I have a Carver preamp/tuner with about 250mV of DC riding on the output of one channel, and is causing thumps when switching sources. It uses a Toshiba 2SC2878 transistor on each channel to mute the output, and I seem to have narrowed down the problem to one of the transistors.
For a situation where low C-E sat is necessary and high gain, I usually pop in a Zetex ZTX1056A, and I did so here. However, after subbing the Zetex trannie for the Toshiba, there is still voltage on the output...less, but still there. I pulled out the Zetex transistor, and moved the Toshiba transistors from L to R channels and soldered 'em back in...the DC followed the transistor, so I know that's the problem.
Why can't I just swap out the C2878 with any decent high-gain transistor? I found the data sheet pretty easy, and unless I'm missing something there doesn't seem to be anything real special about it...
For a situation where low C-E sat is necessary and high gain, I usually pop in a Zetex ZTX1056A, and I did so here. However, after subbing the Zetex trannie for the Toshiba, there is still voltage on the output...less, but still there. I pulled out the Zetex transistor, and moved the Toshiba transistors from L to R channels and soldered 'em back in...the DC followed the transistor, so I know that's the problem.
Why can't I just swap out the C2878 with any decent high-gain transistor? I found the data sheet pretty easy, and unless I'm missing something there doesn't seem to be anything real special about it...
Attachments
DC follows the transistor when swapped from one channel to the next...hence the transistor is the source of the DC.
There is an IC which buffers the output that is just before the circuit that you see. I can remove the IC so that the output leads are only connected to the transistors, and the DC is there.
Two ways of confirming the same thing...that the source of the DC is the transistor. So what's so special about the C2878 that I can't replace it with about any decent transistor?
There is an IC which buffers the output that is just before the circuit that you see. I can remove the IC so that the output leads are only connected to the transistors, and the DC is there.
Two ways of confirming the same thing...that the source of the DC is the transistor. So what's so special about the C2878 that I can't replace it with about any decent transistor?
You are correct...but I must spend at least $25 to place an order, even if it's a $0.50 part.
I just pulled out both transistors...no DC at output. Also, I checked the transisors with the diode tester. Odd readings on both.
The good transistor reads .72V from B to C and B to E, but my meter 'beeps' for a proper diode drop, and I get no 'beep' with these transistors, so this usually means some resistive element in there somewhere.
The bad transistor also measures the same, but I also measure a diode drop when reverse biasing the E to C diode...about .58V, and I get the 'beep' too 😉.
I just pulled out both transistors...no DC at output. Also, I checked the transisors with the diode tester. Odd readings on both.
The good transistor reads .72V from B to C and B to E, but my meter 'beeps' for a proper diode drop, and I get no 'beep' with these transistors, so this usually means some resistive element in there somewhere.
The bad transistor also measures the same, but I also measure a diode drop when reverse biasing the E to C diode...about .58V, and I get the 'beep' too 😉.
Try cleaning all the residual solder from the area and inserting 2 substitute Zetex (which you usually rely on) and see what eventuates - whether you get some DC on both channels or NONE! The cct shouldn't be that sensitive.
Probably the special thing about the 2SC2878 is the high reverse beta.
That way when it saturates hard during muting, it doesn't cause a lot of base current to end up on the output.
That way when it saturates hard during muting, it doesn't cause a lot of base current to end up on the output.
That must be it, I suppose.
When I put in a pair of Zetex transistors, I had offset on both channels... -25mV on one channel, -55mV on the other. With the C2878 transistor in place, offset is 0.1mV.
Looks like I need to order more parts. Again...a case of needing one transistor that is gonna cost $25. 🙁
When I put in a pair of Zetex transistors, I had offset on both channels... -25mV on one channel, -55mV on the other. With the C2878 transistor in place, offset is 0.1mV.
Looks like I need to order more parts. Again...a case of needing one transistor that is gonna cost $25. 🙁
Well, the logic is wrong for a FET, but that's what I'm more used to seeing in a situation like this. A bipolar is odd in this configuration to me.
It still seems that the 2SC2878 has something very odd going on with it. Was hoping someone cuuld put their finger on it...
