Just trying our best, but I can say for my part, I am no expert just a happy amateur. But I do wanna help.
Another thing you might consider, and I do not say this as a besserwisser, but you might want to consider going over your output board solder joints. All of them, including interboard connections (OS/bias boards). You seem to have a few cold joints, and also some flux residue between solder joints. I would redo all cold joints, clean everything properly with isopropanol and a toothbrush, many times, and also look for possible solder bridges. If nescessary through a lens. If other tests prove fruitless.
Just trying to help.
Regards,
Andreas
Another thing you might consider, and I do not say this as a besserwisser, but you might want to consider going over your output board solder joints. All of them, including interboard connections (OS/bias boards). You seem to have a few cold joints, and also some flux residue between solder joints. I would redo all cold joints, clean everything properly with isopropanol and a toothbrush, many times, and also look for possible solder bridges. If nescessary through a lens. If other tests prove fruitless.
Just trying to help.
Regards,
Andreas
Last edited:
Didn't get to try too much tonight, but I did remove the ST_IN2 to 3 connection at the PSU input as suggested. Motorboating still persists.
Here is a recording of the motorboating sound.
http://www.mediafire.com/file/4nr3y4p102sf8qw/PXL_20201210_031802117.mp4/file
Here is a recording of the motorboating sound.
http://www.mediafire.com/file/4nr3y4p102sf8qw/PXL_20201210_031802117.mp4/file
to DronEvil
Hello DronEvil,
what I can hear from your file is no motorboating or hum.
To me it sounds like a regular pulse? 😕
Never heard something like this from an amp.
A cap loading / deloading????
Sorry! I have no idea.
Greets
Dirk
Hello DronEvil,
what I can hear from your file is no motorboating or hum.
To me it sounds like a regular pulse? 😕
Never heard something like this from an amp.
A cap loading / deloading????
Sorry! I have no idea.
Greets
Dirk
If it is a cap, and the sound is there also with a lab PSU, I would look again at your soldering work on the bias and output boards. The only cap in the chain is there, when not using amp PSU and FE is disconnected. Wrt the sound, never heard anything like it. But agree with dirk, sound like something is coupling and decoupling.
And like ZM once wrote: the simplest way for non-technical personnel to find own faults, is to do the big job of desolering and measuring part by part until eureaka, there it was...
Does it play music?
And like ZM once wrote: the simplest way for non-technical personnel to find own faults, is to do the big job of desolering and measuring part by part until eureaka, there it was...
Does it play music?
I will start doing some resoldering and inspection this weekend (time permitting). It does indeed play music, and the music isn't distorted and there's no hum, just that pulsing/putt putt putt sound. The thing that weirds me out is that this happens on both my balanced ba3s. That lead me to believe it was an implementation error on my part (like I installed something wrong the same way multiple times) rather than a cold joint or two. My normal process is to use isopropyl and a brush thoroughly to clean. There is indeed some flux left on a few points where I had to re-solder wires after the chassis was already built. I was being lazy but the time has come to fully disassemble and go over everything it seems. I just REALLY hate doing the chassis work lol. If anyone has any other ideas for what might be coupling/decoupling, I'm all ears! Otherwise, I'll start from scratch and reflow + clean things.
One other question I had was: this probably isn't the TL431s being broken, yeah? Based on my understanding (which is limited) that probably wouldn't be the problem here, but just wanted to check. Thanks again for all the advice!
One other question I had was: this probably isn't the TL431s being broken, yeah? Based on my understanding (which is limited) that probably wouldn't be the problem here, but just wanted to check. Thanks again for all the advice!
Dang, this happens on 2 amps? Probability of two identical bad solder joints is... low?
What do they habe in common?
Could it be (don‘t laugh, I know nothing about cirruitry/debugging but am an ignorant idiot here), like, external noise from the inlet?
What do they habe in common?
Could it be (don‘t laugh, I know nothing about cirruitry/debugging but am an ignorant idiot here), like, external noise from the inlet?
That it plays music fine is at least good to know. I was planning to ask you about the other block, and now you say it has the same problem. I would therefore advise you to wait a little bit before dismantling the thing. It is a tiring and time consuming job, and maybe there are som tests you can do just to rule out a couple of things first.
