We reserve all right in this document and in the information contained therein.

Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

ABB Switzerland Ltd.

UNITROL 1000 AVR TOOL

Date / prepared: 18.05.2012/ Ralf Bachmann Schiavo Date / approved: 23.05.2012. / Rudolf Moeckli
Responsible division: ATPE Document No. Part No. Language Rev. Pages

 ABB Switzerland Ltd 3BHS399489 E01 E B 12

We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

TABLE OF CONTENT
1 USING THE UNITROL 1000 AVR TOOL ..................................... 3
1.1 Settings ............................................................................................................... 3
1.2 Input parameters ................................................................................................. 3
1.2.1 A note on the PID parameter tuning for no-load
operation ............................................................................................. 5
1.3 Results ................................................................................................................ 5
1.4 Analysis ............................................................................................................... 5
1.5 On site verification and tuning ............................................................................ 6
1.5.1 No-load MANUAL tuning .................................................................... 6
1.5.2 No-load AUTO tuning ......................................................................... 7
1.5.3 Parallel to grid AUTO tuning ............................................................... 8
1.5.4 Parallel to grid PF/VAR tuning ............................................................ 9

2 INPUT VALUES VERIFICATION ............................................... 10

2.1 Te ...................................................................................................................... 10
2.2 Xd and XT+Xext ................................................................................................ 11

3 REFERENCES ........................................................................... 12

Document Nr. Lang. Rev. ind. Pages 2

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

1 Using the Unitrol 1000 AVR tool

This tool provides initial settings for Unitrol 1000 AVR PID regulator.

1.1 Settings
Configurable settings for PID parameters calculation:

 Kb: Derivative gain (default value: 3),

 Max.: plotting time – response plotting time,
 Xext: Grid impedance (p.u.) on the base of the generator,
 cos(phi): nominal power factor of the generator.
 Gain reduction: Vp reduction factor (values < 1 generate
conservative settings).

1.2 Input parameters

The parameters of the machine must be introduced in the tab “Input
data”:

Document Nr. Lang. Rev. ind. Pages 3

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

 Xd: Unsaturated synchronous reactance on the d-axis (p.u.),

 Xd’: Unsaturated transient reactance on the d-axis (p.u.),
 Xq: Unsaturated synchronous reactance on the q-axis (p.u.),
 Td’: Short-circuit transient d-axis time constant (s),
 Te: Exciter machine time constant (s),
 XT: Reactance of step-up transformer (p.u.).

Typical values of the open-circuit time constant are given by [1]:

Time Turbogenerators Hydrogenerators
constant Low Avg. High Low Avg. High
0.4 1.1 1.8 0.5 1.8 3.3

If only open-circuit time constants are given for the machine, the short-
circuit time constant can be estimated as follows:

Typical characteristics of generators according to [1] are shown in the

table below:
Parameter Turbogenerators Hydrogenerators
Small units Large units
Low Avg. High Low High Low High
Rating (MW) 300 - 1000 0 40 40 200
1.4 1.6 1.8 0.9 1.1 0.8 1.0
0.23 0.25 0.35 0.25 0.45 0.20 0.40
1.5 1.55 1.6 - - - -

Document Nr. Lang. Rev. ind. Pages 4

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

1.2.1 A note on the PID parameter tuning for no-load operation

Sometimes it is desirable to tune the PID parameters for no-load
operation. In such cases the following work-around is proposed.

1. Set XT = 0 [pu] and the other parameters to the actual values

2. Navigate to the "Settings" tab and set the parameter "Gain
reduction" to 0.65 [pu].
3. Press the "Calculate" button to calculate the PID settings.

1.3 Results
Once the parameters are introduced, the button “Calculate” can be used
to perform the calculation. The resulting parameters are shown in the
tab “Results”.

Remark:
Vp can be adjusted and its influence on the overshoot can be checked
manually in the tab “Analysis”. If the user desires to restore the optimum
Vp calculated by the software, then the button “Restore optimum” can
be used.

1.4 Analysis
The tab “Analysis” shows the expected response of the controlled
system when tuned with the parameters given in the tab “Results”.

Document Nr. Lang. Rev. ind. Pages 5

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

Right click on the graph for the various options, such as zooming or un-
zooming, saving the plot as a figure, or showing the point numerical
values.

1.5 On site verification and tuning

1.5.1 No-load MANUAL tuning

As a first step, the MANUAL regulator should be tuned. The following
procedure can be followed:
1. Introduce “conservative” settings in the “Tune Manual/Ie Limiter”
window of the CMT (with low proportional gain Vp).

