20

0

20

20

40

40

60

60

80

80

100

100

120

B1B2

120

Constant shaft power

(No saving)

PW1

Fig. 6 Pump and power characteristic curves

+ No increase in head even inpart load operation

+ In contrast to throttling thepump pressure remains con-stant when the flow is adjusted

+ Suitable in situations wherelow head is combined withhigh flow rates

+ Well suited if full load oper-ation prevails

– Increased construction costs(bypass circuit)

– No reduction of the powerconsumption in part load op-eration

– In part load operation there isstill excess head

– This method of flow rate ad-justment is uneconomical interms of energy used

Fig. 5 Bypass configuration

Evaluation

Principles

Flow Rate Adjustment by Par-allel Operation of Pumps

If pumps are connected in paral-lel as shown in Fig. 7 their par-tial flow rates are summed.

For the construction of the par-allel operation characteristiccurves, the partial flow rates ofall participating pumps aresummed at several differentpressure levels (between zeroand minimum head). The paral-lel characteristic curve is foundby summing the flow rates at thesame head. In practice it shouldbe taken into account that as theflow rate increases the systemresistances also rise and thus theactual operating point in paralleloperation also lies at this higherpressure level. As a result, theincrease in the flow rate is lessthan originally expected.

8

1 Principles

H [%]

P [%]

Q [%]

Q [%]

Q [%]

Pump characteristic curve System characteristic curve 2(part load)

Excesspump head

Required pump head

Switch-on level

Switch-off level

Power saving

System characteristic curve 1(full load)

160

140

120

80

60

40

20

0

100

80

60

40

20

0

20

20

20

40

40

40

60

60

60

80

80

80

100

100

100

120

BN

B2HA

HE

120

120

Fig. 8 Pump, power and efficiency characteristic curves for one,two and three pumps in parallel operation

Fig. 7 Parallel pump connection

+ Very well suited to flat systemcharacteristic curves with ahigh static head component

+ Good adaptation to part loads

+ High system efficiency

+ Low control cost for pressure-dependent pump operation

+ High operating reliability dueto several pumps (redundancy)

– Increased construction cost(piping, valves, pumps, spacerequirement)

– High switching frequency inunfavourable system designs

– In the event of flat pump/sys-tem characteristic curves pumpoperation is flow-dependent

– Problematic in the event ofhigh inlet pressure fluctuations

Evaluation

Flow Rate Adjustment bySpeed Adjustment

Relationships in thecontinuous speed adjustmentof centrifugal pumps

Unlike the procedures for flowrate adjustment mentioned in thepreceding sections, continuousspeed adjustment permits a con-tinuous modification of the pumpoutput to the system requirementsby changing the pump characteris-tic curve. If the flow rate increaseslinearly, the system resistance(piping characteristic curve) in-creases quadratically. The cen-trifugal pump behaves in a similarmanner. In the event of linearly in-creasing flow rate and linearly in-creasing speed the resulting headalso increases quadratically. As aresult of these relationships evenrelatively small speed changescover a wide working range. Ac-cording to the similarity law thefollowing relationships apply tocentrifugal pumps (see Fig. 9):

Real systems

In practice, it is common to findsystems in which the consump-tion behaviour requires variablethrottling or mixing processes.

The task of continuous pumpspeed adjustment is to cover thecurrent system demand at thelowest possible speed (= cost).

Q2 = Q1 · ( )n2

n1

H2 = H1 · ( )2n2

n1

Flow rate

Discharge head

Power input P2 = P1 · ( )3n2

n1

9

1

H [%]

P [%]

Q [%]

Q [%]

Required head

Powersaving

n = 100 %

n = 100 %

90 %

90 %

80 %

80 %

70 %

70 %

60 %

60 %

50 %

50 %

System characteristiccurve(full load)

160

140

120

100

80

60

40

20

0

100

80

60

40

20

0

20

20

40

40

60

60

80

80

100

100

120

B1

P1

B2

P2

120

Fig. 9

Principles

H [%]

Q [%]

n = 100 %

90 %

80 %70 %

60 %


System characteristic curve(part load)

160

140

120

100

80

60

40

20

020 40 60 80 100 120

B1

Fig. 10 Operation of a variable speed pump at different systemcharacteristic curves

+ Avoidance of excess pressure

+ Soft starting of the pumps viathe frequency inverter

+ See also Chapter 4

+ Protection (wear reduction) ofmechanical components

+ Reduction of hydraulic feed-back effects

+ Power saving

+ Low grid load due to reducedstarting currents

+ Reduction of life cycle costs

– Higher control costs

Evaluation

Flow Rate Adjustment by aCombination of ParallelOperation and Variable SpeedOperation

The division of the flow intoseveral pumps is used in all ap-plications where demand fluctu-ates substantially and where thefollowing requirements must bemet:

• Minimization of power con-sumption

• Reduction of system costs

• Compliance with minimumpump flow rate

A first approximate adjustmentof the pump output to the sys-tem demand takes place by par-allel operation.

The fine adjustment is achievedby infinitely variable speed ad-justment of one or more cen-trifugal pumps.

10

1 Principles

Pump flow rate

Operating range for one variable speed and two fixed speed pumps

Extended operating range if all three pumps are variable speed pumps

Pum

p h

ead

Fig. 11

+ Broad flow rate adjustmentrange (with limited headrange)

+ High control quality

+ Redundancy on the pump side

+ Reduced switching frequency

+ Reduced mechanical load

+ Reduced hydraulic feedbackeffects

+ Low drive energy costs

+ Swapping of the variable speedpump possible

– Limited use in the event ofinlet pressure fluctuations

– Limited working range invariable speed operation

– Medium purchase costs

FI

Grid

Fig. 12 One variable speedpump

Grid

Fig. 13 Several variable speedpumps

Evaluation of one variable speed pump

+ Increased flow rate and headadjustment ranges

+ Use with high inlet pressurefluctuations

+ Low energy use as a result ofoptimal inlet pressure utiliza-tion

+ Large variation possible in theset value range

+ Excellent control quality

+ Full redundancy (pumps andfrequency inverters)

+ Greatly reduced switching frequency

+ Greatly reduced mechanicalloading

+ Greatly reduced hydraulicfeedback effects

+ Extremely low drive energycosts

+ Pump change possible withoutinfluencing the control quality

– High purchase costs

Evaluation of several variable speed pumps

1.1.3Characteristic Curve Conver-sion at Variable Pump Speed

If the pump and pumped fluidare the same, the performancedata for a centrifugal pump invariable speed operation varyaccording to the following mod-elling / affinity laws:

Equation 1

Equation 2

Equation 3

In what follows, the pump char-acteristic curves will be calcu-lated for an example in whichtwo pumps are operated in par-allel (one pump continuouslyspeed controlled, the second op-erated at a fixed speed).

For simplification, we make theassumption of a closed circuitwithout static counterpressure.Using the calculation methodsrepresented the user can alsosolve cases with single pumps oreven multi-pump systems. For adeeper understanding of the hy-draulic interplay between pumpcharacteristic curve and systemcharacteristic curve, we recom-mend that you work through afew systems yourself, followingthe pattern given. For dailywork IT programs convenientlysupport the calculation.

The aim of the following calcu-lation processes is to create apump performance chart whichincludes all important character-istic curves.

• Piping characteristic curve(system characteristic curve)

• Controlled-operation curve

• Pump characteristic curve(nominal speed)

• Affinity parabolas

• Pump characteristic curves(for reduced speeds)

• Pump characteristic curves forparallel operation

• Power characteristic curves,(fixed speed / variable speed)for single pump operation andparallel operation

These results form the basis ofany economic calculation to beperformed.

For the further calculationprocess it is helpful to derive anequation that creates a relation-ship between head and flowrate. To this end, equation 1 issquared and inserted into equa-tion 2 (equation 4).

Equation 4

Rearranging once again givesequation 5.

Equation 5

This equation allows us to cal-culate a second order parabolafrom the origin (Q = 0, H = 0)through a point B2 (H2, Q2) inthe H/Q diagram. The values ofH2 and Q2 are known, since theparabola is to cut through thispoint.

H1 and Q1 are unknown andwill therefore be denoted as Hx

and Qx in what follows.

The flow rate Qx will, depend-ing upon the necessary accuracy,be assumed for several points onthe parabola and Hx then calcu-lated according to the derivedformula.

The value pairs Qx and Hx areshown in table form in what fol-lows for the sake of a betteroverview.

Q1

Q2H1 =

2

H2 ·

Q1

Q2

H1

H2=

2 n1

n2

2

=

n1

n2

P1

P2=

3

n1

n2

H1

H2=

2

n1

n2

Q1

Q2=

11

1Principles

Key:

B: Operating pointH: HeadQ: Flow raten: Pump speedP: Power input at the

pump shaftx: Sought quantity

Indices:

N: Nominal0: At zero flow1; 2: Pump 1; Pump 1 + 2 in

parallel...’: In fixed speed operationW: Leading valueZ: Intermediate points

Calculation of the PipingParabola using Equation 5

The piping characteristic curvein a closed system runs from theorigin to the operating point BN

(full load).

Note:The piping / system characteris-tic curve for open systems withstatic counterpressure is ex-plained in Chapter 1.2.5.

Calculation of the Controlled-operation Parabola

The origin of the parabola isshifted to the level of the setvalue by means of a small expan-sion of the equation (see p. 27,Fig. 55 and p. 50, Fig. 77)

Qx / QNHx =2

(HN - HW) · ( )

Qx / 100 %=2

35 % ·( )

+ HW

+ 65 %Hx

Qx / QNHx =2

HN · ( )

Qx / 100 %= 2100 % · ( )Hx

12

1 Principles

Given Sought

Qx Hx

25 650 2575 56

110 121

Given Sought

Qx Hx

25 6750 7475 85

H [%]

Nominal head

No

imin

al f

low

rat

e

120

100

80

60

40

20

0200 40 60 80 100

BN

Q [%]

HN

QN

Piping characteristic curve

Fig. 14

H [%]

Nominal head

HWN

om

inal

flo

w r

ate

120

100

80

60

40

20

0200 40 60 80 100

BN

Q [%]

HN

QN

Controlled-operation curve

Fig. 15

The controlled-operation char-acteristic curve is a theoreticalcurve along which the operatingpoint should move.

It ensures that from the min-imum to the nominal flow ratethere is always sufficient pump

head available to cover the pip-ing pressure losses and the use-ful pressure at the consumer in-stallation.

The value HW is dependentupon the following influencingfactors:

• Operating behaviour of theconsumer installation

• Similar load behaviour overtime or time-independentload behaviour

• System dimensioning

Pump Selection

A pump is selected that achievesthe nominal head (B’2) at halfthe nominal flow rate.

In addition, the pump character-istic curve must at least intersectthe controlled-operation curve(B1, max) (see also Chapter 1.1.1).

In systems with two pumps(without a stand-by pump) inthe event of the failure of apump at least the system charac-teristic curve must be intersected(B1, fault), since otherwise the re-maining pump will be over-loaded

Calculation of the AffinityParabola by: B’2 (Q’B2, H’B2)

According to the laws of affinitythe operating point B’2 movesalong the affinity parabola whenpump speed is reduced. Thepath of the affinity parabola isfound using the equation below.It yields the working point B2 onthe controlled-operation curve.

Qx / Q'B2Hx =2

HN · ( )

Qx / 50 %=2

100 % ·( )Hx

13

1

Given Sought

Qx Hx

15 925 2535 56

H [%]

Nominal head


HW

Nom

inal

flow

rat

e

120

100

80

60

40

20

0200 40 60 80 100

B’2

B1,max

Q1,maxQN2

B1,Fault

Q [%]

HN

QN


Fig. 16

H [%]

120

100

80

60

40

20

0200 40 60 80 100

B’2

Q’B2

B2

Q [%]

H’B2

QN

Affinity

par

abol

a

Fig. 17

Principles

Calculation of the AffinityParabola by: B’1 (Q’B1, H’B1)

Using the same calculationprocess as before a further point(B1) on the controlled-operationcurve is found.

In many cases it is worthwhileselecting the point B1 at half thepump flow rate.

Pump Characteristic throughB2 at Pump Speed n2

The precise values can also bedetermined by calculation. Forpractical use the read-off valuesare perfectly adequate.

We read off:QB2 = 42 %; HB2 = 71 %

Using the equations given belowwe first calculate the reducedspeed for the operating point B2

from the ratio of heads.

Speed at B2:(QB2 = 42 %, HB2 = 71%)

In the second step the head atzero flow point H0.2 is calcu-lated for this speed n2. This

allows us to draw the pumpcharacteristic curve for n2 withsufficient accuracy.

Head at Q = 0 and n = n2

= H0 · n2 / nNH0.2

2( )

84 % / 100 %= 2120 % ·( ) = 85 %H0.2

n2 = nN ·HB2

H'B2

n2 = 100 · 71 %100 %

= 84 %

Qx / Q'B1Hx =2

H'B1 · ( )

Qx / 25 %=2

115 % ·( )Hx

14

1 Principles

H [%]

120

100

80

60

40

20

0200 40 60 80 100

B’1

Q’B1

B1

Q [%]

H’B1

QN

Affin

ity p

arab

ola

Fig. 18

Given Sought

Qx Hx

10 18,415 41,420 73.625 115.0

H [%]

120

100

80

60

40

20

0200 40 60 80 100

B’2

B2

n2

nN

Q2

B’0

B0,2

Q [%]

H’B2

HB2

H0,2

H0

QN


Fig. 19

Pump Characteristic Curvethrough B1 at Speed n1

The pump characteristic curvethrough the operating point B1

is calculated using the same cal-culation process as before.

Speed at B1:(Q1 = 19 %, H1 = 66 %)

Head at Q = 0 and n = n1

Addition of the Pump Characteristic Curves

The parallel operation charac-teristic curve is found by addingthe flow rates of the two indi-vidual characteristic curves:

Pump 1, fixed speed,with nominal speed nN

Pump 2, variable speed,with speed n2

Starting from shut-off head H0

up to head H0.2 the flow rate isgenerated by pump 1 alone.Pump 2 cuts in at the point B’4as counterpressure decreases.The summed characteristic curveof the two pumps intersects withthe controlled-operation curveat B4 and head H4.

