Dry Type Drive Isolation Transformer too hot? 5

[sparkview]

Dear all,

In a recent '2x500kW Battery Energy Storage project', we have been advised by the manufacturer to install two (2) 'Three Phase Dry Type Drive Isolation Transformer' of which nameplate can be seen below. The project is now commissioned, however, since a few days ago the Transformer is found to be extremely hot.

A thermal imaging camera shows that at the upper core of the Transformer, is at 182 degrees Celsius while the ambient temperature is about 30 degrees Celsius.
Both Transformers also release a smell of "burnt electronics". Can these Transformers be overheating? No hot spots can be observed directly and they only work for about 4-5 hours per day while the batteries are charging/discharging.

In advance many thanks for your support.

With kind regards,

Jairo Vrolijk

 

[itsmoked]

Further info stating the currents would be helpful in making a judgement.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[sparkview]

Hello Keith,

Thanks for the prompt reply. Parameters of the conditions just a few minutes prior to performing the Thermal Imaging are as follows:

Best regards,

Jairo

 

[wroggent]

Also a one-line showing how you've connected these and the system grounding arrangement on both sides would help.

 

[FreddyNurk]

Another thing to be careful of, the currents may be higher than expected due to higher frequency components from the inverter switching.
To me, the readings tend to indicate that the transformer is right on the edge of the allowable temperature rise, and may well exceed it given that its the surface temperature that's shown.

I'm aware of at least one inverter manufacturer that stated that a transformer with a specific rating for high harmonic content, although I don't recall what the specification was. This may play a factor when you measure the current, as not all current clamps will pick it up.

EDMS Australia

 

[sparkview]

Hi,
Below you will find the Single line diagram with system configuration and grounding (also link attached for better resolution).

Best regards,

Jairo

 

[edison123]

It's already exceeding the specified 150 deg C temp rise. The OEM probably bet on the short time operation to see it through the warranty period. Also, I have seen better build quality. You should ask them to explain this.

Muthu

http://www.edison.co.in

 

[Skogsgurra]

Wait till summer arrives. Then the problem may solve itself by setting the transformers on fire. (smile with tears)

Your application probably needs a K-rated transformer because of the current waveforms not being sinusoidal. Do you have access to current waveforms (scope recordings etcetera) that show the waveforms?

For the time being, and before a solution is found, get blowers to cool the transformers. There is a note about minimum distance to walls (on the nameplate). The housing looks rather crowded. Is that distance minimum kept?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[edison123]

Gunnar - The nameplate says it's K rated, which I doubt. Blowers will help if it is not an enclosed space and the hot air gets out of the area.

Muthu

http://www.edison.co.in

 

[Skogsgurra]

Is that what Type K means? There is usually a number associated with K rating. Has that changed?

Anyhow, waveforms would be interesting to see.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Skogsgurra]

Found the data sheet. I can't see anything about K rating there.

https://motorsandcontrol.com/media/...nd_Drive_Isolation_Three_Phase_Data_Sheet.pdf

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[edison123]

Well, it shows a zig-zag secondary. Isn't that normal connection for K rated trafo? I agree K factor no. is missing.

Muthu

http://www.edison.co.in

 

[itsmoked]

I just saw the 182C temp outside the core, I'd only seen the 117C earlier and was puzzled. No wonder it's not smelling so good!!

I could easily see it being another 40C in the center somewhere.

With the currents you're measuring (under the rating) and it's still that hot, harmonics makes the most sense for this high temperature.

I too think you should get a big fan on that soonest as you could be just a few days from meltdown.

You can prove the harmonics with a scope as suggested.
EDIT: Skip the scope, you need a harmonics instrument to read what the harmonic content is and its magnitude, unless, your scope has some power features built into it.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[FreddyNurk]

K-rated was the term I was looking for, thanks Skogs & edison.

EDMS Australia

 

[waross]

K type may not be the same as K rated.
I don't see a zig-zag connection. Those tabs are a way of showing a connection. Also they are at 90 degrees. A zig-zag would be shown at 120 degrees.

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

 

[sparkview]

Hi all,

Thanks for all the very useful information. Does anyone have some pointers to ANSI or IEC standards indicating the parameter limits with regards to Harmonics?
I will try to gather in field Harmonics data and contact the Transformer manufacturer.

Best regards,

Jairo

 

[itsmoked]

Haven't got harmonic limits info but would add that here's a link probably worth reading on troubleshooting transformers most doesn't apply but a few might.

HPS Troubleshooting

Also, there is a good chance they have a fan kit for those style of transformers.

There is also a warning somewhere that hap-hazard application of fans can disrupt critical convection causing 'spot overheating'. So if you apply expedient temporary fan cooling you should probably remove coverings so full area cooling can be achieved verses pointing a fan at the enclosure.

They also define a core "spot temperature" that states:
Spot temp = ambient + rise + 20C

This means the temp in your core must be at least 20C hotter than anything you can see on the outside.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[prc]

First of all, make sure the temperature rise is in core and not on the outside of the winding. The max temperature rise of 150c is of winding and not of core. Of course max core temperature is also limited to 140C normally. Type K seems to be type of the manufacturer (based on efficiency level) and nothing to do with K-rated transformers.K-rated (K4 ,K-9,K-13,K-20.....K-40 ) transformers are covered in C57.110-2008.K-4 rated transformers can stand 16% of fundamental current,3rd harmonics , 10% of 5th harmonics, 7 % of 7th harmonics, 5.5 % of 11th harmonics.K-13 can handle 200 % of harmonic currents of K-4 transformer without exceeding permissible temperatures.

Core will not get overheated due to harmonics in current. It can get overheated due to (1) harmonics in voltage (2) higher V/f is coming on windings due to mismatching with inverter (3)poor quality of core laminations used. In (1) &(2) it is due to over fluxing in core. In (3) it is due to shorting of laminations that cause circulating current and heating in core. Please find out chances for any of above three situations.

Once I faced with a situation of a 10 MVA transformer getting overheated .Within 30 minutes oil temperature was reaching >120C. It was used in a test lab feeding inverters. Once the problem in inverter was corrected, problem disappeared.

 

[electricuwe]

I'm surprised about the neutral connection between inverter and transformer. Large three-phase inverters usually have no output to connect a neutral. The only exception is UPS where part of the load may be only single phase.

The fact that there is no neutral provided is also utilized for calculating the modulation pattern: triplen harmonics are added to the reference signal to allow a larger fundamental. Without a neutral these voltage harmonics will have no impact on current. If then a neutral connection is added, these harmonics can drive also currents.

I'd strongly recommend to check this issue.

 

[sparkview]

Hi, electricuwe,

The only reason an isolation transformer was installed is to provide a neutral point which can be grounded. This was a requirement given by the Battery manufacturer. Details on why is the neutral used for is somewhat unknown to me since most of these details are proprietary. I think it is related to a neutral reference that is required for the inverter unit.
In our power plant, it is not normal to use a 3phase circuit with a neutral connection.

This is the first project in our facility we were required to install an isolation transformer.

Jairo