Faulty SMPS

[dark_hawk]

Hi,

A friend of mine starting in Electronics got a Riden 6006 power supply, I got him a Meanwell LRS-150-48 to power that 6006. He accidentally connected the live 220 AC to  V+ and V-  on the meanwell which caused the circuit breaker to pop of course. Now I'm trying to diagnose the Meanwell power supply for what blew when you connect AC to the output.
When you connect it, the led briefly lights on, and the output voltage starts decreasing.

Only thing I found in the output circuit with a problem is a 200 ohm resistor which was open. Output rectifier diode is fine, TL431 is also fine, I even replaced it to test. All the resistors and zeners in the secondary section are fine. Also the opto isolators are ok.

The voltage at the main primary capacitors is 155V, which I think means that the input circuit up to the capacitors are fine. Mosfet is also not shorted, primary side diodes are also ok.
How do I proceed?

[timenutgoblin]

Mosfet is also not shorted

Which MOSFET are you referring to? On the secondary side there is unpopulated component Q100 (MOSFET) and next to it is a grey coloured box thing. Is that a MOSFET or a schottky diode? There needs to be a rectifier or synchronous switch (MOSFET) on the secondary side of the transformer. I can see at least two secondary windings. U1 is probably a PWM controller for driving primary MOSFET.

[tunk]

Wouldn't really know - have you checked the transformer windings?

[dark_hawk]

Mosfet is also not shorted

Which MOSFET are you referring to? On the secondary side there is unpopulated component Q100 (MOSFET) and next to it is a grey coloured box thing. Is that a MOSFET or a schottky diode? There needs to be a rectifier or synchronous switch (MOSFET) on the secondary side of the transformer. I can see at least two secondary windings. U1 is probably a PWM controller for driving primary MOSFET.

I meant the Mosfet on the primary side. The secondary has two diodes in one TO-220 package, and it's fine.
This page has the schematic of a PSU from the same series and I think the same topology but with different output voltage and wattage rating.
https://www.kirich.blog/obzory/bloki-pitaniya/823-nizkoprofilnyy-blok-pitaniya-meanwell-lrs-350-24.html

[dark_hawk]

Wouldn't really know - have you checked the transformer windings?

The secondary winding of the transformer I checked with an LCR meter at 1khz 79.5 µH and 190µH at 100khz.
That winding which is not connected to anything reads 9.6µH and 21.6µH.
I think the transformer might be ok.

[timenutgoblin]

I think that when the mains AC voltage was applied to the output it underwent a voltage step-up between the secondary and primary sides of the transformer. This would've resulted in there being a much higher (than expected) voltage appearing on the auxiliary winding (next to the primary winding). Check D30 and C35. Hopefully U1 is still okay. The auxiliary winding looks like it powers the PWM controller.

At the moment, you have a One-Shot Monostable Multivibrator. Not very useful...

[dark_hawk]

D30 is fine, so C35. The PWM controller does not have any apparent sign of damage, but that's the only thing I didn't check because I don't know how, and it's a custom meanwell part so it won't be neither easy or quick to source a replacement. 

The strange thing is that before changing that 200 ohm resistor the PSU would start up and work for like 3 minutes before shutting down, not like now where it won't even start up at all!

[magic]

This page has the schematic of a PSU from the same series and I think the same topology but with different output voltage and wattage rating.
https://www.kirich.blog/obzory/bloki-pitaniya/823-nizkoprofilnyy-blok-pitaniya-meanwell-lrs-350-24.html

That's a forward topology PSU, see that big inductor L2 near the transformer output and the arrangement of diodes. Yours is probably a simple flyback and the diodes (or only one of them) are connected directly between the secondary winding and the output capacitors.

That being said, the control circuits may be similar. One optocoupler is driven by TL431 and regulates voltage, not sure what the other is doing yet.

If think the primary switch, transformer and secondary rectifier have to be at least somewhat functional if there is any voltage output.
What's the peak voltage that appears on the secondary before the PSU turns off?
Do you have a scope? (btw, don't connect it anywhere on the primary)

[timenutgoblin]

Are you able to check/test the zener diodes ZD100, ZD102, ZD103? It looks like they're connected between the output of the PSU and the optocouplers for feedback for U1.

I may have found a candidate for U1: NCP1218 (pinout is similar). Check for yourself.

https://www.onsemi.com/pdf/datasheet/ncp1218-d.pdf

[dark_hawk]

This page has the schematic of a PSU from the same series and I think the same topology but with different output voltage and wattage rating.
https://www.kirich.blog/obzory/bloki-pitaniya/823-nizkoprofilnyy-blok-pitaniya-meanwell-lrs-350-24.html

That's a forward topology PSU, see that big inductor L2 near the transformer output and the arrangement of diodes. Yours is probably a simple flyback and the diodes (or only one of them) are connected directly between the secondary winding and the output capacitors.

