diyAudio (
-   AKSA (
-   -   Fixing a descent into hash on AKSA 100 (

seano 23rd January 2012 10:22 PM

Fixing a descent into hash on AKSA 100
My dear old AKSA 100 N+ is being a little frustrating.

Early in the 'listening session'...ususally at random during the first or second disc...the audio output from both channels descends into squirly distorted hash before cutting out into silence.

On rare occasions, it comes back on its own after anything from tens of seconds out to a couple of minutes but it may come back only intermittently before cutting out. I say rare because I rarely let it go on that far.

The way to get it back is simple - flick the power switch. The instant the starts drawing on the capacitors alone...clarity returns. If I flick the switch back to full mains in the couple of seconds before the caps run out then there is no loss of audio - the hash just disappears. I may have to do this three or four times during the first two CDs but after that it's all good.

The system itself is a Cambridge Audio player, Redgum DAC, Yamaha integrated (pre-amp only) and the AKSA into Krix Lyrix floorstanders. Typical mix of budget quality interconnects and speaker cables are 8ga DC power cables less than 2m long.

Any ideas? My thoughts are that it has to be related to the power supply and/or earth or ground loop issues but the temporary nature of the problem has me stumped. :scratch:

I thought initially it might be due to the power supply in the DAC but I can't imagine how that might be possible...but I'm happy to take up suggestions in this regard - bear in mind though that resetting the amp clears the problem!

Without the AKSA in the chain there is no problem but the music doesn't sound as good so there is a problem!!

AKSA 23rd January 2012 11:16 PM

Hmm, Seano, I have a neighbour with that problem too, though the smoke is of a very different nature......

I would say that your power supply rectifiers either have a dry connection - fix with a simple resolder - or there is a problem with the diode that sustains the greatest dissipation, namely the centre diode.

Hope this helps, stay in touch,



seano 23rd January 2012 11:58 PM

Resoldering I can do...the only choice to make is when. Bicycles have become my engineering challenge of choice of late though I'm not sure how successful I am since iseem to spend as much time fixing them as riding them. Returning to audio engineering is like returning to a past life (might be cheaper though).

Is there a test I can perform on the centre diode or should I simply bin it (if the resoldering doesn't fix it)?

Is your neighbour letting the smoke get out, eh? I only do that when I'm cooking pancakes...except the smoke is black while the language is blue!

Bigun 24th January 2012 02:31 AM

This is probably of no help at all, but just in case...

I had a problem with my Lifeforce clones where the sound just faded away after a few seconds to a couple of minutes. If I turned it off, waited and then turned it back on again, the sound returned but then eventually it just went off again. It was mostly one channel, but after the first channel faded it seemed to cause the 2nd one to follow suit. There was a signal going in, but nothing coming out. No smoke (!) or heat or dc voltages or funny sounds, it was just not producing sound.

I never understood the root cause but to this day I'm convinced it was due to parasitic oscillations. I had not installed output inductors and I did not have very good power supply decoupling. My power supply wiring wasn't adequately isolated from the signal input.

The problem started after I introduced some new meaty and long speaker cables.

Later, I rebuilt my Lifeforce clones with new power supply, replaced the signal wiring with shielded coax and added output inductors - the problem was fixed.

edit: p.s. I know the original power supply had no bad diodes or other problems - as it's now running my 3 channel TGM5 amp unchanged and without any problems.

AKSA 24th January 2012 08:28 PM


Did the 10R Zobel resistor on the output of the modules heat up considerably?

If yes, then indeed the problem was oscillation, but not otherwise.


Resolder your centre rectifier diodes, that might fix it. And when it plays up, see if the output capacitors are heating up, indicating passing AC. If not, then the problem is likely a dry joint in the power supply.


Bigun 1st February 2012 10:32 PM

No, I didn't detect significant heating on the zobel resistors. Strange.

AKSA 28th April 2012 11:45 AM


Did you fix this problem?

I should add it's unlikely the neighbour's problem was electronic in nature, I was using the smoking word in the literal sense, if you get my drift....... as I often got his.


seano 29th April 2012 11:36 PM

Not fixed yet. Sitting a little low in the list of priorities. I've disconnected it for the moment.

Looks like my Redgum DAC does have a PS issue as well since it too stutters and drops out for some seconds (and sometimes more than once) early on in the listening session...even without the AKSA in the chain.

Must say I'm beginning to understand how people can end up with cupboards full of unused electronics!!

unmibh 30th April 2012 11:47 AM

Hi Seano,

Why not just redo the soldering, any dull solder is a potential culprit.
I have one aksaphile who ask me to fixed a 55W Aksa and this is all I did for the said unit.
One channel was sounding like distortion and it was because of cold solder.


AKSA 30th April 2012 10:48 PM

Yep, thanks Rom, good advice, Seano. Solder joints cause most of the problems, that is, when the smoke stays inside.......



All times are GMT. The time now is 10:47 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2