Generator.Random Current Fluctuations once Paralleled with the Grid 1

[electrouw]

Good day.

We are in the process of commissioning a 625kVA (at 0.8 PF) natural gas reciprocating synchronous genset. The genset is connected on the LV side of a 1500kVA step up/down transformer.(11kV > 415V). For the most part our plant will be consuming the electricity but at some stages we will be exporting to the grid.

We have been able to take the genset up to 400kW on a resistive load bank with no problems.

Recently we were able to successfully sync to the grid. For some mysterious reason as soon as we are connected to the grid the current in all three phases start jumping. I'm talking from 50A to 700A back to 50A within a second. As a result of this the power factor oscillates between -0.2 and 0.2 centred around 0.95.
The voltage on all three phases is fine at 250V.

It might also be worth mentioning that the neutral current are pretty consistent at 20A. Another thing that might be a reason or a result, is the THD mostly made up of the 3rd order harmonic.

THD of grid alone = 2.5% ( mostly due to 5th order harmonic, induction motors in the plant)

THD of generator on loadbank = 2.5% ( mostly due to 3rd order harmonic)
THD of system ( generator connected to grid ) = 6%
( once again mostly due to 3rd order harmonic)


It would be much appreciated if anyone can shine light on this strange phenomina. Together with the Genset's own protection, I have a Deif GPU-3 overlooking the protection. This unit is not a sync-check device but it does show that the phase angle between the grid and the generator is 0 while synchronised.

I might also add that the throttle actuator is acting very different when connected to the grid as opposed to the load bank. When connected to the grid it constantly jerks between about 20 and 70 degrees ( range is 0 to 90 degrees 90 being fully openned). Could this induce the big variation in current.

As far as I am concerned the system is meant to switch from speed/frequency control to power factor control as soon as we are synched - this could infact not be the case.

I've attached a pdf with the current data, and I can add the voltage and power information if need be.

Thanks is advance!

 

[rasevskii]

As a first step try disabling the PF control function, let the governor run in droop mode only. There are a large number of possibilities for causes to the fluctuation. The PF control however should only affect the excitation level in a slow acting manner to hold a set PF.

What sort of governor or integrated control system do you have?

regards, rasevskii

 

[waross]

Are you running with the governor in isochronous mode? Switch over to droop mode.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[edison123]

Looks like an AVR issue trying to control the pf. Disable the pf mode and see what happens.

Muthu

http://www.edison.co.in

 

[ScottyUK]

Agree with Bill - sounds like you're trying to regulate the speed of the grid in isoch mode. You need to be either in a droop or load control mode.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Kiribanda]

electrouw,

I too agree with other folks. Allow it to run on "AVR MODE" for sometime and see whether the same problem persits.

 

[catserveng]

How about some more info on your controls and system?

Who's engne/generator?

What kind of governing and load share/baseload controls do you have?

What kind of AVR? Does the AVR have the power factor/VAR control mode built in or does it use an external controller? Have you checked the wiring?

You only load bank tested with a resistive load, so you did not qualify voltage droop, is the droop transformer in the generator (or switchboard, sometimes called the cross current transformer) installed with the correct polarity? Your inital description sounds like a droop CT reversed.

After breaker close, what is the fuel control signal doing? Same question for field output from AVR? Is the current and power factor changing but the kW staying more stable?

 

[waross]

There should be no interaction between the AVR and the governor. A fluctuating governor control is a governor issue, not an AVR issue. The AVR may have issues coping with the governor swings but you can't solve governor problems with AVR solutions. Take my word for it. You don't want the long version.
Get the governor stable before even thinking about the AVR.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[catserveng]

Bill is right, when I first read this I only got part of the first thread (my computer is a bit slow today).

What is your load control mode? What governor is installed, and is it using a load share device like a Woodward controller?

Thanks for keeping us honest Bill!

 

[electrouw]

rasevskii,catserveng:

Sorry for insufficient info.

I should have said that we are using a Woodward 2301A Electronic Load Sharing and Speed Control unit along with a ProAct 3/4 Electric Powered Actuator and Driver + Woodward Magnetic Pickups and Proximity Switches.

Both the Engine and the Generator is manufactured by a Chinese Company called Sheng Dong.

As for the AVR and PF controller, it is already installed and some unknown Chinese brand which I can’t spell. We have another one of these Gensets on a different site working a treat though.

waross:

I am almost certain that once the GB closes, the Woodward gets thrown into droop mode. Unfortunately I can’t get to site every day but as soon I can I will ensure that the GB Aux switch is actually wired in correctly and report back to the forum.

edison123, Kiribanda:

Given that the current fluctuations are so big, and the voltages are pretty constant, could it really be the AVR? I might understand the system wrong, but isn't the AVR purely responsible for matching the generator voltage with the grid voltage by regulating the excitation current? Providing the Voltage does stay constant (and the same as the grid) and the current fluctuates a whole lot, this will result in a fluctuation in power delivery which it does indeed. The Fluctuations were too fast to see on the Power Meter on the Panel but recorded with the Yokogawa CW240 we did pick up negative power at some stages. Please help out with explanation.

