[sdiy] Korg DW-8000 power supply help?

Eric Frampton eric at ericframpton.com
Fri Nov 5 21:19:23 CET 2010


Replies in the order I saw 'em...

On Nov 4, 2010, at 5:47 AM, Tony Kalomiris wrote:

> Nice job Eric,

Thank you sir!

> Are you saying C11 was shorted ? Or did you just find it by substitution ?

It was apparently shorted. I followed everybody's advice and metered all the outputs looking for shorts, and the unregulated 11.5V supply was faulty. I figured there weren't a whole lot of parts between the rectifier and the output except that cap (and Harry and Andy suggested I look at the electrolytics first), so I tried removing it, and the short cleared.

> I've seen tantalums short but don't recall this happening with electrolytics.

I've not seen it happen with an electrolytic, either.

> That's the longest time constant I've ever seen - half an hour !

Oh, it gets better. So after a half-hour, the voltage on that reg. hits 4.95, the CPU board comes up, and the instrument springs to life. I reloaded the factory patches, started playing, and could make out sound but it was incredibly distorted and full of DC-type hum. But funny, as I let the machine run over the course of another hour or so, the audio cleared up completely except for the effects board. Thinking I was on to something, I then let it run overnight, came back the next morning, and the effects board audio was completely clean. So the unit seems to be working 100% now by replacing that cap and just leaving it running.

Then, Andy Jury said:

> Glad you got it fixed. You might want to try replacing the 78 series
> regulator that feeds that particular +5v rail.

I plan to leave the unit baking for another couple of days, and if it goes weird then I'll replace it. Otherwise I'll leave it alone.

Then, Peter said:

> I wasn't sure if you just replaced one cap or all of them ?

I started with the one, then decided to replace the 4700uF and the 1000uF as well. I probably should've replaced every electrolytic in the whole box, but at a certain point that piece just isn't worth that kind of time investment.

Then, Harry Bissell said:

> I'd agree. I find it really hard to believe that other caps were "re-forming" to make the voltage
> slowly rise from 3.5V to 5V.  Re-forming usually does not take place politely unless you force it to do so.

I've only read about it - never experienced it first hand - so this is new territory for me.

> Re-forming a dielectric in an electrolytic cap makes the layer thicker and able to withstand higher voltages.
> Its likely that if you applied a higher voltage (5V) and the dielectric could not stand that voltage, it would
> fail shorted. You would be relying on the regulator chip to limit the current in this case. I would not rule
> out what you have reported, just that I'd look elsewhere (almost anywhere...) before settling on this as the
> root cause.
> 
> I'd more suspect another capacitor that may have been shorted (like a tantalum) and that you heated it until
> it opened up. Or some other explanation ???

If a tant opened up, I'd think the unit would've failed instead of working -better-...?

> Check your power supply current carefully on each rail if possible, and check to see if the regulator chips
> are running hotter than you expect.

The +5 rail dedicated to the CPU and Effects board was running warmer than the other +5 regs., but judging by the PSU layout it looked like Korg expected that to happen - they left extra space for heat dissipation.

> That sure is a curious observation. I'd look really hard, or put a lot of hours on the unit before I'd trust that
> your repair is solid...

It was weird to me, too. I figure I'll let it bake for 3 or 4 days solid, and if everything's still working I figure it's probably safe to let go.

Funny, I never intended to be repairing this piece in the first place. I'd never played with a DW-8000 before and wanted to hear what it sounded like, and my friend just got this on a trade with the suggestion that it only needed a fresh battery. So I thought I was just putting a battery in and re-loading the patches and farting around with it - I never meant to get into all of this! But it has been another good learning experience...

> <twilight zone theme plays...>

Indeed!!

e




More information about the Synth-diy mailing list