It still seems that the 2SC2878 has something very odd going on with it. Was hoping someone cuuld put their finger on it...
Hi EchoWars,
The 2SC2878 is designed for high reverse E-B bias. That it why you see it used for muting everywhere. As you discovered, most other transistors leak when installed in that circuit. I think there are an couple other numbers that can be used (don't use a sub book) but I can't remember what they are.
When I had my shop I bought them by the 50's quantity. Changed a lot in many products including CD players. Being warranty for Carver meant we changed quite a few in those too.
-Chris
The 2SC2878 is designed for high reverse E-B bias. That it why you see it used for muting everywhere. As you discovered, most other transistors leak when installed in that circuit. I think there are an couple other numbers that can be used (don't use a sub book) but I can't remember what they are.
When I had my shop I bought them by the 50's quantity. Changed a lot in many products including CD players. Being warranty for Carver meant we changed quite a few in those too.
-Chris
anatech,
Why do you suppose they fail? They are not stressed very much in that muting application!
-- mirlo
Why do you suppose they fail? They are not stressed very much in that muting application!
-- mirlo
Appreciated. I thought it was sooo wierd that I could not just put whatever I wanted in there. Pretty frustrating.anatech said:Hi EchoWars,
The 2SC2878 is designed for high reverse E-B bias. That it why you see it used for muting everywhere. As you discovered, most other transistors leak when installed in that circuit. I think there are an couple other numbers that can be used (don't use a sub book) but I can't remember what they are.
When I had my shop I bought them by the 50's quantity. Changed a lot in many products including CD players. Being warranty for Carver meant we changed quite a few in those too.
-Chris
I'd guess it just comes under the vast heading of **** happens.mirlo said:anatech,
Why do you suppose they fail? They are not stressed very much in that muting application!
-- mirlo
Nothing lasts forever.
anatech said:Hi EchoWars,
The 2SC2878 is designed for high reverse E-B bias. That it why you see it used for muting everywhere. As you discovered, most other transistors leak when installed in that circuit. I think there are an couple other numbers that can be used (don't use a sub book) but I can't remember what they are.
When I had my shop I bought them by the 50's quantity. Changed a lot in many products including CD players. Being warranty for Carver meant we changed quite a few in those too.
-Chris
Echo sorry for digressing:
Chris, I have the same issue in my old Sony pre TA-1000ESD, pops when switching anything and about 1200mV on the outputs!
Any quick fixes?
K-
No problem...perhaps it also uses something like these muting transistors....Echo sorry for digressing:
Hi EchoWars,
In a muting application there needs to be a high reverse voltage to prevent the audio from breaking through on peaks. A normal E-B reverse zener breakdown is on the order of 6~7 VDC. It's not uncommon to see a reverse bias of 10~15 VDC. They will break eventually. I like a relay short to ground for muting, use a signal relay.
That's the one thing I liked about the early Philips CD players.
-Chris
In a muting application there needs to be a high reverse voltage to prevent the audio from breaking through on peaks. A normal E-B reverse zener breakdown is on the order of 6~7 VDC. It's not uncommon to see a reverse bias of 10~15 VDC. They will break eventually. I like a relay short to ground for muting, use a signal relay.
That's the one thing I liked about the early Philips CD players.
-Chris
Hi K-amps,
Can you post or email a schematic of your preamp for me to look at? I'd like to see what's going on there now that you've peaked my curiousity.
I love troubleshooting stuff, sicko that I am. 😉
-Chris
Can you post or email a schematic of your preamp for me to look at? I'd like to see what's going on there now that you've peaked my curiousity.
I love troubleshooting stuff, sicko that I am. 😉
-Chris
Understood. But on this Carver, the DC that I get on the output shows up immediately after replacing the transistor with the Zetex ones, before any signal is applied. So while what you say is true, there is another mechanism at work here.In a muting application there needs to be a high reverse voltage to prevent the audio from breaking through on peaks. A normal E-B reverse zener breakdown is on the order of 6~7 VDC. It's not uncommon to see a reverse bias of 10~15 VDC.
- Status
- Not open for further replies.
- Home
- Amplifiers
- Solid State
- Carver Preamp and the Toshiba 2SC2878 transistor