I have this 30A 5V smps power supply that i was using for a ws2812b led matrix ( 288 led). It worked fine, but then i had some things going on and didnt use it for about a few weeks.
I connected everything again but this time for some reason the supply is outputting exactly 2.5 volts.
I thought something was wrong with the timing capacitor so i pulled up the datasheet of the only ic on board and saw that it has a timing capacitor and a timing resistor(?). so i replaced the capacitor with an identical model and checked the resistance of the resistor which matched with its markings.
Reconnected everything and still have the same problems.
Nothing seems physically damaged on the pcb, the pcb itself was contained in a steel box which was also in another relatively safe container, so very low chance of actual physical damage.
Please note that this didn't fail during operation, i tested it extensively with various brightness for extended periods of time.
The board ( i wasnt sure how to mirror the traces image, so i marked the corners on both sides)
What is the voltage coming out of the wall in your location ?
And have you tried turning the trimpot ? The one next to the screw connections and the power LED .
Maybe it is not making a good connection and therefore not providing the proper reference. Be careful !! Those big capacitors may still hold energy. You should discharge them before handling the board.
Why did you think so? I.e. have you done any measurements that made you to believe this might have been the problem and if so, what were those measurements and the outcome?
If you load the output with e.g. 1A, does the voltage remain stable?
Is the output voltage reasonably clean if you examine it on a scope?
That would be the absolute first thing to check, indeed.
Also, it looks like on the underside of the PCB in that particular corner there is some dried crud, which might be flux residue, but it might also be junk that points towards contamination and bad solder joints. I'd inspect that closely and if necessary resolder that part of the PCB.
I googled a bit and found someone who had the same issue. for that person, changing the timing capacitor worked. well, considering how smps operates, this idea didn't seem too far fetched.
I haven't tested with a load, will do that now.
sadly I don't have an oscilloscope.
I see. Well, there could be any number of things wrong with the unit; the timing capacitor is maybe not an entirely unlikely guess, but neither would it be the first thing that comes to mind.
That's a pity. You could fashion one out of the analog input of an Arduino and having its sketch output its reading frequently to the Serial plotter in the IDE. While it's a bit of a lor frequency scope, it will at least tell you something about what's coming out of that unit and if it looks halfway decent.
I couldnt find low enough load resistor so i took a long wire and twisted it to mitigate inductance.
Terminal voltage shows 2.44 volt with no load, stays around 2.16 with load connected. current measured a stable 3.5 Amps. Twisting the pot does nothing whether or not the load was connected.
edit: Also, cleaned the board the best I could
Ok, that's pretty bad given it's supposed to be a 30A unit. Apparently it fails completely to maintain regulation.
Is there any sign of leakage of the large input capacitors?
You have the choice between:
1 bad scope
2 no scope.
Up to you to do the math.
Honestly given the situation as it's presented now I wouldn't even think of trying to further troubleshoot this without a scope. Hence the Arduino is worth a shot.
Yes, monitor output especially under load. If there's a lot of noise on it, it'll probably show up on the low sample speed of an 'Arduino scope' even if it has a higher frequency. Otherwise consult this for speeding up ADC reads: Arduino Fast(er) Sampling Rate. You could sample in burst mode and visualize the outcome.
If you're somewhat serious about electronics as a hobby, I would heartily recommend getting a somewhat OK oscilloscope.
no they don't need to be configured, they share the same big trace on the PCB. I have 4 terminals, so I bolted the voltmeter probes to two and connected the load on the other two.
yes the selector is set correctly
I meant low enough as in loading it for a few amps, I think the long wire did a splendid job at that. My initial reading of the wire's resistance was 1.5 Ohm but I think that was incorrect for some reason ( oxidation on the wire probably ) and decided to measure the current through it, which was way more stable and gave a believable rating.
also I could've probably used a bunch of resistors in parallel to get lower value resistor but all I have in hand are 1/4 watt resistors so they might burn up.
I am quite serious about electronics, but right now oscilloscopes are far out of my reach, even cheaper computer based ones.
What am I exactly looking for in the output? Say I see that the output is very choppy, what would that indicate? or how would that guide to what part is faulty?
Also, Im not quite sure, but what exactly is the small potentiometer's job in the circuit? As I remember, it didn't significantly adjust the voltage of the output, just minor adjustments, would a faulty pot cause this much difference? also, if it was faulty, shouldn't the ic have regulated the voltage just at 2.44V, but I saw it drop to 2.16V, which as you said, was a failure at regulating.
That depends mostly on the frequency and amplitude. So observe first, then speculate.
It forms a voltage divider with one or two other resistors and those are part of the feedback circuit that allow the TL494 chip to monitor the output voltage.
Yes, that's usually what it does.
It won't cause more difference than its normal adjustment range if it were to operate correctly. It's worth trying, but honestly I think odds are that this is not the cause of your problem.
Correct, so there's something very wrong with this unit.
