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

 

[DanDel]

PMACP, what is the ambient temperature in the room? What is the transformer winding/oil temperatures?

 

[PMACP]

DanDel:

The temperature on the room, at this moment, is about 30ºC, but when the transformers did fail it was greater than this (someone told me that it achieved 40-45ºC). At this moment I don't have the temperature in the winding (its a wet transformer).

PMAC

 

[DanDel]

The ambient temperature is the problem. You need to get some air recirculation going to keep the ambient temp to a reasonable value so the heat from the transformer is dissipated. High transformer operating temp is guaranteed to cause an early transformer failure; I have seen this happen.

 

[PMACP]

DanDel:

Yes, I agree that the temperature may contribute for the problem. However I have some difficulties to accept that it is the only cause of the problem. In fact, two wet transformers have been detroyed, the first six months after being in operation and the second one month after the failure of the first one. Is that possible?

Not that, at the time there was three transformers each one supplied by a syncronous generator and with the secondarys in paralell.

Thanks
PMAC

 

[PMACP]

Dandel:

In the previous post, wher we read "In fact, two wet transformers have been" should be "In fact, two DRY transformers have been"

Thanks
Paulo

 

[ScottyUK]

Dry types typically run blistering hot at the best of times. Class F insulation running at the Class F limit is fairly common; if this is the case then there's very little margin to play with in terms of high ambient temperatures.

Are the vent louvres clear? Is there adequate air circulation? Have you considered forced air cooling?


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

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

 

[edison123]

2 trafos destroyed within one year and no-one investigated the entire system starting form load-study to trafo locations ?

You need to sort out whether these are oil-filled or dry type trafos. There are many contradictory posts.

 

[PMACP]

Hello:

Summarizing the situation:

a)The cogeneration plant had 3 DRY transformers of 0,4/15 kV
b) Each LV side of each transformer was connected to a synchronous generator
c)The MV sides of the transformers are connected to the main grid
d)Each set generator/transformer is connect and disconnect some time a day
e)The first DRY transformer fails after six months
f) the second transformer one month later
g) the room temperature at the time teh transformers fail was 40 to 50ºC

Thanks
PMAC

 

[edison123]

Please post the name-plate details of all the generators & transformers in that co-generation system.

Also, how did the earlier trafos fail ? earth fault, turn-to-turn short, cooked, fire ?

 

[DanDel]

Yes, this can happen from the ambient temperatures you describe.

 

[PMACP]

Hello

I now have new data about the problem.

To facilitate the analysis, I summarized the matte:

a) I have a cogeneration power plant
b) There is three generator-transformer sets (In attach I present two of the systems)
c) The transformers are of dry type and Class F insulation
d) The primary of each transformer is connected in a star connection and linked to a generator. Note that the connection between the primary of the transformer and the generator is made including the neutral conductor.
e) The secondary is connected on a delta connection and linked to a commum bus that allows the connection to the main grid.

Facts:

A few months after putting into operation the power plant, one of the transformers burned and a month latter another had heat problems. The thermal protection did not work.

The temperature in the room at the time would be 45ºC (maybe 50ºC).

One of the transformers has evidences that a electric arc happened between one of the LV poles and the neutral bus on the LV side of the transformer.

Questions:

a) What may have led to the electric arc between a phase and the neutral bus in the LV side of the transformer? Overvoltages?

b) DanDel said before that the room temperature may have led to the failure of the transformers ... with fire? And just the temperature? The issue is that the third transformer had no problems.

c) Any idea about the reason why the thermal protection of the transformers did not work? Well, it may be due to bad connections, but assuming that is not the case, there are other reasons?

d) In the secondary of the transformers (grid side), the voltage has periods of over voltage in the order of 10%. This contributes for the problem?

Thanks

 

[waross]

A transformer with the neutral connected on the wye primary and a delta secondary is prone to circulating currents.
The high ambient temperatures are not helping either.
Float the wye neutral. The delta secondary will keep the voltages balanced. Bring the transformer up with the generator to avoid switching transients if you must energise the wye side first.
Try more ventilation to cool the transformer room, a lot!!
Is the transformer that DIDN'T burn out closer to a door or ventilation inlet?

