Testing mains transformers - UK Vintage Radio Repair and Restoration Discussion Forum

McIntyre · 9th Apr 2005, 10:24 am

I've got a quantity of mains transformers - I know there is a way to test transformers on an oscilloscope for shorts, opens etc - got the test circuit somewhere. Is it possible to place a load (resistive) on mains transformers to test how they would actually work in circuit? If so how would you work out the resistor values for various loads (in mA)?

If a person put this piece of test equipment together in, a box one one could put a AC ammeter in series with a rotary switch to switch on different loads and monitor it - I know a person gets something similar to switch in different loads for the output transformer when doing allignment without the speaker connected. Is there any commercial circuit I could build for the mains transformer? Or can anyone suggest anything about the construction and values if i had to build it up from scratch?

Andrew

Ed_Dinning · 9th Apr 2005, 8:21 pm

Hi Andrew, the standard tests on mains transformers, even vey big ones, is to run them at rated voltage and measure the output voltage at no load as well as the input current .
Input current gives a measure of the goodnes of the core and its losses. It will also show shorted turns if you have previous or similar figures to go by.
It will also give you the turns ratio.
The second test is to short one of the windings through an AC ammeter and slowly wind up the input voltage on a variac until you have the correct secondary current flowing (input volts will be quite low). This is a measure of the copper losses of that winding. It is also a good way to dry out transformers that have been stored for some time.
There is also a Sumpner test that has 2 similar transformers connected back to back and allows all the loses to be calculated.

Ed

Tim · 9th Apr 2005, 9:07 pm

A tip when testing mains transformers of indeterminate type is to use 24 volts AC in place of the usual 240v.
Outputs will of course be a factor of ten less, but effects common to AC(such as shorted turns etc. ) will still show.
Any "disaster" will of course be much reduced.
Regards
Tim

Gordon · 10th Apr 2005, 12:24 pm

Also, for safety, it might be wise to check the insulation resistance from primary to the laminations (and clamps) before connecting any mains voltage. Should be near infinite.
Gordon.

Station X · 10th Apr 2005, 1:38 pm

Quote:

Also, for safety, it might be wise to check the insulation resistance from primary to the laminations (and clamps) before connecting any mains voltage. Should be near infinite.
Gordon.

Is it OK to use a 500 Volt Megger for this test. That's what a Portable Appliance Test (PAT) would do? What about testing the insulation resistance between windings. Is 500 Volts OK here?

Graham.

pmmunro · 10th Apr 2005, 4:09 pm

Graham,

As you'll be testing the transformer with no other components connected, using a Megger type tester at 500V is a good test to apply. (Most modern, battery powered, insulation testers specified as 500V will apply 500V (DC). As the peak of a 230V rms waveform is 325V, this test adds a reasonable, but not excessive, factor of safety). Also remember that the peaks voltages between windings may be in anti-phase when working, doubling the voltage difference.

Many insulation testers in present-day use have 250, 500 and 1000V test voltages and I would not consider it excessive to expect a good mains transformer to be designed to withstand at least 1000V for a period of minutes as a safety test.

A few years ago, I asked a well-known rapid delivery component supplier for a test specification for their clamp type power transformers, which in their usual helpful way they did.

This calls for the following insulation tests at 2800V (DC) and gives minimum figures for the same tests at 500V (DC), shown in square brackets

a) Primary to secondary [5 megohms]

b) Primary to core [7 megohms]

c) Secondary to core [7 megohms]

d) Primary to Primary [2 megohms]

e) Secondary to Secondary [2 megohms]

It also lists flash tests for the same connections, all for 1 minute at the AC rms voltages shown:

a) 4000V b) & c) 2000V d) 1000V and e) 500V

This implies that a 500V test would be a minimum for safety, although a very old transformer might not be designed to such a specification. Possibly you may think it is not worth the risk of using a transformer which can't withstand 500V.

Full load temperature rises of 40 degrees celsius above ambient on full load are not unusual, and will depend on the insulation class and the application. Temperature rises above this value may be permissible but with derating of the power output of the transformer. Given an ambient temperature of 20 deg. C, this would make a fully loaded transformer uncomfortanly hot to the touch.

Someone with more expertise in transformer design will be able to give more detail of transformer testing and insulation classes, or you may find abbreviated version of the appropriate British Standards on transformer manufacturers' web sites.

jim_beacon · 11th Apr 2005, 7:50 pm

Tektronix, in one of their application notes, gave a basic circuit using an op-amp (in theirs, it was a type O plug-in), to give a scope display that would indicate shorrted turns in a transformer.

If anyone is interested, I'll dig it out and scan it.

Jim.

McIntyre · 13th Apr 2005, 2:54 pm

Quote:

Originally Posted by jim_beacon

Tektronix, in one of their application notes, gave a basic circuit using an op-amp (in theirs, it was a type O plug-in), to give a scope display that would indicate shorrted turns in a transformer.

If anyone is interested, I'll dig it out and scan it.

Jim.

I think that this circuit is a bit simpler - sorry for the poor quality. Its an old photostat. But seems to do what the Tektronix does..

jim_beacon · 19th Apr 2005, 9:57 pm

Sorry for the delay,

the scan of the Tek data is here:

www.g1jbg.co.uk/service.htm

and look for the "shorted turns detector" link (nearly at the bottom of the page).

Apologies for the poor qaulity of the scan, it is taken from a dodgy photocopy of the original manual :-(

There is a scan of the full manual on the BAMA site:

http://bama.sbc.edu/

and drill down to the Tektronix pages. It is, however, a very large document to download!

