Counterpoint SA12: What's that smell?

[chromenuts]

What IS that smell? As they say, a picture is worth a thousand words...



This is what I found last night after pulling my Counterpoint amp out of the cabinet and opening it up for a look...toasty resistors on the output stage of the left channel. I also found this...



Ohhhhh...so THAT's what a leaky cap looks like???

I've done the reading as best as I can on this one. Been to Michael Elliot's website many times and searched the threads etc. etc. I've been paranoid for a while I guess, even though my time with this amp has been wonderful up until now. I picked it up about 4 years ago for $500 to my door.

What I'm looking for is some input from the experts as to whether this is definitely a blown output stage, and if so what I should do about it. Is there a SIMPLE way to take measurements with a multi-meter and confirm the damage?

The circumstances of the incident were as follows:

1- Watching a movie...wanted better audio, decided to turn on stereo which I have hooked up to the DVD player.

2- Preamp clicks on...wait 15 or so sec, turn on amp.

3- Amp light cycles green after warmup...release mute on preamp and adjusted levels.

4- Sound is normal...turn to walk away...What's that smell?!!!...Sniff Sniff...is it coming from the cabinet?...a few seconds of uncertainty...mute preamp...shut off amp...shut off preamp.

The sound was still playing from the system when I decided to shut it down. I initially thought the smell was coming from my Audible Illusions preamp, so I cracked it open first. No obvious signs, so I pulled the amp and that was that.

One thing that I am regretting is the fact that I was recently testing some Rotel amps I acquired for a dedicated HT system through the same speakers. I neglected to disconnect the speaker cables running from one of the Rotel amps. I'm wondering if it is possible that the additional cabling and connection added resistance that caused the damage. The speakers used are my old KEF 104/2s that present a 4 Ohm load. None of the wiring appeared to be shorted out, but I am wondering if this happened due to my stupidity.

Assuming the output stage is shot, I'd like some input as to some options I've looked into as far as fixing the amp. Alta Vista is NOT an option as far as upgrades are concerned. I don't have a very good feeling about his used Mosfet's that he pulls from the units he upgrades either, especially considering the $220 he wants for the parts alone and the fact that I may just be setting myself up for another failure even if I could pull off the install. Paying someone else is questionable, even if I could find someone locally whose skills I trust I doubt I could afford it. I have read the information on Elliot's site and a few threads on here concerning the possibility of using Exicon Mosfets as a replacement in the output stage. I happened to find this a few months ago for sale on ebay...



and this...



...which I suppose could be used to address the fact that the power supply has leaky caps.

They seem like a reputable company. I have spent some time on their website, and they offer these as DIY kits and other services including install of these products and a complete overhaul option with other upgrades included...not that it is an option for me financially.

I have contacted them to see if these products would work in my SA12, and I was told they would because the SA12 and SA100 share the same circuit.

My dilemma is this, my skills with electronics are VERY few, and my resources are limited. I still have a pair of ASL Wave 8s sitting on the shelf downstairs because I couldn't solve a humming problem with one of them after an extensive thread of feedback on here a few years ago.

I can follow instructions if they are detailed and I have the required tools. The documentation on these products seems pretty extensive, although it would appear I would have to buy or borrow some tools I have never used such as a variac etc. in order to complete the installation properly.

I love this amp, and I am very sad that it appears that it is more than likely toast right now. I'd like to fix it affordably somehow, and these kits seem like the closest I might get to that goal unless someone else on here thinks we could put together a parts list and a how to that would use these kits as a model for repairing these amps? Thanks.

Kevin

[ramallo]

Hi,

The leak of these capacitors is very common, I had a SA100 and changed them. If your mosfets has died, you only have two options, or get these pulled from other SA12/100 or buy a kit with new circuit.

This eBay circuit have a nice looking, the Altavista kit is a complete and different amplifier, much better than the original but costly (I mounted this kit in my SA100)(Is the same circuit as the Aria amp).

Good luck!

[Tube_Dude]

Originally Posted by chromenuts

One thing that I am regretting is the fact that I was recently testing some Rotel amps I acquired for a dedicated HT system through the same speakers. I neglected to disconnect the speaker cables running from one of the Rotel amps. I'm wondering if it is possible that the additional cabling and connection added resistance that caused the damage.
Kevin

Hi Kevin

Thats the reason you have toasted your mosfets output stage. You had two amps outputs connected together, via the speaker binding posts.