At this pressure level pump 1provides the flow from Q0 toZ’4 and pump 2 provides theflow from Z’4 to B4.

= H0 · n1 / nNH0.1

2( )

76 % / 100 %= 2120 % ·( ) = 69 %H0.1

n1 = nN ·HB1

H'B1

n1 = 100 · 65 %115 %

= 76 %

15

1

H [%]

Pumpcharacteristiccurve

120

100

80

60

40

20

0200 40 60 80 100

B’1

B1

n1

nN

B’0

Q [%]

H’B1

HB1

H0.1

H0

QN

Fig. 20

H [%]

120

100

80

60

40

20

0200 40 60 80 100

B0.2

Z’4Z’4

Z’4Z4

B4Q0

B4

B4B’4

n2

Q [%]

H4

H0.2

H0

QN

Pump 1

Fig. 21

Principles

Determination of AuxiliaryPoints and Intermediate Characteristic Curves

a) Operating point B3 with auxiliary point Z3

Since the operating points BN

and B’4 are quite a long wayapart an additional operatingpoint B3 is placed in between.The point QB3 = 85 % was se-lected with the associated headHZ3. At operating point B3 thepump delivers a flow at reducedspeed, which is represented bythe distance between the pointsZ3 and B3.

For the construction of a pump

characteristic curve for reducedspeed this distance is moved leftto the origin at head HZ3. Theend point is Z3.

We read off:Distance Z3B3 = 26 %.

16

1 Principles

H [%]

120

100

80

60

40

20

0200 40 60 80 100

Z’3Z3 B3

QB3

BN

B’4

Z’3

Z’3

B3

B3

Q [%]

HZ3

QN

Fig. 22

H [%]

120

100

80

60

40

20

0200 40 60 80 100

B’3

Z3

Z’3

B3

QZ3

n3nN

B’0.3

Q [%]

H’B3

HZ3

H0.3

H0

QN

Fig. 23Given Sought

Qx Hx

20 5330 120

c) Pump characteristic curvethrough B3 (Z3) at speed n3

Determination of the speed at n3

H’B3 = 113 % (read off)

(QZ3 = 26 %, HZ3 = 90 %)

Determination of the head at Q = 0 and n = n3

n3 (calculated)

H0 (read off)

H0.3 = H0 ·n3

nN

H0.3 = 120 % · 89 %100 %

= 95 %

2

2

n3 = nN ·HZ3

H'B3

n3 = 100 · 90 %113 %

= 89 %

b) Calculation of the affinity parabola through Z3(QZ3,HZ3)

For the construction of the inter-mediate characteristic curve it isnecessary to convert the pointZ3 to the nominal speed B3. Tothis end an affinity parabola isplaced through the point Z3.

We read off: HZ3 = 90 %.

Qx / QZ3Hx =2

H'Z3 · ( )

Qx / 26 %=2

90 % · ( )Hx

Addition of the CharacteristicCurves of Equally Sized Pumps1 and 2 at Nominal Speed

At a head of 100 %, for ex-ample, the distance to the inter-section with the characteristiccurve of pump 1 is measuredand the same distance markedoff to the right of the intersec-tion.

Using this procedure, dependingupon the accuracy requirement,further points for the summedcharacteristic curve of the twoequally sized pumps 1 and 2 arefound.

Input Power of Two PumpsOperated in Parallel at Nominal Speed

This requires that the inputpower of a pump is known. The total input power in paralleloperation is sought. At point BN

both pumps have an inputpower of P’2. This means thatboth pumps consume the doublepower P’2 x 2. In this mannerthe points P’3 x 2 and P’1 x 2were also found.

Pw = pump input power (shaft power)

17

1

H [%]

120

100

80

60

40

20

0200 40 60 80 100 Q [%]

L2

L1

L2

L1

QN

Pumps 1+2

Fig. 24

H [%]

P [%]w

120

100

80

60

40

20

0

220

180

140

100

60

20

20

20

0

00

40

40

60

60

80

80

100

100

B’1

B’2

Bl l,1 Bl l,3

Bll,3BN

P’ x 22

P’ x 23

P’ x 21

P’2 P’3P’1

Q [%]

Q [%]

QN

Fig. 25

Principles

Input power of pump 1 invariable speed operation

The reduced speeds have beendetermined by the previousstages. Since the input power isknown for fixed speed oper-ation, the input power in ques-tion in variable speed operationcan be calculated.

Pump input power in paralleloperation

(Pump 1 at nN, Pump 2 at n = variable)

Starting from point B3, go hori-zontally left to Z’3, and fromthis point vertically down to P’3.P’3 is the input power of thefixed speed pump.

P3 = P’3 + P3, n3

To determine the proportionalpower of the variable speedpump P3, n3

we use equation 3(page 8). We thus find:

P3,n3 = P'3,nN

·n3

nN

3

Pl,max = Input power as for fixedspeed operation, since thespeed is 100 % = nN

P2 = P'2 ·n2

nN

P2 = 100 % · 84 %100 %

= 59.3 %

3

3

P1 = P'1 ·n1

nN

P1 = 74 % · 76 %100 %

= 32.5 %

3

3

18

1 Principles

H [%]

P [%]w

120

100

80

60

40

20

0

220

180

140

100

60

20

20

20

0

00

40

40

60

60

80

80

100

100

B4

B’4B’3

Z’4Z’3

Z3

Z4

B4

B3

n2

n3

n2

n3nN

BN

P’3

P3

P3 ,n3

P’3

P5

P4

PNP’3,nN

P’4,nN

Q [%]

Q [%]

QN

Fig. 27

H [%]

P [%]w

120

100

80

60

40

20

0

220

180

140

100

60

20

20

20

0

00

40

40

60

60

80

80

100

100

B’1

B’2

B1B2

n2

nN

n1

BN

P’2

P’1

P1

P2

PI,max

Q [%]

Q [%]

QN

Fig. 26

Minimum Set Value in ParallelOperation of Available Pumps

At a given maximum pump flowrate (Caution: motor power re-serve) the minimum target valueto be set can be calculated asfollows:

QN2 - Ql,max

2HW,min = HN -

HN - Hl,max

100%2 - 65%2

HW,min =100% -100% - 80%

·100%2

·QN2

HW,min = 65 %

The further points are deter-mined in the same way

P3 = P'3 + P'3,nN ·

n3

nN

3

P3 = 108 % + 80 % ·89 %100 %

3

P3 = 164.4 %

P4 = P'4 + P'4,n2

P4 = P'4 + P'4,nN ·

n4

nN

3

P4 = 112 % + 52 % ·84 %100 %

3

P4 = 143 %

P5 = 2 · PN = 2 · 100 % = 200 %

19

1Principles

H [%]

120

100

80

60

40

20

0200 40 60 80 100

BNHN

Hw,min

HI,max

BI,max

Q [%]QNQI,max

Fig. 28

Notes: Input power in the eventof a speed changeIf the speed is changed thepoints of a throttled-operationcurve move along second orderparabolas to the other throttled-operation curve.

If the speed is reduced by lessthan 20 % of nominal speed theefficiencies remain almost con-stant. In the event of greater de-viations the efficiency worsensslightly. Since the power require-ment of the pump reduces by a

power of 3 as the speed falls, theslight worsening of efficiency isnot important. In the workedexample no efficiency correctionwas made.

1.1.4Economy Calculation for Infin-itely Variable Speed Adjust-ment Systems with FrequencyInverter

How can the benefits of thepump control systems bedemonstrated? To provide evi-dence it is necessary to know theinfluencing factors and their im-

portance. For the economy of apump system in relation to thepump output these are:

1. The design of the system2. The load distribution over

time of the system3. The pump4. The pump power consump-

tion from the electrical grid

The following shows in moredetail how these factors act.

20

1 Principles

Influences Due to the Designof the System

The operating point of a cen-trifugal pump is always thepoint of intersection between thesystem characteristic curve andthe pump characteristic curve.All control methods thus changeeither the pump or the systemcharacteristic curve.

The system characteristic curvedenotes the pressure require-ment of the system depending

upon the flow rate. It alwayscontains dynamic componentsthat increase quadratically withthe flow rate due to the flow re-sistances – for example in circu-latory systems (heating).

However, it may also incorpo-rate additional static compo-nents, such as differences in geo-detic head or pressure differ-ences caused by other factors –for example in transport systems(pressure boosting). In circula-tory systems the system charac-

teristic curve has no static com-ponents and thus begins at theorigin (H = 0). In practice, toprevent consumer installationsbeing undersupplied, the neces-sary pressure graph lies abovethe system characteristic curve.Its precise path is dependentupon the system in question.

The controlled-operation curve,along which the operating pointshould move, must consequentlylie on or above the necessarypressure line.

Influences as a Result of theLoading of the System overTime

The flow rate Q of a centrifugalpump system can, in the mostextreme case, fluctuate betweena maximum value and zero. Ifwe order the required flow rateover a year according to size weobtain the ordered annual loadduration curve. Its precise pathis dependent upon the system inquestion and can differ from oneyear to the next.

The Figure opposite shows twopossible graphs. The longer theoperating period and the smallerthe area below the curve, thegreater is the potential for pos-sible savings.

[h][∆h]

Q [%]

Nominal flow rate

Minimum flow rate

Op

erat

ing

ho

urs

Stan

dst

ill

1000

1000

1000

1000

1000

1000

1000

1000

760

0 20

6400

8760

40 60 80 100 120

Annual load duration curve I

Annual load duration curve II

Fig. 29Load profile (example): The pump is designed for 100 % flow rate.This output is seldom required in the year. Most of the time a lowerflow rate is required. To save pump drive power the control systemautomatically matches the pump speed to the momentary system de-mand.

The Pump Power Consump-tion from the Electric Grid

Chapter 1.1.3 only addressedthe different pump input powerrequirements of the pumps(shaft power). However, if wewant to precisely determine theelectric drive power saved, thefollowing relationships are alsoimportant:

Electric power consumption,fixed speed operation (PE, u)

The power consumption in fixedspeed operation is increased inrelation to the pump shaftpower (PW, u) by the motorlosses.

Electric power consumption, variable speed operation (PE, g)

The power consumption in vari-able speed operation is deter-mined by the shaft power PW, g

plus the losses of the frequencyinverter plus the motor losses(the motor losses may increaseslightly depending upon the fre-quency inverter type).

The additional losses as a resultof variable speed operation arenegligible, since a power savingis achieved as soon as the flowrate falls below approx. 95 %compared to fixed speed oper-ation (see Fig. 30).

For practical applications it isnot necessary to determine thepower consumption in detail. Itis fully adequate to base calcula-tions upon the pump inputpower (shaft power) in question.This is because, as shown in Fig.30, the absolute electrical powerlosses in variable speed andfixed speed operation are almostidentical.

21

1Principles

Influences of the Pump

The pump can influence the ex-tent of possible savings realizedby pump control in differentways: by the path of its charac-teristic curve, by the differentmotor sizes required and by thedesign of the pump. The graphof the pump input power de-

pends upon the gradient of thehead and the graph of pump ef-ficiency. In general: The steeperthe pump characteristic curve,the flatter the power characteris-tic curve.

The motor size of a pump unithas an influence, since experi-ence tells us that the ratio of in-

vestment to motor size (€/kW)falls as the power increases.

In multi-pump systems (as inour example with 2 operatingpumps) the economy calculationis performed according to thesame way as described in thefollowing sections.

Q [%]0 20

P [%]

40 60 80 100

120

80

40

0120

P E

PE,u

PE,g

PW,u

PW,g

Fig. 30Saved electric powerThe shaded areas in the Figure show that the absolute electricalpower losses in variable speed and fixed speed operation are almostidentical. The higher losses of infinitely variable pump control in thefull load range are compensated in part load operation.

Key:

PW = Pump input power(shaft power)

PE = Electric power consumption

u = Fixed speed

g = Variable speed

∆PE = Saved electric power

A Comparison of Three Sys-tems With and Without SpeedControl

On the following pages we willcompare systems with and with-out speed control:

• In the H/Q diagram

• In the pump input powercurve

• In the savings diagram

The systems shown are closed

circulation systems. The consid-erations and predictions can eas-ily be applied to open transportsystems, such as water supplysystems or waste water systems,for example.

22

1 Principles

1) Throttling configurationwith/without pump speedcontrol

Fig. 31

Diagram: H/Q curve

The nominal flow rate, the nom-inal head and the nominal speedare each marked at 100%. Thepump characteristic curve isdrawn for several speeds in in-crements of 10 % from thenominal speed down. The sys-tem characteristic curve beginsat the origin of the H/Q dia-gram, since it is a closed system,and its path is parabolic. Thegradient of the system parabolais dependent upon the losses inthe pipe network and thus alsoupon the throttling processes ofthe consumer installation. Thefluctuation of the system charac-teristic curve permitted by thepump is limited by the minimumand maximum flow rate.

In practice the required pressurecurve lies above the system char-acteristic curve. To preventundersupply at any consumerinstallation, the pump pressuremust always lie above this curve.

However, for hydraulic and en-ergy reasons the pump pressureshould lie very close to thisboundary. This means that thecontrolled-operation curve,along which the operating point(intersection of the system char-acteristic curve with the pumpcharacteristic curve at the speedin question) moves, should lie aslittle as possible above the re-quired pressure curve.

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

n = 40 %

H [%]

40 60

Pum

p h

ead

80 100

120

100

80

40

0120

Qmax

System characteristic curve (full load)

BN

opt


Required pressure curve

Pump duty limit

Fig. 32

Diagram: Saving

The saving found from thepump input power diagram isnow transferred. The motor effi-ciencies in fixed and variablespeed operation are alreadytaken into account in this dia-gram, as are the frequency in-verter efficiencies in variablespeed operation. At nominalflow rate the saving is, ofcourse, equal to zero or evennegative, but at a reduced flowrate this rises considerably.

