I’m slowly working on a Carver TFM-25 with a blown right channel, working through the schematic and trying to educate myself on the 3-rail system, and have a few questions. Repairing this amp will be mostly an educational experience, and only used for a bedroom stereo. Sonic perfection would be nice to have, but it’s not fully required.
I have built amplifiers off and on since the days of the SWTP Plastic Tiger and Tiger .01, even recently repaired a couple of Adcom 555’s, but I lack the component level skills of knowing what is a good or bad transistor substitution.
I have found one of 2SA1302’s to be open, but due to the number of counterfeits on the market, I would rather replace the PNP/NPN pair with the Hitachi 2SA1943/2SC5200 from Digikey instead. Will this work okay? Should I just get a few of each, match them, and replace them all?
My father bought this amp brand new in 1991 and it worked well until a few of years ago, when he lost the right channel. A thorough cleaning and visual inspection shows the amp in excellent shape, no brown areas on the PCB, no obvious blown or burnt components, and only a few suspect solder joints, which I touched up. The funny thing, is the left heatsink has 4-2SA1302, 1-2SA1301, 4-2SC3281, 1-2SC3280, the right heatsink has 5-2SA1302, 4-2SC3281 and 1-2SC3280. Per the schematic, the left side should be the same as the right side (the correct layout) so it looks like a factory error. From the specs, and to my untrained eye, it looks like the 2SC3281 is just a more robust version of the 2SC3280, but I’m not sure of what the sonic difference should be, or if any of the resistors should be changed for the difference in current between the 2. My question is, can I replace the 2SC1301/2SA3280 pair with the Hitachi 2SA1943/2SC5200 pair? Do I need to changes any resistors? Try to dig up matched 2SA2SC1301/2SA3280 pairs? Try a better substitution?
Last question, I usually make it a habit to change the drivers when I change any outputs, because they are cheap, and it’s cheap insurance. This amp uses 4 pairs of 2SA1370/2SC3467 as drivers and pre-drivers. Digikey doesn’t stock these, but places like B&D, Newark, MCM, etc do. I’m not sure which distributor to trust, or if I should try a possible better substitution from another reliable source.
Sorry for the long winded post, and thanks for your time!
Win
I have built amplifiers off and on since the days of the SWTP Plastic Tiger and Tiger .01, even recently repaired a couple of Adcom 555’s, but I lack the component level skills of knowing what is a good or bad transistor substitution.
I have found one of 2SA1302’s to be open, but due to the number of counterfeits on the market, I would rather replace the PNP/NPN pair with the Hitachi 2SA1943/2SC5200 from Digikey instead. Will this work okay? Should I just get a few of each, match them, and replace them all?
My father bought this amp brand new in 1991 and it worked well until a few of years ago, when he lost the right channel. A thorough cleaning and visual inspection shows the amp in excellent shape, no brown areas on the PCB, no obvious blown or burnt components, and only a few suspect solder joints, which I touched up. The funny thing, is the left heatsink has 4-2SA1302, 1-2SA1301, 4-2SC3281, 1-2SC3280, the right heatsink has 5-2SA1302, 4-2SC3281 and 1-2SC3280. Per the schematic, the left side should be the same as the right side (the correct layout) so it looks like a factory error. From the specs, and to my untrained eye, it looks like the 2SC3281 is just a more robust version of the 2SC3280, but I’m not sure of what the sonic difference should be, or if any of the resistors should be changed for the difference in current between the 2. My question is, can I replace the 2SC1301/2SA3280 pair with the Hitachi 2SA1943/2SC5200 pair? Do I need to changes any resistors? Try to dig up matched 2SA2SC1301/2SA3280 pairs? Try a better substitution?
Last question, I usually make it a habit to change the drivers when I change any outputs, because they are cheap, and it’s cheap insurance. This amp uses 4 pairs of 2SA1370/2SC3467 as drivers and pre-drivers. Digikey doesn’t stock these, but places like B&D, Newark, MCM, etc do. I’m not sure which distributor to trust, or if I should try a possible better substitution from another reliable source.
Sorry for the long winded post, and thanks for your time!
Win
I've used the SA1943/SC5200 (Toshiba) in dozens of Carver repairs and have never had any problems, some people have even noted an improvement with the new transistors (but I have not heard any). I've repaired many Carvers where different transistors were used on the outputs/commutators (I'm talking about the TFM 25/35/45, M1.0t here), but MOST of the time the same transistors are used in these areas. So your case of different transistors is not uncommon. If you use the SA1943 in one side, make sure its compliment is the SC5200 - not the older transistors.
Hi Win,
toddyaudio is correct here. You can use the MJL type parts from On Semi also. They are excellent.
Now, check the areas to the rear just to the left and right of center line. There are some TO-92M type transistors that run hot. The board may be a bit brown here. I have seen those transistors go open or leaky, so just change them.
