Problem with transformers on cogeneration plant 1

[PMACP]

Hi,

I need help to understand a problem I have to solve in a cogeneration plant. The power transformer that do the conncetion between the generator and the network has problems which I can't understand. That transformer fails because it heat excessively.

Some help?

Thanks
Paulo

 

[edison123]

waross.

This is a GSU trafo. The high voltage on the secondary side (which is what the OP says) will not affect the saturation. May be the tap on the HV side is set to produce high voltage.

@OP
This is a trafo which is already overated given that the generator KVA is only 1750 KVA. Why would such a trafo with only 80% load should overheat ?

For such a long thread, I am sorry to tell you that you have not provided any hard numbers so far (voltages, currents, temperatures, frequencies, thermal protection set limits) which would go long way in finding the root cause and solution. Otherwise all this theorizing is just looking for hidden ghosts.

My two cents.

 

[PMACP]

Hello,

At this moment I have some new data (measured). I still treating the information collected.

Note that the collection was made on the LV side of the trasnformer.

The first major revelation of the data is that the currents in the three phases of the LV winding are unbalanced. The unbalance is of 14%. The unbalance on LV voltages is small.

The currents unbalance may result (as much I think) from the existance of some ancillary services that are supplied by the generator before injecting power on transformer (as we can see in the attach figure).

What may be the consequence of the unbalanced currents for the transformer? It may contribute for the heating problem? In fact, the unbalance will produce a homopolar current that tend to circulate on the delta secondary wedding. is that correct? Is there some method to unrate the transformer?

 

[ScottyUK]

I wouldn't worry about the transformer if you have high negative sequence currents, it's the generator rotor you should be concerned about: negative sequence currents cause currents at double the line frequency to circulate in the rotor forging, resulting in serious heating problems within the rotor. You should have an NPS relay, ANSI code 46, to detect the problem and trip the unit before damage occurs. Generally the bigger the machine the lower its ability to tolerate NPS currents as a percentage of rating.


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

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

 

[PMACP]

Scotty;


Thanks for your advise. Until the moment there's no problems relating to generators. How much should be the unbalance in order to achieve a dangerous situation? And concerning the transformers is that some problem resulting from the unbalances?

Thanks
PMCAP

 

[PMACP]

Scotty:

In my previous post I should say that the generator has no current unbalances. In fact, the unbalances are to transformer. As you can see in the scheme I sent in the previous post, the unbalances result because before the transformer there are some ancillary services that are supplied by the generator. The remaining current on each phase is injected on the transformer.

Sorry for the late explanation

PMAC

 

[waross]

You probably have circulating currents in the transformers.
The reason for the small voltage difference is that the circulating currents are acting to correct the voltage unbalance. The circulating current is driven by the voltage unbalance and limited by the transformer impedance, and tends to correct the voltage unbalance. Can you measure the current in the neutral?

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

 

[PMACP]

Waross:

Yes, I may measure the neutral current. I have that value in my office. Tomorrow I'll put the exact value here. However, as I can remember, it is in the order of 50A.

Thanks
PMACP

 

[ScottyUK]

PMAC,

Without knowing much about the generator, 10% NPS current would be enough to be of concern.


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

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

 

[PMACP]

Hi,

I have now some data which is in attach to this post. There are some interesting results. For instance, the power factor is different on each phase. The currentes are unbalanced.

Any new suggestion?

Thanks
PMAC

 

[waross]

At first glance it looks as though a couple of phases are switched in your metering setup. But I may well be wrong. Let's leave this point for now and move on to some other questions.
A couple of unknowns remain:
Actual primary voltage. Can you take the generator and the ancillary loads offline long enough to infer the actual primary voltage balance?
Are the ancillary loads heavier on this transformer than on the others?
What is the nature of the ancillary loads, KVA and PF?
Can you get a meter inside the transformer and measure the actual current in the windings rather than in the leads?


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

 

[PMACP]

Hi

Waross, You think that the phases are switched because of the power behaviour? On yours opinion, which phases may be switched? I think I didn´t make any mistake...

Switch off the loads of ancillary services may not be possible because I think some of those loads corresponds to refrigeration services of the generator. But, i'll check it.

But, you think that those results do not allow to conclude nothing? For instance, the total power os the transformer is 2000 kVA and the actual power is near 1800 kVA. If we derate de transformer due to ambient temperature (40-50ºC), the transformer is overloaded...

Thanks
PMAC

 

[waross]

I thought it was strange that;
R is the lowest current,
R is the lowest kW,
but
S is the lowest KVAR,
T is the lowest voltage.
It doesn't take much load to drop the voltage on a generator due to the relatively high impedance. Normally the AVR corrects, but it can't correct for unbalanced voltages.
With the neutral connected the delta voltages are proportional to the wye voltages and may not sum to zero, hence circulating currents. But the currents tend to correct the voltage unbalances causing them.
I suspect the prime cause of the failures may be high ambient temperatures, but that may be made worse if there are circulating currents. I would ground the generator neutral either directly or through an impedance and float the transformer neutral. I had a unique industrial application where I was able to interrupt the circulating currents by using a "Broken delta" connection. The drop in transformer temperature was dramatic.
But we don't have quite enough data to determine if you do have a circulating current problem.

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

 

[alehman]

You mentioned arcing. Are you switching the transformers on and off or are they brought up with the gens?

 

[PMACP]

Waross and alehman:

Waross, you said: "R is the lowest current,R is the lowest kW,but S is the lowest KVAR,T is the lowest voltage.".

Well, the lowest current and active power on phase R is consistent. Also the lowest voltage on phase R is consistent because that phase has the higher value of current.

alehman: Yes, the tasnsformers are switched on and off once a day. They are switched on at 6 a.m. and switched off at 23 a.m..

Thanks

 

[alehman]

Did the faults happen when they were switched?

How long is the cable between? Any surge arresters or capacitors?

 

[alehman]

And what type of switches/breakers?

 

[PMACP]

alehman:


"Did the faults happen when they were switched?"

I think no, it happen when the trasnformer is operating. But I'll check it.

"How long is the cable between?"

Between the generators and the transformers it should be 10-12 meters.

"Any surge arresters or capacitors?"

NO. The synchronous generators work with a leakage power factor.

Thanks.

 

[alehman]

Can you post photos of the damage from both failed transformers?

 

[PMACP]

alehman:

In attach I present some photos of one transformer.

Thanks

 

[PMACP]

alehman:

Some more photos on attach.

Regards
PMAC