Classé CA-400 power amp, one channel noisy

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jut arrived for service today. One very pristine condition Classe' CA-400 stereo power amplifier with a reported complaint of it being noisy in one channel.

This is a BIG amp, and heavy at 110 pounds. Getting it out of the trunk of my car to the service cart did require a little concentration.

I have not yet powered it up and have to make up a set of adapter speaker cables for it since this demands that I do my testing with speakers that can handle high power.

The design appears to be reasonably straightforward but I'm looking for a schematic before I do anything else. Foreknowledge of the circuit makes for a lower probability of inducing problems while trying to fix them.

Upon removing the top cover....very impressive. Beautifully clean layout and quality. Doesn't look like a service nightmare at this point.

Lots of precautions have to be taken with this beast. Its energy storage is HUGE. I think it'll pull up to 2.4 KW at full tilt.

So, does anybody have the schematic?

There's a JFET in the front end on each channel. I'm not one to jump to conclusions but if the problem is noise in the preamp I'm going to be particularly interested in seeing if the noise is coming from the JFET. Proportionally speaking I've seen more noisy JFETs causing problems than bipolar transistors.
 
It's a recap job. The noise was caps. There is evidence of cap leakage in several spots but it's early stage leakage. Now is definitely the time to do the job.

I've got photos to post...LATER.

This amp is neat and tidy and beautiful to look at with the top cover off.

But it's a NIGHTMARE to work on. And I'm not the kind to complain in nearly every situation.

Well, this is different. The reason is because basically this amplifier was built upside down and all the circuit boards are STACKED.

The power transformer is a solid 60 pounds by itself. That certainly doesn't help!

FORTUNATELY, I missed something important after I'd loosened the transformer and tried to drop it on a block by flipping the amp over. If I'd FAILED to miss something important I'd have almost certainly done massive damage to the circuit board stack.

If you ever have to take on this job, here is the basic guide to it:


Have one beer. Take the edge off. NOT TWO BEERS, this is not going to be a job you should even think about doing if inebriated even slightly. If you're the calm sort, skip the first beer, too.

Pull the top panel. Loosen the hardware and pull the rear panel out a little for a good look into the back of the unit.

At EVERY stage, take photographs and notes.

You'll want to pull the back panel off. Disconnect and remove those parts from the back panel that require the LEAST reassembly. The input buffer board is best left on the back panel, but photograph and identify the three leads at the bottom of the board that you will carefully desolder. It's easier than disconnecting both XLR inputs.

Choose if you want to disconnect the preamp cables at the preamp board or at the input buffer board. I did it at the preamp.

Special word of caution: The PC boards have lightweight traces. They are easy to damage and pull off the board if your soldering technique isn't excellent. Be careful! Nobody wants to have to repair trace damage.

I left the current limiter resistor soldered to the input power harness. I unbolted it from the rear panel. I removed the input power PC board after desoldering the two wires to the fuse holder. This is the simplest way and thus for me it's the preferred way.

With the back panel out of the way, now you will want to remove the side heat sink assemblies and power stages from the chassis. I won't give detailed instructions as you will be able to figure it out. Remember to photograph and document.

You will be needing to unsolder a dozen heavy gauge wires going to the capacitor reservoir boards. Four per group, three groups. Photograph them and note the way they're woven through the heat sink circuit board assemblies.

Of course there's a number of screws on the bottom panel to be released, too.

Upper support rails front and rear have to be removed. For the front rail, you have to pull the top screw that holds the handles on. Be warned, when you remove all front panel handle hardware that big heavy front panel will fall off if you're not ready for it. No, I didn't drop it.

Remove one heat sink side at a time. Take your time and don't hurry.

After both heat sink assemblies are completely removed and so is the rear panel, as well as the front panel after you disconnect the power PC board via its coinnector,


now you will have to contend with the idea that the recap job is going to be easier if you flip this assembly over and separate the circuit board group from the bottom panel.

I haven't completed this step yet.


A super ultra mega special word of warning:

The CA-400 has four power supply reservoir capacitor PC boards. Each contains nine each, 4700 uF 100 volt snap-in Nichicon electrolytic capacitors.

The special warning is that TWO of these boards have the capacitors installed BACKWARDS as compared to the other two capacitor boards.

They are two different boards as configured. Get the polarities or board types mixed up and you are heading for a BAD DAY.

Take photographs and make notes.

Transformer removal:

There is a trick to this. There is what I belive to be the voltage selector board that is attached with one of its screws being secured across the subplate that holds the transformer. To remove the transformer subplate that's going to have to come off. So remove that board and carefully push it out of the way. It's directly connected to the stiff transformer wires so don't manhandle it.

Really think about this part. The last thing you want to do is drop the transformer out when the boards it's connected to are still secured to the chassis!

I don't think there's any getting around the fact that some boards in the board stack will have to be removed and reassembly may be difficult.

I'll be crossing that bridge when I come to it VERY SOON.
 
Well, major disassembly is complete. This is absolutely required for easiest access to all capacitors that will be replaced. (That's all electrolytics, no exceptions.)

It's kind of scary how far I had to disassemble the amp. All that's on the service cart right now is the power transformer and the board group. Bottom plate is off, front and back are off, heat sink units are off, basically it's been scrapped. But to rebuild it.

I'd better get this thing right or I'm going to end up buying it from its owner and if that happens it won't be working.

The capacitor order is close to 350 dollars in parts.
 
Got the parts, began the process of recapping it.

About a third of the 36 reservoir caps show clear signs of electrolyte leakage. It was definitely time to service this amp. Any amp like this of similar age will need this as well.

For future reference, the 820 uF 160 volt caps on the lower center board which I presume is the protection board are snap-in types. All the reservoir caps, all 36 of them, 4700 uF 100 volts, are also snap-ins.

I hate snap-in caps. They have a tendency to grab the solder pads and vias and pull them out no matter what method you use to remove the caps. I had to rebuild two traces on one board when a spur on the cap terminal locked into the pad and nothing I could do could have kept it from doing that.

I upgraded EVERY cap in the amplifier to nothing less than a 105C 5000 hour rating, except in the case where such a cap was simply not available.

The 105C 4700 uf 100 volt caps are taller than the original ones but there's room between the boards for them. It's OK.

I had to make a second cap order to cover a few I missed. I also had assumed those 820s on the protect board were thru hole, which is wrong. I ordered snap-ins because despite my dislike for their tendency to damage traces on removal, they do hold in nicely and allow me to solder them in without having to find a way to prop them up just right so they stay flush to the board during soldering.
 
It's a 2.4 KVA transformer if I have my information right.

As for rail voltage, my meter read 90 volts per rail referenced to ground.

At 90 volts DC that'll support a clean 63.63 volt RMS signal which into 8 ohms is 506 watts at 7.9 amps.

I have no doubt this amp is good for 1KW per channel into 4 ohms and is probably going to be safe to operate into a 2 ohm load, given that it has 24 outputs that have a 200 watt TDP rating.

It reminds me of the classic Krell way of doing things but it's both cleaner in its mechanical design than the Krells I've been inside, and also harder to service than a Krell.

It's definitely working from the same playbook.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.