Diagram: Pump input power

As for the H/Q diagram, atnominal flow rate and nominalspeed the pump input power ismarked at 100 %. Like thepump characteristic curves, theinput powers are also drawn atspeed increments of 10 %.

As the operating point movesalong the controlled-operationcurve away from the designpoint to a lower flow rate, theassociated shaft power of thepump can easily be determined.The intersection of the con-trolled-operation curve with thepump characteristic curve inquestion in the H/Q diagram isextended downwards until thepower curve that corresponds

with the speed is intersected inthe input power diagram. Thesame procedure is followed forall intersection points.

The intersection points can thenbe joined together in the pumpinput power diagram, giving the

shaft power requirements forvariable speed adjustment. Theshaft power saved by the speedadjustment lies between thiscurve and the input power inthrottled operation and at con-stant speed.

23

1

Q [%]0 20

P [%]W

40

Pum

p in

pu

t p

ow

er(s

haf

t p

ow

er)

Pump input power in variable speed operation

Power savingPump input power in throttled operation

60 80 100

120

100

80

40

0120

Fig. 33

Q [%]0 20

∆P [%]E

40

Savi

ng

inel

ectr

ical

po

wer

Electrical power saving

60 80 100

120

80

40

0120

Fig. 34

Principles

2)Throttling configurationwith overflow valve andwith/without pump speedcontrol

24

1

Diagram: H/Q curve

The nominal flow rate, the nom-inal head and the nominal speedare each marked at 100 %. Thepump characteristic curve isdrawn for several speeds in in-crements of 10 % down fromthe nominal speed. The systemcharacteristic curve begins at theorigin of the H/Q diagram sincethis is a closed system. Its path isparabolic and should passthrough the design point (100%) for fully opened consumerinstallations. If the flow throughthe consumer installations is re-stricted, the overflow valveopens and allows the flow rate

that is not required to be dis-charged. This means that thepump almost always works atalmost full power. Withoutspeed adjustment the possiblepressure rise on the pump char-

acteristic curve is limited by theoverflow valve – with the greatdisadvantage that there is an al-most continuous wastage ofdrive energy.


Here too, the pump input poweris set equal to 100 % for the de-sign point. The relatively narrowworking range of the pump de-termined by the overflow valveleads to an almost constantpump input power requirementon fixed speed pumps. For avariable speed pump the bypasscan remain closed, only the min-imum flow rate of the pumpmust be guaranteed.

The pump input power requiredin variable speed operation isdetermined in the same way asfor pure throttling. This meansthat lines are drawn down fromthe intersections of the con-

trolled-operation curves with thepump characteristic curves inthe H/Q diagram until theyintersect with the associatedpower curve (at the same speed)in the power diagram. By con-necting these points we then ob-tain the shaft power requirement

at the modified pump speed.The power saving is the differ-ence between the horizontalcharacteristic of the fixed speedinput power requirement andthe curve of the shaft power re-quired in variable speed oper-ation.

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

n = 40 %

H [%]

40 60

Pum

p h

ead

80 100

120

100

80

40

0120

Qmax

Operating range of thepump

BN

opt

System ch

aracteris

tic cu

rveControlled-operation curve


Pump duty limit

Fig. 36

Fig. 35

Q [%]0 20

P [%]W

40

Pum

p in

pu

t p

ow

er(s

haf

t p

ow

er)

Pump input power forvariable speed operation

Power saving

Idealized representation of inputpower for differential pressure overflow configuration

60 80 100

120

100

80

40

0120

Fig. 37

Principles

Diagram: Saving

This power saving can again beshown in its own diagram. Thisclearly shows that the powersaving potential resulting fromvariable speed operation in asystem with an overflow valve issignificantly greater than is thecase for pure throttling. Q [%]0 20

∆P [%]E

40

Savi

ng

inel

ectr

ic p

ow

er

Electric power saving

60 80 100

120

80

40

0120

Fig. 38

25

1Principles

3) Bypass configurationwith/without pump speedcontrol

Diagram: H/Q curve

The nominal flow rate, the nom-inal head and the nominal speedare each marked at 100 %. Thepump characteristic curve isdrawn for several speeds in in-crements of 10 % down fromthe nominal speed. The systemcharacteristic curve begins at theorigin of the H/Q diagram sincethis is a closed system. Its path isparabolic. The flow rate of thepump is divided into a usefuland a bypass flow rate. Both flow rates can vary from0 – 100 % and always add up to100 %. This means that the sys-tem characteristic curve is alwaysconstant for the pump and thatthe pump operating point alwayslies at the design point. If thepump speed is adjusted to thesystem requirement, the operat-ing point moves downwardsalong the system characteristiccurve in part load operation.

Note:

In this hydraulic system the (dif-ferential) pressure cannot beused as the sole controlled quan-tity. In this case, pump oper-ation is controlled, for example,as a function of the temperaturedifference.

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

n = 40 %

H [%]

40 60

Pum

p h

ead

80 100

120

100

80

40

0120

Qmax

System characteristic curve

BN

opt

Controlle

d-operation cu

rve

Pump duty limit

Fig. 40

Fig. 39


The pump input power at thedesign point is 100 %. If there isno speed adjustment the inputpower remains constant over thewhole flow rate range.

The pump input power withspeed adjustment is found bydrawing lines down from the in-tersection points of the con-trolled-operation curve (identi-cal here to the system character-istic curve) with the pumpcurves at different speeds. Con-necting the intersection pointsyields the shaft power require-ment for speed adjustment.

The power saving varies be-tween the nominal power andthe pump input power at mini-mum speed.

Saving diagram

The saving between the fixedspeed and the variable speedshaft power characteristic curvecan be clearly seen in the savingdiagram. Of the three systemspresented, the potential for pos-sible energy savings is the great-est in this case.

Economy Calculation

Comparison: Throttling config-uration with and without infi-nitely variable speed adjust-ment

This is based upon the H/Q dia-gram (Fig. 43), the power dia-gram (Fig. 44) for the inputpower at the pump shaft, thediagram relating to the saving ofelectric power (Fig. 45) and theload profile (Fig. 46)1). Electri-city costs are taken to be € 0.10/kWh. The annual load durationcurve is converted into rectangu-lar blocks for convenience. In

each case the average flow rateover 1000 operating hours isconsidered. Each average flow

rate can be assigned the savedelectric power from the savingdiagram. In our example,

26

1 Principles

Q [%]0 20

P [%]W

40 60 80 100

120

100

80

40

0120

Pum

p in

pu

t p

ow

er(s

haf

t p

ow

er)

Pump input power forvariable speed operation

Power saving

Idealized representation of inputpower for bypass configuration

Fig. 41

Q [%]0 20

∆P [%]E

40

Savi

ng

inel

ectr

ic p

ow

er

Electrical power saving withspeed adjustment

60 80 100

120

80

40

0120

Fig. 42

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

n = 40 %

H [%]

40 60

Pum

p h

ead

80 100

120

100

80

40

0120

Qmax

System characteristic curve (full load)

BN

opt



Pump duty limit

Fig. 43

approximately 50 % of the flowrate is permanently requiredover 1000 hours; the associatedelectric power saving averages38 %. Multiplying the savedelectric power with the propor-tional operating hours and theprice of electricity yields the sav-ing for the time period in ques-tion. Now we must only add upthe proportional savings. The re-sult obtained is a saving of ap-prox. € 232 per year (basedupon 1 kW consumed nominalpower).

This example was calculatednon-dimensionally to improvecomparability. Following thesame pattern, however, effectivefigures can also be used in thecalculation for each specific ap-plication. For example, if theshaft power required at nominalload is 10 kW, approx. € 2320per year can be saved.

Economy consideration

(based upon 1 kW nominalpower consumption)

Note: The calculation performed here isbased upon a 100% correct piping calcu-lation and pump design. In practice, how-ever, pump power is often greatly over-dimensioned. Consequently, the saving iscorrespondingly even greater.

1) In the saving diagram (Fig. 45) themotor efficiencies in fixed speed orvariablespeed operation are taken into account,plus the frequency inverter efficiency invariable speed operation. The flow rate re-quirement of the pump system over thecourse of a year – arranged in size order –is entered in the load profile (Fig. 46). Thiscurve is called the “ordered annual loadduration curve”. The longer the operatingtime and the larger the area above thecurve, the greater is the possible energysaving potential.

27

1

Q [%]0 20

P [%]W

40 60 80 100

120

100

80

40

0120

Pum

p in

pu

t p

ow

er(s

haf

t p

ow

er)

Pump input power for speed adjustment

Power savingPump input power in throttling configuration

Fig. 44

Q [%]0 20

∆P [%]E

40

Savi

ng

inel

ectr

ic p

ow

er

Electrical power saving

60 80 100

120

80

40

0120

Fig. 45

[h][∆h]

Q [%]

Nominal flow rate

Minimum flow rate

Ordered annual load duration curve

Op

erat

ing

ho

urs

Stan

dsti

ll

1000

1000

1000

1000

1000

1000

1000

1000

760

0 20

6400

8760

40 60 80 100 120

Fig. 46

∆PE : Saved electric powerB : Operating hoursS : Electricity costs∆EE : Saving in electricity costs∆EE = ∆PE · B · S

∆PE B S ∆EEkW h/a Euro/kWh Euro/a

(kW) (kW)

0.23 1000 0.10 23.--0.35 1000 0.10 35.--0.38 1000 0.10 38.--0.40 1000 0.10 40.--0.40 1000 0.10 40.--0.40 1000 0.10 40.--0.40 400 0.10 16.--

Σ 232.--

Principles

1.2Control Principles

1.2.1Definition

Closed-loop control is a processin which the quantity to be con-trolled (e.g. the level in the high-level tank) is continuously meas-ured and compared with a setvalue (desired level).

If the comparison yields a differ-ence between the set value andthe measured actual value of thecontrolled quantity, the manipu-

lated quantity (here: pumpspeed) is automatically adjustedand the control deviation (con-trol error) is rectified. Thisprocess is self-contained, there-fore we speak of a closed con-trol circuit.

28

1

Controller

Z

Set valueadjuster

Levelsensor

Fig. 47

1.2.2Further Control Terminology

Open-loop control

is a process in a system, inwhich one or more quantities inthe form of input quantities in-fluence other quantities in theform of output quantities due tothe system’s inherent properties– open sequence of effect.

Controlled quantity x

is the quantity that should beheld constant

Set value XS

Constant leading value. Setvalue in fixed point control.

Manipulated quantity y

is the quantity with which thecontrolled quantity is influencedas desired (e.g. speed).

Interference quantity z

is the quantity that changes thecontrolled quantity unintention-ally from outside (e.g. variablethroughflow).

Controlled system S

is the part of the system inwhich the controlled quantityshould be held constant (allcomponents between point ofadjustment and point of meas-urement).

Actual value x

is the momentary value of thecontrolled quantity (e.g. differ-ential pressure measured usingfeedback value transmitter).

Leading value w

Variable set value (e.g. managedvia external temperature or flowrate or timetable).

Deviation (Control error) xw = x - w

Deviation from the leading value(from the set value).

Measurement location

is the position in the systemwhere the controlled quantity ismeasured.

Actuator

Is the device that changes themanipulated quantity (e.g.pump, valve).

Actuating drive

Drive of the actuator (e.g. elec-tric motor, frequency inverter).

Feedback transmitter

Converts the controlled quantityinto a standardized electric signal(e.g. 0/4 - 20 mA or 0/2 - 10 V).

Principles

1.2.4Controlled Quantities forClosed Hydraulic Circuits

Differential Pressure Depend-ent Control

If the volumetric flow is variablethe differential pressure is thecorrect controlled quantity. Thefact that pressure changes inwater-filled pipelines propagateat a speed of approx. 1000 m/smeans that a change in differen-tial pressure is reported almostwithout delay. As a result, thepump can quickly react to thedifferent loads by changing itsspeed.

In a closed circuit the pump onlyacts against flow resistances.Geodetic pump heads or thesystem pressure may not betaken into account. These influ-ences can easily be eliminated ifthe differential pressure is usedas the controlled quantity.

In Fig. 49 the flow rate is re-

duced from 100 % to 80 % bythrottling. At fixed pump speedthe operating point moves from

B I to B II. If the measurementpoint is at the pump, the con-trolled-operation curve is a

29

1Principles

Z1 Z2

Measurementlocation

Heatexchanger

Controlled system

X4 - 20 mA

Y:0-10 V

1

Y

5 - 50 Hz2

Feedbacktransmitter

for differentialpressure

Consumerinstallation

(variable loadbehavior =

interferencequantity)

Controller(X ) Ws

Frequencyinverter

1 2

Signal

Fig. 48

Q [%]0 20

H [%]

40 60 80 100

120

100

80

60

40

0120

System characteristic curves

Controlled-operationcurve A


Controlled-operation curve B

BI

BII∆XA

∆XB

BA

AB

BBn1

n2

nN


Pump

Heat/cold

generator

p

Fig. 49

1.2.3Control Terminology Basedupon the Example of PumpControl

straight line that correspondswith the controlled-operationcurve A (see Chapter 1.2.6). Inour example this means that adifferential pressure increase XA

takes place, which is controlledby reducing the speed from nN

to n1. The new operating pointthen lies at BA. For flat pumpcharacteristic curves (e.g. designpoint in the part load range) XA

may be too low for a fault-freecontrol process.

The following measures mayhelp to rectify this:

1. Putting the design point in theback third of the pump char-acteristic curve.

2. Using a pump with a steepercharacteristic curve.

3. Moving the measuring pointaway from the pump into thesystem.

4. Inputting a leading value, e.g.the flow rate (or the externaltemperature in heatingsystems) in addition to thedifferential pressure measure-ment.

