Another power supply 13.8V 20A

[Northy]

I found this one in the bin at an RC car racing meeting, I thought I could maybe use the case for something atleast.
Anyway, I'd like to have a go at fixing it, but this will be my first mains PSU repair.

Could someone point me in the right direction on where to start?
How do these things work?

Thanks,

G

[capt bullshot]

It's a standard crappy switcher.
Do you have any experience with measuring voltages at mains potential?

The mains input is rectified, the two electrolytics on the right hand side are the primary bulk capacitors.

At first, I'd ask you plug it in, turn it on and measure the voltage across these - it's either near zero or some 320V (in a 230V country) or 160V (in a 110V country). If you don't know what I mean, stop here - it's just dangerous. If it's zero, check the fuse, rectifier diodes and other parts on the way from mains input to the capacitors.

Next stage is a (most probably) MOSFET half bridge and the high frequency transformer (the large part with yellow tape in the middle). The MOSFETs chop the rectified mains to some 10kHz, this goes through the transfomer.
The primary MOSFETs are mounted at the right hand heat sink. One could check them with a DMM in diode mode, but I suspect they're OK, because they often fail shorted and let the smoke out. This would be visible here.
Mounted on the other heatsink are rectifier diodes, these rectify the high frequency into the output DC voltage. The toroid is the main storage inductor for "smoothing" the output.
The IC (a standard TL494) is the control circuitry and regulates the output voltage. It's located on the secondary side, so this circuit must have some self oscillating capabilities to enable start up.
You most probably won't find a fault here without a scope, as you'd have to watch if the control circuitry tries to start up.

Did I mention it's quite dangerous to work with DMM and scope on a mains powered circuit? I you're unsure and don't know what I'm speaking of, better stop here.

Some other method would be to test the components in a "safe" (non-powered) state and try to find a defective component. Check the discrete semiconductors, the electrolytics and the red film capacitor. Transformers and inductor typically don't fail here. Maybe you'll find an open resistor or a bad solder joint. Fix what you found, and carefully test the unit. It might explode on you.

If you google for "PC silver box" schematics of the older types, you'll find diagrams of power supplies that work in a similar way.

TBH, I'd just bin the innards and re-use the case. It's a crappy supply.

[George Edmonds]

Hi

I totally agree with Capt Bullshot, bin it before it kills you.  I can see no sign of any primary (high voltage) / secondary (low voltage) insulation barrier, indeed the two appear to intermingle.  NEVER NEVER EVER use a scope around a switch mode power supply unless you want to destroy the scope and possibly blind your self with molten copper, there are up to 1,000V pluses present.

Keep Safe

George G6HIG Dover UK

[floobydust]

Where is the FUSE ?

[ChuckDarwin]

That is some dodgy assembly.  I like the inductor riding on top of the folded over cap.  Salvage what you can and tip it, this is not a safe assembly. 
If you are going have a go at getting it working, fix a few things that aren't going to be the primary problem as well: C3  looks like it may have failed (second photo, bottom).  Lots of components have excessive lead length and may be in contact with other components.  Look at the little cluster of resistors just left of T1 under the cable sleeve in photo 1, for example.   Don't use it after you fix it.

[coromonadalix]

The fuse could be a socket on the rear panel ??

If you want to probe in a psu like this with a scope, you must use an isolation transformer on the scope

Yep c3 seems to have failed

I dont see this psu as a gimmick  its an classic TL494 based design,  until you have full schematics i would not say its a junk,   but i dont like the toroidal coil floating around and over a capacitor,  yes i call that dodgy

And you have coupling transformers between the low power side and the main side, totally safe up to a point.

And yes  you have to know how to work in a psu like this,  if its a 120vac unit  the main dc voltage could be 200vdc and up to 350volts dc, they use a trick to double the ac to dc rectification and create a virtual ground between the 2 main capacitors, see the 2 blue resistors near them, if one of them pops, the psu could not work well.

