If I remember correctly, the logic was that you had replaced all of the suspect front end transistors connected it all back up, and you had high offset voltage and the inability to swing to the rail. With I think oscillation on trying to swing to the rail. This led us to look at the output stage, and you are still seeing that the output stage will not go to the rail.
It does not make sense that it is oscillating, but there is a lot left in place so it is hard to say. I have a feeling that it has something to do with the bias network but again it is hard to say.
When I can't find a reason for a failure I break the problem down further. There is one output that leaks at low voltage and there might be more that leak at higher voltage, in fact it could explain the oscillation. You could hook it all back up, but we don't know if you'll find anything different from the last time you did with all probably good front end parts. Or we could pull the drivers and just test the outputs, I'll give you the step by step if you want to pull the drivers - because there is a bit more to it than just pulling the drivers. It is up to you.
While it may seem like we are moving slowly, one suggestion is to slow down, because the more work you do on the amp other than taking it apart moves you further away from when it was previously working. It introduces more variables obviously.
It does not make sense that it is oscillating, but there is a lot left in place so it is hard to say. I have a feeling that it has something to do with the bias network but again it is hard to say.
When I can't find a reason for a failure I break the problem down further. There is one output that leaks at low voltage and there might be more that leak at higher voltage, in fact it could explain the oscillation. You could hook it all back up, but we don't know if you'll find anything different from the last time you did with all probably good front end parts. Or we could pull the drivers and just test the outputs, I'll give you the step by step if you want to pull the drivers - because there is a bit more to it than just pulling the drivers. It is up to you.
While it may seem like we are moving slowly, one suggestion is to slow down, because the more work you do on the amp other than taking it apart moves you further away from when it was previously working. It introduces more variables obviously.
It might not be a bad idea to check the catch diodes d1 and d2 for leakage as well if they haven't been replaced. Those diode can take a real beating.
David.
David.
Never mind. I see this has already been addressed.
I'm still catching up on my reading.
Pete have you tried duplicating fault conditions in spice to see if you can get the same effects? If that were successful it might give some insight to the problem.
I've never tried it before but who knows.
David.
Hi David, all,
Often it's caused by small particles in the grease that punch through the mica. I've found almost everything you can think of between the transistor and mica (including no mica!). Flattened solder balls are pretty common. Dirt, cut wire and insulation are common. Fingernails are not unheard of. I've even seen common clear plastic wrap used as an insulator (sheesh!).
I always wonder what the devil is going on with some service people! You wouldn't believe how often the grease has been reused (!!!). To avoid these problems, try to use new mica insulators and fresh thermal compound, every single time! There are times on the bench when you just can't believe your eyes.
-Chris
Well, maybe that.
Often it's caused by small particles in the grease that punch through the mica. I've found almost everything you can think of between the transistor and mica (including no mica!). Flattened solder balls are pretty common. Dirt, cut wire and insulation are common. Fingernails are not unheard of. I've even seen common clear plastic wrap used as an insulator (sheesh!).
I always wonder what the devil is going on with some service people! You wouldn't believe how often the grease has been reused (!!!). To avoid these problems, try to use new mica insulators and fresh thermal compound, every single time! There are times on the bench when you just can't believe your eyes.
-Chris
"The arcing is caused from large amounts of back emf off the
speaker at high power output. "
I have found this on amps that have never been repaired before. Once I found a mica with a moon size crater burned in it.
It was first brought to my attention back in the late 70s, manly with the old Phase Linears, by the techs at Van Audio. Back then Van Audio was theee place to go for pro repairs.
I have also found all sorts of foreign material sandwiched between transistors and insulators and in the mounting holes. Gulling from the mounting screws and nuts is a big one. That's why it soooo important to removed all the output and clean up.
I got into the habit of replacing the micas even if they looked good.
David.
speaker at high power output. "
I have found this on amps that have never been repaired before. Once I found a mica with a moon size crater burned in it.
It was first brought to my attention back in the late 70s, manly with the old Phase Linears, by the techs at Van Audio. Back then Van Audio was theee place to go for pro repairs.
I have also found all sorts of foreign material sandwiched between transistors and insulators and in the mounting holes. Gulling from the mounting screws and nuts is a big one. That's why it soooo important to removed all the output and clean up.
I got into the habit of replacing the micas even if they looked good.
David.
I,m looking at the data sheet for Q6 2sa1270. According to the data sheet I'm looking at this transistor has Vce -30V. A rough calculation suggests there is a lot more than that across the CE of this transistor. Is this transistor supposed to operating in breakdown as a zener diode or did I mess up my calculations?
David.
David.
Hi David,
I'm still skeptical about a high voltage punch-through. Outputs will break down E-B once you go beyond about 7 VDC, therefore, how in the world would you generate these high voltages? That, and most higher powered amps do have flyback diodes to discharge any increased load voltages back into the power supply.
Sorry, I just can't accept the high voltage punch-through story. I completely do accept mechanical damage to an insulator that would lead to some arcing. That would be a pure DC supply to grounded heat sink thing that doesn't cause a failure in the channel affected. Having said that, I have seen the plastic TO-3 cases fail where a jet of superheated gas has cut through the transistor case and anything around it. That includes thick steel. On one Carver that was connected to 440 VAC, the TO-3 metal cased MJ15023 transistor suffered a gas jet burn through the top of the case (actually a couple did), and also two burn holes through the lower panel that are spaced about 3/8" or so from the top of the transistor. Colour me impressed! The insulators in all these instances did not fail in any way. A few did get burned badly on the top surface though.
