I have two Peavey CS800 amplifier amps that someone dropped off and never called or came back for. Ten years later, during the pandemic, I now have the time to take a look at them.
These are beasts. I estimate they weigh ~ 65 pounds most of which is a 6x6x7" E-core power transformer, however, they are well built, use quality components, and are easy to service. The size and weight are quite objectionable of course but they are powerful using a +/-80V bipolar power supply.
I'll check the huge filter caps for leakage an reform if possible. I'll remove the crowbar protector. I may replace the few op-amps with modern types. Initially I'll borrow parts from one to repair the other and, if I like the way it sounds, perhaps I'll purchase parts to repair the other.
I'm thinking of buying one of these bipolar switching power supplies that can be had for under $50.00 on ebay to replace the tranformer/rectifier. This would knock off the majority of the weight.
Have any of you tried this?
Also, is PTC protection device a good sub for the triac crowbar circuit?
These are beasts. I estimate they weigh ~ 65 pounds most of which is a 6x6x7" E-core power transformer, however, they are well built, use quality components, and are easy to service. The size and weight are quite objectionable of course but they are powerful using a +/-80V bipolar power supply.
I'll check the huge filter caps for leakage an reform if possible. I'll remove the crowbar protector. I may replace the few op-amps with modern types. Initially I'll borrow parts from one to repair the other and, if I like the way it sounds, perhaps I'll purchase parts to repair the other.
I'm thinking of buying one of these bipolar switching power supplies that can be had for under $50.00 on ebay to replace the tranformer/rectifier. This would knock off the majority of the weight.
Have any of you tried this?
Also, is PTC protection device a good sub for the triac crowbar circuit?
Id expect 50 dollar budget power supplies to not be up to the task.
Find no improvement with op amp updates.
possibly create more issues with stability or dc offset.
Amps are fine as is. With crowbar removed you would destroy expensive speakers
if there was a fault. Instead of a few cheap electronic devices.
Find no improvement with op amp updates.
possibly create more issues with stability or dc offset.
Amps are fine as is. With crowbar removed you would destroy expensive speakers
if there was a fault. Instead of a few cheap electronic devices.
I can't imagine anything more damaging to the sound than putting a PTC resistor in series with the output. The crowbar protection circuit is one of the advantages of these if your speakers are worth more than $2.50. Why do you want to remove it?
Major upgrades I would do would be 1. replace 1/4 phone plug outputs with dual banana jack. Phone plugs pulled partially out by feet is a major source of output transistor failure. Or speakons, although I may not have the drill bits required to install a speakon.
2. If they have a 741 op amp, (rev A B) put in a socket and an OPA604 to cut the low volume hiss. Bypass capacitor around feedback resistor and local P.S. capacitor may be required to eliminate oscillation. If they have a 4558, replace with NJM4560 or MC33078 for same reason, although at low gains not as hissy as 741.
Small electrolytic caps are more likely to fail than the big rail caps. Starting the unit with a light bulb in series with the AC line is enough of a reforming, usually. I have my light bulb socket in a grounded steel box for my safety, as stranded wires tend to fall off the screw of those edison bases. The big rail caps tend to reduce the maximum output with high ESR, which can be easily measured on an 8 ohm 250 w resistor with an analog voltmeter at maximum volume. P=(V^2)/Z.
When Peavey put in a switcher supply (cd800s) they built an entire steel box around it with inductors on the inputs and outputs to eliminate the howl from ultrasonic emmision. Do you have room for that? Peavey copper bound transformers are remarkably hum free.
If you don't like them, they are getting about $250 on ebay today. There is a cs800s with switcher supply today for $170,probably because old units with a switcher supply are so much harder to diagnose & service. Easier to carry though.
Major upgrades I would do would be 1. replace 1/4 phone plug outputs with dual banana jack. Phone plugs pulled partially out by feet is a major source of output transistor failure. Or speakons, although I may not have the drill bits required to install a speakon.
2. If they have a 741 op amp, (rev A B) put in a socket and an OPA604 to cut the low volume hiss. Bypass capacitor around feedback resistor and local P.S. capacitor may be required to eliminate oscillation. If they have a 4558, replace with NJM4560 or MC33078 for same reason, although at low gains not as hissy as 741.
Small electrolytic caps are more likely to fail than the big rail caps. Starting the unit with a light bulb in series with the AC line is enough of a reforming, usually. I have my light bulb socket in a grounded steel box for my safety, as stranded wires tend to fall off the screw of those edison bases. The big rail caps tend to reduce the maximum output with high ESR, which can be easily measured on an 8 ohm 250 w resistor with an analog voltmeter at maximum volume. P=(V^2)/Z.