Wrt TL431: are you able to adjust offset properly? And does the offset stay stable over time? If so, I doubt it has to do with the voltage regulator, but who knows.
One thing I did experience myself, is that the bias board solder joints were a bit fragile. They cant take a lot of heat and reheating, and so care must be undertaken when working the smallest joints. I therefore have ordered new ones and will redo them both. But even though I messed up a few joints, they are working.
I will chime ZM and ask him to take a look at your latest posts.
Regards,
Andreas
Wrt TL431: are you able to adjust offset properly? And does the offset stay stable over time? If so, I doubt it has to do with the voltage regulator, but who knows.
One thing I did experience myself, is that the bias board solder joints were a bit fragile. They cant take a lot of heat and reheating, and so care must be undertaken when working the smallest joints. I therefore have ordered new ones and will redo them both. But even though I messed up a few joints, they are working.
I will chime ZM and ask him to take a look at your latest posts.
Regards,
Andreas
Last edited:
I thought about that too.
Where are you living? Have you tried the amp on a different mains circuit? Or with an isolating transformer?
You did try a lab psu, was that SMPS? If so, it might not be mains related at all.
Hmmm.
Where are you living? Have you tried the amp on a different mains circuit? Or with an isolating transformer?
You did try a lab psu, was that SMPS? If so, it might not be mains related at all.
Hmmm.
I didn’t realize there was a preamp board involved. I guess you should then split the problem by disconnecting the pre and removing power to it. Fix the power amp and get it 100% correct then only reconnect the pre and then get the whole system working so it’s quiet.
(See the Ground Loop presentation - towards the back end there a ‘headphone’ trick you can try - works very well especially if you don’t have a scope. )
(See the Ground Loop presentation - towards the back end there a ‘headphone’ trick you can try - works very well especially if you don’t have a scope. )
Bonsai, thank you. And since your expertise these days is aiding the BA-3-gang, here is the front end/gain stage manual from Nelson P, maybe it sheds some light for you on what errors are definately ours and not Nelsons 😀
http://www.firstwatt.com/pdf/art_ba_3.pdf
And since most are combining it with the complimentary BA-2 output stages:
http://www.firstwatt.com/pdf/art_ba2.pdf (Skip the front end section, it does not apply to the complimentary version).
BTW: many of us are using the stores PSU boards. I, at least, am curious about testing other options when the time is right. Any tips? Could some of the Ovation DIY boards fit the bill?
Another option would be building the PSU point to point. That would be cool, but maybe too advanced just now.
Cheers,
Andreas
http://www.firstwatt.com/pdf/art_ba_3.pdf
And since most are combining it with the complimentary BA-2 output stages:
http://www.firstwatt.com/pdf/art_ba2.pdf (Skip the front end section, it does not apply to the complimentary version).
BTW: many of us are using the stores PSU boards. I, at least, am curious about testing other options when the time is right. Any tips? Could some of the Ovation DIY boards fit the bill?
Another option would be building the PSU point to point. That would be cool, but maybe too advanced just now.
Cheers,
Andreas
Offset jumps around a little bit, between like 0-40mv. It didn't really jump around at all until I connected R5 on the FE to bridge to balanced and then connected FE to OS. When they were not connected and the setup was dual mono things were nice and chill.
Haven't gotten a chance to try another mains. I live in a tiny apartment in Seattle. I have UPS with wave correction feeding all my audio gear, and tonight I'll try without the UPS in the mix.
I have quite alot of other audio gear, like a KGSSHV Stax amp, a class A tube amp I built and some diy sunn clone guitar amps. All on the same mains and none behave this way. Will def try to move to different circuits to experiment though.
There is no active preamp in the mix, just a passive attenuator volume control between the source (a dac) and the BA3.
Bench top PSU is a bk precision 1672
https://www.google.com/url?sa=t&sou...FjABegQIDxAI&usg=AOvVaw0oapSJecC2GgetNv9WPd8l
I will order some replacement bias boards. They definitely had the most re-soldering work done so it is possible that I have done some damage there.
Still at work but will share more findings when I have time to test. Thanks again for all the help!
Haven't gotten a chance to try another mains. I live in a tiny apartment in Seattle. I have UPS with wave correction feeding all my audio gear, and tonight I'll try without the UPS in the mix.
