The (loosely summarized) way I'd trouble shoot a conventional amp is as follows.
Start removing components 1 at a time (logically of course) till the thing comes out of protect. Or stops blowing fuses.
At this point, I can ignore everything else but these few facts.
Both channel commutators are disconnected. Left seems to be OK. Right is burning resistors.
So my usual method of trouble shooting would now be:-
Remove the right side gain (1 leg of R168 and R170) and see if it burns these again (after replacing them of course).
Then try the output 1 leg of R216, 218. See if it stops burning them. Of course then put R168 and 170 back in.
When it stops eating resistors - I assume that last removal step is what did it.
Then I'd test and replace everything in that stage that looks bad.
Is there a reason I cant do that in this case ?
The trouble shooting guide says how to isolate the stages and figure out which stage is bad. However it doesn't help you isolate it down from there.
So left commutator stage is bad, but what all is in that stage and how to test it and bring it to a component level - it doesn't say.
Thanks.
Srinath.
Start removing components 1 at a time (logically of course) till the thing comes out of protect. Or stops blowing fuses.
At this point, I can ignore everything else but these few facts.
Both channel commutators are disconnected. Left seems to be OK. Right is burning resistors.
So my usual method of trouble shooting would now be:-
Remove the right side gain (1 leg of R168 and R170) and see if it burns these again (after replacing them of course).
Then try the output 1 leg of R216, 218. See if it stops burning them. Of course then put R168 and 170 back in.
When it stops eating resistors - I assume that last removal step is what did it.
Then I'd test and replace everything in that stage that looks bad.
Is there a reason I cant do that in this case ?
The trouble shooting guide says how to isolate the stages and figure out which stage is bad. However it doesn't help you isolate it down from there.
So left commutator stage is bad, but what all is in that stage and how to test it and bring it to a component level - it doesn't say.
Thanks.
Srinath.
There is no reason you can't do that 
And this is the kind of thing I would do, doing the same for both channels.
Short across the vbe multiplier (Q112) to force the output stage to draw no quiescent current (assuming the rest of the stage is OK).
I would fit a bulb tester in place of the triac and power it up and see what the state of play was... drawing current (bulb lit) or if the bulb was out then see what the all the rails measured. If the bulb was lit then that would mean fault-finding to where the current draw was.
And then take it from there.
And this is the kind of thing I would do, doing the same for both channels.
Short across the vbe multiplier (Q112) to force the output stage to draw no quiescent current (assuming the rest of the stage is OK).
I would fit a bulb tester in place of the triac and power it up and see what the state of play was... drawing current (bulb lit) or if the bulb was out then see what the all the rails measured. If the bulb was lit then that would mean fault-finding to where the current draw was.
And then take it from there.
I am probably less likely to cut across the triac now that I have seen more smoke from this than I care to ... But I'll do the stage by stage elimination and see.
I am inclined to get the right side eliminated out of the whole picture - as in if I disconnect the gain or output etc, it will stay that way till I get the left to be 100% ... then I'll go on to the right. I am taking the easier ones first.
Cool.
Srinath.
I am inclined to get the right side eliminated out of the whole picture - as in if I disconnect the gain or output etc, it will stay that way till I get the left to be 100% ... then I'll go on to the right. I am taking the easier ones first.
Cool.
Srinath.
What can cause R172 and 174 to smoke >
R220 I believe if it were getting 100 v at one end and 0 load @ the other end (totally points to a dead short to ground (or short into that R172/174 leg) - Grrr have to find it.
Must find solder bridge (or strand of copper to ground) ...
Cool.
Srinath.
R220 I believe if it were getting 100 v at one end and 0 load @ the other end (totally points to a dead short to ground (or short into that R172/174 leg) - Grrr have to find it.
Must find solder bridge (or strand of copper to ground) ...
Cool.
Srinath.
You should always have 60 volts on the right hand side of D126. You should never see 100 volts on the right hand side. If you have 100v on the left side then that is because the transistor switching the 100 volt rail onto D126 has either failed or is being told to turn on. Neither of those things on their own would stop the amp working normally though (except power dissipation would be higher than normal and so it would run hotter)
That was Q140. It was pulled form the circuit. I still had a 100v.
