Anyone repaired a Maplins XM21X PSU?

[JamesF]

I have one of these that hasn't quite died, but certainly has one foot in the grave.  It's producing a wandering open circuit voltage on the output of between 9.5V and 11.5V rather than the intended regulated 13.8V at the output.

I've opened it up and there's not a whole lot inside.  Mostly it's a transformer, a couple of 4700uF electrolytic capacitors, a couple of power transistors (I think) and a lot of empty space.  There's a small circuit board with a few other components on, but that's about it.  None of the electrolytic capacitors appear to be bulging as they sometimes do when they fail.

I'm wondering if a repair might be feasible rather than having to bin it, but guidance from someone who has already done so (if such a person exists) would be very helpful.

James

[Gina]

I've repaired PSUs in the past - analogue ones -some time ago now though.

[Dr_Ju_ju]

The power transistors are probably in a current shunt circuit, so provide the amps as needed. The voltage regulator will be on the small pcb, so I'd start looking there. BTW, large electrolytic capacitors don't need to bulge before failing, have a look to see if they 'wet' i.e. their electrolyte is oozing out... 

[Carbon Brush]

A recognised problem with aged electrolytic capacitors is that they dry out if they have had a long hot life. You don't see electrolyte weeping out.

This happens when you put capacitors in a box with heat generating components like transformers, bridge rectifiers, etc for a long time.

David.

[Gina]

Yes, it could be that the reservoir capacitor isn't holding enough charge.  Check the voltage on it with a DMM or an oscilloscope.

[JamesF]

I'm not sure I'd recognise the reservoir capacitor if it bit me on the, err, fingers, which it might well

I might take some photos just to make sure I'm poking the right bit with my probes...

James

[Gina]

It will be the capacitor connected to the output of the rectifier.  Yes, take some photos, then we can see what's what with any luck.

[Dr_Ju_ju]

I'd stop worrying about the capacitors and concentrate on why it isn't producing a constant voltage when there is no load applied.

To start fault finding, measure the voltage on the output of the Bridge Rectifier, i.e. the first component after the transformer, and will usually have a large electrolytic capacitor across its +- output terminals. It should be a constant voltage under no-load conditions.  If it's steady, then fault lies with the regulator circuit, else it may then be the capacitor, but I've also known transformers\bridge rectifiers to fail.... 

[JamesF]

Here we go.  The entire innards:

The transistors are hiding under here.  I assume the awkward component to get to at the bottom is the rectifier?

What I assume to be the regulator board:

And the filthy great capacitors that nothing else wants to be anywhere near...

James

[symmetal]

Ooh, cobwebs and rust. The construction is not very neat either.

Disconnect from the mains then blow or brush out the debris. The preset potentiometers RV1 and RV2 on the regulator board show signs of rust so there is a good chance the wiper contact or track it connects to is corroded. Give them a squirt of contact cleaner if you have some and adjust the presets back and forth a few times to hopefully clean the track. RV1 probably sets the output voltage. Connect the mains and switch on and measure the voltage on the output terminals and see if it now changes smoothly with RV1. RV2 possibly sets the maximum current limit.

Check the terminals to the board with the two large capacitors are not loose. 

There could be corroded solder joints on the bottom of the regulator board which could also cause problems.

Alan

 

[JamesF]

It has been sitting in the workshop for a few months since it decided it couldn't manage the correct output voltage.  I hope that's where the cobwebs and rust have come from.  Prior to that it was sitting on my desk in the house when it wasn't being used for powering mounts and cameras which, I guess, is also a possible source of the rust when there was a lot of dew.

I have to admit that I wasn't overly stunned by the standard of construction.  Quite a bit of it looks hand-made in a hurry.  It seems an unnecessary amount of work to go to in soldering pins to the PCBs only to solder the wires to the pins, too.  Why not just solder the wires direct to the board?

James

[Gina]

Goodness me!!  That looks like an apprentice job!  A new apprentice at that.  I certainly wouldn't make such a rubbish job!  Nor have I - I've made power supplies in the past plus a heavy duty battery charger capable of 20A for charging tractor batteries.  That was analogue too with gert big heatsinks to cool several power transistors. 😁  Looks like a home made transformer too - it hasn't even got a connection panel.

Yes, the square thing in the bottom is the rectifier.  That is connected to the capacitor board with nuts and bolts.  I imagine the capacitors are wired in parallel but could be wrong.

Edited May 31, 2019 by Gina

[Alien 13]

36 minutes ago, JamesF said:

I It seems an unnecessary amount of work to go to in soldering pins to the PCBs only to solder the wires to the pins, too.  Why not just solder the wires direct to the board?

James

Its bad practice to solder directly to a board and the joint wouldn't survive the rest of the manufacturing process without breaking...

The first things I would do regarding a repair (assuming you have no diagnostic kit) is to resolder the joints on any components that either get hot by design or are large which causes mechanical strain on the joints....

