Servicing Ampeg SVT6-Pro----I screwed up!

[nevetslab]

In doing preventative maintenance on an Ampeg SVT6-Pro, last svc'd Nov 2014, I checked the upper MosFET Output Stage bias, looking at the Source voltages on the device terminals, relative to output buss. I had let it run for 15 min, finding them in the 5-7mV range, which seemed a bit low. I put the small screwdriver onto the bias pot on the main PCB floor below, and began nudging it up while monitoring one of the MosFET's that was at 7mV, and didn't see any change. Increased it a bit more, still no change. I then heard a somewhat loud groan from the power xfmr, turned to look at the Power Analyzer, seeing the mains current had increased from about 1.25A to 1.5A, and began to see a bit of smoke combined with the smell of something bad, as I turned off the power switch.

Wrong bias pot!! Idiot! That's the bias pot for the lower output stage, which is NOT accessible. Not without pulling the main PCB out and cobbling the supply connections to run it outside the chassis. I turned the bias pot all the way CCW, walked the variac back up again, approaching 100VAC, as the mains current was already coming up on 1.4A, and I turned it back to 0V, as I again began smelling the bad news.



And, I was having SUCH a good day before that! Sigh...........

I put the top cover back on, moved it to the check-out bench, and moved the SVT5-Pro to the bench and pulled it's top cover, checked my Inventory Sheet to see when it was last serviced. Nov 2012. Looking close at the top rear panel PCB, seeing solder fractures on the pair of 2-pin headers, I knew this one had to come apart as well, just to deal with the rest of the solder joints, plus I wasn't going to make THAT mistake again.

I pulled this amp apart, sure enough, found what I expected....fractured solder joints on the output jacks, the bridge rectifier....which is mounted to an UNSUPORTED Aluminum Plate which can waggle all it wants in transit.....other connectors. At least while waiting for gear to arrive last week, I had serviced my desoldering iron & tips, found I DID have new tips and it's now sucking solder out of the joints on the PCB as it always had when working properly.....and restored order to the last several years of use.



So, here's two amps that sure need to have a better chassis design. That being, a removable panel on the bottom to expose the lower MosFET output stage so you can monitor the source & gate terminals for bias adjustment, and NOT have to removed the bloody PCB assy, and cobble the AC Mains Primary/secondary connections to the large PCB assy that now has to be run outside the box and not self-destruct from the test set-up. Should be fun.....check and see if I'm still smiling after getting back to the SVT6-Pro.

[nevetslab]

Teardown, rebuild & bias set-up on the SVT6-Pro

Well, after I got thru with the SVT5-Pro, which I'll post details on later, I pulled the SVT6-Pro apart. To get the power amp board out, you have to remove the huge Toriodal Power Xfmr and find all of the screws that lock the power amp assembly into place. I found as I began removing the xfmr primary leads that I had been inside this before, all still clearly marked so I'd know how to put it all back together.

Once I got the bottom heat sink unmounted and removed, I found all but two of the MosFET's were shorted, all but three of the gate resistors were open, all of the P-Ch MosFET's source sense resistors were open, a 13V zener was shorted, and the P-Ch side current limiter xstr was shorted. All of that while never even getting to full CW rotation on the bias pot! It does make me wonder if this amp was on the verge of failure anyway. That ice cream lid is full of all the failed parts.....so beware of what trim pot you get hold of on these amps!!



Fortunately I had fresh Source resistors, gate resistors, source sense resistors, and had just recently screened and batched new N-Ch & P-Ch MosFETs, so I was able to rebuild the lower power amp output stage.

I removed the circuit breaker from the rear panel, and the power switch from the front panel...both being snap-fit into the openings...and plugged them into their appropriate terminal. I found a suitable spacer to stand the PCB assembly vertical allowing me to measure the source terminals as I re-set the bias. I ended up soldering a short bar on the output bus to hang my black probe onto of my DMM, so I could check the MosFET's during bias adjustment, while in clear view of my power analyzer.



I'm just now putting the amp back together. I'll first verify each output section passes signal, so I don't have to tear it all down again if it doesn't.

[nevetslab]

SVT5-Pro Biasing

All is back to normal on this SVT6-Pro. Both channels passed signal just fine open on the bench as it was, outside the chassis. Put it all back together, and working as it should.

The SVT5-Pro, in checking the bias on both halves was likewise a bit tedious, and a balancing act:



As the relays were both pulling in before I pulled the amp apart, I was able to use a phone plug to dual binding post adapter, connecting my black lead to the output side of the power amp channel for reference, then check the source terminals of the MosFET's to see the min/max range, and adjust the bias as needed. The lower half, inaccessible until you gut it and cobble the set up something like this, needed a little increase in bias level.

The upper output stage, accessible, was as far as I wanted to run it, having a bit more min/max range in source voltages across the N-Ch/P-Ch devices. All went back together ok, now with freshly re-soldered connections that made it all worth the trouble.

[J M Fahey]

Ouch!!!

[TimmyP1955]

But it would cost an extra dollar to put that access panel on!

Like the Hughes & Kettner BB600. To change the front panel lamps requires the removal of 40 screws. An access panel would have cut that way down.