Measures 3 and 4 give rise to acontrolled-operation curve witha quadratic path, such as con-trolled-operation curve B, forexample. In the example giventhe differential pressure devi-ation corresponds with thequantity XB. The operatingpoint for variable speed pumpsthen lies at BB with the speed n2.For further details see page 40,“Measuring Location”. Even fora very flat pump characteristiccurve the reduction of the flow

rate to 80 % (in controlled-operation curve B) leads to a deviation (XB - XA), which thenfacilitates a correction (to n2).

Furthermore, measures 3 and 4have additional positive effects:

• The speed n2 is significantlylower than nN and n1

• The excess differential pres-sure, which must be destroyedin the valves, is lower.

• The power consumption fallsmore significantly.

Applications:

Use in circuits with variableflow rate (by throttling at con-sumer installations), e.g. in:

• Two-pipe heating-coolingsystems with thermostaticvalves

• Primary circuit for the supplyof district heating transferstations

• Air conditioning / ventilationsystems

Note:

Do not use in circuits with aconstant flow rate, such as:

• Single-pipe heating systems

• Consumer-side pumps in re-turn mixing / injection typecircuits, in which no throttlingtakes place on the consumerside.

30

1 Principles

In differential pressure controlfor heating pumps we currentlydifferentiate primarily betweenthe control methods ∆p constantand ∆p variable.

Control method ∆P constant

The electronics holds the differ-ential pressure generated by thepump constant over the permis-sible flow rate range at the setdifferential pressure value HS.

Applications:

In two-pipe heating / coolingsystems with thermostatic valvesthat have high load authority(previously gravity systems),generously dimensioned systems(resistance of piping small com-pared to the resistance of thethermostatic valves).

Floor heating with individualroom temperature control.

Control method ∆P variable

The electronics changes the dif-ferential pressure to be main-tained by the pump linearly. Theset differential pressure value isautomatically reduced as thepump flow rate falls. This againoffers the opportunity of reduc-ing energy consumption.

Applications in two-pipe systems:

With thermostatic valves withlow load authority, e.g. tightlydimensioned systems (resistanceof pipe similar to the resistanceof thermostatic valves, in sys-tems with a very long distribu-tion line).

Optimal control method:

Constant differential pressure atthe point of lowest differentialpressure (not easily realizable).

Simple alternative:

Constant differential pressure∆p-c at the pump.

Problem:

If noise occurs at low through-flow despite ∆p-c, ∆p-v can beselected.

∆p-v offers an extension of thecontrol range with additionalsavings potential.

Caution:

Undersupply may occur if ∆p-vis used.

31

1

Q

H

Hmax

Hmin

Hs

∆p-c

∆p-v1 2/ Hs

Fig. 50 Control methods ∆p-constant and ∆p-variable

Principles

Differential Temperature (∆T)Dependent Control

The differential temperature de-pendent control of pumps is de-mand-dependent and independ-ent of the operating point of thepump. Its use is worthwhile insituations where the piping char-acteristic curve is not variable(system sections with largelyconstant volumetric flow).

The temperature differential be-tween supply and return cap-tures the load state (demand atthe consumer installations) dir-ectly.

The following applies:

• Full load output of the pumponly at max. heat consump-tion.

• Automatic speed and flow ratereduction in the event offalling temperature differenceand at the same time a reduc-tion in pump input power.

Due to the transport times of theconveyed fluid long idle timesmay occur, which can impairfault-free control. The use ofadditional measures, such assecondary differential pressurecontrol, allows such systems tobe controlled.

Applications:

Use in circuits with a largelyconstant flow rate and constantor variable supply temperature,e.g.

• On the primary side– Changeover– Injection circuit– Low differential pressure

manifold• On the secondary side

– Return mixing circuit andinjection circuit (withoutconsumer-side throttling)

Note:

Do not use in circulation sys-tems (e.g. in heating systems)with a variable flow rate.

Since during throttling the heat-ing medium in the heating unitcools off more quickly as a re-sult of the longer transit time.The higher temperature differen-tial leads to higher speed, whichhowever leads to the reversal ofthe desired effect, since throt-tling means that less heatingpower is required and thereforea lower flow rate, less dischargehead and reduced speed.

32

1 Principles

TR

TV

p

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

n = 40 %

H [%]

40 60

Pum

p h

ead

80 100

120

100

80

40

0120

QmaxBN

opt

Controlle

d-operation cu

rve

Pump duty limit



Pump

Heat/cold

generator

Fig. 51

Return Temperature (TR)Dependent Control

The return temperature depend-ent control of pumps is generallyused in heating/cooling systemswith heat exchangers that donot use throttling and have aconstant supply temperature.The prerequisite is a load-de-pendent, variable return temper-ature.

In cooling systems the directionof action of the control systemmust be reversed, i.e. at low re-turn temperatures – low pumpspeed, at high return tempera-tures – high pump speed. Thepurpose is to keep the returntemperature largely constant.This achieves a reduction to therequired level of the mass flowto be circulated, particularly inpart load operation. The heatlosses in the return line are re-duced as a result of the reducedreturn temperature. This work-ing principle creates the bestpreconditions for modern con-densing heat generators.

Applications:

Well suited for systems withoutthrottle units and with a con-stant supply temperature.

Note:

• A minimum circulatory flowrate must always be guaran-teed for reliable function (seeFig. 2).

• The operating limits for thecold/heat generator must beadhered to.

33

1

Low external temperatures�

high heat demand�

low return temperature�

high pump speed�

high mass flow rate

High external temperatures�

low heat demand�

high return temperture�

low pump speed�

low mass flow rate�

low power consumption

TR

p

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

n = 40 %

H [%]

40 60

Pum

p h

ead

80 100

120

100

80

40

0120

Qmax


BN

opt

Controlle

d-operation cu

rve

Pump duty limit


Pump

Heat/cold

generator

The following applies to pumpoutput control in heating sys-tems:

Principles

Supply Temperature (Tv) Dependent Open / ClosedLoop Control

The supply temperature depend-ent open-loop control of pumpsis primarily used in heating sys-tems with a constant volumetric

flow and it can be used in al-most any such system. The pre-requisite is a supply temperaturecontrolled by atmospheric con-ditions based upon an automaticmixing configuration or a low-temperature boiler with tempera-ture adjustment option. The

flow temperature is thusmatched to the system load. Thepump speed and thus the flowrate are adjusted according tothe supply temperature:

34

1 Principles

Apart from the supply tempera-ture dependent open-loop con-trol described above there is alsosupply temperature dependentclosed-loop control, in whichthe supply temperature shouldbe held constant. This is the casein heat recovery systems, for ex-ample. In such systems the sup-ply temperature is to remainconstant despite the varying in-coming heat. This means, athigher available heat the pumpspeed increases, at lower avail-able heat the pump speed falls.

Applications:

In all systems, in which the sup-ply temperature is set in relationto the load.

The supply temperature depend-ent open-loop control of thepump flow rate supports this

regulatory function.

Particularly in part load oper-ation, a larger opening stroke ofthe control valve is achieved,which gives rise to better stabil-ity of the temperature controlcircuit.

Note:

A minimum circulatory flowrate must always be guaranteedfor a reliable function(see Fig. 2).

Low external temperatures�

high supply temperature�

high consumption�

high pump speed

High external temperatures�

low supply temperature�

low consumption�

low pump speed�

low pump power consumptionTmin

nmin

nmax

TmaxSupply temperature

Control line

Pum

p s

pee

d

Fig. 53


TA

TV

TV

p

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

n = 40 %

H [%]

40 60

Pum

p h

ead

80 100

120

100

80

40

0120

QmaxBN

opt

Controlle

d-operation cu

rve

Pump duty limit

Controller


Pump

Heat/cold

generator

Fig. 54

1.2.5Controlled Quantities for OpenCircuits

Pressure Dependent Control

Pressure dependent control isparticularly suitable for opensystems with variable volumetricflow. This is brought about byvarious withdrawal rates (throt-tling) at the consumption points.The task of the variable speedpump is to supply sufficientpressure (flow pressure) to theconsumption points. Due to thevarying volumetric flows, vari-able pressure losses occur in thetransport pipes. If the measuringpoint lies at the pump, the con-trolled-operation curve has aconstant (horizontal) path.Good pressure control generatesonly the level of pressure that isrequired in the load state inquestion. This can be achievedby a suitable selection of thepressure measurement locationin the system a long way awayfrom the pump or by intelligentpump control systems (con-trolled-operation curve B).

Applications:

• Water supply systemsDrinking waterPressure boostingFire extinguishing systems

• Industrial processes

• Cooling systems

Note:

The influences of any variableinlet pressure and difference ingeodetic head or counterpres-sures must also be taken intoaccount in the design of pumpsand control systems.

35

1

Min.

p

Q

Nominal head

Nominal flow rate

Nominal operating pointControlled-operation curve A

Controlled-operation curve B

HN

QN

BN

HFL

HFL Flow pressure (desired pressure at consumer installation)Hgeo Geodetic head (from water level to water level)Hdyn Dynamic head component (pipe friction losses)Hstat Static head (head component of the system that is independent of Q)

Hstat

Hdyn

Hgeo


Fig. 55

Principles

Level Dependent Control

If a constant liquid level is re-quired in a tank the level is usu-ally the suitable controlledquantity. Changing the supplyor discharge brings about a leveldeviation. When the target levelis exceeded the speed increases(the pump conveys more), whenthe level drops below the desiredvalue the speed falls (the pumpconveys less). The pump pres-sure is only high enough to com-pensate for the differences ingeodetic head and the frictionallosses. At constant geodetic headand unchanged piping there is aquadratic H/Q curve.

The increase corresponds withpipe friction losses, which risewith the increasing flow rate.The set value Hset is found fromthe desired level in the tank.

Applications:

e.g. in

• Waste water treatment

• Cooling water systems

• Process technology

In the design of the pump/trans-port concept attention should bepaid to the combined effects of,and interaction between, the fol-lowing variables:

Supply, discharge, tank switch-ing volume, pump size, controlspeed.

Note:

In addition to the signal trans-mitter required for system con-trol, devices that protect thetank against overflowing andstop the level from falling belowthe minimum level must also beprovided in the tank.

These protective devices shouldalways be independent of thecontrol system’s signal transmit-ter (in the simplest case bymeans of separate floatswitches).

If there is a danger of blockagesor operating errors it is recom-mended that the pump flow rateis also monitored.

If the total flow rate is split be-tween several pumps (peak loadpumps), particular attentionshould be paid to the operatingmethod of the control system.

36

1 Principles

Min.

Hset

Hstat

Q = var.Inl

Hstat

h

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

H [%]

40 60 80 100

120

100

80

40

0120QN

HN

Controlled-operation curve = system characteristic curve (constant)

opt

Fig. 56 Drainage system

Flow Rate Dependent Control

The objective of this is to holdthe flow rate at a desired value.Interference factors, such asfluctuating inlet pressure or re-sistance (e.g. due to dirty filters)must be compensated. The H/Qcurve / controlled-operationcurve should be a vertical lineon the H/Q diagram.

Applications:

• Water treatment systems

• Cooling processes

• Mixing tasks

• Waste water treatment

Note:

The influences of any inlet pres-sure variations, geodetic headdifferences or counterpressuresmust also be taken into accountin the design of pumps and con-trol systems.

Flow (rate) transmitters shall beselected in accordance with therequirements of the fluid and theexternal operating conditions.

37

1

Q [%]0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

H [%]

40 60 100

120

100

80

40

0120Qset

Hstat


Winter water level

Summer water level

Summer system characteristic curve

Winter system characteristic curve

QTreatment

system

Fig. 57 Water treatment system

Principles

1.2.6Compensation of AdditionalInterference Factors

The task of the control system isto manage the process optimally.

To this end the effects of themain interference factors mustbe offset. In the following twoexamples the throttling behav-iour of the consumer installa-tions represents the main inter-

ference factor. The variable vol-umetric flow results in differentpipe friction quantities. Twopossibilities for offsetting thisadditional interference factor aredescribed below.

38

1 Principles

Compensation by the Selection of the Correct Measurement Location

We will now highlight the influ-ence of the measurement loca-tion on pressure / differentialpressure control using anexample from heating tech-nology. The positioning of themeasuring point has a decisiveinfluence upon the pressure con-ditions and the operating costsof the system. If the differentialpressure transmitter is fitted inthe immediate vicinity of thepump, then at loads less than100 % the system will operateat an excessive differentialpressure. Power consumption isgreater than necessary.

More favourable conditionsexist if the measuring point isfitted a long way from the pumpin the supply network.

System layout

Two consumer installations arepresent, which are each designedfor half the nominal flow rate.The consumers are supplied by apump which is suitably dimen-sioned for the nominal flow data.The figures in the H/Q diagramare shown in non-dimensionalform and relate the design data.Nominal flow rate and nominalhead are each 100 %. The loca-tion for measuring pressure / dif-ferential pressure may lie at thepump or near the consumer in-

stallation, as desired. In the ex-ample it lies close to the pump,between suction and dischargeside (measuring location I), orbetween the feed and returnmanifolds (measuring location II).

In open systems (e.g. watersupply) pressure is measured in-stead of differential pressure.Here too the measuring locationcan be near to the pump (dis-charge side) or near to the con-sumer installation.

Measurementlocation I

Measurement location II

P

V S

Q = 100 %H = 100 %

N

N H = 40 %V,P - V

H = 40 %V,S - P’

Q = 50 %H = 20 %

1

V,1

Q = 50 %H = 20 %

2

V,2

P’

Feed manifold Return manifold

Dp

p=100 % p = 20 %

Dp

2

1

H = 20set

Q [%]0 20

H [%]

40 60 80 100

120

100

80

40

0120

BN

nN


Controlled-operation curve I

Zero flow rate, fixed speed

Excess pressure

Zero flow rate, variable speed I

Zero flow rate, variable speed II

Nominal flow rate, I/II

Controlled-operation curve II

Fig. 58 Influence of measurement location on pressure / differentialpressure control

Key to index:

N = Nominal

V = Feed manifold

P = Pump

S = Return manifold

P-V = Section Pump – feed manifold

S-P = Section Return manifold - pump


Development:

Pressure losses in the pipelinesections are plotted over theflow rate. If the flow rate in-creases linearly the pressure lossincreases quadratically. Para-bolas are formed in the H/Q dia-gram.