Amps hate that...

[chromenuts]

Originally Posted by Tube_Dude

Hi Kevin

Thats the reason you have toasted your mosfets output stage. You had two amps outputs connected together, via the speaker binding posts.

Amps hate that...

Just to clarify...I didn't actually have two running amplifiers connected to the same set of speakers. I simply forgot to disconnect the cables that I ran to the speakers from the second Rotel amp. Whether or not that makes any difference is irrelevant...same outcome I suppose.

I would still like to find a way to confirm the damage and have been reading some of the information related to the installation of the replacement Mosfet kits. They're start up and biasing procedure jolted my memory of a statement I came across on the Alta Vista site related to voltage at the speaker terminals indicating a blown output stage.

From Greenstreet:

"If the DC offset is more than 5 volts, it indicates a fault in the wiring
and the problem must be corrected before continuing. The DC offset will bounce around quite a bit and never truly settle down to a steady measurement, so donʼt panic."

I'm wondering if this is the way to go about confirming the damage?

This is what I sw on Alta Vista:

"On Counterpoint SA-type amplifiers, an output stage that has failed is indicated by a large DC offset at the speaker terminals and fused gate resistors on the MOSFET sockets. Shorted speaker wires is the main cause of such a failure: the high internal losses of the MOSFETs pretty much guarantees that they will fail when their current limits have been exceeded.

Final failure of the entire output stage typically occurs at the next turning-on of the amplifier, when the weak device exhibits a drain-to-gate short. This causes full drain voltage to appear on the gate drive lines, which fuses the gate protection Zener diodes and gate resistors. On some occasions, the muting relay may also be fused, causing the gate drive lines to be shorted to ground. Replacement of all eight MOSFETs per channel (SA-20/220) or all four MOSFETs per channel (SA-12/100), all gate resistors, all four protection diodes and (rarely) the muting relay is indicated."

[VivaVee]

Originally Posted by chromenuts

I would still like to find a way to confirm the damage

The burnt gate resistors are proof enough that the associated MOSFET is fried. These are 1/2 watt resistors and by the time that they have smoked the MOSFET is well and truly deceased

My dilemma is this, my skills with electronics are VERY few, and my resources are limited.

Unfortunately, not a good combination for a tricky amp with expensive bits to replace. Although these amps were quite reliable, you would be working on a 22+ year old piece of kit.

[chromenuts]

Thanks for chiming in Alan. I was afraid that was the case.

Well, perhaps I can figure a way to afford one of these mosfet output kits from GreenStreet.

I have been reviewing the install documentation and I am impressed with the detail. I'm fairly confident if I pick up an inexpensive variac off of ebay as they suggest and the few other items I don't have I would be able to pull it off with some patience.

As far as the power supply is concerned, perhaps I'll just replace the old leaky caps for now...assuming that will save me a bit over the $175 they want for their refurb kit.

[Johnloudb]

Hi, yeah, I had the same experience with my SA-12 just a few months ago. Here's a thread related to that. I see the gate resistor on your amp is burnt. That happens when the MOSFETs die since the drain usually fuses to the gate. That results in DC at the output, and the DC killed one my DIY speakers. I measure like 45 volts DC at the output.

HTGuide Forum - Counterpoint Amp Repair

On turn on there is a relay that shorts the gate resistor to ground to stop turn on thump and let the tubes warm up. So, those gate resistors are then dissipating a whole lot of power, hence the smoke and smell from burnt resistors.

I haven't fixed it yet, but eventually I plan on doing my own output stage. Not sure what transistor to use though. Eliot used carefully matched transistors that were thermally stable, and are not longer available.

I was thinking about paralleling 4 matched pairs (4 p-channels, 4 n-channels) IRF520/IRF9520 with a small source resistance like 0.22 ohms. This would keep the output impedance low, cause Elliot didn't use any feedback on the output stage, so the damping factor would be quite low if using just 2 pairs of output transistors with the added source resistors. I think that would work but, not sure until I try.

