Acoustic 370 oscillation in power stage

[Mars Amp Repair]

hey guys,
I am working on a Acoustic 370 amp. I have replaced a shorted power output coupling cap & also the 8 power transistors. 2 were shorted but because this amp has no ballast emitter resistors, I decided to replace all 6 outputs & the drivers for a better match.

I used Peavey's equivalent of the 2N3055 #70484140. I've also tried 70483180 with identical results.

All other transistors check fine. I have also replaced all the electrolytics on the power board.

The issue is that the power stage oscillates at very HF. If I remove the power stage from the chassis & power it externally (this eliminates the original power supply with its electrolytics as well as the entire preamp stage), it does the same thing

As I have done in the past on some older power amps, I have added a 22pf snubber cap between the base & emitter of the power transistors on both banks to no avail. I can't get the oscillation to stop no matter how I connect a snubber cap

I wonder if anyone has had similar issues with these amps. I'm thinking of adding some emitter resistors just as a precaution...ya never know...that might help, too.

Here's the link to the schemo:
Unofficial Acoustic Control Corporation Homepage -- Schematics

Thanx, glen

[Enzo]

Is R329 22 ohm open in your zobel network across the output? The cap is probably OK, but might as well check it too.


Also check R328 and L301 in parallel. Is the inductor broken off at one end? Measure the resistor in circuit, if it meausres 10 ohms, the coil is broken, if it meausres shorted, that is the coil OK.

C310,311,312 are there for stability. They OK?

Dous your output bus center around 1/2 V-rail? Or is it offset?

Those small lytics C304,305,307 are always suspect at their age.

How about the overall mains draw? Is it elevated? Thinking your new xstrs might have a lower turnon threshold and the amp might be biased hot. If so it would draw a little high at idle. R318,319 set the idle current through Q304, and their ratio could need tweaking. Or they might have drifted, for that matter.


It used to work, don;t re-engineer it.


Here is my scan. NOt great, but it is at least a pdf so you can zoom in easily. PLus a parts list.

[Mars Amp Repair]

Hey Enzo,
Yeah, been through the Zobel & the coil/resistor combo. As mentioned above, I replaced all the electrolytics in the power stage as well as in the preamp. I agree in an amp this old, ya just replace the buggers.

As I mentioned, it still oscillates even when being powered by an external supply sans the preamp assy entirely.

It doesn't draw exorbidant current at idle. It starts oscillation at about 10VDC powering it. I can raise the DC in to its full amount & nothing gets hot. The oscillation is not anywhere near the full power out potential, but because of the oscillation, the amp never produces a decent sinewave or nearly the power it should.

I did jump all those 'pf' value caps to see if there was any reaction there...no help or change.

It did occur to me this weekend while mulling over the issue, that there were only 2 shorted power transistors in the amp, so I could just install a driver & a power transistor from the original bunch in each side to see if that has anything to do with the oscillation.

I violated my own rule & spent way too much time in one shot on this amp. I think the weekend break will bring a different prespective.

BTW, I agree that the older stuff did work at one time, but sometimes given the need to use different components & the mere collective aging of many components, sometimes you just have to get 'creative' to make them comply.

This one could be like the Trace Elliot mosfet amp where I had to add emitter resistors to get even current sharing (and that was with very closely matched mosfets) as well as adding a couple of 22pf caps on the base to emitter on one pair of output transistors to stop oscillation.

That's where I got the thought for this one. It worked like a charm when many were suggesting that the amp was just poorly designed to start with and there was nothing that could be done. The amp is still out there funtioning to this day. Once in a while they just get to me & I have to put in the extra time to see if there is a reasonable solution like that.

Thanx for the input Enzo...I'll let y-all know what comes of this beast.

[Mars Amp Repair]

For some reason my reply to Enzo did not push the thread to the top of the list. Trying this post to see if it will. glen

[Mars Amp Repair]

Here's a quote from a prior Acoustic 370 post by Gbono. I may find it's all created by the different 'modern' power transistors...we'll see.l

"Somewhat unrelated factoid:
I have also seen notes in a 370 service manual concerning the use of only using RCA transistors in the output stages (2N3055 or 2N3773). I'm not sure if they are refering to osc issues with the higher gain bandwidth transistor designs that came along after the RCA homotaxial process?? I have used output devices from On Semi (formerly Motorola) and STmicro without issue but I replaced all the output transistors.

