Triple output power supply blowing main fuse

[partyanimallighting]

Hi Wizards, I've got two filed units with blown power supplies. I've swapped out the power supplies in the units with new power supplies but I want to attempt to repair the blown power supplies. They are 120VAC ~ 24VDC/36VDC/380VDC and the 24VDC leg drives cooling fans, the 36VDC drives the main circuit boards and the 380VDC powers up a ballast for a 7R 230W lamp. I scrapped out the power supplies and both have blown fuses next to the AC voltage input (bottom right hand corner of picture). I bridged the blown fuse with a single strand of wire and powered up and the makeshift fuse blew out immediately so I'm wondering how I can go about repairing this if I can't power it up and I'm not seeing any sort of burnout on the board but my eyes aren't trained to spot these problems like you guys. I'm thinking that the two yellow X2-224K capacitors are some sort of protection and may be blown out? Anyone willing to offer me some advice on this?

[mod edit: removed double pictures]}

 

[Martaine2005]

I think this is the perfect scenario for a series bulb. Put an AC 100W in series with the mains in. You can put it directly across the fuse. The bulb will light brightly indicating a short but wont blow any other components. You can now troubleshoot the ‘live’ circuit.

Martin

 

[Martaine2005]

Incandescent bulb!!

 

[Martaine2005]

Incandescent bulb!!
You can simply remove the X2 caps for testing. They could have shorted the input.

Martin

 

[Harald Kapp]

There seems to be no visible sign of damage.
I think there is a good chance the primary side power MOSFET has a short. That would be, as far as I can judge, the component labeled 8GL28. I can't find a reference what "8GL28" could be. Are there any other imprints on the case of this component?
Does the short circuit vanish if you remove this component?

 

[Harald Kapp]

You can simply remove the X2 caps for testing.

That would be the 2 yellow boxes next to the AC input.

 

[partyanimallighting]

Really? I kind of remember seeing old electricians testing electrical shorts with a bulb and socket with two wires. Here's the funny thing.....I'll have to go buy a 100W incandescent!! Everything I have is LED!! I might have really old stock of effects lighting lamps like 300W 120VAC 64514's lying around somewhere. Do you think they will work?

 

[partyanimallighting]

I'm not an experienced technician and I know this power supply puts out 380VDC so I'm gonna be extra careful running any tests. A large amount of these lighting units were powered up at the same time so I'm figuring the current draw from all the lamps striking at the same time screwed up the power supplies so I'm assuming the two yellow X2-224K capacitors are part of the problem because I'm thinking they're the "safety" caps for the input voltage leg. As for the 8GL28 mosfet, there's no schematic for the unit and no markings at all on the power supply so it's anyone's guess what that component is. The cap is X2-224K MPX/MKP 40/110/56/8 275VAC 305VAC 310VAC.

 

[partyanimallighting]

Martaine2005, can you explain to me how putting the 100W incandescent bulb in series across the fuse terminals will prevent the blowout? Does the bulb create enough current draw to decrease or cancel out the short? Please remember I'm not a trained technician so I would like to understand the scenario better.

 

[Martaine2005]

The series bulb limits the current. Also acting as a resistor, there is now no short detected. I use different wattages for different appliances. A 300W will work too.
The X caps are across the input. Removing them has no consequence other than external interference like clicking through speakers or radio interference. A suppressor of sorts. (If I understand it correctly) of course!.
But you can obviously remove the mosfets and see if the short goes away.
The series bulb saves blowing fuses too.

Martin

 

[Harald Kapp]

As for the 8GL28 mosfet

Can you show a close up photo of this transistor, please? Maybe it is not 8GL28 after all?

 

[partyanimallighting]

Hi Wizards! It's been a while but I don't think an explanation is necessary with what has happened to our world over the past few months. Soooooo......being jobless, I've gotten some free time to start trying to resolve all my little repair problems one by one, so it's back to this power supply. Yes. I know it's been a few months but everything was virtually untouched and I still want to get these power supplies up and running so I've started back doing a little repairing and testing. I tried the incandescent bulb inline to replace the fuse scenario and there were no issues but I'm not sure how to go about doing any further testing as there is still no DC outputs and I'm getting >2VAC and >2VDC on the bridge rectifier. Referencing Martaine2005's post, do I remove the 8GL28 FHA20N60 MOSFET from the power supply, also remove the incandescent lamp from the fuse input, re- bridge and power up to check if the short disappears?