In Fig. 59 the consumer circuitsare connected in parallel, so thatflow rates of equal pressure areadded. This yields the summedcharacteristic curve for the twoconsumers.

In Fig. 60 the consumer circuitsare connected in series. Thismeans that the resistances of theindividual flows are summed.

The final system characteristiccurve is found by adding the

summed characteristic curves ofthe consumer resistances to theresistances of the main circuit.

The required pressure curve

Measurement location II

There must always be a suffi-cient pressure difference be-tween feed and return manifold,so that the consumer installa-tions are always adequately sup-plied. Depending upon the con-sumer installation in question,various load states can exist. Forexample, each consumer instal-lation could be loaded at be-tween 0 - 100 % independentlyof one another.

To ensure that the consumer in-stallations are adequately sup-plied at all times, in the exampleshown the required pressurepath is assumed to be such thatat least the nominal pressure re-quirement of the consumer (hereHset) exists between feed and re-turn manifold.

If both consumer installationsare closed (Q/QN = 0) there isno flow, and thus no flow losses.The pump works at such a lowspeed that the set value (Hset) isjust maintained. If a consumerinstallation is opened there is aflow in the main pipelines ac-companied by pressure losses.However, in order to be able tomaintain the set value, the pumpmust increase its speed and gen-erate more pressure. Since thepressure loss increases quadrat-ically in relation to the flowrate, the controlled-operationcurve II takes on a parabolicshape. The pump generates onlyas much additional pressure as isnecessary to compensate for thedynamic pressure losses thatarise.

This is also particularly clear inthe pressure diagram. The pres-

sure between feed and returnmanifold is constant and in-creases continuously over thelength of the main pipelines tothe pump depending upon theflow rate (up to the nominalflow rate at nominal heads).Pump control (pressure / differ-ential pressure control) is opti-mal if the controlled-operationcurve lies on or only slightlyabove the required pressurepath.

39

1

21

Q

H

Q’1

Q1

, R2 + R2R1 R1

Q’2

Q2

1 2

Q

H

H 1

H’ 1

R2

+ R2

R1

R1

H 2

H’ 2

Fig. 59 Parallel connection Fig. 60 Series connection

Principles

Measurement location I at thepump

If the differential pressure ismeasured at the pump, the tar-get value must be set at thenominal head. In the H/Q dia-gram this means that the pumppressure is constant over the en-tire flow rate range (horizontalcontrolled-operation curve I).

We see that, particularly in lowpart load operation, the gener-ated pump pressure lies abovethe required pressure path.

In the pressure diagram the pres-sure is constant at the pump anddecreases along the mainpipelines depending upon theflow rate.

We see clearly here that despitespeed control the pressure is toohigh at the feed manifold in partload operation. This excessivepressure can have an un-favourable effect upon the con-sumer behaviour. In any case,however, too much pump energyis expended.

40

1 Principles

Compensation by means ofAdditional Measured Variable(Flow Rate)

For various reasons it is not al-ways possible to measure a longway from the pump and near tothe least favourably situatedconsumer installation. This ap-plies to district heating systems,where very long distances haveto be bridged or, for example,systems that are constructed ac-cording to the plan in Figs. 61and 62.

By combining pressure and flowrate detection, both variablescan be measured directly at thepump.

The objective here is to obtain acontrolled-operation curve witha quadratic path (see also Fig.58 – controlled-operation curveII, also called DPC curve).

DPC: Dynamic pressure com-pensation (Pressure con-trol with flow rate de-pendent set value readjust-ment)

Controller

(4 ... 20 mA)

Frequency inverter

Flow ratemeasurement

Differentialpressuresensorp

Feed

Ret

urnHeat/

coldgenerator

Fig. 61 Flow diagram of a heating system

Applications:

• In building restoration withinsufficient system data

• In the event of undersupply invarious load states (by vari-able controlled-operationcurve)

• In the event of long signaltransfer distances

Note:

Modern control systems are cap-able of calculating the optimalcontrolled-operation curve auto-matically.

This requires the following oper-ating data

• Nominal head

• Nominal flow rate

• Pressure requirement of theconsumer installation.

This is also possible withoutflow rate measurement.

41

1

Flowrate

Pressure

Controller

Frequencyinverter

Flow rate measurement

Fig. 62 Flow diagram of a transport system (open)

Principles

1.3Principles of Integral Drive

1.3.1“Intelligent” Integrated Drivesfor Pumps

An integrated drive for pumps isa compact drive system thatconsists of a motor (el. ma-chine), an energy adjusting elem-ent (frequency inverter) and amicrocomputer for open andclosed loop control. Fig. 63shows such an “integral drive”and its process.

42

1 Principles

Micro-computer /controller

Phasecontrol

Energyadjustment

MotorTechnologicalprocess

Process measuringequipment

Set value,higher-levelcontrol system

Functional unit “Integral drive”

Electrical grid

Fig. 63 Integral drive and process

1.3.2Advantages of Integration

The integration of the energyadjusting element (frequency in-verter) and motor is associatedwith many advantages:

• Simple commissioning, sincethe motor and inverter param-eters have already been set inadvance at the factory.

• Cables between frequency in-verter and motor are dis-pensed with, leading to a re-duction in the electrical loadon the motor, less EMC prob-lems

• Integrated control functions,no external control device ne-cessary

• Reduced fitting costs com-

pared to conventional solu-tions

• Significantly fewer wiringerrors

• EMC filter already integratedinto the drive

• Integrated pump and motorprotection

1.3.3Requirements

The following points can belisted as requirements for intelli-gent integrated drive systems:

• Economical fluid transportbased on actual demand as aresult of speed adjustment

• High reliability and availabil-ity

• Mechanical compatibility toIEC standardized motors

• Electromagnetic compatibility

• Simple modification to the ap-plication by on-site parameter-ization option

• Extremely simple operation,local or remote controlled

• Integrated drive protectionand fault diagnosis functions

• Pump-specific control func-tions

• Interfaces for communicationwith higher-level systems(pump control technology)

• Decentralized “intelligence”

The term decentralized “intelli-gence” is used to mean the cap-ability of a pump drive to adaptitself to changed process require-ments. It must be capable ofmonitoring itself and the pump,communicating with the envir-onment digitally, and if requiredboth reacting and acting inde-pendently.

1.3.4Pump-specific Functions

In addition to open and closedloop control, other importantfunctions for integrated pumpdrives include dynamic pressurecompensation, a memory func-tion, minimum flow tripping,dry-running protection, inertia-secure start function and realtime clock functions.

• Due to dynamic pressure com-pensation in variable speedoperation, pipe friction lossescan be compensated when apressure sensor near the pumpis used, so that pressure re-mains constant at the con-sumer installations.

• A memory function recordsthe power input curve with aclosed pump discharge sidethrottle. This data is requiredto activate the “minimumflow tripping” and “dry-run-ning protection” functions.

• Minimum flow tripping (en-ergy saving function) ensuresthat in closed loop controlmode the drive is switched offas soon as the flow rate fallsbelow a preset minimumvalue. This is to avoid wearingout the pump. When demandrises again, the pump isswitched back on automatic-ally.

• If the dry-running protectionfunction is activated, themotor is stopped if the power /speed ratio falls below astored value due to dry-run-ning to protect the pump(mechanical seal) and thesystem goes into fault mode.

• An inertia-secure start func-tion is provided for shaking ablocked shaft loose by gener-ating an alternating torque.

• The internal real time clock ofthe drive means that time-dependent functions, such astime-of-day and day-of-weekprogramming, pump change-over and night-time setbackcan be selected.

An integrated pump drive cangenerally be used as a variable /fixed speed single drive or inMaster/Slave mode.

In Master/Slave mode severaldrives can be operated in paral-lel. In the event of a fault themaster function can be assignedto another drive. The necessarydata exchange takes place via aninternal bus system. No addi-tional external control equip-ment is required for this applica-tion.

Fig. 64 shows a differential pres-sure control setup in a heat sup-ply system in which severaldrives can work in master/slavemode.

43

1

p +

p -Heat generator

Slave drive


Differentialpressure sensor

Set value(internal / external)

Master drive

Furtherdrives

Fig. 64 Pumps with integral drive in a master/slave configuration

Principles

1.3.5Economy / Reduction of LifeCycle Costs

The extra costs for an integralpump drive are, as is the case forother speed-controlled pumpdrives, recouped after just a fewyears due to the power saving.Moreover, an integral drive forpumps offers further savings po-tential.

The following important pointscan be mentioned:

• Very low installation and com-missioning cost

• Low space requirement

• Further power savings possi-ble by minimum flow tripping,time-programming, night set-back

• No downtime due to paralleloperation of pumps

Due to these potential savingsand possibilities offered by amodern integral drive, the inte-gral drive offers clear advan-tages over conventional solu-tions with regard to life-cyclecosts.

44

1 Principles

1.4Principles of CommunicationTechnology

• Modern building managementsystems are used in largebuildings

• Building management systemsprovide “intelligent” input tothe operating systems andtechnical building equipment

• Building management systemscreate open communicationbetween the automation andcontrol systems.

The technical building equip-ment comprises various systems(HVAC = Heating VentilationAir Conditioning), e.g.

• Heating

• Sanitary (supply and disposal)

• Air conditioning / ventilation

• Electric

• Measurement, open andclosed loop control technology

This results in tasks includingthe following:

• Operations management

• Operations monitoring

• System automation

• Energy management

• Maintenance management

• Data archiving

• Operational analysis

Only by planning that covers allsystems can neutral buildingmanagement systems be createdwhich fulfil the expectations ofthe operator.

45

1Principles

Parallel data transfer (previously)

Simultaneous transfer of infor-mation via a large number ofparallel lines

Serial data transfer (new)

Transfer of information via twolines. The information is trans-ferred serially in digital form.

Paralleldata transfer(previously)

Serialdata transfer(new)

S R

S: R:Sender Recipient

Fig. 65 Parallel and serial data transmission

Levels model of automation

Communication in automationis normally defined using thelevels model. The data is trans-ferred vertically via standardized– so-called open – bus systems.

Useful pump data:• Start/stop• Set values / actual values• Speed• Status reports• Fault messages

Functions of the levels and rel-evant bus systems

46

1 Principles

Management level

Control stations, printers,building managementsystems, general IT systems

Automation stationse.g. PLC

Pumps, sensors,fans, etc.

Automationlevel

Fieldlevel

Fig. 66 Levels model of automation

Level Task Bus systems used

Management level • Information management for entire system • BACnet• System monitoring • FND • Parameterization of programs• Data protection• Cost allocation

Automation level • Basic and processing functions • BACnet• Connection of input and output devices • World FIB (France)• Performance of complex control processes • PROFIBUS

• EIB on Automation Net

Field level • Application specific open/closed loop control • BATIbus • Measured value detection and preliminary processing • EIB Konnex• Alarm messages • EHS• Event messages • LON

Example: LONTalk bus system

Master Slave Slave

Vertical communication

Horizontal communication: Master/slave communication

via LON interfaceOpen/closed-loop control, monitoring

via company’s internal bus system

Fig. 67 Communication structure for a pump group

Typical communication structure for a pump group

2System Automation Terminology and Planning Notes

The planning of circuits of allvoltage levels covers the collec-tion of operating conditions andthe specification of the systemconcept and the planning prin-ciples to be used for implemen-tation. The project planningphase represents a period ofintensive co-operation betweenthe principal, his engineeringconsultant, and the contractor.

The operating conditions are de-termined by the environmentalconditions (place of installation,local climatic factors, environ-mental influences), the higher-level electricity supply system(voltage level, short-circuitpower and neutral-point connec-tion), the switching frequency,the required availability, safetyrequirements, and specific oper-ating conditions.

With regard to the equipmentand system costs every measuremust be considered from thepoint of view of necessity andfrom an economic point of view.

Pump automation systems todayare produced to a standardwhich can easily be adapted tofurther-reaching requirementsusing modern CAD tools.

If a pumping task is to besolved, various operating condi-tions have to be taken into ac-count, e.g. acquisition and oper-ating costs, operating reliability(stand-by equipment), processconditions, desired operating be-haviour, etc.

We can proceed with projectplanning in the following order:

a) Flow rate to be split betweenone or more pumps

b) One or more frequencyinverters

c) With / without stand-bypump

d) Pump selection

e) Determine shaft power

f) Specify necessary motorpower

Regarding a)

The majority of costs for aninfinitely variable speed controlsystem are caused by thefrequency inverter.

By splitting the nominal flowrate between more than onepump we can significantly re-duce the purchase costs, whilstthe control convenience remainsbasically the same. Only onepump is operated through thefrequency inverter. Furtherpumps supplied directly fromthe grid cut in and out depend-ing upon demand.

Since the pumps deliver into acommon hydraulic system, thepressure is determined by thevariable speed pump. The oper-ating point moves along the con-trolled-operation curve.

Regarding b)

If technical requirements aresuch that the task cannot rea-sonably be handled by one fre-quency inverter alone, there isthe option of using several dutypumps with several frequencyinverters. This makes it possibleto manage difficult system con-ditions (e.g. greatly fluctuatinginlet pressures, frequent part-load operation) reliably and eco-nomically in addition to achiev-ing increased operating reliabil-ity.

Regarding c)

As can be seen from Fig. 77,75% of the nominal flow ratefor this controlled-operationcurve is achieved with just onepump. In such a case an addi-tional stand-by pump (3rdpump) is only required in sys-tems in which the full hydraulicpower must be available at anytime (e.g. service or process wa-ter supply).