When Peavey put in a switcher supply (cd800s) they built an entire steel box around it with inductors on the inputs and outputs to eliminate the howl from ultrasonic emmision. Do you have room for that? Peavey copper bound transformers are remarkably hum free.
If you don't like them, they are getting about $250 on ebay today. There is a cs800s with switcher supply today for $170,probably because old units with a switcher supply are so much harder to diagnose & service. Easier to carry though.
Last edited:
Yeh, I was bating with the PTC question. That's my opinion too. I see there's a post about using a solid state relay as means to protect speakers. Perhaps I'll pursue that.
I feel a crowbar to be quite a crude method and, as designed, usually takes out many output transistors if activated as it did in the case of the two amps I have. I have four output devices to replace on the first amp at $5.00 each. I'm not sure about the second. I'm skittish about powering it up with crowbar in place. The light bulb suggestion is a good idea.
Would you guys know what the Peavey topology is called whereby only NPN transistors are used - the top half of the waveform driven from the Emitter, and the lower half driven by the collector of a second device?
Advantages/disadvantages?
I feel a crowbar to be quite a crude method and, as designed, usually takes out many output transistors if activated as it did in the case of the two amps I have. I have four output devices to replace on the first amp at $5.00 each. I'm not sure about the second. I'm skittish about powering it up with crowbar in place. The light bulb suggestion is a good idea.
Would you guys know what the Peavey topology is called whereby only NPN transistors are used - the top half of the waveform driven from the Emitter, and the lower half driven by the collector of a second device?
Advantages/disadvantages?
All npn output transistors is called quasi-comp. In early 80's there wasn't a pnp version of MJ15025 so quasi-comp was tougher than MJ15003/15004. Difference in sound to fully complimentary occurs in 2nd or 3rd digit of Harmonic distortion. Nothing anybody could hear on speakers. Speakers have 1-20% distortion. I had quasi comp AX6 versus fully comp CS800s. At 1 W I couldn't hear any difference on SP2-XT speakers.
Cheap upgrade to crowbar protection is put 30 A fuse between rail caps & output board. I did. Definitely disconnect the triac for testing. It only gets installed after dc <200 mv on output and ampputting out rated clean watts.
Note output transistor meltdowns tend to take out drivers, predrvers, op amps, and even 50 v rated ceramic caps. May be on light bulb box for days or weeks. Those little parts don't cost much, I salvaged many from PCAT switcher supplies.
To prevent further upstream damage from future OT shorts, I replaced some jumpers on the output board by 5 A fuses. In the base lines. Also put 3.3 v zeners on the driver side of the fuse to take the surge instead of the driver transistors. However, removing the 1/4 phone plugs gets rid of the most common cause of output transistor meltdown. Also replacing the shutoff switch rated 100C with one rated 70 C. That is on the OT heatsink. Also putting a filter over the fan eliminates the dirt on heatsink problem.
Cheap upgrade to crowbar protection is put 30 A fuse between rail caps & output board. I did. Definitely disconnect the triac for testing. It only gets installed after dc <200 mv on output and ampputting out rated clean watts.
Note output transistor meltdowns tend to take out drivers, predrvers, op amps, and even 50 v rated ceramic caps. May be on light bulb box for days or weeks. Those little parts don't cost much, I salvaged many from PCAT switcher supplies.
To prevent further upstream damage from future OT shorts, I replaced some jumpers on the output board by 5 A fuses. In the base lines. Also put 3.3 v zeners on the driver side of the fuse to take the surge instead of the driver transistors. However, removing the 1/4 phone plugs gets rid of the most common cause of output transistor meltdown. Also replacing the shutoff switch rated 100C with one rated 70 C. That is on the OT heatsink. Also putting a filter over the fan eliminates the dirt on heatsink problem.
Last edited:
Great info inidanajo...and all others too.
Yeh, I have tested all components on the driver board except op-amps. Four output top side transistors went short, a small signal transistor the sets up the bias current went open (unusual), and one of the large reversed protection diodes from rail to power devices went short.
I will try the fuse between rail and caps before I attempt to turn it on to check op-amps.
Yeh, I have tested all components on the driver board except op-amps. Four output top side transistors went short, a small signal transistor the sets up the bias current went open (unusual), and one of the large reversed protection diodes from rail to power devices went short.