2. Operate the generator offline, at reduced voltage (e.g. 50%), in

MANUAL control mode.
3. Apply low amplitude steps (to make sure that overvoltages are
avoided!). Measure the response (Ie) with the oscilloscope of the
CMT. Adjust Vp and Ta until the desired response of Ie is
obtained. Example:

High overshoot
reduce Vp or increase Ta

Document Nr. Lang. Rev. ind. Pages 6

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

Slow response
increase Vp or reduce Ta

Desired response

1.5.2 No-load AUTO tuning

The response in the generator can differ from the expected response
shown in the tab “Analysis”. A fine tuning of the PID regulator should be
performed:
1. Introduce the initial settings in the “Tune Auto” window of the
CMT.

2. With the generator disconnected from the grid, apply a 1% step

using the “Setpoint Adjust” window of the CMT. Measure the
generator voltage overshoot. Larger voltage steps are not
recommended for tuning purpose, because they might cause
internal PID limiters to become temporarily active.
3. Vp can be adjusted until the desired performance is obtained
(10% overshoot is recommended). This is exemplified in the
table below:

High overshoot – reduce Vp

Document Nr. Lang. Rev. ind. Pages 7

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

Slow response – increase Vp

Recommended response
(10% voltage overshoot)

1.5.3 Parallel to grid AUTO tuning

The AVR should now be regulated considering the response of the
loaded generator.
1. Operate the generator in AUTO mode connected to the grid,
loaded below nominal values (e.g. P=40%, Q=30%).
2. Apply small voltage steps (1% or 2%) and measure the
generator voltage (or the reactive power Q, since both will show
the same response). Adjust the proportional gain Vp as to obtain
a fast response with no overshoot. Example:

Overshoot – reduce Vp

Slow response – increase Vp

Desired response

Attention:
In case AUTO tuning is changed compared to
Noload conditions, Noload response must be
tested again.
 Very high overshoot in Noload conditions might
occur.

Document Nr. Lang. Rev. ind. Pages 8

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

1.5.4 Parallel to grid PF/VAR tuning

The PF/VAR regulator should now be tuned considering the response of
the loaded generator.
1. Operate the generator in PF mode connected to the grid, loaded
below nominal values (e.g. P=40%, Q=30%).
2. Apply PF steps (e.g. 0.05) and measure the reactive power Q.
Use the same Ta as for the AUTO regulator. Adjust the
proportional gain Vp as to obtain the desired response.
Example:

Large overshoot – reduce Vp

Slow response – increase Vp

Desired response

Remark:
As starting point for Ta use same as AUTO Ta.

Document Nr. Lang. Rev. ind. Pages 9

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

2 Input values verification

Simple test procedures can be used to validate some of the parameters
used for calculating the PID settings.

2.1 Te
This procedure can be used to validate the exciter time constant:
1. Operate the generator at no load and 50% of nominal voltage.
2. Select open loop control mode.
3. Measure at 50% of nominal voltage.
4. Apply and measure voltage with datalogger
for better accuracy.
The time it takes for Ie to reach 63.2% of its final value is Te.

Te=0.2s

Signals:
white: PWM,
orange: Excitation current Ie.

This procedure can be carried out with the generator at stand-still as

well as rotating.

Document Nr. Lang. Rev. ind. Pages 10

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

2.2 Xd and XT+Xext

This procedure can be used to measure the synchronous d-axis
reactance and the external reactance . The machine has to be
connected to the grid, and the following steps should be followed:
1. In AUTO mode, operate the machine synchronized with the grid,
with active power and reactive power . Measure
, the generator voltage and the generator current
in steady state.
2. Increase the setpoint by 5%. Measure , , and
in steady state.
3. Calculate using the expression below:

( ⁄ )

where the values are in per unit (100% = 1 p.u.).

4. Calculate :

where the values are in per unit (100% = 1 p.u.).

Document Nr. Lang. Rev. ind. Pages 11

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12
 We reserve all right in this document and in the information contained therein. Reproduction, use or disclosure to third parties without express authority is strictly forbidden.
 Copyright 2012 ABB. All rights reserved.

3 References
[1] P. M. Anderson and A. A. Fouad, Power System Control and
Stability, John Wiley & Sons, 2003.

Document Nr. Lang. Rev. ind. Pages 12

ABB Switzerland Ltd 3BHS399489 E01 E B No. of p. 12