A while back I bought a broken CS-800 in hopes of getting some serious power for my Titanic MK-III sub on a college budget. (I got it for $75 after some negotiation with a music store owner after I saw it on his floor 😀)
I got one side working with just a thorough cleaning and since I had just one sub, I simply left channel A disconnected after correcting some safety hazards on the mains wiring. (fused neutral!)
This week I decided to revisit this amp and complete repairs. But why stop at repairs...I'm going to drag it into the 21st century kicking and screaming!
The problems with the first CS800 are numerous, some are cheap and easy to fix and others are inherent in the design (quasi-complementary outputs for example).
Two nasty sins against Hi-Fi are easily rectified for less than $10. First there is a non polarized electrolytic cap in the signal path, second the input stage is an MC1741.....which is basically the same as an LM741. An OPA134 and a metalized poly film cap are nice improvements one can make on the cheap.
Since this is a budget project and I don't have a problem with hum, the soup can sized main filter caps will be unchanged for the time being. It appears that one has been replaced as it's different in appearance than the other. The electrolytics on the drive and input boards are of course being swapped out because it costs hardly anything to do so. (plus, it's good gremlin repellent!)
Assuming everything is OK, just 6 components fixes up the drive board to better than new performance.
An inside view of the chassis...spaghetti city.
The repairman claimed that he had put a lot of time and money into his repair attempt and he wasn't lying! The output stage BJTs are OK! The output BJTs are Motorola with Peavey house numbers but its really 5 pairs of MJ15024. It looks like 6 but on close inspection, the first pair is a driver using MJ15020. The driver stages on the drive board were OK as well, its obviously been repaired as it has a few things from ON Semi...which didn't exist until 1999. The output clamp triacs are also ON Semi devices too. Cool stuff for sure!
I tested all the transistors, diodes, caps and resistors on both power boards and on both drive boards....everything was in order. (Don't ask how long that took....needless to say there are no cold solder joints!)
So with all those parts good, why did I get DC on Ch. B using the Ch A drive board but proper operation when using the Ch B board? The drive boards are identical; in theory it shouldn't matter what side they came from. I think the answer is simple....the repairman used the wrong opamp in the input stage! He placed a dual opamp (NE5532) in place of a single on the ch. A drive board. The pinouts are wildly different but there's no visual indication and nothing to physically prevent the swap. Hopefully that's all there is to fixing that channel.
The zener board has a strange issue...all of the 2K ohm, 5W resistors have failed open circuit. This leaves the plug in expansion ports useless. This is an easy fix although I'd love to know what took out these resistors. As you can see, things have been a bit toasty where the resistors once sat.
More pics....
The most boring task of all...cleaning old thermal compound off.
Completed CS800 module from the top.
Completed module from below showing the drive board.
Input boards with their new caps and clean pots
Power Board, component side
Power Board, solder side....PCBs have come a LONG way! (then again, this was anything but state of the art, even for 1977!)
I got one side working with just a thorough cleaning and since I had just one sub, I simply left channel A disconnected after correcting some safety hazards on the mains wiring. (fused neutral!)
This week I decided to revisit this amp and complete repairs. But why stop at repairs...I'm going to drag it into the 21st century kicking and screaming!
The problems with the first CS800 are numerous, some are cheap and easy to fix and others are inherent in the design (quasi-complementary outputs for example).
Two nasty sins against Hi-Fi are easily rectified for less than $10. First there is a non polarized electrolytic cap in the signal path, second the input stage is an MC1741.....which is basically the same as an LM741. An OPA134 and a metalized poly film cap are nice improvements one can make on the cheap.
Since this is a budget project and I don't have a problem with hum, the soup can sized main filter caps will be unchanged for the time being. It appears that one has been replaced as it's different in appearance than the other. The electrolytics on the drive and input boards are of course being swapped out because it costs hardly anything to do so. (plus, it's good gremlin repellent!)
Assuming everything is OK, just 6 components fixes up the drive board to better than new performance.
An inside view of the chassis...spaghetti city.