I will try the fuse between rail and caps before I attempt to turn it on to check op-amps.
Last edited:
One thing you can do is put a +-80v 1600 W switcher supply in an external steel box. Then you could run 12 ga cables to the CS800, connecting to a high current connector like a jones plug on the outside of the amp. The external supply would need its own fan switch and fuse of course.
Then you would have to carry two boxes instead of one. And this arrangement would not fit in a 19" road case. But nothing would weigh 65 lb. I'd leave the rail caps in the amp, proximity to output transistors matters. Some switcher supplies don't like starting up into a large capacitance. The engineering is up to you.
Then you would have to carry two boxes instead of one. And this arrangement would not fit in a 19" road case. But nothing would weigh 65 lb. I'd leave the rail caps in the amp, proximity to output transistors matters. Some switcher supplies don't like starting up into a large capacitance. The engineering is up to you.
In one unit a couple transistor sockets are broken, ie the insulating nub the extends through the aluminum heatsink is broken off. I can still properly seat a transistor but I would think with +/- 80v power supplies this is a bad thing and could result in arcing to the heatsink. A couple of mica insulators are cracked as well possibly raising the same concern.
I'm thinking of replacing the IC sockets in any case.
Am I being overly cautious?
I'm thinking of replacing the IC sockets in any case.
Am I being overly cautious?
The triacs that are used in both units that impliment "power on" have blown short circuit which means no matter what position the on switch is in the the unit is in fact on.
I could see designing in a triac if you want some small switch or even a computer port to turn on the power. However, a huge power switch is equipped that seams to me should be able to take the turn on surge. Maybe not - your thoughts?
I could see designing in a triac if you want some small switch or even a computer port to turn on the power. However, a huge power switch is equipped that seams to me should be able to take the turn on surge. Maybe not - your thoughts?
The transistor sockets are "unobtainium". They are TO3 albeit with four legs instead of three. The two outermost legs are connected to the threaded screw inserts (collector). Usually there is just one of these. I applaud the presence of the extra redundant collector connection....but.....non-standard. Fortunately they are a laminated part, the bottom lamination attaches the leads very securely and is in tact in every case. The top plastic half of the lamination has the integral tubes that go through the heat sink insulating the legs. This plastic top half is broken in four or five cases. My experience with Peavey has not been great when it comes to vintage pieces of gear. Even if they do have the sockets they will likely charge an arm or a leg for them. So looks like i have to refurbish those rather than replace them....hmmm...do I have the energy? No, looks like I will be making one good unit from two.
The triac is important to save the power switch. They are easily available with 9000 in stock at newark. https://www.newark.com/stmicroelectronics/bta40-800b/triac-800-v-50-ma-40-a-400-a-80/dp/24M0818
$9.50 each.
The socket can be replaced by a mica + ferrule kit via multicomp kit from newark. You have have to drill the holes for the mounting screws larger. However, new T204AA packages now available clear only #4 screws whereas the old TO3 packages allowed #6 screws. Use heat sink compound on the mica washer. The ferrules are used around the screws. The wires are soldered to the legs of the transistor instead of to the socket. Use a hot enough iron to move quickly.
On my PV-1.3k I put a CL-101 NTCR between the power switch and the transformer. That device used to dim the room lights when I turned it on.
$9.50 each.
The socket can be replaced by a mica + ferrule kit via multicomp kit from newark. You have have to drill the holes for the mounting screws larger. However, new T204AA packages now available clear only #4 screws whereas the old TO3 packages allowed #6 screws. Use heat sink compound on the mica washer. The ferrules are used around the screws. The wires are soldered to the legs of the transistor instead of to the socket. Use a hot enough iron to move quickly.
On my PV-1.3k I put a CL-101 NTCR between the power switch and the transformer. That device used to dim the room lights when I turned it on.
The crowbar circuits are fine, as long as the wiring is up to the task. It must support the fault currents long enough for a fuse to open, instead of PCB traces smoking. I think that’s what happened to Indianajo’s PV1.3k. It’s not the feat of engineering that the old CS800 was, by a long shot. Protection boards with speaker relays work well too, if you get one that’s big enough. The little PCB mount ones don’t count - you need an ice cube with 20 amp DC rating per contact, and parallel them. The mistake people (including manufacturers) make with relays is only sizing them for the average musical program current, not the potential fault current they could be called to interrupt. If the contacts eventually get pitted and cause work-function related partial rectification (which you will notice as distortion, but it won’t for many interruptions cycles) you put in another relay which you can just plug into a socket. Fixed in 5 minutes. And small “resettable” faults caused by dropped mics or turntable feedback won’t blow a fuse each and every time.