-Chris
Me too. It's not worth a rework if you need to go back in.
I'm still skeptical about a high voltage punch-through. Outputs will break down E-B once you go beyond about 7 VDC, therefore, how in the world would you generate these high voltages? That, and most higher powered amps do have flyback diodes to discharge any increased load voltages back into the power supply.
Sorry, I just can't accept the high voltage punch-through story. I completely do accept mechanical damage to an insulator that would lead to some arcing. That would be a pure DC supply to grounded heat sink thing that doesn't cause a failure in the channel affected. Having said that, I have seen the plastic TO-3 cases fail where a jet of superheated gas has cut through the transistor case and anything around it. That includes thick steel. On one Carver that was connected to 440 VAC, the TO-3 metal cased MJ15023 transistor suffered a gas jet burn through the top of the case (actually a couple did), and also two burn holes through the lower panel that are spaced about 3/8" or so from the top of the transistor. Colour me impressed! The insulators in all these instances did not fail in any way. A few did get burned badly on the top surface though.
Some factories are filthy places. I have as well.
Had to be from the hot gases leaving the affected part. It's impressive though, isn't it?
-Chris
Hi David,
I'm wondering if that part number is incorrect. A 2SA1370 would make sense, because there will be 2X the supply voltage less about 4 Vbe drops as a rough figure. That part is quite common and has a 200 VDC E-C breakdown rating. A 2SA1270 wouldn't last very many psec in that circuit! Maybe a nsec or two.
-Chris
I'm wondering if that part number is incorrect. A 2SA1370 would make sense, because there will be 2X the supply voltage less about 4 Vbe drops as a rough figure. That part is quite common and has a 200 VDC E-C breakdown rating. A 2SA1270 wouldn't last very many psec in that circuit! Maybe a nsec or two.
-Chris
I hope it's not 2sa1270. I have a pdf copy of the service manual and that is the number printed on the schematic and no it won't last very long.
I just looked at the parts list and it's listed as 2sa1207 which makes a lot more sense.
The manual has a few errors.
David.
The only time I ever saw a hole in a TO-3 that I didn't put in it was in the cover because someone had dropped a handful of drywall screws into the front of a Carver PM1.5
Hi Andrew,
Now, that's cool!
So ..... how did the drywall screws fair?
Carver amps have such silly high energy levels that most people don't really understand what they are all about. Once you start seeing holes burned through TO-3 cases, you begin to get it.
Hi David,
That's a legitimate concern, and one that an experienced tech shouldn't be tricked by. Not unless you get the amp missing parts and you aren't use to them. Another reason why you should fix the problem - then start with anything else you are thinking of doing. Is that a TO-126 part? I haven't looked at the diagram for this question.
-Chris
Now, that's cool!
So ..... how did the drywall screws fair?
Carver amps have such silly high energy levels that most people don't really understand what they are all about. Once you start seeing holes burned through TO-3 cases, you begin to get it.
Hi David,
That's a legitimate concern, and one that an experienced tech shouldn't be tricked by. Not unless you get the amp missing parts and you aren't use to them. Another reason why you should fix the problem - then start with anything else you are thinking of doing. Is that a TO-126 part? I haven't looked at the diagram for this question.
-Chris
The schematic posted here has 2 known errors. The 33.2K on the diff pair should be 33.2, and the 2SA1270 should be a 2SA1207.
Hi Andrew,
Now, that's cool!
So ..... how did the drywall screws fair?
Carver amps have such silly high energy levels that most people don't really understand what they are all about. Once you start seeing holes burned through TO-3 cases, you begin to get it.
Hi David,
That's a legitimate concern, and one that an experienced tech shouldn't be tricked by. Not unless you get the amp missing parts and you aren't use to them. Another reason why you should fix the problem - then start with anything else you are thinking of doing. Is that a TO-126 part? I haven't looked at the diagram for this question.
-Chris
No it's a TO-92
David.
I found a PDF copy of the adcom GAF-555 service manual.
I thought it might be useful for Fred because it has the parts placement and correct orientation, and a parts list that appears to be correct.
David
It won't download because it's too big in size.
I thought it might be useful for Fred because it has the parts placement and correct orientation, and a parts list that appears to be correct.
David
It won't download because it's too big in size.
Last edited:
I simulated it, pulled the diff pair, bias turned down, no output load, 1 K resistor VAS-C to pos rail. ~80V output, no oscillation. Supply is 81 V should see a few Vbe drops, so it looks fine.
Fred has full documentation from Adcom, he has I think 3 revisions of the 555 service manual.
Fred has full documentation from Adcom, he has I think 3 revisions of the 555 service manual.
Hi Pete,
I did the same in pspice with the same finding.
Fred did say the 1 k resistor was beginning to smoke when he tried this for real, but in simulation, with or without T1 closed, there isn't enough power in the 1k resistor to warm it up. So what is drawing so much current through the 1k resistor?
David.
I did the same in pspice with the same finding.
Fred did say the 1 k resistor was beginning to smoke when he tried this for real, but in simulation, with or without T1 closed, there isn't enough power in the 1k resistor to warm it up. So what is drawing so much current through the 1k resistor?
David.