 

[duke37]

A 300W bulb is perhaps excessive, I use a 40W bulb to check old valved radios or transistor amps.
Looks as if the GBU810 is the mains rectifier which feeds the 450V capacitors. Check the resistance across the four diodes both ways in the rectifier. A dead diode would likely be a short.
I wpuld be tempted to remove the rectifier and see if the lbulb does not light. Check resistance from + to earth and - to earth. But where is earth?

 

[partyanimallighting]

Here's an update on testing. I tried a 40W bulb and the voltages on the GBU910 were still low, <2VAC and <2VDC on the relevant legs. So then I followed Harald Kapp's advice and tested the (8GL28) FHA20N60 primary power MOSFET. On the PCB I got continuity across all legs so, I removed it as suggested with the same result, continuity across all legs. So I disconnected the bulb and bridged the fuse with a single strand of wire and powered up. No sparks. No blowouts. No issues. The voltage readouts on the outputs were as follows: 12VDC output ~ 12VDC / 36VDC output ~ 37.5VDC. 390VDC output ~ 160VDC. No pulses, no fluctuations, so I can assume that the MOSFET is the faulty component, correct? Question...the low voltage on the 390VDC leg, is the MOSFET responsible for boosting this output up to the required 390VDC? Will replacing the MOSFET resolve all the problems (based on the correct voltage readings on 2 of the 3 outputs) or may there be other faulty components waiting to baffle me?

 

[partyanimallighting]

No responses to the above questions so here's the update. I managed to source the power MOSFET locally at an exorbitant cost and swapped it out in each power supply and the problem disappeared so it WAS the power MOSFET. I have to say, as a novice, I've learnt a lot just from this attempt and all your assistance so It's on to the next two problematic power supplies so I'll post my issues with these two next. Thanks for all the assistance guys! If anyone has the time, please give me an update on my questions in my previous post. Thanks again!

 

[Harald Kapp]

Good to see you solved this one

 

[partyanimallighting]

It was not possible without your assistance and the others also but I'd still like to get a little feedback to give myself some learning closure re: The low voltage reading on the 390VDC leg with the MOSFET removed, is the MOSFET responsible for boosting this output up to the required 390VDC on that leg only because it does not seem to affect function on the other low voltage DC legs. Their readings remained stable with it removed, the only voltage affected was the 390VDC leg. Just wondering.

 

[Harald Kapp]

Hard to say without a schematic

 

[partyanimallighting]

OK. Thanks for the feedback. Case closed anyway. They've been repaired. Thanks again!! To you and all the others.

 

[partyanimallighting]

Well, I'm back. As I updated before, I replaced the FHA20N60 MOSFET with what was available in my little island, G20N60 and the power supply powered up fine with the correct output voltages. So I tried installing it into a Platinum 5R moving head to test a while ago. This is not the original unit that the power supply was removed from, but a similar model from the same company. This unit powers up via 390VDC (powers the ballast to strike the lamp) and 36VDC and 24VDC (not 12VDC). Now the mainboard of the unit receives both 36VDC and 24VDC but I decided to test the power supply in this unit anyway, figuring that the 24VDC would be driving the DC fans. The unit powered up fine, quick lamp strike, no error messages and all motors and fans functional. So, no issues with the 12VDC
instead of 24VDC......or so I thought. When I tried to open the shutters for lamp output, the power supply died. So I pulled it out and checked it....no burnouts, just a blown fuse. So I scrapped it out and replaced the fuse and powered it up. Fuse blows, same problem like before. Now, based on my above post, I now have a basic idea how to retest the power supply to see if it's the G20N60 MOSFET that blew out but I can't find a datasheet for FHA20N60 to compare anyway. My question is, would the MOSFET not having the exact part number be an issue? Maybe this part is incorrect? Or would the current draw difference between the 12VDC supplied from the power supply and the 24VDC required by the moving head cause the blowout? I'm taking into consideration that the unit had a faultless startup and lamp strike and the power supply only failed when I opened the shutter. Any views on this please? I do have 390/36/24VDC power supplies but.........they also need repairing so it looks like it's time for a new post so I can get some assistance to repair these other power supplies.