Page 9 of 10

Re: Amps - Whats in em and why?

Posted: Mon Nov 11, 2019 7:48 pm
by Lurcher300b
Here is a tail of a odd failure of a pair of monoblocks I had just built and was running in. Working fine, sounding great, suddenly, they both go into protection. Hmm, well, they both failed, and stay failed, so for them both to go, it must be a common source. Must be excess voltage out of the phono stage. Is that broken?, no, that's fine. Take amp cover off, one of the rail fuses has blown (+ve). Take other amp top off, the other rail fuse (-ve) has blown. Replace, fuse, try on bench, fuse blows again. Disconnect power supply, power up amp on bench supply (current limited, assuming it will let me see where the failure is). Works just fine, thrash amp in every way I can think trying to break it, nothing. Hmm, maybe a bridge rectifier has gone. Shouldn''t have, but maybe. Try power supply with no amp connected, works fine, change bridge rectifier anyway thinking it may be failing under load. Nope still works fine on its own. Connect amp, it fails. Only thing different is the amp is powered from the mains when it fails and amp is connected to power supply and the chassis is connected to earth, but is floating when on bench supply. Eventually Track down problem, one of the TO220 driver transistors has shorted past its mica washer and shorted one of the power rails to earth. Cause of the problem was I forgot to widen the hole around the fixing screw in the heatsink, so there was not much insulation to keep the two apart. Replace mica and drill out hole, its working fine.

But its a mechanical fault, my fault, but still mechanical, so why did the other one go at exactly the same time? Repeat process with second amp, exactly the same problem, but this time in the driver transistor for the other rail. Same cause, not opening the hole enough, but why the hell would they both go at the exact same time?

The only theory I have is that one failed, shorted +ve rail to ground, blew fuse, shorting the rail to ground will have briefly pulled up the earth to a +ve voltage. Other amp, having its ground pulled up, will have seen a larger voltage between its case and its -ve rail (ground moving up, so further away from -ve rail). Increased voltage caused the insulation to pick that moment to fail under increased voltage.

It sounds unlikely, but its the only reason I can think they both went at the same time (well one a few miliseconds after the other maybe), and the other clue was that one went in the +ve rail, and the other in the -ve one.

In both cases the protection circuit disconnected the loudspeaker, shut the power supply down and flashed the leds on the control board indicating a DC issue.

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 10:06 am
by NSNO2021
Lurcher300b wrote: Wed Nov 06, 2019 9:06 pm
So 690 views and 40 odd posts and I'm still no more educated than when I started (Bar the one sensible post)
All you will get from your question is other peoples opinions, if you want facts, then you will need to spend five years learning electronics to understand the answers, by which time you will not need the question answering anyway.
Amen to that 🙌 I am just starting out again re hi-fi for want of a better phrase and every answer I read prompts multiple questions. As a result I accept that I am on a huge learning curve, that I will make mistakes.and those mistakes will cost me in time and money. I will be happy if I minimise the number and cost of the mistakes so I arrive at a sound I am happy with sooner rather than later .

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 10:37 am
by karatestu
If you go down every possible road and try every permutation out it will take a very long time. You will however be extremely qualified in knowing which YOU think is best. I am guilty of always wondering what something else will sound like when really i should be finishing the diy stuff i have started and concentrating on listening to music.

Many things in audio are a compromise, trading off one thing for another (especially speakers). What is best ? Only you will know that.

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 11:19 am
by savvypaul
I'm mildly amused at the 'bill and coo' over photos of component innards that seems to be the norm on some forums and user groups. I doubt that 99% of the people commenting have a reliable understanding of what they are looking for...and the 1% who do have a proper understanding know that a photo alone cannot give them all the information they need to reach any deep conclusions.

As Nick (Lurcher300b) recently commented (elsewhere)..."Electrons don't get OCD"

Sometimes, a 'little' knowledge is a dangerous thing.

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 11:23 am
by Geoff.R.G
savvypaul wrote: Tue Nov 12, 2019 11:19 am I'm mildly amused at the 'bill and coo' over photos of component innards that seems to be the norm on some forums and user groups. I doubt that 99% of the people commenting have a reliable understanding of what they are looking for...and the 1% who do have a proper understanding know that a photo alone cannot give them all the information they need to reach any deep conclusions.

