# Need help diagnosing 24v power supply. both NPN's test "OL" between all pins

#### bigjoncoop

Joined Feb 1, 2019
134
I've been using this 360w 24v DC power supply for quite a while and it's worked well. from what I can tell online it's decent we built for what it is. All the sudden today I go to use it and there is no DC output. so I do a quick search online and how to troubleshoot this type of power supply... it has 2 - d13009k NPN high voltage fast switching power transistors. I found a few tutorials on how to test npn transistors and no matter which pins I probe, my multimeter just shows "OL" overload in diode / beep mode. Even swapping red and black test leads. I am testing them in circuit. at first I thought maybe I was doing something wrong since both of them are showing the same thing. so I checked out a couple YouTube videos on checking npn transistors and there should be some readings between certain pins even when still in circuit.

so basically my main question is is it possible that something else is wrong to make me have these reading while testing these transistors? Or is it possible both of them are dead?

if you guys say that they're dead from the readings I'm getting I will replace them with some npn power transistors from a different power supply with equivalent ratings. just wanted to make sure that something else wouldn't be causing these readings and if I replace them if they'll get blown the same way...

and if you guys have any other advice exactly how to go about testing to determine what might be wrong if it's not these transistors I would appreciate it.

#### AlbertHall

Joined Jun 4, 2014
10,771
The transistors are probab;y deceased.
Is the fuse still intact?

#### bwilliams60

Joined Nov 18, 2012
1,361
As @AlbertHall has eluded to, the transistors are probably toast and will need to be replaced. The trick is to find out what else dies with it and chances are good that there afe more issues. SMPS power supplies can be tricky. Get a good schematic and do a lot of reading. You may be in for an adventure. Good luck.

#### dl324

Joined Mar 30, 2015
12,212
I found a few tutorials on how to test npn transistors and no matter which pins I probe, my multimeter just shows "OL" overload in diode / beep mode. Even swapping red and black test leads.
So you're saying that you checked both junctions with polarity both ways? CE should read open.

#### bigjoncoop

Joined Feb 1, 2019
134
The transistors are probab;y deceased.
Is the fuse still intact?
yeah. fuse isn't blown. and everything visually looks good. also the fan doesn't try to spin now. if thats any help. and the one thing I remember off top of my head is, at the bridge rectifier diode I got 42v on each leg. between each leg and neutral...

also, I've never pulled more then 10 amps from this psu. and the last time I used it , I do remember a faint hot electronic smell...

#### Dodgydave

Joined Jun 22, 2012
9,837
That PSU uses a TL494 chip, with a 28V supply to self oscillate, and is also the fan supply, it uses Q2,Q3, transistors to pulse the small pulse transformer Tr1 which drives Q1,Q4 Large power transistors to pulse the output voltage transformer Tr2, which is regulated by pins 1,2 using the Yellow preset near the output terminals.

I would check the DC first across the Large capacitors C2,C3 ,around 350V to see if that's ok, if not then the fuse, the mov and the voltage selection switch.

Last edited:

#### MisterBill2

Joined Jan 23, 2018
7,689
It appears that the transistors are still in place in the circuit board and still connected to the circuit. If they were unsoldered then you might be a blt to check them again without removing them. And if you are using the meter probes on the top side of the PCB there is a chance that you are not making contact with the transistor leads. So the very first thing would be to do the check from the underside of the circuit board to be certain that you are making an adequate contact. In several instances I have not been able to make adequate contact when probing from the top of a PCB.

#### bigjoncoop

Joined Feb 1, 2019
134
@Dodgydave and @MisterBill2 and anyone else

so I removed the circuit board from the case and I'm still measuring nothing in diode mode between all three pins on each of the two npn power transistors. and I took voltage measurements throughout the board starting with the big fat capacitors to the npn transistors both sides of each Transformers. I'm reading 120 volts at the capacitors, on each of the three pins on both transistors, and on the front side of the small transformer in front of the transistors on all 3 pins. Starting on the other side of that small transformer the voltage drops down to 78v A/C again on all 3 pins. And from that point forward on all the components and traces all the way to the DC output terminals I also am getting 78v A/C on both positive and negative rails.... Beside 13v one one pin of the big transformer. (Detailed picture included)

I've included a picture with my all my voltage readings marked.

All these readings are with one of my voltmeter test probes on the mains input neutral.

So something is definitely not right if I'm getting 78v A/C between mains neutral and 41v A/C between mains positive, on the D/C outputs. But it seems like that may be a good indication of what might be wrong Hopefully.

