-KVar in a Shaft generator DECS 200

[ETO1]

Hi everybody,

this is my first time here, I hope you can advise me.

months ago we had a problem in one generator coupled to a main engine, the engine was tested and generator not online, so we had high voltage alarm, nominal is 690V and went to 900V, on the DECS 200 I got LOSS OF SENSING VOLTAGE, we opened the generator and found the SENSING TRANSFORMER (690V/110V) and the DAMPING RESISTOR burned, so we received new ones, not the same but the same characteristics, when installed the new things we tested and we got only 655V, so we tested the transformer on workshop and if we apply 690V on the primary we get 116V on the secondary, so we decided to increase the setting on the DECS 200 on the section..

SYSTEM PARAMETERS/TRANSFORMERS.....GEN PT SEC = 110V ...we changed to 116V

So we tested and now we get 690V and KVar is positive, when put in parallel with another generator the KVar is negative but we do not get any alarm, when we put three generators online KVar is more negative and we get alarm on the IAS(REACTIVE LOAD SHARING ERROR on the computer in control room) the other two generator Kvar is normal. When we put four on line we do not get any alarm and KVar is positive but way less than the other three generators.

We got in contact with hyundai and recommended to change this parameter


SYSTEM PARAMETERS/TRANSFORMERS.....GEN PT PRI = 690V ....TO 655V (volts we get with the new transformer) and
SYSTEM PARAMETERS/TRANSFORMERS.....GEN PT SEC = 110 (keep it in 110V)

We tested and we got the same results....

do you think that this problem could be caused by the ratio of the SENSING TRANSFORMER?

here it is a table of parameters we meassure with two generators...GENERATOR 3 is the one with the problem.

Another thing that is making me noise...when megging, and inspecting the generator, diodes, etc, I disconnected some cables that goes to the DECS 200, when connecting again I notice one thing...that the cables PGU, PGV, PGW that goes to the DECS 200 were "interchaged PGU and PGW" and I put "interchaged" because I am pretty sure that they cannot swap. Anyway before testing I look in the numbers printed on the cables and 1, 2, 3 correspond to the same position in the AVR and the PGU, PGV, PGW according to the diagram....I think the person that put the labels was wrong...here it is a picture to explain better....

in diagram

PGU go to C2
PGV go to C3
PGW go to C4

that is why I paid attention to the number printed on the label not the yellow labels that I think they are wrong.

Anyway...what could happen if PGU and PGW are swap?

I hope you can help me guys...

thank you

J.G.

 

[waross]

When running alone, the KVAR is determined by the load.
When running in parallel the KVAR is determined by the load but the KVAR sharing is determined by the relative voltages of the parallel sets.
I note that the load seems to be changing between taking readings.
Cross current compensation adds an internal bias to the actual set point voltage of the AVR.
Do you have cross current compensation?
Do you have a switch on the switchboard to enable/disable cross current compensation?
When the units are in parallel how much change is there when cross current compensation is switched in and out?
If the wires that you swapped are three phase voltage sense wires, then they may be interchanged with no effect when the unit is operated alone. When cross current compensation is active interchanging sense wires will interfere with the operation of the cross current compensation. Depending on which phases are interchanged, the compensation on one set may be forced in the wrong direction. That may cause the effect that you are seeing.

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

 

[xnuke]

The wires you show swapped are only providing three-phase power to the unit, so their order shouldn't matter. Something else is wrong, but I can't put my finger on it yet. Too many questions arise:

Why is Gen 3's sensed voltage so much higher than that seen by Gen 1? Is it causing the AVR to underexcite Gen 3 while it tries to lower sensed voltage? This could account for negative var output.

Are the generators directly coupled? It sounds like they are on your post.

How are the AVRs set up to operate in parallel? One isochronous, one droop? Both droop? Var mode? pf mode?

Is the paralleling CT connected with proper polarity, and connected to the correct DECS inputs for the CT current rating? Is the reactive current compensation working properly, or are the negative vars due to circulating current?

If in droop, do all generators have the same droop value set?