Jim.

9th Apr 2005, 10:24 am	#1
McIntyre Retired Dormant Member Join Date: Sep 2004 Posts: 0	Testing mains transformers I've got a quantity of mains transformers - I know there is a way to test transformers on an oscilloscope for shorts, opens etc - got the test circuit somewhere. Is it possible to place a load (resistive) on mains transformers to test how they would actually work in circuit? If so how would you work out the resistor values for various loads (in mA)? If a person put this piece of test equipment together in, a box one one could put a AC ammeter in series with a rotary switch to switch on different loads and monitor it - I know a person gets something similar to switch in different loads for the output transformer when doing allignment without the speaker connected. Is there any commercial circuit I could build for the mains transformer? Or can anyone suggest anything about the construction and values if i had to build it up from scratch? Andrew

9th Apr 2005, 8:21 pm	#2
Ed_Dinning Dekatron Join Date: Jan 2004 Location: Newcastle upon Tyne, Tyne & Wear, UK. Posts: 8,195	Re: Testing mains transformers Hi Andrew, the standard tests on mains transformers, even vey big ones, is to run them at rated voltage and measure the output voltage at no load as well as the input current . Input current gives a measure of the goodnes of the core and its losses. It will also show shorted turns if you have previous or similar figures to go by. It will also give you the turns ratio. The second test is to short one of the windings through an AC ammeter and slowly wind up the input voltage on a variac until you have the correct secondary current flowing (input volts will be quite low). This is a measure of the copper losses of that winding. It is also a good way to dry out transformers that have been stored for some time. There is also a Sumpner test that has 2 similar transformers connected back to back and allows all the loses to be calculated. Ed

9th Apr 2005, 9:07 pm	#3
Tim Dekatron Join Date: Jan 2003 Location: Bradford on Avon, Wiltshire, UK. Posts: 3,310	Re: Testing mains transformers A tip when testing mains transformers of indeterminate type is to use 24 volts AC in place of the usual 240v. Outputs will of course be a factor of ten less, but effects common to AC(such as shorted turns etc. ) will still show. Any "disaster" will of course be much reduced. Regards Tim

10th Apr 2005, 12:24 pm	#4
Gordon Pentode Join Date: Oct 2003 Location: Bournemouth, Dorset, UK. Posts: 131	Re: Testing mains transformers Also, for safety, it might be wise to check the insulation resistance from primary to the laminations (and clamps) before connecting any mains voltage. Should be near infinite. Gordon.

10th Apr 2005, 4:09 pm	#6
pmmunro Octode Join Date: Jul 2004 Location: Dundee, UK. Posts: 1,813	Re: Testing mains transformers Graham, As you'll be testing the transformer with no other components connected, using a Megger type tester at 500V is a good test to apply. (Most modern, battery powered, insulation testers specified as 500V will apply 500V (DC). As the peak of a 230V rms waveform is 325V, this test adds a reasonable, but not excessive, factor of safety). Also remember that the peaks voltages between windings may be in anti-phase when working, doubling the voltage difference. Many insulation testers in present-day use have 250, 500 and 1000V test voltages and I would not consider it excessive to expect a good mains transformer to be designed to withstand at least 1000V for a period of minutes as a safety test. A few years ago, I asked a well-known rapid delivery component supplier for a test specification for their clamp type power transformers, which in their usual helpful way they did. This calls for the following insulation tests at 2800V (DC) and gives minimum figures for the same tests at 500V (DC), shown in square brackets a) Primary to secondary [5 megohms] b) Primary to core [7 megohms] c) Secondary to core [7 megohms] d) Primary to Primary [2 megohms] e) Secondary to Secondary [2 megohms] It also lists flash tests for the same connections, all for 1 minute at the AC rms voltages shown: a) 4000V b) & c) 2000V d) 1000V and e) 500V This implies that a 500V test would be a minimum for safety, although a very old transformer might not be designed to such a specification. Possibly you may think it is not worth the risk of using a transformer which can't withstand 500V. Full load temperature rises of 40 degrees celsius above ambient on full load are not unusual, and will depend on the insulation class and the application. Temperature rises above this value may be permissible but with derating of the power output of the transformer. Given an ambient temperature of 20 deg. C, this would make a fully loaded transformer uncomfortanly hot to the touch. Someone with more expertise in transformer design will be able to give more detail of transformer testing and insulation classes, or you may find abbreviated version of the appropriate British Standards on transformer manufacturers' web sites.

11th Apr 2005, 7:50 pm	#7
jim_beacon Retired Dormant Member Join Date: Mar 2003 Location: Bishop's Waltham, Hants, UK. Posts: 939	Re: Testing mains transformers Tektronix, in one of their application notes, gave a basic circuit using an op-amp (in theirs, it was a type O plug-in), to give a scope display that would indicate shorrted turns in a transformer. If anyone is interested, I'll dig it out and scan it. Jim.

19th Apr 2005, 9:57 pm	#9
jim_beacon Retired Dormant Member Join Date: Mar 2003 Location: Bishop's Waltham, Hants, UK. Posts: 939	Re: Testing mains transformers Sorry for the delay, the scan of the Tek data is here: www.g1jbg.co.uk/service.htm and look for the "shorted turns detector" link (nearly at the bottom of the page). Apologies for the poor qaulity of the scan, it is taken from a dodgy photocopy of the original manual :-( There is a scan of the full manual on the BAMA site: http://bama.sbc.edu/ and drill down to the Tektronix pages. It is, however, a very large document to download! Jim.