A strange problem with new chipamp build

[PJN]

Hi Guys,

I'm in the middle of a chipamp.com 3886 build, so far the PSU functions well, but when I put power to the assembled amp boards something shorts out. The amp boards are pretty simple and I've triple checked it for possible solder bridges, and miss-installed components. I checked the amp board power traces with my DMM and could find no obvious problems/shorts. Everything looks fine and no smoke or anyother sign of a problem with the amp boards or components is evident. I even removed one of the amp chips and found that the board was still drawing a pretty heft load even though it didn't blow the fuse. Has anyone had an experience with bad amp boards causing symptoms like this. Does anyone have any advice on other diagnostics that I might try.

Thanks,

PJN

[Andy5112405]

How are your soldering skills ? You could have cooked the LM3886 with excessive heat.

[Arty]

that is rather unlikely, surprising what they can withstand.
Picture of the board, good quality so we may spot shorts, accidents, whatever, pcb error would help a lot more. Allso picture the PSU too.

[Andy5112405]

Try measuring the PSU on its own with a resistive load. If there is any AC that could explain the high current.

[PJN]

Hi Guys,

I've built several successful projects before without problems, a couple of RevC's, a descrete amp, a couple of preamps and two DAC's so my soldering skills aren't terrible. I'll post a couple of pics. The PSU puts out a steady +/- 34vdc at no load, I'll try measuring it with a load.

Thanks,

PJN

[sregor]

Do you have 3886 with metal tabs and are you using a mica or Silpad insulator?. The metal tabs are connected to the negative power supply pin, and will short out to the heat sink if no insulator> Just a possibility ....

[PJN]

Hi Guys,

Here are a few pics of the test rig and boards. I put things together with stuff that I had lying around in my parts bin, but all parts values matched the BOM. I did use mica insulators between the chip and the heat sink and the spreader bar and the chip, and the heat sinks are mounted on a board.

PJN

[multisync]

that bracket that bolts the chip to the heatsink is it plastic or metal and is it isolated from the chip as well? With the powersupply unpluged and everything put together ohm out the +/- rails to ground, you should see 0 ohms and the caps slowly charging from the ohm meter. check for any electrolytic installed backwards

[quikie22]

have you connected the supply correctly? I once accidentally switched V+ and V- and drew a lot of current and then blew up some electrolytics.

Alternatively, one of the electrolytics is failing under high voltage... and shorting through.

[sregor]

Does your transformer have 2 separate windings or is it a center tapped winding? The center tapped ones have to be wired differently and might draw current once the 2 grounds are connected together. (Use a single bridge instead of two bridges I think)