Regarding d)

The required flow rate QN atthe nominal operating point caneither be achieved using onepump or by two or more pumpsin parallel operation. It shouldbe taken into account in thepump selection that the pumpsinvolved must be capable ofintersecting the controlled-oper-ation curve. In practice, exces-sive safety factors in the calcula-tion of the pipeline resistancesmean that the pumps may oper-ate outside the permissible oper-ating range.

47


Regarding e)

Fig. 77, shown below, can beused to determine the pumpshaft power requirement.

Based upon the greatest possibleflow rate of the base load pump(intersection: controlled-oper-ation curve with pump charac-

teristic curve at max. speed) thenecessary shaft power is deter-mined by drawing a line downto the pump power characteris-tic curve.

Regarding f)

For the necessary nominal motorpower (P2) safety factors of 5 -

10 % should be included in thecalculation due to tolerances inthe pump and system character-istic curves and additional mo-tor losses due to frequency in-verter operation.

48

2 System Automation Terminology and Planning Notes

2.1General Electrical Notes

Power Supply System Types

TN-C system

The neutral point of the voltagegenerator is directly earthed.The housing of the connectedoperational equipment (controlcabinets, motors, etc.) is con-nected to the neutral point viathe combined neutral and pro-tective conductor (PEN).

TN-S system

As above, but protective con-ductor PE and neutral conductorN are laid separately.

TT system (common in France)

The neutral point of the voltagegenerator is directly earthed.The housing of the operationalequipment is connected to dedi-cated earthed electrodes, whichare independent of the earthingof the voltage generator.

L1

L2

L3

PEN

RB

Fig. 68 TN-C system

L1

L2

L3N

PE

RB

Fig. 69 TN-S system

L1

L2

L3N

RARB

Fig. 70 TT system

Earth Leakage Circuit Breakers(ELCB)

Earth leakage circuit breakersdisconnect all poles of operatingequipment within 0.2 s as soonas an electric shock hazardoccurs due to an insulation fault.

ELCBs are designed for differentnominal leakage currents.

Designs with tripping currentsof 30 mA also act as personnelprotection. At greater trippingcurrents protection against firesignited by earth leakage currentsdominates.

Devices with rectifier circuits(e.g. frequency inverters), inwhich direct leakage currentscan occur in the event of a fault,may not be operated behindELCBs. In these cases so-calleduniversal ELCBs (for all types ofcurrent) with a higher trippingcurrent are used.

Power System Dependent Pro-tective Measures

Power system dependent pro-tective measures are protectivemeasures using protective con-ductors. The protective conduct-or (PE) is connected to the in-active bodies of the electricaloperating equipment. Protectiveconductors and PEN conductorsare marked in green/yellow.

In power system dependent pro-tective measures, line-side over-current protective devices switchthe power off and a fault mes-sage is generated in the IT sys-tem.

Ambient Temperature

In compliance with the relevantDIN and VDI provisions the fol-lowing simplified system classifi-cation can be made:

• Ventilation devices and sys-tems for aeration and deaera-tion, e.g. if the permissibleambient temperature is higherthan the (maximum) externaltemperature.

• Cooling devices and systemsfor pure heat removal, e.g. ifthe permissible ambient tem-perature is less than or equalto the (maximum) externaltemperature.

• Air conditioning devices andsystems for the air condition-ing of rooms if certain roomclimatic conditions must beadhered to in addition to heatremoval (temperature, mois-ture, air quality, etc.)

Starting Method (StartingProcess) for Squirrel-CageMotors

Squirrel-cage motors (asyn-chronous three-phase motors)have high starting currents. Inorder to prevent disruptive volt-age fluctuations, power supplycompanies prescribe certainstarting methods for high powermotors. For motor powersabove 4 kW certain starting pro-cedures are necessary for three-phase motors.

Star delta starting

When starting using a star deltacircuit the starting current andstarting torque of three-phasemotors are reduced to a third of

the value for delta operation.This is the most common start-ing method. (It is less well suitedin the event of a high load mo-ment of inertia and small motormoment of inertia due to amarked speed reduction in theswitchover pause).

Soft starting

It is possible to soft start three-phase asynchronous motors us-ing fully electronic soft start de-vices. In this process the start-upcurrent, and thus the start-uptorque, is deliberately influencedby the voltage dosing (phaseangle control).

We can differentiate between:

• Soft starters with adjustablerun-up period (current limit-ing only possible by increasingthe run-up period)

• Soft starters with adjustablemaximum start-up current(the start-up time is auto-matically adjusted via themoment balance betweenmotor moment and loadmoment).

• Soft starters with combinedadjustment of run-up periodand max. start-up current.

49


2.2Control Functions

A measured-value transmitterinstalled in the system suppliesthe momentary actual value tothe controller. This continuallycompares actual value and setvalue and progressively correctsany deviations. The controlfunction is only ensured if theproper direction of control ac-tion is set for the controller.

When selecting a controller, itshould be noted that the direc-tion of control action can be se-lected. We can differentiate be-tween two options:

Direction of control action forthe controller:

1. PositiveIf the set value is exceeded thespeed falls (e.g. in pressure con-trol)

2. NegativeIf the set value is exceeded thespeed rises (e.g. in inlet-side levelcontrol)

Controlled Quantity [Set Value]

• Pressure [bar]

• Differential pressure [bar]

• Flow rate [m3/h]

• Level [m]

• Differential pressure con-trolled by external tempera-ture [bar]

• Differential pressure con-trolled by flow rate [bar]

• Differential pressure con-trolled by internal flow ratefunction (only possible for asingle duty pump) [bar]

• Temperature [°C]

• Differential temperature [K]

• Temperature, combined withdifferential pressure [°C]

• Differential temperature, com-bined with differential pres-sure [K]

• Pure open-loop control mode,signal from external controller(external peak load)

50


Selection of controlledquantity

The objective of the followingconsiderations is to find a con-trolled quantity that permits thepump to be adapted to systemrequirements by changing thespeed.

The fluctuating system demandrepresents the main interferencefactor. The task of the controlsystem is to cover the system de-mand despite these interferencequantities and to largely elimin-ate the negative effects of theinterference quantities, e.g. un-desired pressure increases.

To this end the measurement lo-cation of the controlled quantityshould be located as near as pos-sible to the place where theinterference arises (e.g. represen-tative consumer installation).This ideal case is in practice hin-dered by the following prob-lems:

a) Large distances between con-trol system and ideal (repre-sentative) measurement point.

b) Branched systems with alter-nating or not very pro-nounced points of lowestpressure reading (representa-tive consumer installations).

These problems too can besolved by proven optimizationfunctions (see OptimizationFunctions)

51

2

Specify controlledquantity

Controlled quantity selected:

. . . . . . . . . . . . . . . . .

yes

no

(open)

no

(constant)

yes

Possible controlled quantities



Differential pressure. . . . . . . . . . . . .

TemperatureLevel. . . . . . . . . . . . .

Temperature- Supply temperature- Return temperature- Differential temperature- . . . . . . . . . . . . . . . .

Closedpipe system

Pipingcharacteristic curve

variable

Fig. 71

The two most important decision-making criteria for the selection ofthe controlled quantity are based on the following questions:

1. Is the piping system a closed circulation loop or an open flow?

2. Is the piping characteristic curve variable or constant, i.e. does thesystem function at a constant or variable flow rate?

Once these two questions have been answered, an important prelim-inary decision has already been made for the most favourable con-trolled quantity.

In open conveyance systems the most common controlled quantitiesare- pressure- level- flow rate

In closed circulatory systems these are- differential pressure- temperature

The controlled quantities are described in detail in what follows.

System Automation Terminology and Planning Notes

Set Value / Set Value Switching

In systems that do not imposeany great requirements on thecontrolled quantities, fixed valuecontrol with a fixed set value isoften used. However, since evenin such simple systems the loadbehaviour can change greatly,set value switching between twobasic set values permits a simpleadaptation to the system load.

Criteria for switching can be:

– Manual pre-selection

– Signal from the process (limitvalue detector)

– Time-dependent

Optimization of the Con-trolled-operation Curve by:

• Internal variable (only possi-ble for a single duty pump)in a– linear relationship– quadratic relationship

• External input in a – linear relationship– quadratic relationship

(DPC curve; flow rate de-pendent set value readjust-ment); flow rate transmit-ter required

The objective of the optimiza-tion functions is to reduce thepump flow rate as far as possi-ble, whilst still supplying allconsumer installations suffi-ciently.

The duty limits of the pumpsover the entire range of the char-acteristic diagram should betaken into account and – as faras possible – utilized.

Control using set value readjust-ment:

In addition to the controlledquantity, further variables aretaken from the controlled sys-tem that act upon the control in-put. This allows the controlled-operation curve to be easilyadapted to the system require-ments (the piping characteristiccurve). Excess pressure, particu-larly in part load operation, andcases of inadequate pressure atfull load, can be reliably avoidedin this manner.

Parameter set switching

Switching to the 2nd ParameterSet

For further adaptation to thesystem requirements there is theoption of switching over to asecond controller (with sepa-rately adjustable parameters). Itis thus possible to addresschanging system behaviour.

52


0 20

n = 100 %

n = 90 %

n = 80 %

n = 70 %

n = 60 %

n = 50 %

40 60

Pum

p p

ress

ure

[%

]

80 100

120

100

80

40

0

Quadratic

Constant(reduced)

Linear

Full loadPump flow rate [%]

opt

Fig. 72 One of the controlled-operation curves shown can be useddepending upon load behaviour.

Monitoring the Pumps and theHydraulic System in the Auto-matic Operating Mode

Even in the planning phase it isimportant to provide a suitablereliability concept for the entiresystem. The objective of this isto limit faults and, as far as pos-sible, to maintain the function ofthe system. Impaired functionalgroups are switched off and,where available, replaced bystand-by groups or emergencyfunctions. The most importantmonitoring functions for electricand hydraulic limit values areexplained in what follows.

Excess current monitoring

Basic protective function for anelectric motor against thermaloverload in direct operation onthe power supply system. Cur-rent-dependent protective de-vices that monitor the tempera-ture of the motor winding indir-ectly by means of the currentflowing in the supply line. Thisgenerates a current-dependentpicture of the heat buildup inthe motor. Over-current relays(bimetal) with a protective andback-up fuse or an over-currenttrip in a motor protection switchare used. An over-current tripbehind a frequency inverter (FI)remains inactive since the FIlimits the output current to avalue below the tripping current.Therefore, in the worst case ablocked motor is supplied atnominal current and overheatsdue to lack of cooling.

Thermistor type motor protec-tion

Greatly increased protection isoffered by temperature measure-ment in the motor to be pro-tected.

PTC thermistor detectors fittedin the motor winding directlymonitor the temperature of themotor winding. When the nom-inal response temperature of thePTC thermistor is reached its re-sistance increases sharply, andthe motor is switched off. Indi-vidually, klixon or other tem-perature monitors based uponbimetallic technology are alsoused for motor protection. Ifcontinuous temperature meas-urement is also desired, PT 100sensors can also be used.

Dry-running protection

To protect the pumps, monitor-ing takes place to establishwhether sufficient pumpingmedium is present. This can bedetermined by various measur-ing procedures. If the value fallsbelow the set limit value the sys-tem is completely shut downand the corresponding messageprovided. The system can be re-started manually or automatic-ally, depending upon the safetyrequirements. In any case, thefault message should be savedfor the operating personnel.

Flow monitoring

Flow monitors are used to pro-tect the pumps against overheat-ing due to zero delivery. Flowrates that fall below the limitvalue briefly are unproblematicfor the pumps and are not takeninto account (e.g. during run-ning up and running downprocesses).

In accordance with the maingoal of the reliability concept –maintaining the operation of thesystem where possible – differ-ent reaction modes can be se-lected depending upon require-ments. The requirement to pro-tect people takes precedenceover all protective and emer-gency functions.

53


Measuring Equipment

Particular care should be takenin the selection of the measuringequipment (transmitters). Theoperating reliability of an au-tomation system stands or fallsby the fault-free functioning ofthe detecting element. If the sys-tem operator uses special detect-ing elements with a high level ofsuccess for certain conditions ofuse, then the same type of detec-tors should be planned for newsystems to be constructed. If noown values based upon experi-ence are available with regard todetectors, then the supplier ofthe automation system shouldalso supply the required detect-ors. This allows the functionaland warranty problems to begreatly reduced.

In general, the following factorsshould be taken into consider-ation in the selection and fittingof detecting elements:

• Nominal operating pressure,

• Permissible max. pressure,

• Temperature limits for the am-bient temperature and thetemperature of the fluidpumped,

• Auxiliary power supply eithervia remote supply (measure-ment line) or via a local grid-fed supply device,

• Electric signal transmission fornormal applications,

• Type and maximum length ofthe measurement signal cable(number of cores, cross-sec-tion, protection type),

• Optical signal transmission forspecial conditions of use, e.g.:

large distance (1 km), environ-ment with severe EMC inter-ference, explosion protection.

The fitting location should beselected such that

• turbulence,

• air pocket formation, and

• dirt

cannot impair the measurement.

Differential pressure (pressure)

Common measurement prin-ciples:

• Piezoresistive measurementbridge on crystal membrane

• Inductive distance measure-ment of a metal membrane

For detecting elements it mustbe ensured that the sum of themaximum measuring pressureand the static system pressureremains below the permissiblemaximum pressure.

For high fluid temperatures thepressure measuring cable shouldbe sufficiently long to allow thefluid to cool off. The measure-ment connections shall bearranged such that no depositscan enter the measuring cable(e.g. laterally or at the top of themain pipe).

Measurement parameters:throughflow / flow rate


• Magnetic inductive measure-ment transducers

• Ultrasonic measurement trans-ducers

Note:

The electronics usually consistof two components, a trans-ducer with a measurement headand an evaluation device.

Minimum flow velocities arenecessary for these measurementprinciples. Therefore a continu-ous measurement right down toa flow of zero is not possible.Both the minimum and max-imum flow rate are decisive inthe selection of a flow ratetransducer.

