I have been working on a NAD 2200, which had a number of fried components centered on Q308. I replaced the damaged components and then when the protection clicked off thought, excellent all solved.
I then did my usual recap (which I done many times) and had the amp ready to bias . Then I discovered the problem.
The right channel has 90v (both +ve and -ve) on the collector of each of the output devices, whereas it should be 60v. There is a detector stage Q322/Q324 which switches the rail between 60v and 90v to achieve the class G operation and the additional headroom that provides. ("Power envelope").
I have replaced D330 and D332 and Q328 and Q326 (KSC2690 and KSA1220 respectively). I have even replaced Q330 and Q332 as these had been earlier replaced as well.
The voltages on the detectors Q324/Q322 measure as they should, but I am at a loss as to why 90v is going to the collectors of all the output s (Q316/Q320/Q314/Q318).
I have replaced Q306/Q308 and Q302/Q304 although all these tested fine as did Q334 and Q336 which I also replaced.
I then went for my "go to" problem areas with these amps and replaced D312/D310 and ZD302/ZD304 and D322/D318 and D320/D324 as these can all cause issues.
So I am now flummoxed, as the only way I can see the 90v rail getting to the collectors is through the switching circuitry Q326/Q330 for the +ve rail and Q328/Q332 for the -ve rail. The collectors of each of those should be at 60v (+ve and -ve) not the 90v I have now.
I did resolder any connection on the PCB that looked suspect and checked there is an earth to the power supply as well.
Any guidance, thoughts, insights or wisdom would greatly appreciated, as I keep bashing my head into this wall.
Thanks for reading, and circuit attached
Peter
I then did my usual recap (which I done many times) and had the amp ready to bias . Then I discovered the problem.
The right channel has 90v (both +ve and -ve) on the collector of each of the output devices, whereas it should be 60v. There is a detector stage Q322/Q324 which switches the rail between 60v and 90v to achieve the class G operation and the additional headroom that provides. ("Power envelope").
I have replaced D330 and D332 and Q328 and Q326 (KSC2690 and KSA1220 respectively). I have even replaced Q330 and Q332 as these had been earlier replaced as well.
The voltages on the detectors Q324/Q322 measure as they should, but I am at a loss as to why 90v is going to the collectors of all the output s (Q316/Q320/Q314/Q318).
I have replaced Q306/Q308 and Q302/Q304 although all these tested fine as did Q334 and Q336 which I also replaced.
I then went for my "go to" problem areas with these amps and replaced D312/D310 and ZD302/ZD304 and D322/D318 and D320/D324 as these can all cause issues.
So I am now flummoxed, as the only way I can see the 90v rail getting to the collectors is through the switching circuitry Q326/Q330 for the +ve rail and Q328/Q332 for the -ve rail. The collectors of each of those should be at 60v (+ve and -ve) not the 90v I have now.
I did resolder any connection on the PCB that looked suspect and checked there is an earth to the power supply as well.
Any guidance, thoughts, insights or wisdom would greatly appreciated, as I keep bashing my head into this wall.
Thanks for reading, and circuit attached
Peter
Odd that both independently operating switching circuits fail to work right, but seem ok.
At this point I would start looking upstream of the switching circuits to find one problem
that affects them both, maybe in the area of Q336 and the associated clamp diodes?
I haven't seen this circuit before, though. Does the bias circuit Q334 and pot still work ok?
Is the overall voltage gain still on spec?
At this point I would start looking upstream of the switching circuits to find one problem
that affects them both, maybe in the area of Q336 and the associated clamp diodes?
I haven't seen this circuit before, though. Does the bias circuit Q334 and pot still work ok?
Is the overall voltage gain still on spec?
Last edited:
@dbxdx5
I did the recap as part of repairing that channel, as often the problem is one of the smaller caps failing.
@rayma
I have tested the diodes D314/D316 and they are both good, as are R320,R#@!, R324, R326,R328 and R316 and R314.
The bias circuit seems to work, the pot brings up the bias, resulting in a higher current draw. Q336 measures 0v on base and emitter and 90v on collector.
I should have added the voltages on the attached circuit are the reference voltages for a working unit. Q330/Q332 and Q326/Q328 all have +ve and -ve 90v on their collectors respectively, whereas they should be 60v. The voltages on Q322 and Q324-the detectors are correct , 90v on the base and collector (or thereabouts) and 60v on the emitter.
I think you maybe right, that the problem is upstream. The emitters of Q216/Q218 both measure +/- 5.8v as they should.
thanks
Peter
I did the recap as part of repairing that channel, as often the problem is one of the smaller caps failing.