Probably other and maybe better combinations to try on the output stage. I was just trying to keep the capacitance and transconductance figures on par or better that Elliot's output stage.

Anyway, good luck with fixing the amp.

[Johnloudb]

Good to know about the Green Street kit. I hadn't fully read this this thread earlier when commenting. I don't think the price is bad at all considering what you're getting. Though like you I was hoping for just a simple and inexpensive fix for my amp.

[anatech]

Hi chromenuts,
Looking at the picture there, it doesn't look like those gate resistors are really toasted as I've seen them in the past. At this point, I would just stop and do some investigating.

You need a half decent meter, digital preferred. Check to ensure the filter capacitors are discharged so you don't get a nasty surprise. The next step is super simple, check the speaker fuses, they should have blueish plastic covers over them. Make sure both fuses are good for the next test, remove anything you had to change when you're done the next step. Stick a meter probe into the red speaker jack, then set your meter to resistance function and touch the filter capacitor for each supply polarity, one positive and the other negative. Also, check to the ground (the connection to both capacitors). You should read really high resistance for all tests. Let us know what you get. Also, if you measure the ground connection (same one you just probed) and move the probe from the speaker jack to each side of the discoloured resistors, you should measure a very low resistance (a short, depending on how good the probes are). Again, let us know what you find.

Without knowing exactly what that new assembly you found on Ebay is (a schematic with values), I wouldn't touch it for love nor money. The original designer isn't an option either - unfortunately. Alan is extremely knowledgeable on these amplifiers, and I do have a fair amount of experience with them as well. Understand that there are two different basic versions of the driver section (the last tube). The really good news is that with lower supply voltages, your version is far more reliable than the SA-100. This matters. Be aware that there are design choices made on the main PCB that impede the sound quality as it sits as well. I've been redesigning it, but I'm not ready yet. The improvements in sound are already pretty stunning, while retaining the characteristic sound of Counterpoint. That's the hard part when redesigning something.

-Chris

[chromenuts]

Originally Posted by anatech

First off...thanks once again for taking the time to offer any guidance and feedback. That goes out to all you guys....it is really appreciated.

You need a half decent meter, digital preferred.

OK...well all I have to work with is your standard run of the mill Craftsman digital multi meter that I use for working on cars and what not...still it should do.

Check to ensure the filter capacitors are discharged so you don't get a nasty surprise.

Yup...made a discharging rig for the caps on the amps I was trying to fix a few years back.
The next step is super simple, check the speaker fuses, they should have blueish plastic covers over them.

Checked every fuse on the board with the tester for continuity...all good.

Make sure both fuses are good for the next test, remove anything you had to change when you're done the next step. Stick a meter probe into the red speaker jack, then set your meter to resistance function and touch the filter capacitor for each supply polarity, one positive and the other negative.

OK...the left channel (with the burnt resistors) shows nothing ( I set the tester to 2000K Ohms) to either polarity on both caps. Did the same test (on the same setting) to the right channel for both caps and I show about 350 across the board.

Also, check to the ground (the connection to both capacitors). You should read really high resistance for all tests.

Same results as above. Nothing to the grounds on the left channel with toasted resistors and 350 to all ground points when checking the right channel.

Let us know what you get. Also, if you measure the ground connection (same one you just probed) and move the probe from the speaker jack to each side of the discoloured resistors, you should measure a very low resistance (a short, depending on how good the probes are). Again, let us know what you find.

Clamped onto the ground bar of the supply caps (confirmed continuity of ground) and tested resistors. I had to adjust the tester as it seemed like it was measuring something out of the set range. The ends of the burnt resistors that are soldered to the individual Mosfets appear to have a resistance that increases (as if a capacitor or something is charging in the circuit). If I hold the probe in place it seems to level out at about 220 Ohms for this channel. The other side of the burnt resistors that are connected via the small teflon covered rail that also connects to the diodes measures about .5 Ohms. I repeated this test for the Right channel using the same scale on the meter. The side of the resistors that are connected to the Mosfets immediately shoots up to 400 Ohms with no fluctuation. The other side of the resistors are connected via the bar to the diodes measures .5 Ohms like the Left channel did.

Kevin