K-rated transformers 2

[caporter]

Is there a good resource for learning where and when to apply K-rated transformer and chosing the appropriate K factor? My training in is micro-electronics but I've been given 4 building electrical systems. Several areas here have non-linear loads that make up >50% of the transformer load & we just had a failure of a 75KVA 460-208 delta wye transformer. The transformer load is 100% SCR controlled test ovens which got me thinking harmonics took it out.

 

[Skogsgurra]

The k rating is based on harmonics contents of the load current. Since frequency influences heat, the harmonics number squared is part of the calculation. 100 A at 180 Hz heats a lot more than 100 A at 60 Hz.

Fluke has a good paper on this:

http://support.fluke.com/find-sales/download/asset/2403202_a_w.pdf

Most transformer manufacturers give their view. This for instance:

http://www.powertransformer.us/transformerkfactor.htm

Question about your failed transformer: Did it overheat? Or did something else happen?


Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[caporter]

It overheated and shorted but not necessarily in that order. The picture below show the center winding. Testing with a DMM show all phases shorted together and to ground.

The transformer feeds a 225a 208v circuit breaker panel with 11 circuits. The loads are mostly 26A single phase environmental chambers. Before my meter died the current harmonics read ~75% triplen, ~50% 5th, ~33% 7th and ~15% 9th on a single load. I'll read the pdf from Fluke and, after the batteries recharge, get better readings.

 

[alehman]

If correct, that seems like pretty high harmonic distortion. Did you notice any overheating particularly related to the neutral connections? Neutrals should probably be oversized, maybe 2x, for this load. If possible, you may want to reconnect the ovens to operate on 208V, line-to-line, instead of line-to-neutral. That will eliminate the triplen harmonics.

Alan
----
"It’s always fun to do the impossible." - Walt Disney

 

[trosepe]

What is the input voltage to these environmental chambers?

120v, 208v or 120/208v? These can all be single phase.

What was the temperature rise rating of the transformer?

Can you take a photo of the transformer nameplate?

 

[caporter]

All chambers are 120Vac, 26A, single phase.

Transformer is a standard Square D and the name plate is on the link.

 

[trosepe]

With only about 35 KVA of connected load, the 75 KVA transformer should have been OK. Our standard specification is for 115 deg C rise transformers for upto 25% non-linear loads and K13 rated past that. I have used this specification in telecommunication facilities, which have mostly inverter loads, for 15 years without a problem.

The transformer you have shown in your photo is 150 deg. C rise which is not the best for non linear loads. The high load losses and the additional heating from the harmonics push the thermal ratings.

 

[electricuwe]

With single phase SCR-loads you are not only having a harmonics problem, but also a problem with the neutral, as the triplen harmonics on the neutral will be in phase there.