Stray losses in transformers 2

[freefallingbody]

Hi all,

I am trying to estimate operating efficiency of a 50 Hz, 11kV/433V, 1000 kVA distribution transformer. I have measured no load losses, which I understand is essentially the core losses. Now, from the measured winding resistances on primary and secondary sides and the load current measured, I can estimate the copper losses also.

Problem is, can I assume a stray loss value?. In motors, I know I can take values from IEC recommended values as per IEC 34-2. My question is whether there is any such values for transformers?

(If I conduct a short circuit test, I can measure the total load losses, inclusive of copper losses and stray losses. But I don't think it is possible.)

Can you guide me?

Regards

Dinesh

 

[dpc]

The applicable IEC or ANSI test standards will provide information on the approved methods of loss and efficiency measurements.

An actual load test would provide the best data on unit losses, but these are expensive to conduct.

The core (iron) losses and winding (copper) losses are generally all that is considered. Unlike a motor, the transformer has no friction or windage losses to consider.

However, for forced cooling, you do need to consider the power required for operation of pumps and fans. Some test standards may not take these into account when determining transformer efficiency.

 

[skiier]

For such a large and expensive transformer could the cost of installing 2 wattmeters not be incidental or negligible?

Why not just hire someone with the equipment and measure your loads. Maybe install a couple of chart recorders to really get an idea of things.
A wattmeter in front and behind your transformer would tell you a ton of useful information.
Why not rent some portable clamp-on Amprobe type wattmeters and measure instead of estimating. You will find out exactly what you want to know and possibly more.
Just a thought.

 

[busbar]

For ANSI evaluation, IEEE C57.123 …Transformer Loss Measurement seems to devote a lot more detail accounting for external-jumper losses compared to stray losses. Given basic physics, IEC is probably similar.

 

[jbartos]

Suggestion: Visit

http://leeh.ee.tut.fi/transformer/tra211.htm

http://www.liebert.com/support/whitepapers/documents/sl_24200.asp

for:
Transformer load losses consist of winding I2R losses plus stray losses. Using UL1561 test methods, stray losses are assumed to be primarily winding eddy current losses for transformers 300 kVA and smaller. For example, a transformer having winding I2R losses of 2000 watts and 60 Hz stray losses of 100 watts would, with a K-20 rating, be required to dissipate the 2000 watts of I2R losses plus 20 times the 60 Hz stray losses of 100 watts for a total load loss of 4000 watts without exceeding the maximum winding temperature rise. The result is a larger, more expensive transformer.

http://ieeexplore.ieee.org/xpl/abs_free.jsp?arNumber=109859

etc. for more info

 

[freefallingbody]

Thank you guys for your time and the references!

I still haven't figured out how to guess stray losses in transformers.

IEEE recommends a reduced voltage short circuit test on H.V side using a single phase supply and measure load losses, which when extrapolated to load currents can give me the total load losses under actual load conditions. Then I can estimate copper losses. The difference of the above two will give me stray losses.

I was looking for a short cut, without conducting the above short circuit test.

Direct input & output measurements may lead to more than 100% efficiency ( !) taking into instrument uncertainties.

IEC or IEEE standards is silent on the transformer stray losses issue. They suggests methods to estimate stray losses and not keen on giving absolute ( or %) values as they did in motors.

Copper development Association booklet mentions that 'stray losses which were 10% of the load losses is now only 5% due to better manufacturing CAD practices'.

 

[busbar]

IEEE C57.12.90-1999 §9 mentions stary losses in load-loss calculations, but doesn’t seem to elaborate on measurement.

 

[electricuwe]

skiier,

installing two wattmeters wouldn't be useful because large transformers have very high efficiency. The measurement error would easily lead to totally wrong results.

 

[ScottyUK]

For a relatively small transformer such as the one you describe, you could carry out the S/C and O/C tests which will give you enough information to get pretty good estimates of overall efficiency and the distribution of losses.

Approach the manufacturer and ask them for copies of the open-circuit and short-circuit test results. They might not have the data for your exact transformer, but should have a type-test result.

Failing that, for a typical transformer of this size with, say, 4% impedance, you could do the S/C test from the LV side using a 3ø primary injection set capable of delivering 20V @ 1500A per phase. A powerful injection set for sure, but certainly not beyond the realms of possibility. You will probably have to approach a company specialising in testing of protection systems or similar.

If you apply the short circuit to the LV windings, then depending on the transformer impedance you may be able to test from the HV side. I assumed 4% impedance as typical. Using this value of impedance, the voltage required to the HV side to circulate rated current in the windings is about 440V at 53A. This should be readily achievable from a mobile generator.

It is quite practicable to arrange direct control of the test generator's field from a DC source, thus giving you control of the generator output voltage and therefore the test current. You can therefore plot the entire short-circuit characteristic from a few volts up to the voltage correesponding to full rated current.

If you use this method, check everything twice, then walk away from it and check it again later. Take time to make all your connections properly, especially those on the generator field circuit and those shorting the LV winding. If you are not 100% happy, don't do the test. High power tests are not the place for mistakes.


Hope this helps.






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

Start each new day with a smile.

Get it over with.