As far as lightening the amp, you could start with the power supply. A 1500W 80-0-80 switcher that’s really up to the task is likely to cost close to a thousand bucks so not really practical. Buy something cheap on E-bay and it will either be a ticking bomb or go into current limit at low frequencies where you need to draw 3.14X the rating at the top of every cycle. A 1000A 58-0-58 Antek toroid is only 100 and will weigh less than the EI. They will easily support 1200VA continuous and 2000 intermittent for 10 minutes at a time so it is big enough. Modern 15000 uF snap-in caps will also save weight and the caps might eventually need replacement anyway.
As far as lightening the amp, you could start with the power supply. A 1500W 80-0-80 switcher that’s really up to the task is likely to cost close to a thousand bucks so not really practical. Buy something cheap on E-bay and it will either be a ticking bomb or go into current limit at low frequencies where you need to draw 3.14X the rating at the top of every cycle. A 1000A 58-0-58 Antek toroid is only 100 and will weigh less than the EI. They will easily support 1200VA continuous and 2000 intermittent for 10 minutes at a time so it is big enough. Modern 15000 uF snap-in caps will also save weight and the caps might eventually need replacement anyway.
See, this guy is thinking. As long as there are no stairs involved, this will make it very lightweight.
wg_ski - I agree. I went the Antek torroid route on a two hundred watt SS guitar amp with an E-core and knocked 10lbs off the total weight.
Indianjo - tx for the kit suggestion. I was able to find a similar socket though from Keystone - 4600. It doesn't have the fourth leg but does include the a very similar plastic insulator laminate that I plan to steel and apply to the existing Peavey socket pin laminate. No need to drill a hole for the fourth leg since that is only a feature on the bottom laminate.
Perhaps putting an NTC thermistor in series with the original triac to save the switch would work. I did this on a massive off-line SMPS and it worked well enough to get the inrush down reasonable.
Has anyone designed an inrush limiter using a solid state relay shorting out a power R after a time period?
I know those things can take quite a lot of amps.
Indianjo - tx for the kit suggestion. I was able to find a similar socket though from Keystone - 4600. It doesn't have the fourth leg but does include the a very similar plastic insulator laminate that I plan to steel and apply to the existing Peavey socket pin laminate. No need to drill a hole for the fourth leg since that is only a feature on the bottom laminate.
Perhaps putting an NTC thermistor in series with the original triac to save the switch would work. I did this on a massive off-line SMPS and it worked well enough to get the inrush down reasonable.
Has anyone designed an inrush limiter using a solid state relay shorting out a power R after a time period?
I know those things can take quite a lot of amps.
An AC mains rated solid state relay IS a triac (with optically isolated driver circuitry). Whether you make the relay or buy it doesn’t matter. Big ice cube relays work for soft start resistor bypass just as well as triacs. They don’t have to make or break large currents - when it bypasses the resistor, the start up surge has passed. It just needs to handle rated maximum load (whatever the amp draws with sine wave load - which will be MORE than what’s stated on the back panel. Look it up in the service docs).
What you do need to consider when using a bypassed resistor for soft start is what happens if the relay fails to engage. You need to sense this condition and either shut it down entirely, mute the audio, or just keep it from running hard. Some amps (cheap modern Crown comes to mind) don’t have any provision, and the resistor can catch fire.
When you start adding protection circuitry to an amplifier, it can quickly balloon and become more complicated than the amp itself. And sometimes add additional points of failure.
What you do need to consider when using a bypassed resistor for soft start is what happens if the relay fails to engage. You need to sense this condition and either shut it down entirely, mute the audio, or just keep it from running hard. Some amps (cheap modern Crown comes to mind) don’t have any provision, and the resistor can catch fire.
When you start adding protection circuitry to an amplifier, it can quickly balloon and become more complicated than the amp itself. And sometimes add additional points of failure.
wg_ski, Good point about the current a start up relay needs to endure. - makes sense.
I appreciate your insight about protection circuitry. It reminds me of software. Only 10% percent of the code is dedicated to the function and executing typically while the remaining 90% addresses the "what ifs". This 90% raises the odds a bug will manifest itself.
I appreciate your insight about protection circuitry. It reminds me of software. Only 10% percent of the code is dedicated to the function and executing typically while the remaining 90% addresses the "what ifs". This 90% raises the odds a bug will manifest itself.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.