07n90e equivalent

[snipersquad100]

Hi, I cant find a 07n90e mosfet anywhere. It's needed to repair a power supply. I'll link the datasheet down below.
Is there anything else I could use.
tried a tip3055, blew up in my face lol.

[T3sl4co1l]

Afraid of what other things you'll screw up (or have already screwed up) if you blatantly tried a 2N3055 without thinking...

[snipersquad100]

I was a silly thing to do. I was lucky mind, lent over the psu while I turned the power on, BANG jump a mile lol

[snipersquad100]

Thanks AcHmed99 I'll have a look for one of those, And I'll stand clear when I switch it on lol.

[wraper]

tried a tip3055, blew up in my face lol.

Yes you are lol  putting low voltage BJT instead of the HV mosfet  . Anyway, what was driving it, likely is already dead. So you will need also change PWM controller IC, resistor(s), diode and anything else what was between that IC and mosfet gate. Current shunt resistor between source and ground. Maybe other pwm controller surrounding parts. For example you can use 2SK1120, 2SK3878, 2SK2613.

[snipersquad100]

Not sure what was driving it, its in the shed at the mo. If im right it was on the primary side of the transformer just after the bridge rectifier. The psu is a 13v 32amp.

[wraper]

Yes it is primary side, there is likely 8 pin (maybe more) IC nearby. BTW before testing it again, highly suggest to attach 100-200W incandescent or halogen light bulb in series with any mains wire (except ground lol). That will save you one more explosion if something is wrong.  Bulb will light up, no explosion and likely all parts will survive. If all ok, or not ok but not so wrong that it will cause explosion, bulb won't light up (maybe just for a half second before capacitors get charged). Btw how did you check that mosfet was dead?

[snipersquad100]

Good idea, will do that.
On the mosfet I wasn't getting any reading on the diode setting of dmm between gate and drain. all secondary fets, diodes checked out ok. Even thou that mosfet wasn't shorted being open circuit or no reading at all between gate and drain made me suspect that. probably a red herring.

[dannyf]

That's likely a 7a 900v n-channel mosfet. Fairly easy to find replacement.

[wraper]

Good idea, will do that.
On the mosfet I wasn't getting any reading on the diode setting of dmm between gate and drain. all secondary fets, diodes checked out ok. Even thou that mosfet wasn't shorted being open circuit or no reading at all between gate and drain made me suspect that. probably a red herring.

Why should it read something? This is mosfet, not BJT. Such mosfets usually fail with source-drain short. It is likely OK, but other things likely are no longer OK after that tip3055 experiment. And if you just solder it back very likely it will just blow again as controlling circuit most likely is damaged now. As of primary reason of PSU not working very likely was dead capacitor on PWM controller IC power, something another an that IC power or just faulty IC. As well it might be shorted diode on secondary side which prevented PSU to work normally.

[wraper]

So basically it was a minor fault but now became a real PITA. Post a pictures, preferably high resolution so we can help you to diagnose if there anything left that still can be saved. As you may notice on MOSFET symbol on this picture, gate is not electrically connected to other terminals, basically it is isolated by very thin insulator layer from them.

[snipersquad100]

I see what you mean.
I'll take some picks in the morning. I've ordered a 2SK3878, once it comes and is fitted and if it still don't work ill prob stick it in the parts donation bin. That's if there any parts I did'nt blow up lol

[wraper]

I see what you mean.
I'll take some picks in the morning. I've ordered a 2SK3878, once it comes and is fitted and if it still don't work ill prob stick it in the parts donation bin. That's if there any parts I did'nt blow up lol

Likely it won't work, even if nothing more were blown (very unlikely). You still need to repair first fault which was before the explosion. You can still use the old mosfet if you did not discard it yet.

[snipersquad100]

sorry for late reply, been away for a few days. I've put the mosfet back in, I have noticed that the green wire is giving me 3.6v but nothing happens when I connect it to ground. The ic the green wire goes to "U6" is CD7510 I cannot get a datasheet on this. The date code is 1401G so im thinking it might be a new chip.

