DG Sizing of Generator
DG Sizing of Generator
DG Sizing of Generator
This Document and all contained herein are proprietary of ROLTA INDIA LIMITED and is subjected to confidentiality restrictions
between ROTLA INDIA LIMITED and the Recipient. Copyright Reserved.
Subject Page no
8. EMPIRICAL FACTORS USED IN SIZING FOR MOTOR LOADS.................................................7
9. LOAD ANALYSIS...........................................................................................................................9
12 REQUIRED CALCULATION INPUTS........................................................................................15
13 REQUIRED CALCULATION OUTPUTS....................................................................................15
This design guide for Diesel – Generator Set (DG SET) covers the following aspects.
Standard DG ratings and voltage levels
Sizing of DG set
Parallel operation with the grid
Neutral earthing
2. Applicability
3. References
DG sets upto 1000 KVA are rated for 415V and those above 1000 KVA are rated for 3.3 / 6.6 /
415V : 10 / 25 / 50 / 75 / 100 / 125 / 150 / 200 / 250 / 300 / 400 / 500 / 650 / 750 / 800 / 1000 KVA
3.3/6.6/11 KV : 1460 / 2200 / 3000 / 5000 / 6500 / 8000 / 12000 KVA
When DG sets are used for continuous duty i.e. more than 6000 hours per year, it is recommended
that engine of either low speed or medium speed are selected for getting high service life. DG sets used for
emergency purpose can be of high speed type.
The above considerations are true only for the conventional loads like pump / fan motors, heaters,
lighting etc. However sizing of DG for feeding non-conventional loads like Arc Furnace and Rolling Mills is
very complex and not considered here.
6. DEFINATIONS:
The most common type of industrial load characterized by a high initial starting kVA at low P.F. and a
fairly high peak kW during run-up. Motor output usually expressed in kW so that:-
Columns (1) to (6) of Fig. 1 list typical full load performance figures for induction motors from 1.1 to
1000kW output.
The method used for starting squirrel cage motors is most important. The simplest method id D.O.L.
(direct-on-line) starting at full voltage and a typical motor would drawn an initial starting kVA of about
6.5 times the full load value, and a possible peak input kW during run-up of about 3.25 times full load
value. A motor supplied from a generating set and started in this way would probably produce serious
transient voltage dip problems and might, under some conditions, require an oversize engine to power the
run-up.
To reduce the demand on the supply during starting, cage motors frequently use different forms of reduced
voltage starting, the two most common being start-delta and auto-transformer starting. Both the initial
kVA and the peak kW are substantially reduced but the motor develops reduced torque, which prolongs
the run up time. Columns (7) to (12) of Fig. 1 show initial start kVA and peak input kW for typical cage.
motors with three methods of starting, D.O.L., auto-transformer with 75% tap and start-delta.
Used where high starting torque is required with a moderate starting current. Starting involves full voltage
applied to stator and series rotor resistance which is progressively cut out during run up. Starting kVA is
typically between 1.25 and 2.0 times rated input and starting power factor is high, typically 0.9 so that
generating set sizing considerations usually centre on engine power rather than transient voltage dip.
Columns (13) to (18) of Fig.1 show the kVA and kW input to slip ring motors during starting for three
starter settings 1.25, 1.5 and 2.0 times full load input. Rated load input is as for cage. motors and can be
taken from columns (1) to (6).
voltage control instability may also occur. Because of this, generators supplying non-linear loads should
be generously rated so that the kVA rating should be at least twice the full load kVA input to the rectifier
or thyristor.
Regenerative Loads
Cranes and hoists sometimes use electrical regeneration as a means of braking. The effect of returning
power to a generating set is to reduce the engine power requirement and produce an increase in speed.
Small levels of regeneration will be absorbed in the losses of the set but if it becomes appreciable, the
speed may rise to an unacceptable extent since the governor can have no effect in limiting the speed under
such conditions. To guard against this the net regenerated power should not exceed 20% to 25% of the
generating set kW rating, the lower figure for sets above 200 kW, the larger figure for small sets of say 20
kW rating. If this cannot be achieved then a ballast load should be connected across the generator
(activated by a reverse power relay) intermittently rated but capable of absorbing the maximum
regenerated kW’s.