I believe there was a modification to reduce the standing current in Q202 and 204 (repeat for the other channel). Look for over heated resistors and overheated capacitors as well (small ones in that area). I think it's important to find out what the mods were and execute them. If memory serves, there was also another mod for the bias transistor near the outputs (there are two per channel, one adjustable and the other fixed).
-Chris
toddyaudio is correct here. You can use the MJL type parts from On Semi also. They are excellent.
Now, check the areas to the rear just to the left and right of center line. There are some TO-92M type transistors that run hot. The board may be a bit brown here. I have seen those transistors go open or leaky, so just change them.
I believe there was a modification to reduce the standing current in Q202 and 204 (repeat for the other channel). Look for over heated resistors and overheated capacitors as well (small ones in that area). I think it's important to find out what the mods were and execute them. If memory serves, there was also another mod for the bias transistor near the outputs (there are two per channel, one adjustable and the other fixed).
-Chris
Thanks for the responses guys!
Toddyaudio,
I'll get a few pairs of 1943/5200 to replace the blown outputs, and currently unmatched 1302/1380 pair.
Chris,
Yes, I have a copy of Service Bulletin #TFM-25-2, dated 4/15/92, which details the modifications, "To prevent components from discoloring the PCB and to improve circuit reliabilty."
It shows to replace the 4 pairs of 2SA1370/2SC3467 predrivers/drivers spaced 3/8" to 1/2" above the PCB, and replaces a few resistors with higher values. It also replaces a few resistors with the same value, but in 'Flame Proof'.
Two of the 1370/3467 pairs are the TO-92M, Q202, Q204 you're thinking about. After all my research I expected to see the PCB very brown in that area, but it only has the slightest discoloration. I'll do a further check on the other components in those areas. This amp really is in much nicer shape than I expected, but I'll still replace them and do all the mods.
I don't have a copy of the bias transistor mod, so I'll need to research that a little further.
I guess reading all the counterfeit posts, buying components has me a bit gun shy, besides places like Digikey, Newark, etc, where's a good, reliable source for the 1370/3467 pairs and the 1/4W flame proof resistors?
Thanks again for the help, it's greatly appreciated!
Toddyaudio,
I'll get a few pairs of 1943/5200 to replace the blown outputs, and currently unmatched 1302/1380 pair.
Chris,
Yes, I have a copy of Service Bulletin #TFM-25-2, dated 4/15/92, which details the modifications, "To prevent components from discoloring the PCB and to improve circuit reliabilty."
It shows to replace the 4 pairs of 2SA1370/2SC3467 predrivers/drivers spaced 3/8" to 1/2" above the PCB, and replaces a few resistors with higher values. It also replaces a few resistors with the same value, but in 'Flame Proof'.
Two of the 1370/3467 pairs are the TO-92M, Q202, Q204 you're thinking about. After all my research I expected to see the PCB very brown in that area, but it only has the slightest discoloration. I'll do a further check on the other components in those areas. This amp really is in much nicer shape than I expected, but I'll still replace them and do all the mods.
I don't have a copy of the bias transistor mod, so I'll need to research that a little further.
I guess reading all the counterfeit posts, buying components has me a bit gun shy, besides places like Digikey, Newark, etc, where's a good, reliable source for the 1370/3467 pairs and the 1/4W flame proof resistors?
Thanks again for the help, it's greatly appreciated!
Hi inttech1,
Nice! The values are not on the schematic. Priceless. I don't have the parts list for that one either.
Measure both of them. They act as a fuse between the 30 VDC rails and 15 VDC regulators. I don't know if they are also using them to split the heat power between the actual regulator pass element or not.
It is possible those resistors are simply overheated. They may function fine, you'll have to measure them to be sure.
-Chris
Nice! The values are not on the schematic. Priceless. I don't have the parts list for that one either.
Measure both of them. They act as a fuse between the 30 VDC rails and 15 VDC regulators. I don't know if they are also using them to split the heat power between the actual regulator pass element or not.
It is possible those resistors are simply overheated. They may function fine, you'll have to measure them to be sure.
-Chris
Hi Chris,
Thanks for your reply. Amongst other issues with this amp, including the need for the factory Service Bulletin upgrade, and blown outputs on one channel, this was the last thing I was trying to get resolved.
I believe the resistors should be identical for both positions, but when you remove them they measure very differently.
From what I can make out, the bands on both are the same from what I can make out.
I really don't know which one is closer to the actual.
One measures about 22 ohms and the other is over 43.
Perhaps someone owns one and could tell me the next time they remove the cover for cleaning, what the band colors are....
I really appreciate your input!
Ron
Thanks for your reply. Amongst other issues with this amp, including the need for the factory Service Bulletin upgrade, and blown outputs on one channel, this was the last thing I was trying to get resolved.
I believe the resistors should be identical for both positions, but when you remove them they measure very differently.
From what I can make out, the bands on both are the same from what I can make out.
I really don't know which one is closer to the actual.
One measures about 22 ohms and the other is over 43.
Perhaps someone owns one and could tell me the next time they remove the cover for cleaning, what the band colors are....
I really appreciate your input!
Ron
Hi Ron,
I believe these are different values.