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

 

[PMACP]

Waross:

Thank you for your tip. Please, help me to clarify some matters:

a) when you said that "the transformer with the neutral connected on the wye primary and a delta secondary is prone to circulating currents", are you referring to the primary or to the secondary? i.e. the circulating currents happens on the primary (LV) or on the secondary (MV)? Have you some book or paper that allows me to better understand that issue?

b) What may be the consequences of floating the wye neutral? Well, in fact the MV side is connected on dealta wich should ensure the balance between voltages and so, the primary voltages will be also balanced. As well, the primary currents will remain balanced and so the neutral wire is not necessary. Is that correct?

c) Do you believe that the high voltages that happen on the secondary has no influence? Why the evidences that a electric arc happened between one of the LV poles and the neutral bus on the LV side of the transformer?

Thanks
PMACP

 

[edison123]

I think the transformers are poorly manufactured (premature failures, flash-overs, non-working RTD's point to that) with insuffcient clearances.

Please post a few photos of the transformers here.

 

[DanDel]

What kind of thermal protection do you have?

 

[waross]

In distribution systems a wye;delta transformer bank is problematic.
If the wye neutral is connected, a small primary voltage unbalance will result in a heavy circulating current in the delta.
If the wye neutral is left floating, there may be 173% transient voltages when the bank is energized. These transients will be avoided if the transformer is connected to the generator before the generator is started.
this may not be an issue, but it is well to be aware of the possibility.
I would look first at ambient temperatures.
I would look second at the possibility of saturation. At 110% Voltage you don't have much safety factor left. If high voltage is saturating the transformer it will burn out very quickly. Do you have taps on the transformer that may be used to better match the voltages?
Another possibility is poor workmanship. I have been aware of several transformer failures that were attributed to lightning strikes and poor grounding. I found that hard to believe as the transformers were fed by long underground feeders and the grounding was up to north American standards.
I later encountered an installation done by the same crew. The main cables were so loose that had the installation been energized, it would have soon failed. Of course the failure would have been blamed on lightning. (The customer pays for lightning damage.)
To my mind it is unlikely that over-voltages on the 15 kV side would cause a flash-over on the low voltage side without damaging any generators. More likely that a loose connection was arcing internally and the arc spread out or tracked to engulf the neutral bus. (But I have found over the years that almost anything is possible.)

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

 

[PMACP]

Hi,

Thank you waross, edison123 and Dandel for yours good tips.

DanDel, I'm not sure what thermal protection is installed (I'll check). However, a protection, no matter its type, should act, right?

Waross, I think that the circulating currents may be the problem. Well, I'm not sure but It is a strong issue. I'll made some measures namely concerning currents, voltages, harmonics and neutral currents.

Waross, hen you said "a small primary voltage unbalance" you are thinking on a unbalance of what magnitude? 5%, 10%?
Since the primary is supplied by a synchronous generator, is that likely that a unbalance with that origin may occur?

Waross, concerning saturation you said "I would look second at the possibility of saturation. At 110% Voltage you don't have much safety factor left. If high voltage is saturating the transformer it will burn out very quickly. Do you have taps on the transformer that may be used to better match the voltages?" Have you some idea that allow mw to understand if the saturation is ocurring? Perhaps measuring the harmonics and, if the 3º harmonic is present?

Thank you
PMAC

 

[waross]

Circulating currents.
Circulating currents are limited by the transformer impedance so if the percentage voltage unbalance approaches the transformer percent impedance you are in trouble.
Saturation.
At the saturation point the current rises disproportionately to the voltage. A small voltage increase results in a large current increase. You need to know the saturation curve of the transformer and the history of overvoltage surges on the 15 kV side.
If the voltage on the 15 kV side is normal then a 5% voltage surge for a few minutes is no problem.
If the voltage on the 15 kV side is 10% high already, then a 5% voltage surge for a few minutes may destroy a transformer by over heating.
When a transformer is fully saturated, a 1% increase in voltage may result in a current increase of over 5000%.


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

 

[slavag]

Hi.
Intresting thread.
All possible, but from my point of view, PMACP you MUST strat from trafo mnf.
Not once, I see troubles with dry trafos and usually it's what Edison saied:
"I think the transformers are poorly manufactured (premature failures, flash-overs, non-working RTD's point to that) with insuffcient clearances."
I suggest, are some no name company and w/o any type-test.
of course ambient temp., overvoltages increase reasons of faults, but isn't primary reason.
Best Regards.
Slava.
Sorry, I hope, you don't work for this company .

 

[PMACP]

Hi slavag:

Thank you for your opinion. Well, the company that manufactured the transformers is recognized on the market. I'll not say the company, at least until understand the problem. Note, I'm not saying that the transformers has no problems. Is that possible to test the transformers? How?

Thanks
PMACP