Power supply unstable except when active load attached (but not enabled)

[uoficowboy]

I have a Shenzhen Mastech HY3003D-2 that is acting... funny. If you measure the output with a scope, it has about a 10V peak to peak sawtooth waveform at about 8.5KHz (on top of the DC). However, if I attach a active load, specifically a BK Precision 8500 to it, the oscillation stops and it acts like a normal supply - it handles loads gracefully as well. But if I connect a resistor to it, it just oscillates. Further, the load stabilizes both channels at the same time. As in, even though the load is only connected to one channel, both channels stabilize (and they're set to be independent). I'm measuring this with a grounded scope. I've tried grounding either side of the output, and no change.

Anybody have any ideas as to what is going on here?

[Jay_Diddy_B]

Hi,

The load probably has a RC network connected across the input.

Try connecting an electrolytic capacitor across the output of your power supply and the load resistor. If the removes the oscillation open up the load and look for a loose capacitor/bad joint on the output capacitor.

The output capacitor is required for control loop stability.

Good luck !!

Jay_Diddy_B

[uoficowboy]

Hi,

The load probably has a RC network connected across the input.

Try connecting an electrolytic capacitor across the output of your power supply and the load resistor. If the removes the oscillation open up the load and look for a loose capacitor/bad joint on the output capacitor.

The output capacitor is required for control loop stability.

Good luck !!

Jay_Diddy_B

Looks like you're right - it has a 470uf capacitor that measures out as about 100nf on the output. Ouch.

Oddly enough, only one channel has a shrunken capacitor. The capacitor on the other channel (that did not exhibit the problem as much, but still exhibited it) measures out at about 490uf. But I suspect the ESR is horrendous or something similar, so I'll replace it anyways.

Thanks for the help!