More fun with 60s electronics: noisy, oscillating Scully 280 - FIXED (mostly)

[sircletus]

It's been a banner week here between the noisy Altec and now-silent Lang.  So...on to the Scully.  Awhile ago, a friend of mine dumped a Scully 280 head amp into my lap and asked me to take a look at it for use as a mic pre, since that's a thing now.  "High noise floor," said the attached note.

Previous owner had attached a guitar pedal-style 2.1 coax barrel cable to the power terminals on the Amphenol connector, so I whipped up a regulated -24V power supply (with a 7924) in a Hammond box with an output jack to mate with the barrel connector.  Not the best connector, for sure.  All internal voltages measure where they should, FWIW.

So here's what it does:  "PSSSSHHHHHHHHHH" on mic input as you turn up record level.  Passes sound, amplifies mic.  At about 7 or so on the dial, it starts oscillating.  Audible frequency range.

Here's where it gets fun: tapping unit, shaking it, thumping it changes the oscillation.  With the proper "technical thump" oscillation goes away!  PSSSSHHHHHHHHH is still there, though.

So where should I look first?  One thing that sucks about these is that the playback and record cards are in a a spot-welded subframe that CANNOT be opened, making probing specific areas of each card to locate the noise source next to impossible without a card extender.

Other facts:

- all components, including electrolytics, are original.
- putting the mode switch into playback mode reduces noise dramatically, making me think the mic pre/record card is the problem.

I have a VERY small supply of otherwise unobtainable transistors; certainly not enough to replace all of them.  I might be able to replace everything on the input card.

Any thoughts?  Run and hide?  Kill it with fire?

While I would at this point prefer the latter, it's not mine, it's my friend's.

 

[abbey road d enfer]

- putting the mode switch into playback mode reduces noise dramatically, making me think the mic pre/record card is the problem.

Actually, the record amp should not be left unloaded. It can deliver very high gain when unloaded. Also, bias voltage can radiate everywhere. You should disable the oscillator and the last stage of the record amp.

I have a VERY small supply of otherwise unobtainable transistors; certainly not enough to replace all of them.  I might be able to replace everything on the input card.

Semi conductors is probably the last thing you need to replace.

 

[ruffrecords]

Look for dry joints, cracked tracks and broken small value compensation caps that should manage the open loop response so it does not hoot.

Cheers

Ian

 

[sircletus]

Thanks for the responses, gentlemen!

I actually already pulled the oscillator card before I first posted. Toned the oscillation down a bit.

Also found a cracked joint on the base of the 2N257 in the second half of the record amp.  I'll

What's interesting to me is that many folks, including a different friend, have used these as mic pres completely unmodified: no disabling anything, no replacing components.

I'll definitely check the other stuff and kill the second half of the record amp.  I'm going to pull input and output decoupling caps to get it out of circuit.  I hope that's enough.

So what do you think the source of the horrible hissing is?  For some reason I assumed failing transistors. 

 

[sircletus]

Long story short:

The source of all the noise was the very first transistor, Q853.  I replaced it and the ol' beast works like a charm.  Mostly.  While I was at it, I went ahead and recapped the entire box, with the exception of the large can capacitors.  I'll get to those later.  Oh, and at the recommendation of abby road d enfer, I went ahead and decoupled the second stage of the record amp.  No measurable change, but it was an easy thing to do.

ONE PROBLEM REMAINS:  sporadic oscillation at high gains, both at mic and line level.

Any remaining ideas where to look?

I'm otherwise calling this fixed.

 

[mjrippe]

These have a ton of gain - somewhere around 70dB IIRC - so will probably never be used wide open.  But finding the source of oscillation may have beneficial effects at lower gains. 

 

[sircletus]

But finding the source of oscillation may have beneficial effects at lower gains. 

Yep, that's why I'd still like to track it down.  It even changes based on physical interactions: give it a good thump and the oscillation will change!

 

[mjrippe]

Yep, that's why I'd still like to track it down.  It even changes based on physical interactions: give it a good thump and the oscillation will change!

Well, percussive response often indicates a bad solder joint, dirty contact, or similar.

 

[sircletus]

Well, percussive response often indicates a bad solder joint, dirty contact, or similar.

Yeah.  Given how much of this is point-to-point, I'm debating whether or not a deep dive is worth the time or potential cost to my sanity.  I'll update the thread if I find anything.