A system has four 336 kW natural gas gensets A, B, C and D operating grid parallel.
Grid import set at 100 kW.
Load varies around 800 kW + 150 kW
With drop add percentages and building load are selected so that 3 units A, B and C run, the result is a very smooth balance grid/DG as defined by grid import steady = 100 kW.
Here is the problem:
With drop add percentages and building load are selected so that all 4 units run, the result is a widely swinging grid import level. The value of grid import runs wild. For example at every second sample time we see -15 kW to 0 kW to 65 kW to 5 kW to 45 kW 110 kW to 140 kW to 70 kW to etc. etc.
Also during the above 4 unit scenario the PF of the generator bus also runs wild from -0.98 to -0.4 to past 0 to 0.4 to 0.9 to etc. etc.
We have only seen the above fluctuations when "D" is the 4th generator to come on to the bus.
An upcoming test will let another generator, say "A" be the forth generator to come onto the bus.
The paralleling equipment is the Cummins Digital Master Control and the generator controllers are 3100.
Any early thoughts as to if the problem is the control system causing the fluctuations or if it may be a problem with generator D controls?
One link that I liked is below.
http://cr4.globalspec.com/thread/57066/Alternators
for paralleling 2 or more
1- RMS line voltages must be equal
2- The generator to be paralleled must have the same phase sequence. If the phase sequence is different, then even though one pair of voltages is in phase the other two pairs of voltages are 120 degree out of phase. If the generators were connected in this manner there would be no problem with phase a but huge currents would flow in phases b and c damaging both machines.
3- Generator output phase angles must be the same
4- The oncoming generator must have a slightly higher operating frequency as compared to the system frequency. This is done so that the phase angles of the incoming machine will change slowly with respect to the phase angles of the running system.