Have you tried to manually control the excitation in Gen 3? If not, you might try that while in parallel and see how it behaves.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[waross]

PGU Power-Generator or Potential-Generator?
The Cross Current compensation CT must be installed on B phase in the correct polarity.
Which terminal on the DECS 200 does the PGU wire connect to.

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

 

[ETO1]

Hi,

we have for MAN engines and each engine is coupled to one generator HYUNDAI. so we can put 1,2,3 or 4 online depends on the load.

WAROS,

The load changes a little bit due to fan, or cooling systems or pumps that may run or stop while I was taking the readings.
In every GEN/DECS 200 we have a CCT (2500/5A, 40VA,60Hz, 1.0 CL) on phase B of the GEN connected to CTB (B2, B3) on DECS 200.
No cross current compensation switch on MSB.
And the "swap cables" are not the sensing voltage,are the ones that goes from the PMG of generator to the INPUT POWER (C2,C3,C4) on DECS 200.

XNUKE
The same thought about the cables, that are just for power the unit, but still had the doubt.
The sensing voltage is higher on GEN3 cause we got a SENSING TRANSFORMER from another provider and it is supposed to be 690/110 but when applying voltage we get 690/116...that is why we modify:

SYSTEM PARAMETERS/TRANSFORMERS.....GEN PT SEC = 110V ...on the DECS 200, we changed to 116V , before modifying this we got only 655V from the generator.

Each generator has its main engine couple.
The CT is connected ok.
All four DECS 200 has the exact same parameters except DROOP
GEN1=2.5% GEN2=3.5% GEN3=2.5% GEN4=3.7% (and the one we just modify because of the SENSING TRANSFORMER on GEN3)
The FIELD V on the generators are these (taken one generator online at a time and almost same load)
GEN1=15V GEN2= 19.95V GEN3=37V GEN4=35V

this question "How are the AVRs set up to operate in parallel? One isochronous, one droop? Both droop? Var mode? pf mode?" I will answer later I need to check it.

Before the problem on gen 3, I was told that all four generators work properly kvar ok, I saw the parameters on a screen shot. So I think the parameters on the DECS 200 are set ok according the maker.

Thank you guys for the feed back

 

[ETO1]

WAROSS

PGU Power-Generator or Potential-Generator?
The Cross Current compensation CT must be installed on B phase in the correct polarity.
Which terminal on the DECS 200 does the PGU wire connect to.

-----

From the P.M.G (permanent magnet generator) comes these three wires PGU, PGV, PGW and this goes to the DECS 200 to the INPUT POWER (C2,C3,C4).

 

[waross]

Thanks for the information. Do check that the sense connections and the connections from the cross current CT have been connected correctly.

DROOP:-
There may be a droop setting on the governor. This is similar to the proportional band on the P of a PID controller.
Then there is the actual droop of the generator, the change in speed or frequency from no load to full load.
These are not always the same.
If the throttle linkage is such that the full movement of the governor is not used, the governor droop will not match the actual generator droop.
If the actual generator droops are not the same, the generators will not share the load in proportion to their capacity as the load changes.
If they are set to share properly at 25% load then they will not share properly at 50% or 100% load.
With the sets set to share properly at low load, the 3.7% droop set will be about 67% loaded when the 2.5% droop set is fully loaded.


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

 

[FreddyNurk]

waross has already mentioned it, but if it was all working before the sense transformer died, there's no chance the sense voltage connections were remade incorrectly?

There is, of course, the possibility that the incorrect sense voltage reading is the cause of the issue, and that changing the setpoint will result it, but I'd check this first:

The table indicates 'Woodward' as a voltage measurement device. There's no chance its using a bias signal to the Basler and inadvertently 'hiding' the incorrect voltage?

No chance the Woodward is using the same sense PTs but doesn't have the scaling factor applied, throwing out the voltage when in parallel?

Are you using some kind of load sharing device (e.g. a Woodward) for paralleling?

EDMS Australia

 

[waross]

I may have misspoken in relation to the droop settings.
My comments were in relation to governor droop.
If the droop settings you have described are AVR droop settings then please disregard my remarks in relation to droop.


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