Transistor sub help needed

Geek · 30th September 2007, 07:18 AM

Hi folks,

I have a gorgeous CM Labs 35D in for repair.

The transistors are all made by RCA, but are house marked for CM labs and their numbers are of course, like CM1 - CM12 and obvious stupidity like that.

Pic attached of output transistor.

I've done the usual site search and crossref, ST, NTE, blah, blah...

Anybody have an idear?

Thanks!

wayne chou · 2nd October 2007, 10:01 PM

Hi Peter,

Don't know if you got this reply on the Bozak forum, but now that I see the pic, the CM 10 on the multi component board in the front is a high voltage medium power germanium transistor (at least Vce 120v), which you will probably not find today. But a silicon pnp sub will do - the only thing is that the bias current will not be right if you replace it with a silicon device and you would have to make up the difference of about 0.3 volts in the bias circuit.

Before we get into any modifications, how do you know that it is at fault? Have you measured whether it is open or shorted? Simple test - take a non-digital ohmmeter on the lowest scale, and with the pins facing you vertically place one of the leads on the base (lower one) and the other to either the other pin (emitter) or to the case (collector). Repeat the readings with the leads on the meter reversed. There should be a big difference on the readings with the meter leads one way vs the reading the reversed way. If so, the transistor is probably still good.

Just finished repair of a 35D for Ed Coleman. But I just sent it back today. If I had it opened up, I could tell you what you could use. In any case, before you are done, you should replace all the small electrolytics, because as they get leaky and old, they don't function as they should, and some of the original MEC black plastics with epoxy ends were not very good to begin with, as we found out later. The Seimens were a lot better. And the input caps should be replaced with lower leakage tantalums.

If the relay doesn't click in, the relay coil may be burnt up from the two 3.3K 2w resistors underneath that usually gives a burned area on the board below. This is because at about 3 watts into the carbon resistors, they generally overheat and with a negative coefficient resistance past a certain temperature, they even reduce in value tending to overheat the coil, burning it up.

Also, it could be one or more of the outputs could be shorted, which prevents it from powering it up. Simple check with an ohmmeter to see if emitter to collector are low ohms, and that each output transistor is okay by checking diode polarity from base to emitter and base to collector for each, as before with the CM10. Connector is provided at each output transistor to help with this simple test. If all is good there, then it should power up at least 80 volts or so, so then the relay should kick in.

If it still stays pretty low, there is a short somewhere. Or maybe the big 250 ohm, 20w resistor gave up. Before you even apply power again, check whether the relay coil is still alive. If so I would immediately remove the two carbon 3.3k 2 watt resistors (if you measure less than 1.5k, it's in trouble). Sub with a single 1.8k, 5 watt cement wire wound from Mouser or Digikey. If the coil is dead, you will want to get another 24 v relay with approx 600 ohm coil. I assume you want to keep the protection circuit alive and functioning.

Get back to me to see what you have found. I can try to talk you through the rest of it, when I know more.

Wayne Chou

Geek · 2nd October 2007, 11:57 PM

Hi Wayne,

I'm Gregg, fixing this for Peter

I'm interested because I found this (see pic) in there too, which has already been replaced and may have caused the trouble in the first place.

I haven't had time to poke around much more or test each part individually.

Thanks for stopping by, it's greatly appreciated!

wayne chou · 3rd October 2007, 09:10 PM

Hi Gregg,

Don't recognize this one - on the heat sink, right? Subs for anything on the heat sink one could use 2N3773's. But if they have been cooked, it might have affected the drivers before them, as well as the transistor on the same heat sink.

I would do the simple ohmmeter test, as stated previously, on the driver transistors as well, and compare from one channel to the other, before unsoldering everything, even if they are being loaded by other circuit elements. If they don't compare favorably, then unsolder them to check more thoroughly.

Note previous post and check relay, replace two watt resistors underneath, and all small electrolytics, as well as tantalums for input.

Keep me posted,

Wayne

wayne chou · 4th October 2007, 12:31 AM

Don't know if you got the other message.

In the beginning I thought it looked like the TO-3 on the circuit board. But now that I see it is the output transistor on the heat sink. Scratch the comments about the PNP germanium (CM11), though we might have to address that issue if it had been damaged.

Wayne Chou

Geek · 8th October 2007, 10:43 PM

Hi Wayne,

Thanks for the info!

Got time to look at this today and am a little confused...

Q18 is toast. Open on every pin (that's the no-name sub). Q19 is OK, BUT it tests as a PNP than an NPN the schematic shows

Now going to poke around the driver board...

** edit **

Q16 toast. B-E shorted, B-C is a thermistor.

Cheers!

wayne chou · 9th October 2007, 12:00 AM

Okay,

Gregg - As far as I know there are no CM numbers exceeding CM13, which is a TO-3 PNP germanium device on the board. The other numbers are news to me. I see you can get around the simple test stuff, but I can't follow the devices you mention. Please elaborate.

Wayne

wayne chou · 9th October 2007, 12:00 AM

PS>

If it is on the schematic, please post a pic of that, because I can't relate to what you are saying without it. Thanks, W

Geek · 9th October 2007, 12:55 AM

Hi,

Not CM - Q as in part number on the schematic.

See attached.

Cheers!

wayne chou · 9th October 2007, 02:18 AM

Oh okay,

The CM 10 is an output transistor on the heat sink, and a good equivalent is a 2N3773. I think in those days, we pulled special voltages from the line before it became the 2N3773. The other driver is on the board, and I don't recall if it is a TO-66 package. If so, I was just recommending a more robust sub to someone else - a 2N4233 which has higher breakdown voltages, and better speed. It wasn't available in those days.

