Power supply is blowing/schorching!

[soma_hero]

I put together a power supply for my dual LM3886 amp. It's an antek 22VAC, 400VA toroid with dual secondaries. I posted the power supply schematic (i made my own design off a few to incorporate a preamp in front of the power amp) and got the ok so I constructed it.

Now I made a PCB for it, etched it and got everything wired up in the case (not yet connected to the amplifier) for a preliminary test. Turned it on and I see a flash on the circuit board. Measure the voltages and from one secondary powering the positive rail I'm getting a good supply of +30V and +15V for my pre. The other secondary has some blown traces around the diodes. I relay the traces with some wire, fire it back up, and now traces blow on the other secondary!

I know the traces need to be bigger to accommodate the high current, but they're by no means small and they are getting burnt up. Should I redesign the PCB with bigger traces or can I just keep turning it on and replacing traces as they blow?

Is it not necessarily the current, but the huge rush from the massive transformer?

What should I do to cure my issues?

Thanks all,
Mike

[pacificblue]

Originally posted by soma_hero
I relay the traces with some wire, fire it back up, and now traces blow on the other secondary!

There is too much current-flow in you circuit. The traces of the first acted as fuse. After replacing them with bigger wires, the traces of the second work as fuse.

Originally posted by soma_hero
I know the traces need to be bigger to accommodate the high current, but they're by no means small and they are getting burnt up. Should I redesign the PCB with bigger traces or can I just keep turning it on and replacing traces as they blow?

You wrote that you are testing without an amplifier connected, so there should be next to no current flowing at all. You have a short somewhere. Don't replace the traces with wires until you found the fault and corrected it.

Originally posted by soma_hero
Is it not necessarily the current, but the huge rush from the massive transformer?

That affects the primaries, no the secondaries.

Originally posted by soma_hero
What should I do to cure my issues?

1. Install adequate fuses.
2. Use a light bulb tester.
3. Find the fault. I. e. check for shorts, for wrongly mounted components, etc.
4. Correct it.

[impsick]

make sure you got the diodes correct. that could be confusing sometimes.
unless your using two bridges

[ThyDntWntMusic]

post a good clear, reasonably high res pic of the PCB.

both sides please

[soma_hero]

thanks for the quick replies guys,

I'll post some pics of the PCB later today if I find some free time at work.

[soma_hero]

Here it is, single sided. Schematic:




And layout:




I just realized in the schematic the secondary numbers don't match with the board. Arbitrary anyways.... I separated the grounds for the positive and negative rails and have them connected by a trace at the end of the board where the off-board connections will be made. This way I can cut the trace and isolate the grounds if they need to be connected at the star ground for hum reasons.

I originally intended to use MUR860 singles, but ordered the dual package on accident. So I redid the board to accommodate the dual packages.

This is the board I have constructed, I'll go over it when I get home to make sure there are no solder bridges or flux bridges or anything like that. I'll find the problem I'm sure...

[ThyDntWntMusic]

where is the track burning?

what about photos of the actual board?

[Redshift187]

Your drawings don't indicate it, but you are using zener diodes for D9 and D10, right?

[pacificblue]

Originally posted by soma_hero
I originally intended to use MUR860 singles, but ordered the dual package on accident. So I redid the board to accommodate the dual packages.

In the layout the AC2/negative rail circuit has the diodes the wrong way round. If you turn all electrolytics and the Zeners in the upper half of the board by 180°, the PSU should work. If the electrolytics are not broken from being the wrong way round, that is.

Once you have the dual diodes, why don't you connect both negative pins of them?

[soma_hero]

I"m confused.

the PSU schem at chipamp.com and the CarlosFM show the diode bridges connected differently. I followed the wiring for the chipamp.com version.

Why are these two different?

Yes D9 and D10 are 15V zeners.

And the traces have been blowing around the diodes. I'll take a picture tonight.