Capacitive Loads
Although not normally a factor in sizing generating sets, mention should be made of the effect of
capacitive loads on generating sets. Such loads may result in an overall leading power factor condition
which tends to self excite the generator and in extreme cases may lead to excess terminal voltage which
the AVR cannot control. Special consideration must therefore be given to such cases.
V = Pu
For slip ring motors the starting kVA is restricted by the external starter resistance in the rotor
= FLkVA x K x E
a. The value given to E will depend on the application and particularly the motor load inertia. For
squirrel cage motors the maximum value will be 0.5 and other values can be found with the aid of
Fig.2.
For slip ring motors maximum kW occurs at the instance of starting and to allow for the high starting
power factor a value of 0.9 should be assigned to E for this class of machine. The same value can be used
for squirrel cage motors using special “soft start” techniques. where the run of characteristics obtained
from thyristor controls is not unlike that of a slip ring motor.
9. LOAD ANALYSIS
Figure 3 shows a tabulation, which is the analysis of a load comprising a number of separate elements. At
the top of the sheet the load elements are itemised in the order in which they will be applied to the
generating set. In the absence of other data the first item should be any passive load such as heating or
lighting followed by the dynamic loads (i.e. motors) starting with the machine which imposes the largest
starting load and working down in descending order. This is only a general rule and some re-arrangement
of order may prove advantageous after completing the first assessment.
The next lower section of the load analysis table lists the demand on the set as the successive load
elements are added.
At each application the initial kW is shown, the effective peak kW during run up of the motor and the final
kW after the load element is up to speed and running at its stated load. The sequence is repeated until all
the elements have been added as shown for the sample installation in Fig. 3. At the bottom of the table are
shown three key figures which will be used to determine the recommended plant.
Maximum Total Peak kW
This is the highest value in line reference 8 of the table and represents the maximum load the engine must
handle during the process of connecting the load elements.
Maximum Final kW
This is the final figure in line reference 9 of the table and is the expected total load when all the load is
connected and running at stated value. Normally this figure would represent the plant kW rating.
After analysing the load as described above it is worth checking to see whether a different sequence of
applying the load elements would be beneficial. Clearly it cannot affect the final kW but it could result in a
lower maximum total peak kW which frequently determines engine size.
XL = Pu load reactance
RL = Pu load reactance
Total circuit
impedance
Z = ZL + X (added vectorially)
I = E = 1.Pu
ZZ
VT = IZL.Pu
Transient Voltage
Dip,
= E – VT = 1 – VT.Pu
KVA = P * LF
PF * η * DF
Where,
P = Connected KW on DG
remains within the new steady state speed band. For G1 governing
8 seconds maximum for the load changes defined for transient
speed variation limits.
Speed Droop
Speed change between zero and rated load. For G1 governing, 5 percent of rated speed.
Alternator Voltage Control Characteristics
The electrical load on a generating set is usually fluctuating, sometimes over wide variations and to
maintain a substantially constant output voltage an automatic, fast acting regulating system must be used.
In modern brushless alternators this is achieved by the use of a solid state AVR (automatic voltage
regulator) which adjusts the exciter field current to compensate for load, speed and temperature changes.
As in the case of the engine speed governor, a voltage change must occur before AVR action is initiated
and a measurable time elapse before the corrective process is complete.
BS.4999, part 40 (and other International Standards) cover voltage regulation and recognizes several
different grades. Grade VR2.23 is normally chosen for general purpose use in generating sets. Voltage
regulation performance can be defined by three factors.
Steady State Limits
Limits between which the terminal voltage will lie for any fixed load between no load and rated load at
rated power factor. For Grade VR2.23 steady state limits are 2.5 percent but limits of 2.0 percent or
even 1.0 percent are frequently specified for special application.
Transient Limits
Initial voltage change immediately following the sudden application of a specified low power factor load
or the sudden removal of rated load. For Grade VR2.23 the figure is 15 percent voltage dip for application
of 60 percent rated kVA at 0. To 0.4 lagging and 26 percent overshoot for removal of rated load at 0.8 p.f.
(Different transient limits are frequently specified for the application of specified loads such as starting of
motors).
Recovery Time
Time between point of sudden load application and the point at which the voltage returns to 97 percent of
rated voltage. For Grade VR2.23 maximum recovery time for load application specified in (b) above is 0.5
second.