I don't think you would have to go back further than that as far as the damage is concerned. Try that and I think you should be up and running, but don't forget the cap and resistor upgrades. Bias at the positive end of the output cap should be about 47 volts or so - adjusting for symmetrical clipping. Compare with the good side. Keep me posted.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
transistor substitution help needed! Thanks	crippledchicken	Parts	18	4th July 2008 12:08 PM
ZTX653 transistor or sub needed	thereminhead85	Parts	4	26th April 2006 06:01 PM
Transistor specialist needed	portlandlay	Pass Labs	18	28th November 2005 10:51 PM
low noise transistor needed	Ricren	Parts	15	16th November 2004 12:40 PM
Transistor pin-layout needed	Jennice	Parts	2	1st November 2003 09:19 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

2nd October 2007, 10:01 PM	#2
wayne chou diyAudio Member Join Date: Oct 2007	Hi Peter, Don't know if you got this reply on the Bozak forum, but now that I see the pic, the CM 10 on the multi component board in the front is a high voltage medium power germanium transistor (at least Vce 120v), which you will probably not find today. But a silicon pnp sub will do - the only thing is that the bias current will not be right if you replace it with a silicon device and you would have to make up the difference of about 0.3 volts in the bias circuit. Before we get into any modifications, how do you know that it is at fault? Have you measured whether it is open or shorted? Simple test - take a non-digital ohmmeter on the lowest scale, and with the pins facing you vertically place one of the leads on the base (lower one) and the other to either the other pin (emitter) or to the case (collector). Repeat the readings with the leads on the meter reversed. There should be a big difference on the readings with the meter leads one way vs the reading the reversed way. If so, the transistor is probably still good. Just finished repair of a 35D for Ed Coleman. But I just sent it back today. If I had it opened up, I could tell you what you could use. In any case, before you are done, you should replace all the small electrolytics, because as they get leaky and old, they don't function as they should, and some of the original MEC black plastics with epoxy ends were not very good to begin with, as we found out later. The Seimens were a lot better. And the input caps should be replaced with lower leakage tantalums. If the relay doesn't click in, the relay coil may be burnt up from the two 3.3K 2w resistors underneath that usually gives a burned area on the board below. This is because at about 3 watts into the carbon resistors, they generally overheat and with a negative coefficient resistance past a certain temperature, they even reduce in value tending to overheat the coil, burning it up. Also, it could be one or more of the outputs could be shorted, which prevents it from powering it up. Simple check with an ohmmeter to see if emitter to collector are low ohms, and that each output transistor is okay by checking diode polarity from base to emitter and base to collector for each, as before with the CM10. Connector is provided at each output transistor to help with this simple test. If all is good there, then it should power up at least 80 volts or so, so then the relay should kick in. If it still stays pretty low, there is a short somewhere. Or maybe the big 250 ohm, 20w resistor gave up. Before you even apply power again, check whether the relay coil is still alive. If so I would immediately remove the two carbon 3.3k 2 watt resistors (if you measure less than 1.5k, it's in trouble). Sub with a single 1.8k, 5 watt cement wire wound from Mouser or Digikey. If the coil is dead, you will want to get another 24 v relay with approx 600 ohm coil. I assume you want to keep the protection circuit alive and functioning. Get back to me to see what you have found. I can try to talk you through the rest of it, when I know more. Wayne Chou

3rd October 2007, 09:10 PM	#4
wayne chou diyAudio Member Join Date: Oct 2007	Hi Gregg, Don't recognize this one - on the heat sink, right? Subs for anything on the heat sink one could use 2N3773's. But if they have been cooked, it might have affected the drivers before them, as well as the transistor on the same heat sink. I would do the simple ohmmeter test, as stated previously, on the driver transistors as well, and compare from one channel to the other, before unsoldering everything, even if they are being loaded by other circuit elements. If they don't compare favorably, then unsolder them to check more thoroughly. Note previous post and check relay, replace two watt resistors underneath, and all small electrolytics, as well as tantalums for input. Keep me posted, Wayne

4th October 2007, 12:31 AM	#5
wayne chou diyAudio Member Join Date: Oct 2007	Don't know if you got the other message. In the beginning I thought it looked like the TO-3 on the circuit board. But now that I see it is the output transistor on the heat sink. Scratch the comments about the PNP germanium (CM11), though we might have to address that issue if it had been damaged. Wayne Chou

8th October 2007, 10:43 PM	#6
Geek diyAudio Member Join Date: Sep 2004	Hi Wayne, Thanks for the info! Got time to look at this today and am a little confused... Q18 is toast. Open on every pin (that's the no-name sub). Q19 is OK, BUT it tests as a PNP than an NPN the schematic shows Now going to poke around the driver board... edit Q16 toast. B-E shorted, B-C is a thermistor. Cheers!

9th October 2007, 12:00 AM	#7
wayne chou diyAudio Member Join Date: Oct 2007	Okay, Gregg - As far as I know there are no CM numbers exceeding CM13, which is a TO-3 PNP germanium device on the board. The other numbers are news to me. I see you can get around the simple test stuff, but I can't follow the devices you mention. Please elaborate. Wayne

9th October 2007, 02:18 AM	#10
wayne chou diyAudio Member Join Date: Oct 2007	Oh okay, The CM 10 is an output transistor on the heat sink, and a good equivalent is a 2N3773. I think in those days, we pulled special voltages from the line before it became the 2N3773. The other driver is on the board, and I don't recall if it is a TO-66 package. If so, I was just recommending a more robust sub to someone else - a 2N4233 which has higher breakdown voltages, and better speed. It wasn't available in those days. I don't think you would have to go back further than that as far as the damage is concerned. Try that and I think you should be up and running, but don't forget the cap and resistor upgrades. Bias at the positive end of the output cap should be about 47 volts or so - adjusting for symmetrical clipping. Compare with the good side. Keep me posted.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)