The repairman claimed that he had put a lot of time and money into his repair attempt and he wasn't lying! The output stage BJTs are OK! The output BJTs are Motorola with Peavey house numbers but its really 5 pairs of MJ15024. It looks like 6 but on close inspection, the first pair is a driver using MJ15020. The driver stages on the drive board were OK as well, its obviously been repaired as it has a few things from ON Semi...which didn't exist until 1999. The output clamp triacs are also ON Semi devices too. Cool stuff for sure!
I tested all the transistors, diodes, caps and resistors on both power boards and on both drive boards....everything was in order. (Don't ask how long that took....needless to say there are no cold solder joints!)
So with all those parts good, why did I get DC on Ch. B using the Ch A drive board but proper operation when using the Ch B board? The drive boards are identical; in theory it shouldn't matter what side they came from. I think the answer is simple....the repairman used the wrong opamp in the input stage! He placed a dual opamp (NE5532) in place of a single on the ch. A drive board. The pinouts are wildly different but there's no visual indication and nothing to physically prevent the swap. Hopefully that's all there is to fixing that channel.
The zener board has a strange issue...all of the 2K ohm, 5W resistors have failed open circuit. This leaves the plug in expansion ports useless. This is an easy fix although I'd love to know what took out these resistors. As you can see, things have been a bit toasty where the resistors once sat.
More pics....
The most boring task of all...cleaning old thermal compound off.
Completed CS800 module from the top.
Completed module from below showing the drive board.
Input boards with their new caps and clean pots
Power Board, component side
Power Board, solder side....PCBs have come a LONG way! (then again, this was anything but state of the art, even for 1977!)
That was quite a challenge you took on.
Well done.
Other than the big caps the only other thing I might change is the fan.
Well done.
Other than the big caps the only other thing I might change is the fan.
The fan is going to be changed as it does not run. For home theater service, the amp may be run passively provided that the top is not installed. With no signal the heatsinks barely get warm if the amp is working right.
Beware that open heatsinks on a CS 800 can be dangerous as the transistor cases are at rail voltage and the heatsink is grounded.
A small 12V transformer and rectifier could be fit so one could use one of the many 12VDC fans made for computers. 60CFM should be more than enough for almost any application. 100CFM fans would only be needed for live performance.
Swapping an OPA134 for a 741, TL071, TL081 or other 741 pin compatible device may require a small modification. If your amp uses the offset null pins on the 741, you will have to devise a way to swap pins 5 and 8. In most cases this involves only running a jumper as one of the pins is NC, the other offset null.
The CS800 does not use the offset null pins so no rewiring was required. 😀
Beware that open heatsinks on a CS 800 can be dangerous as the transistor cases are at rail voltage and the heatsink is grounded.
A small 12V transformer and rectifier could be fit so one could use one of the many 12VDC fans made for computers. 60CFM should be more than enough for almost any application. 100CFM fans would only be needed for live performance.
Swapping an OPA134 for a 741, TL071, TL081 or other 741 pin compatible device may require a small modification. If your amp uses the offset null pins on the 741, you will have to devise a way to swap pins 5 and 8. In most cases this involves only running a jumper as one of the pins is NC, the other offset null.
The CS800 does not use the offset null pins so no rewiring was required. 😀
Last edited:
Channel B is up.....channel A would be if I had just 5 more mica insulators.
On the bright side....the S/N ratio is audibly improved. The hiss is GONE. 😀
Who's bright idea was it to put a 741 in this thing? Morons! (Then again....we are talking about 70s Peavey engineering here...big, tough and dumb!)
On the bright side....the S/N ratio is audibly improved. The hiss is GONE. 😀
Who's bright idea was it to put a 741 in this thing? Morons! (Then again....we are talking about 70s Peavey engineering here...big, tough and dumb!)
Peaveys are like diesel trucks! they just run and run and run! the Armageddon will come and there will be Cockroaches and Peaveys left! LOL!
I have rebuilt many many of these things! good job!
I have rebuilt many many of these things! good job!
Channel A is having problems!
I get output and it sounds decent but it will not give even 1% of full power before lighting the overload light. If its not shut down immediately, two resistors on the drive board burn and the overload light stays on regardless of gain. The overload light will not light at any gain setting if there is no signal.
These appear to be in series with the rail supply to the drive board. They are 4.7 ohms.
I tested EVERY semiconductor on both boards....all test good. I did a resistance check across all resistors and compared it to the working drive board on channel B, all were in tolerance. No shorts on caps either.