ScottyUK:

As I mentioned I will ensure that the genset does go over to droop mode. Thanks. Just on what you said before "You need to be on a droop or load control mode".
Do you mind educating me a little. Where does droop/load control mode and PF control come in to play. Is it totally different controls independent of each other? I.e. the droop looks after matching the load and the generator output by manipulating the throttle and the independent of that the PF control is taken care of by the AVR manipulating the excitation current? A clarification on my understanding of those to controls would be much appreciated.

Another question which no one has mentioned before...

Is it possible that the harmonics could have an influence as described? I guess what I am trying to get at is whether the harmonics can cause theses fluctuations or will these fluctuations actually cause the harmonics. What is a typical THD on generators this size?

Thanks for your input so far guys!

 

[waross]

Read my lips, hear my words. Fix the governor first, then worry about any remaining issues.
Droop mode is very similar to proportional band in instrumentation systems.
A standard droop governor will be set to Hz+3% at no load.
That's 61.8 Hz for a 60 Hz system. The speed will drop or "droop" to 60Hz at full load.
On manual control the speed is set to 60 Hz before synchronizing. Once online, the speed control is advanced. At a setting corresponding 101% or 60.6 Hz, the governor will be open 1/3 and the set will be picking up 1/3rd of rated load.
At a setting of 102% or 61.2Hz the set will be picking up 2/3rds of rated load.
At a setting of 103% or 61.8 Hz the set will be picking up 100% of rated load.
Note well:
The grid is locking the set at 60 Hz and niether the speed or the frequency of the generator will change. This is a simple, dependable way to use a proportional band controller to control loading of a gen set that is sync'ed to the grid.
In a stand alone set, the AVR sets the voltage. In a paralleled set the AVR sets the power factor. The voltage is controlled by the grid and the AVR and the generator output can not overcome the grid.
BUT, as the load on an islanded genset increases the terminal voltage drops because of internal impedance of the generator and the AVR increases it's output to compensate.
With the machine on the grid, the AVR will see the constant grid voltage and will hold the excitation constant. As your wild governor pumps the throttle, the real power output of the machine will fluctuate just as wildly. As the AVR is seeing a constant voltage, its output will not change and the machine will become under-excited as the load increases. Depending on the AVR setting the generator may become over-excited as the load decreases. That is why your power factor is swinging.
Once you get the governor working in droop mode, you may see the power factor dropping as the governor is advanced and the load increases. Then is the time to check the AVR. it may be in droop mode instead of power factor mode.
Governor> Droop GOOD.
AVR> Droop BAD. Use PF control.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Kiribanda]

electrouw,
My apologies.I think what I said was wrong.Since the voltage is remian constant as it is connected to the grid,I too suspect that the governor is "HUNTING".
Therefore,as Waross said,first fix the governor.It will solve the problem.

 

[catserveng]

I second Bill's suggestion, fix the governing first, if you have the actuator swinging that much,it needs to be fixed. One thing to check with that setup, hook up an analog meter to the actuator output of the 2301A, if the output of the 2301A is fairly steady and the actuator is still slamming around you may have a power supply issue with the ProAct driver. 2301A actuator output should be from 0-8VDC.

On a 2301A, as long as the CB AUX contact is open you're in Droop, reference Woodward publication 82312, in the version I have the droop setting explanation is on page 18.

http://www.woodward.com/pubs/pubpage.cfm?web_section=power&SrchCond=PUBLICATIONS

So you're only using a 2301A in droop and no load control module like a Generator Loading Control?

You will likely find that you will need to dummy up the governor dynamics using the Proact actuator and driver when paralleled to the grid as compared to the settings you had for running standalone on the load bank. A gas engine will typically need some actuator compensation, not familiar with the Chinese engine, but most engines I have worked on with a Proact I start with 20% actuator compensation, and gain and reset at about mid position. Paralleled to grid you may see the Gain as low as 20%.

Mike L.

 

[nawao]

electrouw
As catserveng said:
"You only load bank tested with a resistive load, so you did not qualify voltage droop, is the droop transformer in the generator (or switchboard, sometimes called the cross current transformer) installed with the correct polarity? Your inital description sounds like a droop CT reversed."

When paralleled to the grid, both, governor and AVR must be on droop mode.
Invert the polarity of the droop CT and tell us what happen.