Alan

Edited May 31, 2019 by Alien 13

[Gina]

Unfortunately, the rest of it is full of bad practice!

[pete_l]

9 hours ago, symmetal said:

Ooh, cobwebs and rust. The construction is not very neat either.

That was exactly my reaction. The cobwebs aren't too much of an issue, but the rust screams a warning. As you say, it is likely the two presets have corroded.

Personally, with a rusty transformer like that, I'd toss the whole thing as it's a hazard. If the transformer core has rusted then there is a chance that moisture has got into the windings. That could compromise the insulation of the primary and cause mains to appear on the low voltage side.

I'm sorry to say that if this was mine, even though I have all the test equipment needed to diagnose and fix something like this, it would go straight in the bin for safety reasons.

[Gina]

Yes, I agree.  Certainly could be a hazard and not worth trying to repair.  Chuck it and buy a decent one.  Good quality power supplies are not expensive.  I prefer to repair where possible but Pete has a good point - that transformer has not been properly made and the windings may not be properly insulated.

Edited June 1, 2019 by Gina

[Stub Mandrel]

The best advice was to to wiggle the voltage adjustment pots a few times and see if it can be reset to 12V.

The split bobbin transformer will ensure you aren't exposed to dangerous voltages if anything fails.

If the mains winding fails it will just stop working.

The construction may be ugly, but as long as the box is earthed it should be safe.

 

[Dr_Ju_ju]

I agree with Neil, it looks worse that it actually is, for its age.....   

The transformer former maybe a tad rusty on the outside, but I've seen worse, in supposedly pro gear, but as long as its doing its job, then ignore it. If it bothers you that much, change it for a new toroidal transformer, as long as it has the same specifications for output voltage and current.....

The two 4700uF caps are common components, so could be replaced if needed, and I suspect the board they are on can take multiple caps, as they would be wired in parallel.

But as everyone else has pointed out the 2 pre-set pots look like they've come to the end of their useful life, personally I'd just replace them, but instead of using open-frame construction pots, I'd use enclosed and\or even multiturn pots, which will give finer control.

 

[Gina]

Could you take a photo of the other side of the transformer James?  It looked to me as if the mains connections were on the same side as the secondary but maybe not.

[Stub Mandrel]

17 minutes ago, Gina said:

Could you take a photo of the other side of the transformer James?  It looked to me as if the mains connections were on the same side as the secondary but maybe not.

You can see the mains leads disappear behind the transformer, they will be wired to the bottom coil on the far side.

The second pair of wires from the low-tension side of the transformer go to terminals marked 'AC' on the control board. I don't know what these do but I suspect they function in a different version with a fully populated board.

[JamesF]

32 minutes ago, Stub Mandrel said:

The second pair of wires from the low-tension side of the transformer go to terminals marked 'AC' on the control board. I don't know what these do but I suspect they function in a different version with a fully populated board.

That did confuse me, I have to admit.  I couldn't see an obvious reason for an AC feed to a mere DC regulator board.

I shall have a play with the unit over the weekend when I have time and do as people have suggested regarding testing/cleaning up.  In truth, if it has to go in the bin then it will go in the bin after I've scavenged a few bits that might come in useful, even if that just turns out to be the fuse holder, switch, terminals and heatsink.  I've already replaced it thanks to a used purchase from another SGL member, but I saw it on the bench the other day when I had a clear-up and thought it would be preferable if it could be made to work again rather than just scrapped.  Our "disposable" society does pain me sometimes :)

James

[Stub Mandrel]

I think it;'s probably to detect whether its plugged in or not, if the board is built up as a battery charger.

[JamesF]

6 minutes ago, Stub Mandrel said:

I think it;'s probably to detect whether its plugged in or not, if the board is built up as a battery charger.

Ah, yes, that might well make sense.

James

[Oddsocks]

15 hours ago, symmetal said:

Give them a squirt of contact cleaner if you have some and adjust the presets back and forth a few times to hopefully clean the track.

It's perplexing how often I see this recommended across the web and even written in published manuals.

On the REME sponsored courses I did in Germany back in the late 70's early 80's we were told never to use any type of contact cleaner or solvent on skeleton carbon presets as it softens the carbon bonding agent on the track and within a very short while the preset would fail completely. The general recommendation was to replace a defective preset if possible but if desperate we were instructed to gently lift the wiper from the track and slide a piece of typewriter paper under the wiper, release the wiper back on to the paper and wiggle the preset back and forth a few times to clean the wiper contact, then slide the paper away.

We were told that contact cleaner is only for metals, never for carbon or carbon composites, at least, that was the instruction then and I've not seen any changes to the construction of skeleton presets that would suggest it is now acceptable to squirt solvents on to them.

[Oddsocks]

15 hours ago, Gina said:

Goodness me!!  That looks like an apprentice job!  A new apprentice at that.

I don't even think it was the apprentice Gina, probably one of the warehouse operatives filling in while the apprentice was on day-release.