Corsair RM1000e - Help identify 5VSB Zener diode

[hobostove]

Hey there guys. I've got a Corsair power supply that's come across my bench dead. I usually don't mess with power supplies, but it's a slow week so I'm giving it a whirl.

Taking it apart I heard a screw rattling around inside, and when I got it apart I found a couple blown components near what I think is the 5VSB area.

There's a 22ohm resistor that's melted and open, but it still has the bands.

The other part is a zener diode and it's top went to jesus, so I can't read the marking.

It looks like the diode connects the source and drain of a TNY284 through a couple of 2M resistors and a transformer winding.

Does anyone know what a good value for this zener would be, and what it might be doing?

Any help much appreciated!


TNY284 Datasheet: https://www.mouser.com/datasheet/2/3...et-1511375.pdf

[прямо]

It's most likely 1N4148W.

[hobostove]

1n4148 is a regular diode. Isn't ZD83 supposed to be a zener?

[Per Hansson]

Have you ruled out that it isn't TVS diode P6KE150A (VR1) or Zener diode 1N5254 (VR2) from the example schematic you posted? (By studying the actual circuit closer/drawing it up more it should be "easy" to see).

[c1q3]

I don't know will it help or not.

This is from CM6502UHH PFC controller typical application circuit and it has TNY227P.

Resistor is 2M (r45 + r46). On your photo something is behind.
My wild guess it is sort of hz15-2. Enable pin should be connected to optocoupler.

[c1q3]

don't take me serious. I'm just studying.

[hobostove]

This is a big help. I'm away from it until Monday, but I'll answer the questions I can remember.

To optoc? Yes

20 oh? Bands on resistor are red-red-gold-gold-white. Which I think means 2.2ohm. (see picture)

Source to gnd? I don't think so. I checked for shorts, and I think source wasn't connected to ground.

[c1q3]

I made a mistake on pic above. Its not 200M its 2M. But overall it fits.

btw I cant find what is red-red-gold-gold-white. We need some1 to explain.

[c1q3]

changed 200m to 2m.

[Per Hansson]

I think the resistor is a 2.2Ω fusible resistor, it looks like it did it's job and is no longer a resistor

https://www.engr.siu.edu/staff/spezi...standscode.pdf

[c1q3]

I think the resistor is a 2.2Ω fusible resistor, it looks like it did it's job and is no longer a resistor

https://www.engr.siu.edu/staff/spezi...standscode.pdf

so the third golden ring is like blank space in numeral. Interesting.

[Per Hansson]

I could be wrong of course, these "special" resistor bands take a bit of guessing: the best way is to have more of the schematic available.
Right now I can't see where it connects to so I have no way to make a better educated guess.
But if it is before or after the source or drain for example then it is acting as a protection if the MOSFET in the TNY284 gets stuck on...
(And if you look at the top-side photo above it seems the PCB around the drain leg is discolored so I'd say this is what has happened).
And welcome to Badcaps btw and good job so far with the circuit

[hobostove]

Incredible work c1q3, thank you very much. Thanks to you too Per Hansson. I will gather more info and confirm the drawings above when I'm in front of it tomorrow morning

[hobostove]

It looks like the resistor is a 2.2ohm fusible. I couldn't find the exact one, but this is close: https://www.digikey.com/en/products/...52-2R2/2167125

That resistor is connected between 380V and Drain, so yup, it looks like it went pop as intended.

Source is connected to GND, and ZD5 connections match, so I think you found it! This looks like an OK replacement: https://www.digikey.com/en/products/...C15-7-F/755469

I can't find any shorts, so I'm going to order the diode, the resistor and a new TNY284 and hope for the best. Will report back in a week or so.

Thank you both!

[Per Hansson]

Measure the optocoupler in diode test mode with your multimeter: since the failed ZD5 zener is parallell to it (talking about the example schematic that c1q3 posted here).

[hobostove]

Parts came in today, and I replaced the 2.2o resistor, the zener, the flyback IC, and the optocoupler. Still no life so I suspect something is open up stream. Going to poke around a bit more

[hobostove]

Just found main cap voltage is only 167V when powered up. Shouldn't it be 300V+? I'm on 120AC here, so maybe ~170V is normal?

[прямо]

Just found main cap voltage is only 167V when powered up. Shouldn't it be 300V+? I'm on 120AC here, so maybe ~170V is normal?

That is normal. 120V AC / 0.707 = 169V DC.

It will be boosted by the PFC controller chip when the PSU is starting such as when you jumper PS_ON to ground.

If the boosting doesn't happen, check the supply voltage of the PFC controller chip. I don't know what chip it is, but the usual voltage is around 15V DC. Also, when you check this voltage, because the PFC controller chip is located at the primary side, the black lead of your multimeter should be at the negative pin of the bridge rectifier.

Be very careful though when poking around the high voltage side.

If the PFC controller chip supply voltage is present and around 15V DC, then the next logical step is to check the supervisor chip fault protection (FPO) pin. Consult the chip datasheet to see if it's an active high or an active low logic signal. If the FPO status is telling you that PSU is in protect mode, this explains why no pimary voltage boosting is happening and why the PSU won't start. Therefore, you have to disable the protection in order to force the PSU to start and proceed with diagnosing what is triggering the protection, usually it is because of an under voltage or an over voltage on one or some of the output rails.

But more importantly, is the 5VSB present?

[hobostove]

5VSB is behaving strangely. It's fluctuating between 3-3.5V. It's a couple seconds from peak to peak

[прямо]

5VSB is behaving strangely. It's fluctuating between 3-3.5V. It's a couple seconds from peak to peak

Replace the 5VSB output caps.