Mains transformer failure without obvious signs

[Jolly 7]

Recently one of my multivolt wallwarts died. Initially I had suspected that the LM317T regulator inside might have failed, but replacing it made no difference. I was then surprised to discover that it was the primary winding of the mains transformer that has become open. I didn't feel it getting hot or see any smoke coming out, so was wondering how this could have happened ? Has anybody else experienced the same ?

[wave solder]

In some of these there is a small thermal fuse in the primary winding that just goes open circuit for no apparent reason. I have had it happen a few times.

[martin.m]

In the 1980s Solavox (ITT chassis) TVs sold by Comet used a small mains transformer to power the remote control circuitry and customers usually left their sets plugged in permanently. A common fault was the transformer primary going open circuit resulting in a "dead" set. I don't know what caused this though.

[Restoration73]

When the unreplaceable thermal fuse in my 3 month old DAB tuner failed, I requested
and received a new transformer. Failure was caused by disc ceramic across the bridge
going short circuit (nasty 50V ceramic, should have been 100V).
The manufacturer had of course fitted an external fuse in the primary winding, but there
is little point in a T2A fuse when the primary current was 43mA rms. I fitted a T250mA fuse and have had another 15 years of service.

[Argus25]

If you look at the side of the transformer bobbin, sometimes its possible to see the thermal fuse wires. I have seen one where the fuse has blown, and been shorted out for a repair, but an additional fuse is added in series with the primary, for a small wall wart 100mA fuse was added, as I recall.

[Jolly 7]

I attempted to locate the fuse but could not find it even after stripping the primary terminal coverings down to this level. There is still an open circuit which makes me think that the problem is too deep-seated for repair.
Secondary windings shown for interest.

[Terry_VK5TM]

I'm guessing cheap chinese wall wart -> no thermal fuse in these things, any safety protection relys on luck.

[trobbins]

Sometimes the measured capacitance of each terminal of the open-circuit winding can help 'deduce' where the open-circuit is located.

As an example, if a layer-wound primary winding was sandwiched between the core and a secondary winding, then capacitance of a primary terminal to core, and to secondary, could be noticeably different depending on the primary terminal tested.

A wire break at or very close to a terminal would likely give a very low capacitance from that terminal to anything else. That test confirmed where the break was in one vintage transformer for me, and allowed an easy fix.

Of course if the break is embedded in the winding (eg. a hot-spot type of failure), then it won't be a practical fix option (except as a rewind), and capacitance readings could be very indecisive.

[Argus25]

Originally Posted by Terry_VK5TM

I'm guessing cheap chinese wall wart -> no thermal fuse in these things, any safety protection relys on luck.

...or the thin wire of the primary acting as a fuse !

Actually, a rewind is quite easy on a split bobbin former.

[joebog1]

Argus!! is a $3 transfomer that is MOST likely dipped worth repair?
Tim!! is a $3 transfomer even worth testing?

It will be "so close" to what is available off shelf for about $10 from ( In Australia) from various suppliers, it simply isnt worth the effort. In fact a look at the local Saint Vinnies or Salvation Army shop will most likely yield 10 of each wall wart required for 50 cents or perhaps a $1 each.

Sorry guys

Joe

[kalee20]

Originally Posted by wave solder

In some of these there is a small thermal fuse in the primary winding that just goes open circuit for no apparent reason. I have had it happen a few times.

I had a small cheap wall-wart go O/C for what I presumed was an internal thermal fuse dying. No evidence of it having become over-hot before doing so. I didn't investigate - would have been nice to do so, but there were more pressing things to do! It was definitely a mains transformer / rectifier / capacitor jobbie, not an SMPS.

To answer your question, though, you're not the only one for this to happen to.

[Jolly 7]

Originally Posted by joebog1

Argus!! is a $3 transfomer that is MOST likely dipped worth repair?
Tim!! is a $3 transfomer even worth testing?