@rayma
I have tested the diodes D314/D316 and they are both good, as are R320,R#@!, R324, R326,R328 and R316 and R314.
The bias circuit seems to work, the pot brings up the bias, resulting in a higher current draw. Q336 measures 0v on base and emitter and 90v on collector.
I should have added the voltages on the attached circuit are the reference voltages for a working unit. Q330/Q332 and Q326/Q328 all have +ve and -ve 90v on their collectors respectively, whereas they should be 60v. The voltages on Q322 and Q324-the detectors are correct , 90v on the base and collector (or thereabouts) and 60v on the emitter.
I think you maybe right, that the problem is upstream. The emitters of Q216/Q218 both measure +/- 5.8v as they should.
thanks
Peter
thanks- The challenge is that with the 90v on the output collectors, the bias keeps rising and Q308/Q306 start getting very hot very quickly as do their associated resistors, as the application of this voltage for the additional output is meant to be for very short periods to provide +6db additional headroom.
So I can only have the amp powered for short periods to check voltages etc.
I will pull and test the upstream transistors and double check again that I have not made a mistake in replacing a part. Though why this is happening on both rails still has me confused.
I have tested all the output transistors with the PEAK DCA75 and they all test good.
thanks again
peter
So I can only have the amp powered for short periods to check voltages etc.
I will pull and test the upstream transistors and double check again that I have not made a mistake in replacing a part. Though why this is happening on both rails still has me confused.
I have tested all the output transistors with the PEAK DCA75 and they all test good.
thanks again
peter
so I have replaced all the small transistors-just in case- and some more diodes, replaced R374 and R308 and still 90v on that channel. Checked the bias resistors, they are fine.
It would seem that either Q336 and associated circuitry or Q334 and bias circuitry could cause both rails to go to 90v rather than 60v. I even read Rod Elliot's excellent piece on Class G.
So now I am at a total loss- so please any guidance is greatly appreciated.
Peter
It would seem that either Q336 and associated circuitry or Q334 and bias circuitry could cause both rails to go to 90v rather than 60v. I even read Rod Elliot's excellent piece on Class G.
So now I am at a total loss- so please any guidance is greatly appreciated.
Peter
While you fault find the rest of the circuit, to prevent the amp blowing up, pull Q330 and Q332, or at least disconnect their collectors. The amp should still work, just with the rails limited to 60V
thanks @jaycee- I did remove Q330 and Q332 at one point, and still had the 90v, there is also 90v (+ve and -ve) on Q328/Q326- both of which have been replaced- that said I will try again tomorrow as you suggest.
I would assume I can remove Q328 and Q326 as well and see if the voltage drops to 60v?
As far as I understand the only other connection of the rails is D330 and D332?
Thanks again
Peter
I would assume I can remove Q328 and Q326 as well and see if the voltage drops to 60v?
As far as I understand the only other connection of the rails is D330 and D332?
Thanks again
Peter
so some "progress".
I removed Q332 and Q328, and the voltage on the collectors of Q316 and Q320 were at 60v.
Both Q328 and Q332 are new, and test fine- so as Q324 voltages are correct, which is the detector, seems to be functioning?
So why are Q328/Q332 being turned on so that they apply the 90v to the collectors?
I removed Q332 and Q328, and the voltage on the collectors of Q316 and Q320 were at 60v.
Both Q328 and Q332 are new, and test fine- so as Q324 voltages are correct, which is the detector, seems to be functioning?
So why are Q328/Q332 being turned on so that they apply the 90v to the collectors?
So splitting the problem into parts.
The -ve rail is OK (i.e. 60v on the collectors of Q316/Q320) with Q328 and Q332 removed.
The +ve rail still has 90v on the collectors of Q314/Q316. (Q330 and Q326 installed-both new)
So I removed Q332, the detector for the +ve rail- and no joy still 90v on the collectors.
From this I am surmising that the detectors are operating correctly, which my implication suggests that Q336 and associated circuitry is also OK (Q336, R326,R328(both fried when i got the amp),R324,D316,D314 are all new and R320,R560 test OK)
The implication is the bias circuitry- at the collector of Q304 I have -1.68v and at the collector of Q302 I have +2v- which is too high. R374 and R376 have both been replaced, as has Q334 and SVR302. R348 and R350 both measure OK. I will replace D324 and D320 and see if that makes a difference
Again any assistance is greatly appreciated as my head is hurting from bashing it against this brick wall
peter
The -ve rail is OK (i.e. 60v on the collectors of Q316/Q320) with Q328 and Q332 removed.