[SeanB]

8 year or so old PC power supply with passive PFC, likely a 200W ( at best) unit. You can go buy a new 400W unit for cheaper than the repair on this one, and it will at least come with some warranty. I see CrapXcon inside, so those likely are close to dead in any case.

[snipersquad100]

Voltage on chip U6
green wire not grounded                 green wire grounden
pin1   0v                                                0v
2       0v                                                0v
3      4.6v                                              0.2v
4      3.8v                                              0.5v
5      0v                                                 0v
6      0v                                                 0v
7      4.6v                                              4.6v
8      0v                                                  0v
green wire runs to pin 4 via a resister.

[snipersquad100]

its 2 weeks old, only reason I haven't returned it is because ive cut some wires. I was using it to run a bitcoin miner. it's a 730watt supply.

[wraper]

its 2 weeks old, only reason I haven't returned it is because ive cut some wires. I was using it to run a bitcoin miner. it's a 730watt supply.

730W?  . That thing don't look like true 730W at all, no PFC. Brand/model? Isn't it Codegen by chance? As I understand standby power is working? What is voltage on violet wire? U6 likely is supervisor IC http://www.hardwaresecrets.com/datasheets/WT7510_V2.311.pdf. I see there are two IC's near the smaller heatsink, likely fault lies somewhere there. "ON" command goes through one of the three optocouplers, two others are for standby and main voltage feedback.

[wraper]

Should be Leadman LP-8867 series aka Echostar 680W aka Blue Star 650W utter crap. What current rating for 12V rail it has?

[snipersquad100]

The model is a Artic red+ AD-E750AE-A5/A6
Acording to the "sticker see pic" its got 4 12v channels rated 32amp,54 amp max, but all the 12v wires are all connected to the same trace on the pcb.
I traced the green wire thou the opto to U1 "KA3843 smps controller" output off that goes to the gate of Q1 via a 33ohm resister, the transister I thought was gone in thr first place.
I did some voltage checks, i had 1.5v coming out of U1 output but nothing in the gate of Q1. I checked the resister and it was 2 odd mega ohms. I replaced that resister and now im getting the voltage on gate of Q1 when green wire in grounded, 330v on drain but nothing coming out thou the emitter.
Im having 5v out of the vilot wire.

[SeanB]

NFW is that a 650W power supply, more likely 200W on a good day with a following wind. The transformer is similar in size to ones I see in 200-250W switchers, and it will never deliver 3 times the power, let alone the rectifiers and the capacitors on the secondary ( let alone the primary side) being rated to actually handle that and not explode.

Consider the primary side is pretty much plain with no PFC aside from that passive PFC inductor ( and that can be a fake as well in many cases, copper is a lot more expensive than steel laminates and a 4 turn wire through with a lot of tape padding) and the main switching transistors are going to explode with 3A average through them. It will do as a 200W PSU, but anything more is not going to work.

Is there a supply to the SMPS controller , which is a pretty generic part, and if so is it oscillating. The drive likely is being carried over by the 3 optoisolators, so if they are not working, or if the main transistors or the components around them are faulty you will get the output you have as one transistor is not turning on when needed or is dead short and has blown the other open.

[snipersquad100]

im not here to compare dick sizes, I just want to fix it.
I've already found a fault with Q1, no out put when theres a gate voltage present.

[wraper]

http://www.cclonline.com/product/78081/PSU-507/Power-Supplies/Artic-Red-750W-Power-Supply-Unit/PSU0324/ OMG 18 pounds for 750w  . Half decent 400W (which will outperform this one) cost a little bit more. I would be surprised if it can continuously supply current specified even for one 12V rail. Suspect there is no more than one PCI-E X6 and no PCI-E X8 connectors at all what says than you cannot connect even one mid end GPU to it. KA3843 = UC3843 almost certainly is dead. If there is diode or zener connected to the gate of the mosfet, it also must be replaced.

[dannyf]

OMG 18 pounds for 750w

Last millennium's technology today,

[snipersquad100]

There was one pci-e6 and one pci-e8 lol so I chopped the pci-e8 for another pci-e6 to run a antminer s1. It did it nonstop for a week so hats off lol.