It will be "so close" to what is available off shelf for about $10 from ( In Australia) from various suppliers, it simply isnt worth the effort. In fact a look at the local Saint Vinnies or Salvation Army shop will most likely yield 10 of each wall wart required for 50 cents or perhaps a $1 each.

Sorry guys

Joe

In my opinion, as enthusiasts we sometimes like to give electronic equipment a second chance, no matter how cheap they may be to replace with new. In addition, if repairing something like an old transformer is successful, it could be used for another project if not a standard wall wart. And then there is the satisfaction of keeping something going albeit in a different shape or form that might have been of personal value. In my case, the wallwart was quite versatile and had a range of 1.5 to 12V. It served me faithfully for over ten years before failing.

[duncanlowe]

Sometimes the thermal fuse is right up against the copper windings. It's there to break the circuit if the transformer gets too hot, not (by design at least) to open on overcurrent. I too have seen this failure on transformers, and recently on the motor in my bathroom extractor fan.

[G6Tanuki]

I've autopsied a couple of these where the failure has been in the few mm of the very fine primary winding emerging from the transformer and soldered to one of the tags. Could this be caused by some sort of metal-migration effect (in the same way as solder would erode copper soldering-iron bits)?

Disassembled and stripped/rewound, these little wall-wart transformers can find a second-life as push-pull driver- and output-transformers for 'classic' couple-of-Watt audio-amps with a pair of OC81/AC128/AC153 or something like two BFY51 if you want to go silicon. Biggest problem is getting them to come apart if they've been resin-dipped.

[AC/HL]

Originally Posted by duncanlowe

I too have seen this failure on transformers, and recently on the motor in my bathroom extractor fan.

The fan in our "new" house was the same. The fuse was accessible and I managed to find one of the same ratings on eBay.

[turretslug]

Originally Posted by G6Tanuki

I've autopsied a couple of these where the failure has been in the few mm of the very fine primary winding emerging from the transformer and soldered to one of the tags. Could this be caused by some sort of metal-migration effect (in the same way as solder would erode copper soldering-iron bits)?

I've cogitated on this syndrome, too, and wondered about metal-migration- it wouldn't take much extent/depth to compromise these very fine indeed windings. Or maybe a bit of flux residue eventually corrodes the wire- it could be that something fairly aggressive gets used to ensure that an effective joint gets made to mild steel bobbin pins/tags- they're more likely to be worried about short-term than long-term dependability. Or simply that the tension on the short slip of free wire is a bit too high as the end is wound around the tag and heat/cool cycling eventually fatigues it- some of these little transformers really bake in their plastic jacket.

Annoying when it happens though.

[Skywave]

In my experience, small mains transformers are often wound with a rather mean turns per volt on the primary - and using very fine wire. So even if a primary fuse is fitted (often not), chances are that with an overload, even a relatively small one, the primary will immediately go O/C.

Al.

[Argus25]

Originally Posted by joebog1

Argus!! is a $3 transfomer that is MOST likely dipped worth repair?
Tim!! is a $3 transfomer even worth testing?

Joe

True, in theory, that is if you can get one with the same sized lamination stack that fits and secondary windings. Also most elcheapo wall warts only have a single output voltage. If you have an appliance that needs more than one voltage its a real headache. Have a look at the wall wart I had to re-wind for this project on pages 4 & 5:

http://worldphaco.com/uploads/VOTRAX...TS_HERO_JR.pdf

[Jolly 7]

This is the regulated voltage selector circuit of my old wallwart. Here are some pictures with the heatsink removed. The LM317T was screwed on to it and thermal paste had been used. The switch contacts form part of the double sided PCB itself and the voltages were selected by switching in/ out various resistor combinations. The six holes on the PCB are where the six secondary wire tappings from the burnt stepdown transformer were connected.

[joebog1]

I see your point Terry, but it's not often a supply such as yours is required. In that case I would have done something similar, although truth be told I wouldnt have bothered or tried to repair it.

I would not have charged the customer.

Joe