The +ve rail still has 90v on the collectors of Q314/Q316. (Q330 and Q326 installed-both new)
So I removed Q332, the detector for the +ve rail- and no joy still 90v on the collectors.
From this I am surmising that the detectors are operating correctly, which my implication suggests that Q336 and associated circuitry is also OK (Q336, R326,R328(both fried when i got the amp),R324,D316,D314 are all new and R320,R560 test OK)
The implication is the bias circuitry- at the collector of Q304 I have -1.68v and at the collector of Q302 I have +2v- which is too high. R374 and R376 have both been replaced, as has Q334 and SVR302. R348 and R350 both measure OK. I will replace D324 and D320 and see if that makes a difference
Again any assistance is greatly appreciated as my head is hurting from bashing it against this brick wall
peter
@rayma- check all the caps you mentioned and they all measure good- they are all new WIMA FKP.
In my dreams(yes I dream about fixing amplifiers), it occurred to me that what ever the problem is may have been the one that caused the amp to originally fail. So although I have measured each resistor and diode, replaced each cap and transistor- I am wondering if there is a fault on the PCB. I have cleaned up any solder joints that look problematic, but am wondering if there is a hairline crack somewhere (an earth?) that my hours of staring at the board has missed?
Thanks for reading
Peter
In my dreams(yes I dream about fixing amplifiers), it occurred to me that what ever the problem is may have been the one that caused the amp to originally fail. So although I have measured each resistor and diode, replaced each cap and transistor- I am wondering if there is a fault on the PCB. I have cleaned up any solder joints that look problematic, but am wondering if there is a hairline crack somewhere (an earth?) that my hours of staring at the board has missed?
Thanks for reading
Peter
You might want to double check for trace damage from when R326 and R328 went "poof." I'm working on another 2200 that had the corresponding resistors (R325 and R327) burn up, and the trace damage was significant.
So I looked very carefully and tested the continuity of R326/R328, and the bias circuit, but to no avail.
When I got the amp Q316/Q320 were not fitted- I expect they were fried as R368 was open circuit.
As far as I can see the common elements to both rails being 990v rather than 60v is eiether the bias circuit or the circuitry around Q366. As the base of Q316/Q320 is about -0.5v and Q314/Q318 is about +0.5v and all the emitters are 0v, I am fairly sure the output transistors are fine (I did replace Q316/Q320 with new onsemi transistors to be sure.
all thoughts welcome-my head hurts
When I got the amp Q316/Q320 were not fitted- I expect they were fried as R368 was open circuit.
As far as I can see the common elements to both rails being 990v rather than 60v is eiether the bias circuit or the circuitry around Q366. As the base of Q316/Q320 is about -0.5v and Q314/Q318 is about +0.5v and all the emitters are 0v, I am fairly sure the output transistors are fine (I did replace Q316/Q320 with new onsemi transistors to be sure.
all thoughts welcome-my head hurts
The over current circuit is connected to the 90V rail via C313, so maybe a fault in
that circuit or C313 could force both bus switching circuits to be latched up at once.
Remember that both the positive and negative switching circuits are triggered on,
even though they operate independently of each other. The switching is triggered
by capacitor coupled signals, so a capacitor could be shorted and hold the HV on.
Too bad the manual doesn't have a theory of operation section, just a block diagram.
that circuit or C313 could force both bus switching circuits to be latched up at once.
Remember that both the positive and negative switching circuits are triggered on,
even though they operate independently of each other. The switching is triggered
by capacitor coupled signals, so a capacitor could be shorted and hold the HV on.
Too bad the manual doesn't have a theory of operation section, just a block diagram.
Last edited:
@rayma- I wish there was a theory of operation, I have read a few texts on this configuration, but they tend to describe what rather than how. Did you mean C312?
there is an imbalance in that D310 is +2v and D132 is -1.68- they both should be 1.77v, which suggests that Q302 and surrounding components may not be operating correctly. The voltage at the base of Q212 and the preceding Q202/Q205/Q210/Q214 are all as they should be.
I will try and delve further into that area
Thanks
there is an imbalance in that D310 is +2v and D132 is -1.68- they both should be 1.77v, which suggests that Q302 and surrounding components may not be operating correctly. The voltage at the base of Q212 and the preceding Q202/Q205/Q210/Q214 are all as they should be.
I will try and delve further into that area
Thanks