[Enzo]

If you back up to the index you came from, it will show the same old order. Refresh the page and it will be in position.

I don;t remember any Acoustic amp I worked on that was sensitive to that. I don;t have any RCA transistors.

[gbono]

You can search this site for references to output transistor selection for a 370.
unofficial acoustic control corp forum - Message Board - Yuku

You can get the RCA version of the 2N3055 here:
Electronic Surplus Inc. - Welcome to Electronic Surplus Inc- Offering obsolete IC's, discretes, motors, relays, switches, pots, and much, much more! Always looking for your excess inventory. Electronic Surplus Inc. - Welcome to Electronic Surplus Inc

good luck -

[Mars Amp Repair]

thanx Gbono & Enzo,
Great info from all as ususal...I'll chime back in when I make some advancement...glen

[JerryP]

I have a service manual PDF for the 370 but it's 10MB and I can't upload. Shoot me an email if you'd like it. It has some added notes and transistor info.
jerry@fjamods.com
Jerry

[Mars Amp Repair]

Well,
it appears all the talk about the 'homotaxial' & 'epitaxial' transistors really is true. This amp was apparenty designed with low freq response (slow) transistors in mind.

Only 2 of the six power transistors were shorted, so I put the other 4 back in & I just happen to have 2ea 2N3737 original transistors I replaced the 2 shorted ones with. I placed those transistors in the driver positions so I'd have the 6 power transistors all original & matched.
Voila, no more oscillation. I'm thinking with the next one of these I get, since I won't have anymore original homotaxial transistors, I'll look more into adding some pf capacitors throughout the power stage to quell it's desire to oscillate.

as I mentioned above, the 22pf between the bass & emitter of the power transistors here did not help at all.

Jerry for the manual & especially the one page that mentioned the problem with subbing 'fast' transistors in this amp & Enzo as always for your input. glen

[gbono]

You know, Mouser wants $3.00/ea for an ON Semi 2n3055 and the RCA Homeotaxial parts from Electronic Surplus are $1.29. I think I'll stock up on some of the RCA parts since I have a 370 waiting for repair. Also noticed on the On web site that the 2n3055 has been discontinued.

I may try and see if a complete set of higher gainbandwidth output transistors (i.e. On Semi, etc) can be used, in the 370, with some sort of compensation scheme.

Don't have the 370 schematic in front of me but it appears that the 2N3773 (also obsolete) would be a better choice for the 370 PA section. BVceo is 140V vs 60V for the 2N3055.

I worked for Fairchild a long time ago and remember seeing power transistors which would exhibit a "latch-up" if you pushed the device into breakdown (BVceo). There was a region of "negative resistance" (in the output - CE) which would cause many different oscillation issues depending on the application. I'm not saying this is the problem you encountered since i believe most "modern" BJT processes should not have this problem.

[Enzo]

On Semi is usually more expensive at Mouser, I don;t think they are directly conected. Allied always has better prices on Moto/OnSemi. I usually look in Mouser first for most things, but ALLied gets my Moto buys.

Allied Electronics - Electronic Components Distribution

I haven't been to the On site, but they discontinued a lot of parts, but the parts are still available now made in the G version (lead free). Just went. Oh I see, it is the 2N3055H version that is obsoleted. The 2N3055 is still an active part.

[gbono]

WOW, that's a big difference in price between Mouser and Allied. I usually buy connectors and passives from Allied but I'll check out their semi's from now on. Almost a full time job trying to manage the best price from the different distributors ;-)

[Enzo]

Mouser is usually the best price on most things, but specifically Moto/OnSemi, Allied beats them totally. But Allied won;t have any of the 2SCxxxx types that Mouser does.

I have a want sheet on a legal pad, and I make columns to the right, one for Allied, one for MOuser, one for Digikey. I don;t use Digigkey as much, but they do have a bazillion caps and a lot of connectors.

SOmetimes it isn;t a matter of price. Like if I mainly need a stock of something, but something else I need is not available from the preferred supplier, then instead of placing two orders with one supplier, I might pay a higher price for a couple transistors and save it back on not paying two shipping fees.

[gbono]

Enzo: excellent technical skills AND he does a good job at supply chain management too