Peavey Deca 1200 Digital Amp Repair Class D amplifier

[Kansasnut]

OK im new to this site...I would like some help if anyone wants to chime in,.I have fixed a few amps in my time but this is a monster...
Status...
it is a 600watt per channel stereo class d amp...A channel is working ...B tries to work but is distorted. Power supplys check out ok.If I disable the ddt (peaveys protection compression circuit)and test across dummy 8 ohm load I can see distorted positive sine...I recently read up on class d on how they pulse width modulate the signal with high freq square wave to drive the fets...getting access to put a probe on this is very difficult
Anybody have any personel exsperiance with these? I will continue on thru the fog and post updates periodicallly

 

[Mikebits]

I don't suppose you could post a schematic. Of course do not violate any copyright material. Only post if allowed

 

[Kansasnut]

Peavey thought it would be ok...they don't charge for it

 

[tomizett]

Hi Kansasnut, sounds interesting!
Firstly, can I ask what you mean by a "distorted positive sine"? Is that to say just the positive-going half of the sine wave, or a sine that is positive (in phase) with what you are feeding in? What kind of distortion are you seeing (perhaps a photo of your scope trace?).

With class D the most common failures stop the amplifier working completely (usually letting the smoke out in the process). If we're lucky here then you've got a fault on the input side, before the amp propper. Can you identify some input circuitry and try and probe the signal before it is fed into the modulator, see if it's clean there?

I'm pushed for time now (lunchtime!), but will have a browse this evening and see if I can turn up any info.

 

[Mikebits]

I was able to locate the owners manual, but not the schematic, so without that I am afraid I would only be making wild guesses.

 

[tomizett]

Me too... I did stumble accross a place which claimed to be selling a soft copy, but I don't know if I'd trust that.
There is discussion on another furum here:
https://music-electronics-forum.com/t25516/
which seems to suggest that they will send you a manual if you just ask, and also that Peavey themselves will repair for a reasonable price (but where's the fun in that?).

It certainly looks like an interesting bit of kit. I've never come accross one before, and neither had my coleague (who'se been in audio all his working life). Perhaps they never sold (m)any in the UK?

 

[Kansasnut]

 

[Kansasnut]

Tom I don't think that's old Peavey info..I don't thing they repair this one anymore. I can send you schematic if you want. I have DDT out of ckt so i can see distortion. I have been trying to trace analog before it gets to digital. See above top half of wave clipped...not at low input level though.Input at 30% looks clean on output...does that mean FETS working?

 

[tomizett]

On the plus side, if you're getting output then we can be pretty sure that the output FETs (if they are FETs!) are working. On the other hand, faults where "it's working a bit" can sometimes be a lot harder to track down then "it's completely dead".

So you say you're getting clipping on the top half of the wave, as per your right-hand drawing, only on one channel and only above a certain output level? First things first, have you checked your power supplies. Measure the DC power rails on both the output stages (these should appear on the output devices and main PSU caps), and check to see if they are the same. Check low voltage power rails around op-amps and other ICs on the intput stage. It's also worth putting a 'scope on these rails and checking that they are free from ripple. It looks like this amp has a switching power suppy - is this right? Is there any sign of damage, overheating etc, and are any of the capacators bulging or showing signs of leakage?

What where your findings from probing the analogue/input stages - where you able to find the signal and see if was clean or not?

And yes, if you have the schematics it'd be very useful if you could post them here... I've got very little idea how a class D amplifier of this era might work.

 

[Mikebits]

That almost looks like deadtime distortion, but I imagine the modulator uses the same triangle gen for both channels. Hmm, yeah I agree with above, check all supplies with a meter and a scope.

 

[Kansasnut]

thanks Mike,Tom....there is 2, 1k power resistors in parallel(500 ohms) on 80 volt ps cap board and one of them actually smoked a bit so I shut down testing....but...I measured them and it reads 500 ohms....also I cant get over the fact that one channel(left side) is working fine....did you get the schematic mike , ? I will send to tom if you want it.,I will try and go thru pv procedure on this next it checks out those lv supplies...it has me leave main switch off and hook a separate ac line to lv primary transformer ,thesessupplies are not switchers...I will try and get back sunday night

 

[tomizett]

there is 2, 1k power resistors in parallel(500 ohms) on 80 volt ps cap board and one of them actually smoked a bit so I shut down testing

That could be a clue! I'd check and see what those resistors are connected to, in case it's related to the fault.

As long as you've not signed a non-disclosure agreement (NDA) with Peavey, you should be able to attach the schematics to your post - that way everyone can see them, including people cleverer than me!