The outputs check OK as well and there are no apparent shorts. With no load, the magic smoke is released even faster.
What else should I check?
Schematic shown below....burned resistors circled in red
I get output and it sounds decent but it will not give even 1% of full power before lighting the overload light. If its not shut down immediately, two resistors on the drive board burn and the overload light stays on regardless of gain. The overload light will not light at any gain setting if there is no signal.
These appear to be in series with the rail supply to the drive board. They are 4.7 ohms.
I tested EVERY semiconductor on both boards....all test good. I did a resistance check across all resistors and compared it to the working drive board on channel B, all were in tolerance. No shorts on caps either.
The outputs check OK as well and there are no apparent shorts. With no load, the magic smoke is released even faster.
What else should I check?
Schematic shown below....burned resistors circled in red
There must be a short or near short if the resistors are burning out.
A first check would be the electrolytics after the resistors as these age first.
A first check would be the electrolytics after the resistors as these age first.
I know there's a short...but I cannot find it!
The electrolytics on the boards are new! They test OK.
The electrolytics on the boards are new! They test OK.
Have you put one in the wrong way around ?
Try pulling the IC to see if that has blown.
You mean the zenered supply that feeds the input stage opamp? Yes.... +16V and -15.8V Seems reasonable. (spec is +/- 16V)
Measurements are tricky...turn on, glance at meter, turn off.
Im going to go for a walk and think this crap over...be back in an hour. If anyone has any weird suggestion, Im open to it.
Im going to go for a walk and think this crap over...be back in an hour. If anyone has any weird suggestion, Im open to it.
The triple diode in the bias circuit is connected with a Molex connector, correct? Connector backwards? Peavey "usually" makes these connectors idiot proof but it's been awhile since I was in a CS800.
Craig
Craig
ANy of them backwards, those e-caps? Any chance a couple of those two pin molex cables have swapped places? For those 4.7s to burn they really pretty much need to be shorted together on their left ends.
They put a 5534 in place of the 741 after a while.
This is the CS800 A series. I'd post a clean pdf version of the schematic, but the tiny file size limits here prevent it. The file is 330kB. Customer service at Peavey will send it to anyone upon request.
They put a 5534 in place of the 741 after a while.
This is the CS800 A series. I'd post a clean pdf version of the schematic, but the tiny file size limits here prevent it. The file is 330kB. Customer service at Peavey will send it to anyone upon request.
Triple diode molex is keyed. It's impossible to reverse.
All connections and e-caps verified as correct...same as channel B which operates fine (LOTS of power and amazingly low noise for a big amp this old...I cant wait to get stereo!)
All connections and e-caps verified as correct...same as channel B which operates fine (LOTS of power and amazingly low noise for a big amp this old...I cant wait to get stereo!)
If I do that I get no early warning of impending doom.....
That little red light comes on just a couple seconds before smoke.
I need to isolate the VAS and OPS.....my theory is simple: VAS faults rarely kill the OPS, on the other hand, OPS faults often kill the VAS.
Connect the suspect drive board to a known good power board?
There is a possibility that there's a marginal transistor...one that works fine in the tester and breaks down at rail voltage. Why do I have a bad feeling its a leaky transistor?
That little red light comes on just a couple seconds before smoke.
I need to isolate the VAS and OPS.....my theory is simple: VAS faults rarely kill the OPS, on the other hand, OPS faults often kill the VAS.
Connect the suspect drive board to a known good power board?
There is a possibility that there's a marginal transistor...one that works fine in the tester and breaks down at rail voltage. Why do I have a bad feeling its a leaky transistor?
Last edited:
My trick is working!
Suspect VAS from Ch. A placed on known good OPS from Ch B..... no overload light, no smoke and a happy subwoofer. 😀
Now going for a 1 hour burn in with this combo to verify VAS integrity.
Ch. A power board is getting a 2nd inspection. There's a fault somewhere.
Suspect VAS from Ch. A placed on known good OPS from Ch B..... no overload light, no smoke and a happy subwoofer. 😀
Now going for a 1 hour burn in with this combo to verify VAS integrity.
Ch. A power board is getting a 2nd inspection. There's a fault somewhere.