There was an upgrade to allow the pre-driver transistors to run at a lower temperature. You could also upgrade them to a different case style (TO-126 or equiv.). The changes where some resistor value increases in the emitter circuits I believe.
-Chris
I believe these are different values.
There was an upgrade to allow the pre-driver transistors to run at a lower temperature. You could also upgrade them to a different case style (TO-126 or equiv.). The changes where some resistor value increases in the emitter circuits I believe.
-Chris
) They are 56 ohm flame proof resistors.
Hi Bigred,
Right you are.
I found a note to that effect in my stuff. That's the official line at least, but this could be wrong. It wouldn't be the first time! On other Carver amplifiers, these are not the same value and reflect the different current draws on these supplies.
Now, the different values Ron is measuring would suggest that they are indeed different. Some one should really look inside a working amp for the proper answer
-Chris
Right you are.
I found a note to that effect in my stuff. That's the official line at least, but this could be wrong. It wouldn't be the first time! On other Carver amplifiers, these are not the same value and reflect the different current draws on these supplies.
Now, the different values Ron is measuring would suggest that they are indeed different. Some one should really look inside a working amp for the proper answer
-Chris
Well considring the circuit is a mirror image with one side -15 and the other +15 supply, I'm certain they are the same. The parts list shows them as being the same part number as well. However it certainly wouldn't hurt for someone to confirm with an actual unit because lords knows I have come across many an error in service manuals over the years.
Hi Bigred,
My main concern here are the obvious differing values that Ron is measuring. That sets off alarm bells in my head.
-Chris
That may very well be, however Bob Carver has been known to compensate for the differing current draw from each 15 V supply. This has been historically his normal way of doing things. Just look at an M1.5t for example.
I can confirm that.
No shortage of examples there!
My main concern here are the obvious differing values that Ron is measuring. That sets off alarm bells in my head.
-Chris
Thanks Guys,
I logged on again today to see that I needed to catch up on the replies!
Even the burnt color bands on the resistors that were removed, seemed to suggest that they were identical.
But, like Chris says, it may not be the right guess. 56 ohms is even further from the 20 odd and 40 odd ohms that I have been able to measure.
If all else fails, I will install the 56 ohm types and measure the voltages and observe how things are behaving.
Maybe someone else may drop by with the "for sure" answer!
I logged on again today to see that I needed to catch up on the replies!
Even the burnt color bands on the resistors that were removed, seemed to suggest that they were identical.
But, like Chris says, it may not be the right guess. 56 ohms is even further from the 20 odd and 40 odd ohms that I have been able to measure.
If all else fails, I will install the 56 ohm types and measure the voltages and observe how things are behaving.
Maybe someone else may drop by with the "for sure" answer!
Hi Ron,
I strongly feel that your 22 ohms resistor is probably not a 56 ohm. I've never seen one drift so far down from overheating. The 43 ohm one may have dropped that low in value over time being overheated. I still am concerned that it may be anywhere in the 39 ohm to 56 ohm (on the outside).
The values may affect how these rails collapse when the amp is turned off to avoid thumps and bangs. There is a high degree of engineering that went into these products. More than most people give them credit for.
I wish I had one I could look at for you. Can you look on Google for images of the PCB? Possibly a member at http://www.carveraudio.com/index2.htm Someone may be able to look under the hood for this information.
-Chris
I strongly feel that your 22 ohms resistor is probably not a 56 ohm. I've never seen one drift so far down from overheating. The 43 ohm one may have dropped that low in value over time being overheated. I still am concerned that it may be anywhere in the 39 ohm to 56 ohm (on the outside).
The values may affect how these rails collapse when the amp is turned off to avoid thumps and bangs. There is a high degree of engineering that went into these products. More than most people give them credit for.
I wish I had one I could look at for you. Can you look on Google for images of the PCB? Possibly a member at http://www.carveraudio.com/index2.htm Someone may be able to look under the hood for this information.
-Chris
Hi Chris,
Yes, I did investigate the Carver site intensely. I was able to get a lot of information from the forums, but not all that I was looking for...eventually someone will have the answer, and at the moment I'm still waiting for the output transistors to arrive.
I will move forward with the repair based on the arrival of the outputs, as I'm sure my customer will be looking for his unit back.
I have seen resistors drift over time myself, but do agree that the drift is usually not that large. On one occasion though, I did have a 100k ohm unit change to about 10k ohms....I think I saved that one, and may still have it in the shop somewhere.
Cheers!
Yes, I did investigate the Carver site intensely. I was able to get a lot of information from the forums, but not all that I was looking for...eventually someone will have the answer, and at the moment I'm still waiting for the output transistors to arrive.
I will move forward with the repair based on the arrival of the outputs, as I'm sure my customer will be looking for his unit back.
I have seen resistors drift over time myself, but do agree that the drift is usually not that large. On one occasion though, I did have a 100k ohm unit change to about 10k ohms....I think I saved that one, and may still have it in the shop somewhere.
Cheers!