 

[Kansasnut]

https://www.maximintegrated.com/en/app-notes/index.mvp/id/3977
above is a great link to what is going on in this amp...I am presently digesting it

 

[Mikebits]

Hi KC, yes I got your email, and thanks. I did look at your schematics briefly, There is a lot going on with your amp (circuit wise). I currently have a deadline I have to make, so I hopefully will have more time later this week, and will devote some time looking over what you have. Without a doubt this problem will be a good learning exercise. I did notice there is quite a good trouble shooting guide, This should get you quite far.
Thanks

 

[Mikebits]

Hi KC, any luck with your amp? Have you gone through the trouble shooting guide? Almost forgot about ya

 

[AudioSystemsEngineering]

Hi, just landed here via google. I have a lot of these DECA1200's because they're so cheap to repair in house. I must stress to anyone else attempting to repair this amp. DON'T poke around with scope probes with the unit fully powered up ! - you'll kill the output FETS.
As stated in the service man - this uint has 2 seperate power supplies. locate the small mains transformer near the front.. disconnect it from the tag board and connect it to a mains supply.
leave the manis input on the back panel disconnected.
NOW you're safe to poke around
you should be able to detect a nice 500kHz square wave at the Main fet gates.
if one is missing, you've most likely got a dead fet. there is a small zener across the gate of each FET. this should be replaced at the same time as the fet.
check the 4R7 fusible gate resistors.
while you've got the FETs out, check each of the UR820 ultra fast diodes for short.
also check the output filter choke hasn't burnt (seems to mainly happen on DECA724).
----
9 times out of 10, this will repair a dead channel.
----
if you're still missing a gate drive, check the opto isolators.
then the small switching PSU that lives on the output card.
one of the fet gate drives floats, so its driver is powered by this tiny supply. - you'l see some 4pin dil fets (something like an IRFD 1Z3) these drive that supply - check them, check the 4R7 fusibles around them.
----
Once you've got good FET gate signals, you can power the unit up with ~20 to 40VAC into the mains in on the back panel.
The main switching supply should start to work giving ±20-40V DC (otherwise, fix that... it's usually just the FETS).
If the PSU rails are good, put your scope onto the FET outs - each of the big 4.7R power res.
you should see nice big square waves.
Its OK to measure this side of the FET with the full AC input - (just don't touch the gates or their drivers).
as you increase the supply voltage, keep an eye out for overshoot - should be about 20V max, otherwise replace the associated UR820 (snubber).
the output square wave should have a noticable 'step' in it.
----

 

[schmitt trigger]

Hi, just landed here via google. I have a lot of these DECA1200's because they're so cheap to repair in house. I must stress to anyone else attempting to repair this amp. DON'T poke around with scope probes with the unit fully powered up ! - you'll kill the output FETS.
As stated in the service man - this uint has 2 seperate power supplies. locate the small mains transformer near the front.. disconnect it from the tag board and connect it to a mains supply.
leave the manis input on the back panel disconnected.
NOW you're safe to poke around
you should be able to detect a nice 500kHz square wave at the Main fet gates.
if one is missing, you've most likely got a dead fet. there is a small zener across the gate of each FET. this should be replaced at the same time as the fet.
check the 4R7 fusible gate resistors.
while you've got the FETs out, check each of the UR820 ultra fast diodes for short.
also check the output filter choke hasn't burnt (seems to mainly happen on DECA724).
----
9 times out of 10, this will repair a dead channel.
----
if you're still missing a gate drive, check the opto isolators.
then the small switching PSU that lives on the output card.
one of the fet gate drives floats, so its driver is powered by this tiny supply. - you'l see some 4pin dil fets (something like an IRFD 1Z3) these drive that supply - check them, check the 4R7 fusibles around them.
----
Once you've got good FET gate signals, you can power the unit up with ~20 to 40VAC into the mains in on the back panel.
The main switching supply should start to work giving ±20-40V DC (otherwise, fix that... it's usually just the FETS).
If the PSU rails are good, put your scope onto the FET outs - each of the big 4.7R power res.
you should see nice big square waves.
Its OK to measure this side of the FET with the full AC input - (just don't touch the gates or their drivers).
as you increase the supply voltage, keep an eye out for overshoot - should be about 20V max, otherwise replace the associated UR820 (snubber).
the output square wave should have a noticable 'step' in it.
----

what can I say: one of the most useful posts in a looong time

 

[tomizett]

Agreed - fair play to ASE for posting that information unprompted.
Welcome to ETO...