[studeb]

Quote:

Originally Posted by Beefy /img/forum/go_quote.gif Finally got around to making a sound recording of the buzz my PSU makes on power up. The file attached is the best I can do with a webcam mike and Head-Fi's file size limit - hope it is acceptable to give an idea of what is going on. I still can't find anything else wrong, the PSU works as normal when the buzzing stops, and the buzzing doesn't happen when the PSU is warm......

i can not extract it either

 

[Beefy]

Quote:

Originally Posted by runeight /img/forum/go_quote.gif I mean the power inlet terminal on the PS board.

No, definitely coming from somewhere between the caps and the heatsinks. And it is two separate noises, seeming to come from each rail. The negative rail almost always stops buzzing first; in the file I posted it was a rare case when they stopped together.

Also, getting up so close and personal with the amp, I just noticed that the LEDs on a particular rail are a touch brighter while the noise is happening on that rail.

Quote:

Originally Posted by les_garten /img/forum/go_quote.gif I couldn't get the MP3 to extract for some reason.

Sorry, re-done with legacy compression.

 

[WilCox]

Quote:

Originally Posted by runeight /img/forum/go_quote.gif This is helpful. If you can try listening to the bias section near the power inlet and see if the noise is coming from there.

A piece of surgical tubing or even a straw makes a nice makeshift stethoscope to find noisy components.

Speaking of problems, I am curious how much offset Stax headphones can tolerate. My eXStatA SS has about a 12-15V balance missalignment in one channel upon power-up the settles to less than 1V after 20-30 minutes. (The other channel is <1V from the start.)

Will this harm my phones?

 

[studeb]

Quote:

Originally Posted by Beefy /img/forum/go_quote.gif Sorry, re-done with legacy compression.

That works, same sound as mine.
i played with the lid, as long as the lid is within one inch it will buzz.
and when that buzzing stops, i can hear the transformers buzz. not sure if they buzz the whole time not.

 

[les_garten]

Quote:

Originally Posted by Beefy /img/forum/go_quote.gif No, definitely coming from somewhere between the caps and the heatsinks. And it is two separate noises, seeming to come from each rail. The negative rail almost always stops buzzing first; in the file I posted it was a rare case when they stopped together. Also, getting up so close and personal with the amp, I just noticed that the LEDs on a particular rail are a touch brighter while the noise is happening on that rail. Sorry, re-done with legacy compression.

WOW! That's quite a buzz you got there!

 

[Beefy]

Quote:

Originally Posted by les_garten /img/forum/go_quote.gif WOW! That's quite a buzz you got there!

That file is plus 24dB

It is just loud enough to hear with the lid on.

 

[runeight]

Quote:

Originally Posted by Beefy /img/forum/go_quote.gif No, definitely coming from somewhere between the caps and the heatsinks. And it is two separate noises, seeming to come from each rail. The negative rail almost always stops buzzing first; in the file I posted it was a rare case when they stopped together. Also, getting up so close and personal with the amp, I just noticed that the LEDs on a particular rail are a touch brighter while the noise is happening on that rail. Sorry, re-done with legacy compression.

And I think you said that you've tried grounding all of the heatsinks?

The difference in LED brightness is a normal variation because the CCS settings are inexact, but close enough.

This is, obviously, an oscillation, but it is an audio frequency oscillation which is, in itself, peculiar.

The fets could be oscillating (even with the gate stoppers), but their freqs are usually much higher.

The shunt regs themselves could be unstable for some reason, but their gain is not that high and there really isn't any global nfb as there would be if there were a differential error amplifier. There is gain, but not a huge amount relative to other types of regs.

Furthermore, there are 1u caps on the outputs to stop oscillation.

Etc. etc. etc.

So my only useful thought is, if it changes when the top cover is moved closer/farther then the only things on the board that could have any significant capacitative coupling to the top are the heatsinks and, therefore, grounding these should stop the problem.

 

[gilency]

I have a question for Mr Cavalli:
Glass Jar Audio sells some of your designs as a kit. Do you think it would be possible for them or somebody else to eventually sell them that way?
There are plenty of people who would love to get the eXStata.
Would that be a way to increase the availability of this wonderful amp?

 

[sachu]

Quote:

Originally Posted by gilency /img/forum/go_quote.gif I have a question for Mr Cavalli: Glass Jar Audio sells some of your designs as a kit. Do you think it would be possible for them or somebody else to eventually sell them that way? There are plenty of people who would love to get the eXStata. Would that be a way to increase the availability of this wonderful amp?

I think it has been talked about before and that Jeff from glass jar has indicated that he would not be able to take on more kits at this point.

 

[runeight]

I have been in touch with GJA recently because of the other of my amps that Jeff carries. Without speaking for Jeff, I would guess the GJA is saturated with the products that it now sells.

Given all of my other committments at this moment I am thinking it may be a while before these boards come out again. Furthermore, I have been working on determining if other of my amps designs which have not been made available on this forum are good candidates for commercial products. Taking this step would be a significant undertaking. Thus, any free time that I have now, which is slim, will probably go into this investigation and, obviously, cannot be discussed on the diy part of headfi.

However, in the original thread posted by ericj there were other ideas proposed for stat amps. Maybe someone else can come up with a good, cheap design in the meantime?

Edit: I see that Sachu beat me to a reply. He usually does.

 

[Keithpgdrb]

Sachu, you may want to re-word your signature about commissioned builds. I dont even know you, and it rubbed me the wrong way. I'm sure your a great guy, but that signature is pretty harsh. Just trying to keep the Karma good.

 

[gilency]

Thank you Sachu and runeight. I look forward for any other electrostatic ventures Mr Cavalli may have. Either DIY or commercially available.

 

[sachu]

Quote:

Originally Posted by Keithpgdrb /img/forum/go_quote.gif Sachu, you may want to re-word your signature about commissioned builds. I dont even know you, and it rubbed me the wrong way. I'm sure your a great guy, but that signature is pretty harsh. Just trying to keep the Karma good.

I will consider it...but I have good reason to keep it strongly worded.

 

[Beefy]

Quote:

Originally Posted by runeight /img/forum/go_quote.gif And I think you said that you've tried grounding all of the heatsinks?

That was someone else, but I can ground the PSU heatsinks easily enough.

Quote:

The difference in LED brightness is a normal variation because the CCS settings are inexact, but close enough.

Just focusing on a single rail though, the LED is brighter while the buzzing is happening. It dulls a touch when the buzzing stops. So perhaps the buzzing indicates extra current moving through the shunt. Though I suppose this could be an issue with regulation, or the FETs supplying extra current.

Quote:

The shunt regs themselves could be unstable for some reason, but their gain is not that high and there really isn't any global nfb as there would be if there were a differential error amplifier. There is gain, but not a huge amount relative to other types of regs.

I recall reading that the 'Max' modification increases gain at some point. Could this be unique to Max builds?

Quote:

So my only useful thought is, if it changes when the top cover is moved closer/farther then the only things on the board that could have any significant capacitative coupling to the top are the heatsinks and, therefore, grounding these should stop the problem.

That was somebody else; lid on or off doesn't seem to bother my build. But I will ground the heatsinks to see if this helps.

 

[pabbi1]

Quote:

Originally Posted by Keithpgdrb /img/forum/go_quote.gif Sachu, you may want to re-word your signature about commissioned builds.

Sadly, his wording is appropriate, and might could be stated even stronger. It is what it is, to keep everyone in 'good standing'.

For you hybrid guys, there are 3 five tube sleves of matched construction on eBay. Price has increased a wee bit.