As Nick (Lurcher300b) recently commented (elsewhere)..."Electrons don't get OCD"

Sometimes, a 'little' knowledge is a dangerous thing.
Sometimes rusty knowledge is equally dangerous! The difference is that what is rusty can be cleaned up but what isn't there in the first place...

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 11:29 am
by Geoff.R.G
Lurcher300b wrote: Mon Nov 11, 2019 7:48 pm Here is a tail of a odd failure of a pair of monoblocks I had just built and was running in. Working fine, sounding great, suddenly, they both go into protection. Hmm, well, they both failed, and stay failed, so for them both to go, it must be a common source. Must be excess voltage out of the phono stage. Is that broken?, no, that's fine. Take amp cover off, one of the rail fuses has blown (+ve). Take other amp top off, the other rail fuse (-ve) has blown. Replace, fuse, try on bench, fuse blows again. Disconnect power supply, power up amp on bench supply (current limited, assuming it will let me see where the failure is). Works just fine, thrash amp in every way I can think trying to break it, nothing. Hmm, maybe a bridge rectifier has gone. Shouldn''t have, but maybe. Try power supply with no amp connected, works fine, change bridge rectifier anyway thinking it may be failing under load. Nope still works fine on its own. Connect amp, it fails. Only thing different is the amp is powered from the mains when it fails and amp is connected to power supply and the chassis is connected to earth, but is floating when on bench supply. Eventually Track down problem, one of the TO220 driver transistors has shorted past its mica washer and shorted one of the power rails to earth. Cause of the problem was I forgot to widen the hole around the fixing screw in the heatsink, so there was not much insulation to keep the two apart. Replace mica and drill out hole, its working fine.

But its a mechanical fault, my fault, but still mechanical, so why did the other one go at exactly the same time? Repeat process with second amp, exactly the same problem, but this time in the driver transistor for the other rail. Same cause, not opening the hole enough, but why the hell would they both go at the exact same time?

The only theory I have is that one failed, shorted +ve rail to ground, blew fuse, shorting the rail to ground will have briefly pulled up the earth to a +ve voltage. Other amp, having its ground pulled up, will have seen a larger voltage between its case and its -ve rail (ground moving up, so further away from -ve rail). Increased voltage caused the insulation to pick that moment to fail under increased voltage.

It sounds unlikely, but its the only reason I can think they both went at the same time (well one a few miliseconds after the other maybe), and the other clue was that one went in the +ve rail, and the other in the -ve one.

In both cases the protection circuit disconnected the loudspeaker, shut the power supply down and flashed the leds on the control board indicating a DC issue.
That is an excellent example of intelligent troubleshooting, something I have found rare amongst some, so called, professional engineers. I had one try to tell me that a cracking noise from a speaker was the, passive, speaker failing! Swapping the channels over soon disproved that idea. it turned out to be a capacitor in a signal processing unit, which unit failed some time later.

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 11:53 am
by Lurcher300b
That is an excellent example of intelligent troubleshooting, something I have found rare amongst some, so called, professional engineers.
I have spent much of my working life fixing software and hardware, you can't fix a bug unless you know whats causing it. To do otherwise is to just cover it up and possibly give yourself twice the number of problems.

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 11:54 am
by Lurcher300b
I am guilty of always wondering what something else will sound like when really i should be finishing the diy stuff i have started and concentrating on listening to music.
I have found the solution to that is never to start a project unless you know exactly what flaw in existing kit its meant to solve. "better" is not a precise enough goal.

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 12:55 pm
by Daniel Quinn
You sell your wares. It will necessarily involve a different ethos.

Re: Amps - Whats in em and why?

Posted: Tue Nov 12, 2019 2:10 pm
by Lurcher300b
Daniel Quinn wrote: Tue Nov 12, 2019 12:55 pm You sell your wares. It will necessarily involve a different ethos.
I disagree, knowing what you don't like about what you have and what you want to improve is a good way of decising if you are just looking to buy something or make something DIY. Not doing that just leads to endless box swapping.