The nominal diameter of thetransducers is often smaller thanthe pipe (lower costs, higher vel-ocities). The fitting positionshould be selected such thatthere is no air pocket formationor turbulence in the measuredsection. For magnetic-inductivemeasurement transducers, aminimum conductivity of thepumped fluid is necessary. Con-ductive deposits in the measuredsection (e.g. magnetite in circu-latory systems) can lead tomeasurement errors. The ultra-sonic measurement principle issensitive to contamination of theconveyed fluid by solids.

54


Flow monitors


• Calorimetric

• Flow paddle

Note:

Flow monitoring devices are pri-marily used as limit signal trans-mitters for monitoring and con-trol purposes (dry-running pro-tection, minimum flow detec-tion).

The simple flow paddle is moresensitive to contamination andpressure surges in the pumpedfluid.

Level detectors


• Capacitive

• Hydrostatic pressure

Note:

Level detectors that function ca-pacitively require a pumpedfluid with certain properties(high dielectric constant, pos-sibly conductivity).

They react sensitively to depositson the electrodes. In the event ofhigh levels of contamination ofthe fluid, dynamic pressuremeasurement using the bubblercontrol process has proved itselfwell.

Temperature sensors:

Measurement Principle:

Temperature-dependent resist-ance change

Note:

Submersion sensors have rela-tively long response times (slowreaction to temperaturechanges).

The design should be basedupon the planned insulation ofthe pipe (sensor length).

Documentation

The costs and level of complex-ity of the system are of decisiveimportance to the content andscope of the documentation. Forsmaller systems or manufac-turer’s standard systems, massproduced documentation is nor-mally sufficient.

For customized systems or largeprojects a description is re-quired. Particularly when plan-ning large objects it is frequentlynecessary to draw up the docu-mentation according to the plan-ning stage and project progress.

In addition to manufacturers’delivery times, the principal’sapproval phases should also betaken into account in the overallscheduling.

Installation

• Assembly (electrical andmechanical) of the set upcontrol cabinet modules takesplace in-situ.

• The laying and routing ofcables and lines for the powersupply, motors, detectors andcentralized instrumentationand control connections takesplace on-site.

The assembly of all componentsinto a functional system must becarefully planned due to the nu-merous interfaces. The followingsummary includes the most im-portant electrical tasks.

Assembly (electrical and mech-anical) of the set up controlcabinet fields in-situ takes place:

Site-supplied – by the bidder– see separate

quotation

Electrical connection of:

Type Qty. Cross-

section

Power cable …… …… ……Motor cable …… …… ……Sensor cable …… …… ……

Connection of all cables at thecontrol cabinet

• Power supply Quantity / terminals

• Motor cableQuantity / terminals

• Sensor cableQuantity / terminals

• …………………………………

55


Connection of all cable connec-tions to

• Motor Quantity / terminals ……

• Sensor Quantity / terminals ……

• ............................

Cable laying takes place

– By the customer

– By the bidder

Supply and laying in customer’scable routes

• Power supply

• Motor supply lead

• Sensor cable

Supply and laying of electricalcables including connection andfastening materials

• See separate quotation

Commissioning

The initial commissioning in-cludes commissioning and func-tional testing of the (electricallyand hydraulically) properly in-stalled system and the supply ofthe handover log.

The system must be preparedsuch that all load ranges and op-erating states can be tested.

Estimated time required:

……………………………………

• Costs in accordance withKSB’s terms and conditions forinstallation

• Costs for this are included inthe quotation price

Each additional commissioningperiod required due to circum-stances for which KSB is not re-sponsible will be subject to anadditional charge in accordancewith the attached KSB terms andconditions for installation.

Additional costs such as accom-modation, daily travelling andallowances will be charged atcost.

Extended commissioning

Test operation, instruction andoptimization

Test operation of all controlcabinets including the activationof all interlocks and protectivedevices

This also includes the perform-ance of the necessary acceptancetests and the instruction of theoperator’s operating personnel.The operator shall bear in mindthat commissioning and test op-eration cannot occur directlyafter the end of fitting and thatit is possible that not all func-tional tests can be performedsequentially.

The costs will be charged on thebasis of time and expense in ac-cordance with KSB’s terms andconditions for installation.

Training of operating personnel

Complex systems or those withhigh availability requirementsrequire well trained operatingpersonnel. The following train-ing points are important for asafe operation of the system:

Process interactions and proced-ures, system operation, reactionto fault, fault detection andrectification. This training al-ways takes place after the com-missioning and acceptance ofthe system.

56


3Project Planning Examples

3.1System Description

The system is a district heatingnetwork. It comprises 26 heattransfer stations with a differen-tial pressure requirement of18 m at the transfer points. Theheat transfer stations are con-nected via heat exchangers. Theprimary side output is adjustedby atmospheric conditions via athrottle fitting. The district heat-ing system is designed for a sup-ply temperature of 130 °C witha return temperature of 80 °C.The planned new construction isbased upon a maximum heatoutput of 47 MW. The flow rateof 861 m3/h is calculated fromthe maximum values of heatoutput and temperature differ-ential.

At this flow rate the piping sys-tem calculation results in a pres-sure loss of 24 m to the most re-mote heat transfer station.

Due to the differential pressurerequirement of the heat transferstations of Hw = 18 m and themaximum line losses of HT =24 m a pump head of HN = 42 mmust be met (provided).

57

3

Boiler

Fig. 73 System diagram

3.2Calculation of the Piping Characteristic Curve

(see also page 12 Fig. 14)

Once the nominal flow rate(QN) and line losses (HT) havebeen determined the pipingcharacteristic curve can be con-structed.

Qx / QNHx =2

HT · ( )

Qx / 861 m3/h= 2100 % · ( )Hx

H [m]

Q [m /h]3250

50

4240

3025

18

105

430.5Q =N/2 Q =N

H =N

H =w

500

Piping characteristic curve(full load)

750861

1000

Fig. 74 Piping characteristic

Given Sought

Qx [m3/h] Hx [m]250 2500 8.1750 18.2861 24

Project Planning Examples

3.3Further Steps in AccordanceWith the “Project Planning Se-quence Plan”

(please refer to page 71)

58

3 Project Planning Examples

�

Determine controlled quantity

Piping system closed

Consumerwith throttling

behaviour

Possible controlled quantities– differential pressure– . . . . . . . . . . . . . . . . . .

Controlled quantity selected– differential pressure

Calculation of the controlled-operation curve (see 1.1.3)

Detailed explanations on page 50

“Piping system closed” Closed circulatory heating system

Output adjustment in heat transfer stations by means of externaltemperature controlled supply temperature in the consumer circuit

Since the consumer behaviour determines the system resistance, thedifferential pressure is the correct controlled quantity.

Controlled-operation curve or required pressure path

H [m]

Q [m /h]3250

50

4240

3025

18

105

430.5Q =N/2 Q =N

H =N

H =w

500 750861

1000

Fig. 75

Hx = (HN - HW) · (Qx / QN)2 + HW

= (42 m - 18 m) · (Qx / 861 m3/h)2 + 18 mHx

Given Sought

Qx [m3/h] Hx [m]0 18

250 20500 26.1750 36.2861 42

Due to the very extended system train a differential pressure meas-urement at the point of lowest pressure is not realizable (costs, worstpoint determination).

Since a flow rate measurement is available in any case (from energymetering), the differential pressure in the boiler house is also meas-ured. The use of DPC, which is available in the modern KSB controlsystems (see page 40), provides a good alternative to measurement atthe point of lowest pressure.

Due to the output data and the typical load behaviour (frequent op-eration at part load), the total flow rate should be split between twopumps.

For reliability reasons a third, equally sized pump is installed (pumppower: 3 x 50 %). The stand-by pump is of course fully integratedinto the automation concept (emergency changeover, alternatingpump operation).

Full load point (new system design)

The nominal flow rate is divided between two identical pumps. Forthis nominal flow rate (QN) (see Fig. 76) a nominal head (HN) is re-quired. For the individual pumps (P1), this means that at half thenominal flow rate (1/2 · QN) they must still achieve the nominalhead (HN).

Part load operation

In part load operation care should be taken to ensure that the re-spective operating points of the variable speed base load pump lie onthe so-called controlled-operation curve. The controlled-operationcurve is specified such that at least the specified pressure is achieved.For reliable operation of the pump it is important that the given

59

1

�

�

Specify measurement location


Flow rate split between one ormore pumps

With / without stand-by pump

Pump selection

At the pumpwith set valuereadjustment

Atthe

pump

Inthe

system


Measurement location selected:at the pump with DPC

The next step is to determine therequired differential pressurecurve.

To prevent undersupply at anyof the heat transfer stations theminimum differential pressuremust be 18 m. This means thateven at minimum consumption ≅ 0 m3/h this minimum pressure

Hw ≅ 18 m must be maintained.As the flow rate increases thepiping losses are added to thisvalue (see also piping character-istic curve).

The origin of the parabola ismoved to the level of the setvalue by a small expansion ofthe formula.

pump characteristic curve (at nominal speed) cuts the controlled-operation curve. In the example, this would be the point B1,max.

From the relationships described for part load operation it is clear thatthe individually operated base-load pump requires its maximum shaftpower (Pw, max) at the point B1, max. This value can be read off thepump diagram in question or calculated using the power formula.

Caution! The maximum shaft power of the base-load pump isgreater than is the case for QN2.

The shaft power of the motor to be selected must be at least as greatas the max. shaft power of the pump (Pw, max) plus safety factors.The safety factors take into account tolerances in the characteristiccurves plus additional losses in the motor due to the frequency in-verter. A 5 % safety factor is used in these calculations. Differentsafety factors are recommended depending upon the frequency in-verter type and size.

For our example:From the power diagram: Pw, max = 78.2 kWPower safety factor 5% = 3.9 kWRequired motor power: P2 = 82.1 kWNext possible standardized motor size = 90 kW

In Fig. 77 all important results of the project planning are broughttogether. At this point the hydraulic selection of the pumps and thespecification of the operating behaviour in variable speed operationis concluded.

60


�

�

Specify required motor power

Determine shaft power

H [%]

HW

120

100

80

60

40

20

0200 40 60 80 100

BI,N

BNBI,max

QI,maxQN/2


Q [%]

HN

QN

Pump 1

Fig. 76 H/Q diagram, schematic representation

A range of further conditions and requirements remain to be takeninto consideration in the complete project plan. These are listed inthe tender text for the system components.

For the pumps: • Fluid type• Materials• Temperature• Pressure, etc.

For the control system: • Electrical equipment features• Mechanical design, etc.• Control technology interface

Special system requirements: • Necessary safety precautions– Emergency stops– Overpressure protection devices–Emergency power supply, etc.

• Process conditions– Starting process according to time

program– Control quality (tolerances)– Manual intervention option, etc.

61

3

�

�

Special requirements

H [m]

P [kW]

1.2

1.0

0.8

0.6

0.4

0.2

0

2.2

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0

140

130

120

110

100

90

80

70

60

50

40

30

20

10

0.2

0.2

0.4

0.4

0.6

0.6

0.8

0.8

1.0

1.0

1.2

1.2

1.3

1.3

1.4

1.4

1.6

1.6

Controlled-operation curven ~ 100 %

n ~ 90 %

n ~ 80 %

n ~ 70 %

n ~ 60 %

Parallel operation P1 + 2(n )N

P1 + 2

P1

Power consumption invariable speed operation

Throttled operation

Throttled operation

1.8

1.8

2.0

2.0

2.2

2.2

2.4

2.4

2.6

2.6

0

0

250 500 750 1000

Q/Qopt

Q/Qopt

Q [m /h]3

P/P opt

PW,max

Saving

H/Hopt

Qmax

P1(n )N

4240

3025

18

105

H =N

H =w

opt

430.5Q =N/2 Q =N 861

Fig. 77 H/Q diagram and power characteristic curve


Automatic pump control brings a technical system into the requiredoperating state (e.g. demand-dependent). The system should monitoritself, avoid critical operating states and put itself into a safe state inthe event of faults.

The general method of functioning and the interaction of the indi-vidual components and assemblies are represented in the system de-scription.

The tender text specifies the required system, operation and powerdata. In addition, the tender text includes a precise technical specifi-cation of the devices and components used and, where applicable,the desired commercial conditions.

For economy considerations, the electrical power requirement oftwo fixed speed pumps is compared with the power requirement invariable speed operation. The latter system operates with a variablespeed pump and a fixed speed peak load pump. The district heatingsystem that underlies this example has consumer installations withthrottling characteristics.

A comprehensive description of an economy calculation study is in-cluded in the fundamentals section (Chapter 1.1.4). Furthermore,the calculation of the pump characteristic curves for operation withspeed adjustment is also described in this section.

Procedure:

The power consumption characteristic curves in controlled oper-ation are determined graphically using the H/Q diagram and thegiven controlled-operation curve (Fig. 80).

For our practical example the following permissible simplificationsare made:

• The saved shaft power Pw is set equal to the saved electric effectivepower

• The various switching limits for the peak load pump in fixed andvariable speed operation are not taken into consideration

• Using the load profile, certain load states are assigned annualoperating times and in the power diagram the associated powerconsumption savings.

A standardized load profile can be found in the approval specifica-tions for the German environmental “Blauer Engel” label for heat-ing circulators (RAL-UZ105)

62


�

�

System descriptionTender texts

Economy calculation

63

3

H [m]

P [kW]

140

130

120

110

100

90

80

70

60

50

40

30

20

10

0


n ~ 90 %

n ~ 80 %

n ~ 70 %

n ~ 60 %


P1 + 2

P1

Power consumption invariable speed operation

Throttled operation

Throttled operation

250

250

500

500

750

750

1000

1000

Q [m /h]3

Q [m /h]3

PW,max

Saving

Qmax

P1(n )N

4240

3025

18

1050

H =N

H =w

opt

430.5

430.5

Q =N/2

Q =N/2

Q =N

Q =N

861

861

[∆h]

Q [%]

Nominal flow rate

Minimum flow rate

Ordered annual load duration curve

Op

erat

ing

ho

urs

Stan

dsti

ll

1000

1000

1000

1000

1000

1000

1000

1000

760

0 20

6400 h

8760 h

40 60 80 100 120

Fig. 78 Characteristic curves for the economy calculation

�

�


The results are summarized in the economy calculation table (seebelow). The following influencing factors are relevant to the calcula-tion of payback periods:

• New system/old system (modernization)• Competence of the switchgear• Company-specific calculation methods• Price of electricity inc. additional costs

Depending upon the influencing factors a payback period between1.8 and 2.9 years should be expected.

