[REPAIRED] Cheap 150W 48V SMPS repair (clicking noise)

[dominicM]

I have a cheap chinese 150w 48V power supply which failed after little use. It is actually the second psu like this one to fail in the same way so I would like to find the cause of the failure. No visible damage or any noise/smell at failure time. The psu just keeps making clicking noise with output in the mV range. I also have a similar 24V 150W supply that I used as a comparison. The voltages are the same as the 24V psu up to R4 (~330V). Measuring voltages across the 2 electrolytic caps near D4 give 1V on the left one and 13-15V on the right one. 24V psu was a stable 15V for the latter. Both caps measure 10uF after desoldering (have no replacements so soldered them back in). Both diodes measure about 0.5V or OL depending on direction. All resistors are within spec also. Is it likely that the IC itself failed on both psus? Found no shorts and am out of ideas at this point, any suggestions?

[retiredcaps]

In my limited experience, a clicking noise suggests one of the following:

1) a dried up startup electrolytic cap with high ESR that is tied to the SMPS IC VCC pin

2) a shorted component on the secondary side like a diode

3) a shorted SMPS IC - lookup the datasheet and see if VCC is shorted to GND

[dominicM]

In my limited experience, a clicking noise suggests one of the following:

1) a dried up startup electrolytic cap with high ESR that is tied to the SMPS IC VCC pin

2) a shorted component on the secondary side like a diode

3) a shorted SMPS IC - lookup the datasheet and see if VCC is shorted to GND

It being a cheap Chinese PSU there's no  documentation and the IC markings are scratched off. I tested all diodes and there are no shorts. Is it likely that the cap is dried up/high ESR and still be within capacitance spec? There are also no shorts between any of the IC pins.

[retiredcaps]

Is it likely that the cap is dried up/high ESR and still be within capacitance spec?

Yes, a capacitor could measure within specification with regards to capacitance, but have high ESR.

Some people suggest warming up the capacitor with a hair dryer to lower its ESR initially so the SMPS IC can start.  Of course, long term, lots of heat kills the capacitor.

[Seekonk]

There is a winding on the primary side that powers the chip once it starts.  There is a high ohm resistor that feeds it so it initially turns on.  The clicking means it is not getting powered by this winding.  The winding may be open or the diode on it is bad.  Since it tries to start the cap and zener are probably good.

[dominicM]

Will give it a shot, should the caps be heated before applying AC power or it doesn't matter if it's already clicking away when caps are heated? I can probably salvage some caps, can I replace the 10uF caps with lets say 100uf caps of same voltage? Which ones are worth replacing first/most likely to be faulty in your opinion?

[dominicM]

Will give it a shot, should the caps be heated before applying AC power or it doesn't matter if it's already clicking away when caps are heated? I can probably salvage some caps, can I replace the 10uF caps with lets say 100uf caps of same voltage? Which ones are worth replacing first/most likely to be faulty in your opinion?

The D1 and D2 are both OL / 0.48V.
R1 measures 0.9 (R value is 0.15).
R3 measures 95k.

L2 is actually replaced (with 22ohm) though I think it's the wrong value and I replaced it with the original since. The original resistor is brown, brown ,black, silver (or maybe red, red) which is 11ohms but it measures 0.7. Perhaps it's just my multimeter unable to measure it  properly because the other failed 48V psu and the working 24V psu have identical values and measurements.

[retiredcaps]

Which ones are worth replacing first/most likely to be faulty in your opinion?

I can't tell from the photo, but either C2 or C3 will be your startup capacitor.

[retiredcaps]

L2 is actually replaced (with 22ohm) though I think it's the wrong value and I replaced it with the original since.

Components that start will L are usually an inductor.  Why did you replace it with a resistor?

[dominicM]

L2 is actually replaced (with 22ohm) though I think it's the wrong value and I replaced it with the original since.

Components that start will L are usually an inductor.  Why did you replace it with a resistor?

I assume it was marked wrong on the pcb because all 3 psus have same marking and have a resistor there and not an inductor. It has 4 bands - brown, brown, black, silver.

[SaabFAN]

There are small inductors that have the same ring markings as resistors.
Sure it is a resistor?

[dominicM]

Which ones are worth replacing first/most likely to be faulty in your opinion?

I can't tell from the photo, but either C2 or C3 will be your startup capacitor.

Ok, I swapped out C2 and C3 caps between working 24v psu and clicking 48v psu and still same, 24v is working but 48v is clicking so the caps must not be the issue.

[dominicM]

There are small inductors that have the same ring markings as resistors.
Sure it is a resistor?

Yes, it does look like it's an inductor after all   Resistance reading was close to marked bands so I didn't even consider it at the time.

[strawberry]

Replace D2

[SeanB]

Resolder all the pins on the transformer, you probably have a dry joint on the feedback winding. The one joint does look pretty dodgy, and resoldering them will at least fix any dry joints there. Also resolder R15 and R7, the joints there look terrible, and they are also a part of the feedback.power supply for the chip.

[dominicM]

Replace D2

No change, what was the reason for this suggestion specifically?

[dominicM]

Resolder all the pins on the transformer, you probably have a dry joint on the feedback winding. The one joint does look pretty dodgy, and resoldering them will at least fix any dry joints there. Also resolder R15 and R7, the joints there look terrible, and they are also a part of the feedback.power supply for the chip.

Done with no change and I really don't see how a dry joint could be a cause of 2 psu's failure with no physical stresses

[strawberry]

Replace D2

No change, what was the reason for this suggestion specifically?

Sometime they use wrong parts.

Once I had bias winding broken.

[xavier60]

The ticking means that C2 is being charged up to startup threshold voltage for the IC but not being sustained. For the 48V output rail to be so low, there could be a short on it.

[dominicM]

The ticking means that C2 is being charged up to startup threshold voltage for the IC but not being sustained. For the 48V output rail to be so low, there could be a short on it.

I replaced C2 and C3 so the caps are fine and tested it to make sure - no shorts.

[dominicM]

The ticking also persists after ac power is removed for about a 2 minutes slowing down until it stops. Thought I might add that in case it provides any clues to someone.

[xavier60]

Did you also check the 48V output for a short circuit? There can be can be other reasons like an open circuit in the secondary circuit.

[xavier60]

It is also possible to have shorted windings in the transformer. This can be difficult to prove.

[_Wim_]

Replace D2

No change, what was the reason for this suggestion specifically?

Do not replace this one by a "normal" diode. You need to use a ultra fast recovery type in many circuit to work correctly. Also ensure is has a high enough reverse breakdown voltage. I had 2 SMPSes failing because the diode reverse breakdownvoltage was serverly limited (got to hot probably), but measured still ok with a multimeter.

Some more info which type to select can be found here (altough I just picked one I had lying around from a previous project):
https://www.maximintegrated.com/en/app-notes/index.mvp/id/849
 

[kxenos]

Is U2 LM431? Can you swap it with a good one?