Transient reactance of generators is effective whenever load changes occur. On applying a sudden load it
behaves as a series reactance, absorbing some of the generated voltage and
causing the temporary voltage dip at the alternator terminals. Transient reactance is usually expressed in
per unit terms (although sometimes percentage) related to the rated load impedance. Thus a 0.25 Pu (or 25
percent) transient reactance would be numerically equal to a quarter of the impedance of the rated load
circuit. Since transient reactance is purely reactive whereas the load impedance usually comprises both
resistance and reactance it follows that the effect of transient reactance on voltage dip will depend on the
load power factor and zero power factor lagging will produce the greatest transient voltage dip for a given
load current.
12 REQUIRED CALCULATION INPUTS:
The following is a summary of the required inputs to the calculation to satisfy this procedure:-
a. Generator Bus Design
Loading kW and KVAR
(Motor)
kW and KVAR (Non Motor)
b.Profile for Load Changes
c. Permitted limits for voltage and frequency under steady state and transient conditions.
(1) (2) (3) (4) (5) (6) kVA kW (9) (10) (11) (12) (13) (14) (15) (16)
1.1 1.5 0.8 0.75 1.81 1.47 (7) (8) 6.52 3.26 3.98 1.99 2.26 2.04 2.72 2.44
1
11.8 5.88
3 4 0.8 0.81 4.46 3.70 16.1 8.03 9.81 4.91 5.58 5.02 6.69 6.02
3
30 15
7.5 10 0.8 0.85 10.6 8.82 38.2 19.1 23.3 11.7 13.2 11.9 15.9 14.3
3
68.9 34.5
15 20 0.8 0.885 20.7 16.9 74.5 37.3 45.5 22.8 25.9 23.3 31.1 27.9
2
135 67.3
30 40 0.8 0.905 40.4 33.1 145 72.7 88.9 44.4 50.5 45.5 60.6 54.5
2
263 131
50 67 0.8 0.925 65.1 54.1 234 117 143 71.6 81.4 73.2 97.6 87.9
3
423 212
80 107 0.8 0.935 101 85.6 364 182 222 111 126 114 151 136
5
656 328
100 134 0.8 0.94 125 106 450 225 275 137 156 141 187 169
5
812 406
125 167 0.8 0.942 156 133 562 281 343 172 195 175 234 211
5
1014 507
150 200 0.8 0.945 185 159 666 333 407 203 231 208 277 250
6
1202 601
170 228 0.8 0.945 209 180 752 376 460 230 261 235 313 282
6
1358 679
200 268 0.8 0.945 246 212 886 443 541 271 307 277 369 332
6
1599 800
250 335 0.8 0.945 308 265 1109 554 678 339 385 346 462 416
6
2002 1001
300 400 0.8 0.945 373 317 1343 671 821 410 466 420 559 504
5
2424 1212
400 536 0.8 0.945 498 423 1793 896 1096 548 622 560 747 672
5
3237 1618
500 670 0.8 0.945 630 529 4095 2047 2268 1134 1386 693 787 709 945 850
4
600 804 0.8 0.945 756 635 4914 2457 2722 1361 1663 832 945 850 1134 1021
4
800 1070 0.8 0.945 1020 847 6630 3315 3672 1836 2244 1122 1275 1147 1530 1377
3
1000 1340 0.8 0.945 1290 1058 8385 4192 4644 2322 2838 1419 1612 1451 1935 1741
2
.50 (a)
.45
(b)
.40
(c)
.35
Load Elements
Line Ref A B C D E F G H
Max value of Total Eff. Peak kW (max value of 651 Sizing Recommendation
line ref 8)
Cummins KTA 3891
Max value of Final kW (max value of line ref 9) 556
Max value of Start kVA (max value of line ref 6) 886 577 kW Cont/635 kW O/L
DG sets upto 500 KVA solidly earth system is recommended. The earthing is provided so as to minimise
the damage to stator winding in the event of an earth fault. If DG is envisaged to operate in parallel with the grid,
it is recommended to provide solidly earth system irrespective of KVA rating. In case of isolated operation of
DG set, for rating above 500 KVA, resistance earthing is considered for limiting the earth fault current to
maximum full load current of the DG set.
DG sets of 3.3 / 6.6 / 11 KV are always provided with resistance grounded system. The earth fault current
is limited to maximum full load current of the DG set.
The value of neutral resistor is given by
R = VL
√3 * IL
Where,
VL = Line to Line RMS voltage in Volts
IL = Rated Current in Amps