R27 seems to have heated a lot or faded with time   i would change it and use longer leads to lift it away from the pcb

Any branding on the case   the numbers on the pcb  doesn't give anything on google

[janoc]

If you want to probe in a psu like this with a scope, you must use an isolation transformer on the scope

Good way to kill yourself, especially a newbie who doesn't really know what they are doing.

Get a proper high voltage differential probe instead of isolating your scope and potentially getting lethal voltages on the exposed scope terminals.

[coromonadalix]

ah  i'm still alive  loll  doing this now and then   if you know how to do that

Anyways  critisizing is not constructive to help the OP with his psu

The OP must be aware that danger is lurking in the psu  and any wrong doing could be hazardous

Once the warnings are done, what can we do to help

just trash it ?   if OP want to repair it ?   you have many results on the web involving tl494 psu designs / schematics

[Northy]

Thanks for all the feedback and help so far!

A bit of background about me. I've been an electronic engineer for 19 years, but pretty much always hand held or low power stuff. I'm just starting to potentially move into the world of doing a bit of design that involves mains voltages on PCB's so that's why I'm starting to look at the rules/regulations.
One of the ideas I had was to have a look at this broken PSU that I've had in the garage for a few years! I've also got a bench lab PSU in there somewhere that needs repairing when I find it.

It's interesting that people are less than impressed by the design/construction. I'm no expert in mains PSU design obviously, but it's didn't look 'nice' to me when I took the lid off. However, I think if you cut open most China wall warts they'd have similar types of construction wouldn't they?
Interesting also because a few years ago this was a pretty popular PSU in the RC racing world (probably only UK). These days most people seem to use a converted server power supply. I've also got a box of those I rescued out of a skip!

I don't have a scope at home, but I plan to change that soon.

Thanks all.

G

[janoc]

ah  i'm still alive  loll  doing this now and then   if you know how to do that

Anyways  critisizing is not constructive to help the OP with his psu

I think I have said pretty clearly what is the correct way of doing things there. Or maybe I understand the idea of constructive criticism wrong?

The risks of isolating the scope and why is that a bad idea have been discussed on this forum in the past, including a case of an engineer who got hurt doing that, despite taking a lot of precautions.

See e.g.:
https://www.eevblog.com/forum/projects/floating-scopedut-how-to-explain-to-my-boss-that-this-is-wrong/

https://www.eevblog.com/forum/testgear/floating-scopes/

If you want to use an isolation transformer, use it on the DUT, not the scope. If you don't have a differential probe, use two channels of the scope and a math function to subtract them. Then you can probe the circuit without worrying about frying your scope or shorting something out.

The OP must be aware that danger is lurking in the psu  and any wrong doing could be hazardous

That's not in dispute. But explicitly giving dangerous advice to someone who probably doesn't know any better, like floating a scope, is a different matter.

Once the warnings are done, what can we do to help

just trash it ?   if OP want to repair it ?   you have many results on the web involving tl494 psu designs / schematics

Given the "quality" of the construction of that supply, that would be trash for me. Having a schematic for something doesn't mean it is worth fixing - how can you be sure that the transformers have adequate insulation given the "quality" of the rest? There doesn't seem to be any  low/high voltage barrier neither, as mentioned by others already. No way to fix that, regardless of any schematic.

[capt bullshot]

This (the component tester of an old Hameg scope) might be a way to debug the supply safely:

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/msg3266896/#msg3266896

[floobydust]

In a perfect world you would have a scope, isolation transfomer, HV probes, schematic and a full understanding of how it works.
You can do alot without even powering the unit up.  Look for cold solder joints, test the usual suspect components. People do it all the time using just a multimeter.
The power supply is model FG-114 by ChenghuiXing Electronics (Huizhou) Co., Ltd. Many components are crooked and not fully seated, so this is hand assembled in the gutters of guangdong. I would make sure components are not touching or shorting to each other.