,

Last edited:

#### Dodgydave

Joined Jun 22, 2012
9,837
You can't measure the DC output with the Neutral as a Common, .

If the DC supply is 240V, that sounds like the Transistors are faulty if you're getting open circuit on B/E , B/C terminals,.

Test the diodes D1,D2 , D7, D8,to test your meter.

(The pulse transformer Tr1 is there to isolate the chip from the mains input.)

#### MisterBill2

Joined Jan 23, 2018
7,689
The valid reference point for measuring on the output side of the supply is the output common terminal.
The supply IS an isolated supply, which means that the output is isolated from the mains side. That means that all the readings past the isolation must be against the output side common. So the 78 volts AC readings are not useful. That big transformer is the isolation device.
AND, with the two power transistors reading open, it does seem like they have failed.

#### bigjoncoop

Joined Feb 1, 2019
134
OMG... DAMN!!!

ok. yeah I'm 100% sure both of those transistors are open circuit. After work ill replace them and cross my finger they don't fail when I power it up

#### bigjoncoop

Joined Feb 1, 2019
134
@Dodgydave
@MisterBill2

I Haven't been able to find equivalent transistors in my stuff to replace the blown ones. And for the price of 2 -13009 series transistors is damn near half the cost of the entire power supply. I only paid $17 from Amazon . I need to place order @ Jameco for some other items but they have$20 minimum order. And of course no 13009 trannys.... But they do have one that is very close and maybe another . I've posted a question about if I could use these trannys I found, and someone moved it to another category and it hasn't received any replies.

in the post I've attached screenshots of the specs of the 13009 and the 2 I found on Jameco. If you wouldn't mind taking a look and see if I could use one of them I would really appreciate it. The one has a similar max voltage but the other is lower but keep in mind im only feeding the power supply @120v. below is the link to the other post... FYI the 2 trannys in Question are - BUV48A by STMicro & FJA4313OTU by Onsemi. The voltages and current seem to work but I'm unsure of the other values

#### MisterBill2

Joined Jan 23, 2018
7,689
My experience with Jameco was that they were far more expensive than quite a few other companies. Some others have challenged that, but that was my experience. Plus they had no quantity packages, so if you needed 30 ICs all the same you got 30 of them packaged individually and priced that way.
I followed that link and saw the specs and the ST transistor looks fairly good. But here is a question: Is the power supply a switcher or a linear regulator. Linear regulators are less critical applications. But the pictures show a switch mode supply and so the replacement devices need to have a few more qualities.

But given the specifications of those transistors, do you have any idea why they failed? Those werem certainly high power devices and fairly rugged, so if they were destroyed it must have been something serious. You would not want to destroy a second set right away, I am sure.
One interesting option might be to ask ST semi for two sample transistors. Sometimes that works.

Last edited:

#### bigjoncoop

Joined Feb 1, 2019
134
@MisterBill2 no idea why it failed. Umm I did definitely push it hard often. It was advertised as 15a... And I definitely pulled really close to that regularly for few hours at a time. And those trannys only rated 12a...

No sign of any damage anywhere... So even that StMicro won't work ? If not maybe I'll just buy another psu for \$17. At least now I know its only rated 12a

#### AlbertHall

Joined Jun 4, 2014
10,771
Those transistors are on the primary side of the supply, switching high voltage into the transformer. So they handle high voltage at a lower current than the output side of the supply. 12A transistors does not mean the output is limited to 12A.

#### Irving

Joined Jan 30, 2016
1,070
If both are toast it's likely something else is toast too. The usual culprits are electrolytic capacitors going soft but there could be many other options including the main controller chip. I doubt this is going to be an economic repair; by all means do it for the learning experience but expect to waste a few parts... And make sure you discharge the big caps each time you turn it off to work on it!

#### MisterBill2

Joined Jan 23, 2018
7,689
If both are toast it's likely something else is toast too. The usual culprits are electrolytic capacitors going soft but there could be many other options including the main controller chip. I doubt this is going to be an economic repair; by all means do it for the learning experience but expect to waste a few parts... And make sure you discharge the big caps each time you turn it off to work on it!
Like I said: in a switch mode supply all of the parts are far more interacting. AND most component failures do not result in any visible damage. If a snubber diode fails open there is no clue except that the high voltage spike can damage other parts. So checking the other parts for correct function before replacing the transistors makes sense.

#### Dodgydave

Joined Jun 22, 2012
9,837
This is part of the primary circuit with the push pull transistors Q1, Q4 , on the heatsinks. They are a self oscillating pair, then the TL494 takes over.

Q2 ,Q3, pulse the transformer Tr1 when the chip takes over, .