I have quite alot of other audio gear, like a KGSSHV Stax amp, a class A tube amp I built and some diy sunn clone guitar amps. All on the same mains and none behave this way. Will def try to move to different circuits to experiment though.
There is no active preamp in the mix, just a passive attenuator volume control between the source (a dac) and the BA3.
Bench top PSU is a bk precision 1672
https://www.google.com/url?sa=t&sou...FjABegQIDxAI&usg=AOvVaw0oapSJecC2GgetNv9WPd8l
I will order some replacement bias boards. They definitely had the most re-soldering work done so it is possible that I have done some damage there.
Still at work but will share more findings when I have time to test. Thanks again for all the help!
Maybe I’m confused- that’s the power amp driver board - I don’t think you can run your amp without that. Just ignore my earlier post.
Thanks for appreciating the tips. Not much expertise, but we have to help each other. Sounds to me like DC is leaking from your FE to your bias boards. That could be broken FE output caps, or some other connection error. Something is amiss. I would check speaker output offset without the FE connected, just to rule out voltage reg error.
It is interesting that the offset started drifting after that change you did on the FE. Did the motorboating/clipping sound appear then, or was it there regardless?
You could be facing to different issues, but let’s hope it is just one.
I read your first post again. Did you get your offsets corrected? You should set both OS individually so you have 0VDC between speaker + and gnd and 0VDC between speaker - and gnd with inputs shorted to gnd.
Unfortunately the coupling sound happens without the FE connected.
I was able to get bias/offset set nicely after I un-bridged R5 (which was my mistake in my original post many months ago). When I set each channel individually everything was fine.
Unfortunately I didn't think to connect the amp to speakers before bridging it to balanced config. I probably should have used RCAs, hooked up speakers, and done a test before bridging, but I'm an idiot and I didn't. 🙁
I was able to get bias/offset set nicely after I un-bridged R5 (which was my mistake in my original post many months ago). When I set each channel individually everything was fine.
Unfortunately I didn't think to connect the amp to speakers before bridging it to balanced config. I probably should have used RCAs, hooked up speakers, and done a test before bridging, but I'm an idiot and I didn't. 🙁
You are no idiot. Time is a constraint and sometimes shortcuts are taken. That does not mean you are an idiot. You will get this right! And probably learn a lot along the way.
6L6 wrote the 5U 400mm can accommodate 6 deep with a little creativity. How creative, and how to do it? Anyone? Just realized I have both the boards and fets needed.
Regards,
Andy
Regards,
Andy
One PCB on the UMS with the Mosfets on top row.
2nd PCB on chassis floor, mosfets on bottom row of UMS.
Or
Both rows of Mosfets on the UMS where you’d expect them, PCBs mounted on 90deg brackets to UMS.
2nd PCB on chassis floor, mosfets on bottom row of UMS.
Or
Both rows of Mosfets on the UMS where you’d expect them, PCBs mounted on 90deg brackets to UMS.
Sounds feasable. Thanks, Jim!
I just might do that.
Also, I realise now my fets probably arent mismatched per se. They were matched at a lower voltage, and with 1ohm sr, whilst I use 0,47 ohms. So guess your assessment of reducing voltage, is a good idea. So long as I use Nelsons 0,1v matched fets, would you recommend me going the 1ohm route?
One fet is very deviant in current draw. The others are well within 8 percent bracket, but I guess reducing rail voltage and increasing sr’s, would make then more even. Does that equal less distortion? Other benefits, besides the obvious of less likely blowing up?
I guess going 6 deep more than makes up for reduced rail voltage and increased source resistors (of course, 1ohm being the spec at 6 deep).
Hope all is well.
Regards,
Andy
I just might do that.
Also, I realise now my fets probably arent mismatched per se. They were matched at a lower voltage, and with 1ohm sr, whilst I use 0,47 ohms. So guess your assessment of reducing voltage, is a good idea. So long as I use Nelsons 0,1v matched fets, would you recommend me going the 1ohm route?
One fet is very deviant in current draw. The others are well within 8 percent bracket, but I guess reducing rail voltage and increasing sr’s, would make then more even. Does that equal less distortion? Other benefits, besides the obvious of less likely blowing up?
I guess going 6 deep more than makes up for reduced rail voltage and increased source resistors (of course, 1ohm being the spec at 6 deep).
Hope all is well.
Regards,
Andy