They are drawing too much current when the rectifiers are connected to the caps and the voltage is up. One channel blew the fuse. It really acts like I'm shorting the PS, but I can't figure out how.
I'm using the standard First Watt F5 (or other PP amps) power supply (having trouble finding a good clean link) with all the included thermistors and the usual Antek transformer. (Transformer and fuse are smaller since they are monos.) The only differences are that I'm using single caps (33k) instead of pairs, and for some reason, I put the bleeder resistor (which is way too high a value) on the second caps rather than the first.
(The signal side, which isn't even connected right now, is per the schematic too, with no P3s. Some of the resistors are overrated because I had them, but the values are all correct.)
Again, it's really hard to tell from the pics. Have you tested the psu without the amp pcb connected? It's all hooked up in the pics so we don't know.
Both channels doing the same thing?
I would test voltage at the secondaries unconnected (be careful), then voltage at the rectifiers without caps connected. It will be unsmoothed so it won't reflect the actual rail voltage but it will let you know if there is a short.
Then connect the caps with amp pcb unconnected. If you have 24v ish you are good and the fault is in the amp boards.
If it's the amp boards I'd check the resistor values...I think the small peter Daniel boards are marked different than the diyaudio layout. Also, check that the mosfets are isolated from case but I bet you already did that.
You say your bleeder resistor value is too high? What are using?
Both channels doing the same thing?
I would test voltage at the secondaries unconnected (be careful), then voltage at the rectifiers without caps connected. It will be unsmoothed so it won't reflect the actual rail voltage but it will let you know if there is a short.
Then connect the caps with amp pcb unconnected. If you have 24v ish you are good and the fault is in the amp boards.
If it's the amp boards I'd check the resistor values...I think the small peter Daniel boards are marked different than the diyaudio layout. Also, check that the mosfets are isolated from case but I bet you already did that.
You say your bleeder resistor value is too high? What are using?
Test the build in stages.
Power on the transformer alone via the Mains Bulb Tester. Measure the secondary voltages.
Power Off. Tell us the result.
Power on the transformer alone via the Mains Bulb Tester. Measure the secondary voltages.
Power Off. Tell us the result.
I apologize if I am somehow not being clear enough.
With the transformer connected to the rectifiers but the rectifiers not connected to the caps, the voltage looks good. (appropriate based on the position of the variac--it would of course be high/low if I cranked it all the way up with no load).
When I connect the rectifiers to the caps, it acts like there is a short: almost no volts on the rails, and it will blow the fuse if I put some voltage into it.
It does not matter whether or not the signal board is connected to the power supply.
And what I didn't say before but should have: both channels act the same.
Last edited:
Are you certain that each + rectifier block's terminal goes to the + capacitor terminal?
If the connections to the electrolytics were reversed, you'd see similar behavior.
Last edited:
Oh man. that would require quite a lot of disassembly. And what are the odds that both transformers would have the same problem? And what could that problem even be, given the symptoms?
Don't overlook anything. Sometimes it is the strangest thing that isn't working properly.
Test the transformer alone, then connect just one secondary to a bridge, test, then to the capacitors, test. Repeat with the second secondary. If you don't use a methodic approach you'll have a hard time figuring it out.
I believe we might need more pictures.
We can see a green and blue wire from one of the bridge going to the first capacitor but there are also 2 more wire (also blue +green) coming from the second bridge, are they connected ? We can't see much.
If you don't connect the capacitor what is the DC voltage across each bridge ? I wonder if your cap. are rated high enough, I believe they are 25Vdc.
Keep us posted.
BR,
Eric
+1
You must assemble and test in stages.
Use a Mains Bulb Tester to prevent damage from miswiring.
If the PSU starts acting up with the capacitor bank connected, you should maybe check which two wires from the rectifier bridges meet at the common ground connection.
That said, that amp looks like a real proper fire hazard. Stay safe.
That said, that amp looks like a real proper fire hazard. Stay safe.
Well Paul, you've got a really tricky one there!
About the only thing not mentioned is faulty caps, or incorrectly aligned wrappers, unlikely as that may be - they look like 33,000uF/25V - slow charge? how long did you connect them for? Scraping the barrel here, I know!
As far as I know, no 'snap-in' caps have negative terminal connected to case so that's a doubtful cause - a shorted cap?
On photo 2, does that thermistor connected to the common supply Gnd buswire go to a chassis connection with that earth wire from the IEC mains socket tucked away underneath?
Betsy K's going well.
About the only thing not mentioned is faulty caps, or incorrectly aligned wrappers, unlikely as that may be - they look like 33,000uF/25V - slow charge? how long did you connect them for? Scraping the barrel here, I know!
As far as I know, no 'snap-in' caps have negative terminal connected to case so that's a doubtful cause - a shorted cap?
On photo 2, does that thermistor connected to the common supply Gnd buswire go to a chassis connection with that earth wire from the IEC mains socket tucked away underneath?
Betsy K's going well.
Is it possible that I clamped the caps too tight and damaged them internally? Maybe I'll try and scare up some other caps I can wire in. The caps are nice Panasonics that I pulled from their original packing, so failures in both amps seem unlikely.
Yeah, the thermistor under the big ground goes to a chassis ground directly below it.
I'll try to tear them down further. But I'm going to be traveling for most of the next couple of weeks. Thanks for all your help, folks!
Paul
Yeah, the thermistor under the big ground goes to a chassis ground directly below it.
I'll try to tear them down further. But I'm going to be traveling for most of the next couple of weeks. Thanks for all your help, folks!
Paul
Are you sure you have the rectifiers to cap bank wired properly? I really like to see the rectifier under the inrush thermistors... Can you unbolt the transformer and lift it up and take a photo?
Also, can you verify that the filter resistor bank on the side nearest the IEC isn't touching the heatsink?
Also, can you verify that the filter resistor bank on the side nearest the IEC isn't touching the heatsink?
Last edited:
Possible but not probable. In the old days they used to squeeze electrolytics
to raise the capacitance.
😎
- Status
- Not open for further replies.