64


Key:∆PE : Saved electrical powerB : Operating hoursh/a : Hours per yearS : Electricity costs∆EE : Saving on electricity costs∆EE = ∆PE · B · S

∆PE B S ∆EEkW h/a Euro/kWh Euro/a

19 1000 0.10 1,900.--15 1000 0.10 1,500.--21 1000 0.10 2,100.--26 1000 0.10 2,600.--26 1000 0.10 2,600.--28 1000 0.10 2,800.--

25.5 400 0.10 1,020.--

Σ 14,520.--

Economy calculation

�

Result:

Under the assumed conditions, annual electricity

cost savings of approx. € 14,500 can be

expected for the circulatingpumps.

Depending upon the influencing factors a paybackperiod between 1.8 and 2.9

years can be expected.

Reasons for Pump Automation and Control4

4Reasons for Pump Automationand Control

4.1Operational Reliability

a) Protective measures e.g. • Pump changeover for equaldistribution of pump operatinghours

• Ensuring minimum pumpflow rate

• Monitoring of pump charac-teristic diagram to avoid imper-missible operating states

Benefits:• Higher system availability

b) Fault changeovers e.g. •Changeover to– stand-by pump– stand-by frequency inverter–mains operation

Benefits:• The pumping task continues to

be fulfilled

c) Monitoring e.g. •Limit values for– lack of water– temperature– controlled quantity•Switching off the system

Benefits:• Protecting the system against se-

vere damage

4.2Improving Operating Behaviour

a) Holding process data constant e.g. •Differential pressure in dis-trict heating systems

•Pressure in pressure boostingsystems

•Level in sewage works•Flow rate in water treatment

Benefits:• Process optimization ensures

uniformly high quality

b) Reducing pressure surges e.g. •In water supply systems Benefits:• Higher operating reliability• Reduction of shocks, noises and

material destruction

c) Reduced switching frequency e.g. •In water supply systems Benefits:• No “fluttering”

d) Reduced flow noise e.g. •In heating element thermo-static valves

Benefits:• Greater comfort

65

4

66

4

4.3Increasing Product Quality

a) In machine tools e.g. •Constant pressures in cooling lu-bricant systems

•Lower transfer of heat from thepump into the cooling lubricant

Benefits:• More accurately dimensioned

workpieces

b) Mill trains e.g. •Adapted flow rates and pressuresat the nozzles

Benefits:• Higher quality steel profiles

4.4Reducing Operating Cost / Life-Cycle Cost

a) Effective requirement optimization

e.g. •Adaptation of the pump output tothe requirement profile of the sys-tem

Benefits:• Reduced investment cost• Saving in electricity costs

b) Reduced drive powercosts

for •Over-dimensioned pump output•Systems with mainly low-consump-

tion operating times•High motor power

Benefits:• Saving in electricity costs

c) Protection of systemcomponents

e.g. •At throttling elements, pipelinesand pumps

Benefits:• Saving in maintenance costs

d) Reduced energy loss ofthe system

e.g. •In district heating the overflowvalve seldom trips in, therefore lessheat is lost into the earth

Benefits:• Saving in fuel costs

e) Modified flow rates for •Water shortage by reduction of theset value characteristic curve

Benefits:• Protection of water reserves

4.5Improving System Information

a) Pump operating data e.g. •Capture•Summarization•Assignment•Evaluation•Display

Benefits:• Determining weak points• Information on operating se-

quences• Optimization of the system• New findings that feed into plan-

ning of new systems

b) Process information e.g. •Evaluation of sensors•Storing of measured values, fault

data, etc.•Operating statistics•Fault detection and diagnosis•Trend recognition

Benefits:• Reduction of the inspection and

servicing cost• Early detection of damage

c) Data transmission via bus system

e.g. •Start / stop•Set value / actual value•Fault•Status

Benefits:• More information• Simpler transmission• Simpler processing

Reasons for Pump Automation and Control

5An Overview of AutomationConcepts

Depending upon the pumpingtask requirement and the operat-ing conditions to be observed,various electrical and hydrauliccircuit concepts may be the mostfavourable solution. These rangefrom one variable speed pumpto several equally or differentlysized pumps with one or morefrequency inverters. Hydraulic-ally, the options range frompumps connected in parallelthrough pumps connected inseries to a combination of both.

Figs. 79 and 80 show anoverview of the most commoncircuit options. Some parallelconfigurations are briefly de-scribed on the following pages.

67

5

Fixed speed Variable speed

1Frequency

inverter

2Frequencyinverters

Severalfrequencyinverters

Fig. 79 System diagram “Parallel connection of centrifugal pumps”

P1,2

H1,2

H2

Qmax,2

Qmax,1,2

Qmax,1,2

Qmax,1H1,2

H1

P2

P2

P1,2

P1

P1

Fixed speed Variable speed

1Frequency

inverter

2Frequencyinverters

Fig. 80 System diagram “Series connection of centrifugal pumps”

An Overview of Automation Concepts

5.1Parallel Connection of IdenticalPumps with One Frequency Inverter(one pump in variable speedoperation on an alternatebasis)

The flow rate is split betweenseveral equally sized pumps, re-sulting in:

• Good adaptation to demand

• Simple layout with regard to

– electrical system

– control technology

– hydraulic system

Each pump unit can be operatedboth through a frequency in-verter and on the 50 Hz grid.This results in:

• Operating reliability

• Equal distribution of operat-ing hours by alternating thepumps in variable speed oper-ation

• Fixed speed peak load pumpsprovide a cost-effective in-crease of the flow rate

• Steady controlled-operationcurve even in parallel oper-ation with fixed speed pumps,since the pump system pres-sure is determined by thevariable speed pump

If the pump output is reduced –by reducing the speed – only agreatly reduced shaft power isrequired. This effect is alsoachieved after peak load pumpshave cut in.

68

5 An Overview of Automation Concepts

M5M4M3M2

Set value

Grid

Process

Transmitter

Automationdevices

Powercomponents

Pump units

Frequencyinverter

Pressuretransmitter

M1

Fig. 81 “Parallel connection of centrifugal pumps”

H [%]

P [%]

120

100

80

60

40

20

0

240

220

200

180

160

140

120

100

80

60

40

20

0

0.2

0.2

0.4

0.4

0.6

0.6

0.8

0.8

1.0

1.0

1.2

1.2

1.3

1.3

1.4

1.4

1.6

1.6


n ~ 90 %

n ~ 80 %

n ~ 70 %

n ~ 60 %


P1 + 2

P1

Variable speed operation

Throttled operation

Throttled operation

1.8

1.8

2.0

2.0

2.2

2.2

2.4

2,4

2.6

2.6

0

0

Q/Qopt

Q/Qopt

Saving

Qmax

P1(n )N

opt

Fig. 82 Power diagram for the “parallel connection of centrifugal pumps”

5.2Parallel Connection of IdenticalPumps with Two FrequencyInverters(2 pumps in variable speed op-eration on an alternate basis)

The flow rate is split betweenseveral equally sized pumps.Each pump unit can be operatedon each of the two frequency in-verters and on the 50 Hz grid.

This results in:

• High operating reliability

• Starting current of the variablespeed pumps limited tonominal current.

• Electrically and hydraulicallysoft pump changeover pos-sible.

Due to the option of operatingtwo pumps by means of two fre-quency inverters, the followingmain hydraulic advantages areobtained:

• Greatly increased adjustmentrange

• The pump limit curve Qmax P1is doubled to Qmax P1+P2

• Soft hydraulic running up andrunning down of the variablespeed pumps and greatlydamped operating behaviourof the fixed speed pumps.

Further savings are possiblecompared to operation with justone frequency inverter. Themain reasons for this are:

• Greater adjustment range, e.g.greater utilization of inletpressures

• Better part load efficiencywith more than two operatingpumps.

69

5

M5M4M3M2

FI I

FI II

M1

Set value

Grid

Process

Transmitter

Closed-loop control

Powercomponents

Pump units

Pressuretransmitter

Open-loop control

Fig. 83 System diagram “Parallel connection of centrifugal pumps”

1.21.11.0

0,8

0.4

0

2.4

2.2

2.0

1.8

1.6

1.4

1.2

1,0

0.8

0.6

0.4

0.2

0

0.2

0.2

0.4

0.4

0.6

0.6

0.8

0.8

1.0

1.0

1.2

1.2

1.3

1.3

1.4

1.4

1.6

1.6

P1 + 2(n )N

P1 + 2

P1

P1 - 3(n )N

P1 + 2

P1 + 2

System curve

1.8

1.8

2.0

2.0

2.2

2.2

2.4

2.4

2.6

2.6

0

0

Q/Qopt

Q/Qopt

Qmax

Qmax

H/Hopt

P/P opt

P1(n )NP1(n )limit

P1


Throttled operation

Throttled operation

Saving




5.3 Parallel Connection ofNon-Identical Pumps(of 3 main pumps and one ortwo base-load pumps; onepump of each group with vari-able speed option)

The total flow rate is divided be-tween a low-load pump and sev-eral main load pumps. Such sys-tems are usually used in installa-tions with sharply fluctuatingconsumption. This applies bothto systems with consumptionfluctuations based upon the timeof year, for example, and also tosystems with frequent consump-tion changes in a daily workcycle.

The system can be designed suchthat an identical stand-by pumpis available for the low-loadpump. For moderate reliabilityrequirements, this pump can bedispensed with. The supply isthen taken over by the mainpumps in emergencies.

The low-load range is coveredby a small pump. As a result, thefollowing should be achieved:

• Better pump efficiency

• Minimum flow rate to themain pumps ensured

• Reduction in switching fre-quency at low load

70

5

M5M4M3M2

Process

FI I FI II

M1

Set value

Grid

Transmitter

Closed-loop control

Powercomponents

Pump units

Pressuretransmitter

Open-loop control

Fig. 85 System diagram “parallel connection of centrifugal pumps”

H [%]

P [%]

120

100

80

60

40

20

0

240

220

200

180

160

140

120

100

80

60

40

20

0

0.2

0.2

0.4

0.4

0.6

0.6

0.8

0.8

1.0

1.0

1.2

1.2

1.3

1.3

1.4

1.4

1.6

1.6


Base load pump

n ~ 100 %

n ~ 90 %

n ~ 80 %


P1 + 2

1.8

1.8

2.0

2.0

2.2

2.2

2.4

2.4

2.6

2.6

0

0

Q/Qopt

Q/Qopt

Variable speed base load pump

Fixe

d s

pe

ed

ba

se l

oa

d p

um

p

Qmax

P1(n )N

opt

P1


Throttled operation

Throttled operation

Saving



5.4Further Electric ConfigurationConcepts from the KSB Prod-uct Range

The optimal solution, from acontrol point of view, is one inwhich each pump/motor is as-signed a frequency inverter. Thedisadvantage of this is the higherequipment costs and a greaterspace requirement. For certainapplications (e.g. district heatingwith two supply and returnpumps each) this is the best so-lution.

71

5

M4M2M3

Grid

FIs

M1

P1 P3 P2 P4

Fig. 87 Each pump is assigned a frequency inverter


72

yes

Determine controlled quantity

Closedpiping system


behaviour

no (open)

Possible controlled quantities Differential pressure. . . . . . . . . . . . . . . . .

Possible controlled quantitiesPressure. . . . . . . . . . . . . . . . .

Possible controlled quantities Temperature– Supply temperature– Return temperature– Differential temperature– . . . . . . . . . . . . . . . .

Possible controlled quantities TemperatureLevel. . . . . . . . . . . . . . . .


. . . . . . . . . . . . . . . . . . . .


Measurementlocation

Neargenerator

At the pump with set value tracking

Nearconsumer installation


. . . . . . . . . . . . . . . . . . . .



Inthetank

Neargenerator

Nearconsumer installation


. . . . . . . . . . . . . . . . . . . .



At the

pump

At the pump with set value tracking

In the

system

Measurement location selected:

. . . . . . . . . . . . . . . . . . . .

Split flow rate between one or more pumps

With / without stand-by pump

Pump selection

Determine shaft power

Specify required motor power Special requirements

System descriptionTender texts

Economy calculation

yes yes

no (constant) no (constant)

Level Temperature


behaviour

Overview of Project PlanningSequence

Fax order form “KSB know-how” series

KSB Know-howVolume 01

Water Hammer

KSB Know-howVolume 02

Boa-Systronic

KSB Know-how

SelectingCentrifugal Pumps

At your request, we will be pleased to send you all “KSB know-how” brochures previously published as

well as any volumes to be published in the future. All we need is your address and confi rmation below.

Company address or stamp:

Company:

Attn.:

Street address:

Post or ZIP code / City / Country:

Please send me the following technical brochures: (tick where applicable)

... copy and fax this form to:

Fax: +49 (62 33) 86 34 39

KSB Aktiengesellschaft

Johann-Klein-Straße 9

67227 Frankenthal

Tel.: +49 (62 33) 86 21 18

Fax: +49 (62 33) 86 34 39

www.ksb.com

KSB Aktiengesellschaft • Bahnhofplatz 191257 Pegnitz (Germany) • www.ksb.com

2300

.024

-10

11/

06

k

ub

bli

Sub

ject

to

tec

hn

ical

mo

dif

icat

ion

wit

ho

ut

pri

or

no

tice

W e l o o k f o r w a r d t o h e a r i n g f r o m y o u .

Your local KSB representative:

Pump Control/System Automation - KSB

Documents

Transcript of Pump Control/System Automation - KSB