Anyway I got R172/174 replaced and R220 is still burnt. I desoldered Q122 and Q124.
I powered it up and the resistors didn't burn.
But the left side Q112 - little thermal sensor mounted on the heatsink sparked a few seconds later.
The Left side - You know the side that was the original problem, which I removed the commutator section and its still removed and I have the whole section inactive and the thermal protector should in reality turn on or off and not get a voltaged fed to it - yea that transistor sparked - or something like that.
I am pretty much now thinking I have a metal hunk somewhere moving around and getitng caught and causing havoc.
I will find it and it will start acting a bit more rationally. Not a solder bridge - a lot worse, a moving invisible solder bridge.
I will look much more carefully and now I'll look at the whole amp. Previously I looked @ the left, then left was acting better, and right went to hell. I also keep the amp on its side when I am not working on it. Making me think gravity dictates what acts up ... I'll find it ...
Yea my fault ... clumsy ... desoldering and resoldering and I've been through a few irons and tips this time, and I've used a iron and a solder sucker mostly, who knows if it drops stuff when you sit it over the part you need to suck, the desoldering irons keep hot, and hence are easy to spit the solder out.
Cool.
Srinath.
Anyway I got R172/174 replaced and R220 is still burnt. I desoldered Q122 and Q124.
I powered it up and the resistors didn't burn.
But the left side Q112 - little thermal sensor mounted on the heatsink sparked a few seconds later.
The Left side - You know the side that was the original problem, which I removed the commutator section and its still removed and I have the whole section inactive and the thermal protector should in reality turn on or off and not get a voltaged fed to it - yea that transistor sparked - or something like that.
I am pretty much now thinking I have a metal hunk somewhere moving around and getitng caught and causing havoc.
I will find it and it will start acting a bit more rationally. Not a solder bridge - a lot worse, a moving invisible solder bridge.
I will look much more carefully and now I'll look at the whole amp. Previously I looked @ the left, then left was acting better, and right went to hell. I also keep the amp on its side when I am not working on it. Making me think gravity dictates what acts up ... I'll find it ...
Yea my fault ... clumsy ... desoldering and resoldering and I've been through a few irons and tips this time, and I've used a iron and a solder sucker mostly, who knows if it drops stuff when you sit it over the part you need to suck, the desoldering irons keep hot, and hence are easy to spit the solder out.
Cool.
Srinath.
Work with what you know to be definite problems.
So you had a 100v with Q140 out. The only place that should see 100 volts in that case is the actual source of the 100v rail in the PSU. If you are measuring 100v on the output stage anywhere other than there then there has to be some short or something.t
Seeing things sparking doesn't sound too good.
So you had a 100v with Q140 out. The only place that should see 100 volts in that case is the actual source of the 100v rail in the PSU. If you are measuring 100v on the output stage anywhere other than there then there has to be some short or something.t
Seeing things sparking doesn't sound too good.
I now think I know why he called me on the sharp @ audiokarma.
If you look back, this thread started out as a opamp question. I posted the opamp question here but on carver forum I started way earlier about modding the amp. But basically opamp -> the entire amp is how this went.
If I was dealing with a complete amp especially a relatively well known one, I'd post on audiokarma.
If its a single component (opamp, power chip, trafo winding) etc etc I have a question on I post on here.
Not to say I dont blend those 2 and I have received excellent advice from both sites on both types of topics ( both types ha ... all types of topics ).
I know a good bit about Jap bikes form the 80's and 90's. I'll definitely pay it forward/backward if anyone had a question on those. PM me if you dont want to post here. I'll see what I can do.
Cool.
Srinath.
If you look back, this thread started out as a opamp question. I posted the opamp question here but on carver forum I started way earlier about modding the amp. But basically opamp -> the entire amp is how this went.
If I was dealing with a complete amp especially a relatively well known one, I'd post on audiokarma.
If its a single component (opamp, power chip, trafo winding) etc etc I have a question on I post on here.
Not to say I dont blend those 2 and I have received excellent advice from both sites on both types of topics ( both types ha ... all types of topics ).
I know a good bit about Jap bikes form the 80's and 90's. I'll definitely pay it forward/backward if anyone had a question on those. PM me if you dont want to post here. I'll see what I can do.
Cool.
Srinath.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.