That being said, the control circuits may be similar. One optocoupler is driven by TL431 and regulates voltage, not sure what the other is doing yet.

If think the primary switch, transformer and secondary rectifier have to be at least somewhat functional if there is any voltage output.
What's the peak voltage that appears on the secondary before the PSU turns off?
Do you have a scope? (btw, don't connect it anywhere on the primary)

The peak voltage with a multimeter when it turns on is 37V. I have a borrowed scope, and tried to capture the turn on event with it, but I have limited knowledge using it.

[dark_hawk]

Are you able to check/test the zener diodes ZD100, ZD102, ZD103? It looks like they're connected between the output of the PSU and the optocouplers for feedback for U1.

I may have found a candidate for U1: NCP1218 (pinout is similar). Check for yourself.

https://www.onsemi.com/pdf/datasheet/ncp1218-d.pdf

The zeners are fine as diodes, don't know if their output is fine or no, but they are not shorted.
I checked every pin in the PWM controller against each other, there is no short. I don't know how to test it further.
Their is also this strange sound when being turned on, like a spark or something, I think coming from the transformer. Maybe it blew when it got ac across it's secondary. But it gives a µH reading at the primary and secondary and not a dead short.

[magic]

Do you know if the secondary capacitors are alright?

I recently learned about a useful troubleshooting technique: disconnect power input, apply a voltage slightly below and then above the rated output (48V?) to the secondary and watch what happens.

Does the PSU draw significant current (indicating that something can't withstand the rated voltage and breaks down)?
Is the TL431 circuit working correctly?
Is any of the optocouplers getting turned on by the control circuitry?

[dark_hawk]

Do you know if the secondary capacitors are alright?

I recently learned about a useful troubleshooting technique: disconnect power input, apply a voltage slightly below and then above the rated output (48V?) to the secondary and watch what happens.

Does the PSU draw significant current (indicating that something can't withstand the rated voltage and breaks down)?
Is the TL431 circuit working correctly?
Is any of the optocouplers getting turned on by the control circuitry?

Great idea.
The secondary capacitors are fine, the three are in series and reads 352 uf and 0.02 ESR in circuit. 120uF 63v rated.
I applied 47V to the output, it draws around 7ma. The ref leg of the TL431 reads around 2.4v. The first optocoupler has a 1.13v and the second one has 0.76v.

[magic]

The one with 1.13V may be conducting. (This can be confirmed with certainty if there is a resistor in series with it).

The second optocoupler (not the one controlled by TL431) on the schematic of that other PSU permanently shuts down power to the PWM chip on the primary. If yours is the same then this can totally explain the problem. Keep the secondary powered and find out what's going on.

[dark_hawk]

Something very weird happened.
This is the resistor I changed. It is a 200 ohm resistor If I'm not mistaken.
Out of circuit it reads in the mega ohms. On a whim I put it back and now the PSU is working!
Problem is, I know it will shut down in a while. But that at least rules out the PWM controller and the transformer, diodes and mosfets.

Edit: The PS works even with that resistor completely removed!

[magic]

Isn't it in series with one of the optocouplers? Then if you remove it, the opto will never turn on.
Problem is, you don't know what kind of functionality you have just disabled
Draw a schematic of the circuit involving the two optos: all the resistors, the three zeners and where the TL431 cathode is connected.

[dark_hawk]

This is the first schematic I ever draw, so there might be some errors.
The resistor causing the issue is R156 200

[TheBay]

For what a genuine Meanwell LRS-150-48 costs, is it really worth messing around with?
Putting over 4x the rated output voltage in to the secondary side is never going to end well, even if you do get this working there is a very high chance that a lot of components are either damaged or weakened and can fail at any time. The insulation of the transformer windings could be compromised as well.

[magic]

It looks like emergency overvoltage protection.
If regulation fails for any reason, the two zeners break down and current starts to flow through U3. This may trigger a shutdown of the primary or something like that.
There is an error on the schematic: the 200Ω goes to the positive side of U3 and the negative is grounded - this is visible on one of your closeups.

Conclusion: the zeners are damaged and break down too early. Replace them with some combination that breaks down near 50V.
Living dangerously is an option too, but if something goes wrong a secondary overvoltage could damage a lot of components and the equipment being powered as well.

BTW, beware that there is TL431 and TL432. Same thing, but different pinout.

[dark_hawk]

You're absolutely correct. ZD100 and ZD103 are SOD-123 marked M1 which corresponds to 30V zeners. They break down at 18v which makes the OVP trigger at 36V. Changed them with 30V zeners and the PSU is working now.

Thank you.