Gates Radio SA-39 limiter restoration and modification

[lassoharp]

http://bit.ly/1arTLNl

Recent eBay listing of a Gates SA-39B with a replaced (output?) transformer.
Anyone know whether this is a big change in sonic character?

That is a replaced power transformer, not output.  Output looks original.  The power transformer should have no effect on how the unit sounds.

 

[dustbro]

Hey guys,
Looking for some tips from the Gates gurus. I have an SA-39 that has a 60hz hum. Can't quite track it down.
So far I have replaced C10-C11, tweaked R43 to 275V, re tubed everything except the 6L7's and lamp,  and verified that r44 was within spec.
After adjusting R43, i can no longer get my meter to zero. At the minimum value, the pointer is sitting around +2.  I also noticed that adjusting the meter zero pot I get audible "scratchy" noise at the output. Should I be hearing that??
Tapping on any tube makes the hum go away, but as soon as signal is fed thru the unit, 60hz comes back.
Any thoughts?

 

[QNote]

Hey guys,
Looking for some tips from the Gates gurus. I have an SA-39 that has a 60hz hum. Can't quite track it down.
So far I have replaced C10-C11, tweaked R43 to 275V, re tubed everything except the 6L7's and lamp,  and verified that r44 was within spec.
After adjusting R43, i can no longer get my meter to zero. At the minimum value, the pointer is sitting around +2.  I also noticed that adjusting the meter zero pot I get audible "scratchy" noise at the output. Should I be hearing that??
Tapping on any tube makes the hum go away, but as soon as signal is fed thru the unit, 60hz comes back.
Any thoughts?

I'll chime in, since I currently have many hours into a 39B.  Post your current readings for J1 through J6 as per the manual (using a 1/4" TS plug and current meter).

 

[andre tchmil]

late night knock on the door.
Can anyone tell the cosmetic  difference  between a SA-38 and SA-38 A version ? and the specific build date between the two ?

 

[emrr]

Recessed meter, then non-recessed.  Don't recall the changeover year.  Early 1950's. 

 

[andre tchmil]

thanks

 

[buildafriend]

Hey guys,

I'm working on breathing the life back into a somewhat neglected SA-39. Model Number M-2996A. First thing was the meter wouldn't zero. I changed R47 from 220 ohms to 150 ohms and now it zeros. The thing I'm trying to tackle is a little bit of DC bouncing happening that's visible on my oscilloscope while changing settings and what might be some sub-audible oscillating that is just slightly noticeable inside of the sinusoid lines. I think retubing and recapping are in order. I also pulled out and replaced a burned up looking wire on the HT line, post choke. R46 seems to still be in spec. I haven't looked at R44 yet. It would be nice if I could confirm I'm using the exact correct schematic.

I was wondering if anyone could share some test points with me so I can see what you're getting at J1 through J6. I'm also curious what voltages you get at the NE-45 and how anyone using a zener there might be doing with it at this point in time. What zener voltage and wattage did you guys end up going with? I might give that a shot.

Any advice would be appreciated.

All the best,

 

[buildafriend]

Here's what I've got while applying a +4dBu signal:

J1 3.83 VDC @1.72mADC and 0.014 VAC
J2 3.83 VDC @1.55mADC and 0.014 VAC
J3 0.00 VDC @0.98mADC and 0.00 VAC
J4 0.00 VDC @0.85mADC and 0.00 VAC
J5 13.43 VDC @36.3mADC and 0.010 VAC
J6 13.53 VDC @35.4mADC and 0.002 VAC

And at the regulator bulb I have 77.7 VDC with no AC component

 

[Rob Flinn]

I had some sub audio oscillating on an SA39 I was fixing.  It was caused by not having very well matched valves in the balanced part of the circuit.  Looking at the voltages you have measured it appears that your valves are not too well matched.....

 

[buildafriend]

How tight would the cathode current need to be for the tubes to be matched considerably well?

Fixed typo*

 

[Rob Flinn]

I think J5 & J6 would be alright, but J1 & J2 and J3 & J4 are not close enough.

Also, if you haven't worked it out there is very high V on the meter terminals.  I normally put some insulating tape over them & the PSU switch when I work on these to save myself having an afro hairstyle.

 

[emrr]

The 6H6 and 0.1 timing cap can both be the culprit, along with all the various grounding possibilities of the overall unit.  Lose the side chain and look at it again.  ALWAYS do this with any vari-mu.  The 'best' static current matched 1612's may not be the best in action, since it's about the match over the current curve.  I've seen ones with bad max current match which worked best out of a batch at 3-5 dB GR.  Match for several conditions with a half dozen or dozen and then go by ear after that, also use something like kick drum or piano with low bass and look at waveform distortion on the attack in your DAW. 

 

[buildafriend]

Like you stated these all seem to vary a bit in their design. One of the transformers here looks like an add on, I can tell by the non factory cut out. I'll post a pic soon. I have to lock down where the timing cap is. How is your zener holding up? What zener voltage did you go with? It's looking like the main tube characteristic measurement tools that I'll be able to use are the jacks for measuring cathode current, a hickok 532, and whatever else I end up observing with couple of DMMs and an RMS meter.

All the best, 

 

[buildafriend]

I normally put some insulating tape over them & the PSU switch when I work on these to save myself having an afro hairstyle.

My forearm worked that out before my brain did. The tape found it's way there before this wise warning  ;D

 

[buildafriend]

For what can be expected from such an old design, I have it working well at this point. I only wish it could catch LF transients faster and that it had a hair less hum. While having practical expectations it works great and shines. While slamming snare tracks it misses the transient. Of course those are extreme settings.. and tubes aren't always all that fast. It  would benefit from good matching to help common mode and from DC heaters. Self noise exposes some 60 cycle to the ears and to the spectrum analyzer with the output cranked.

 

[Rob Flinn]

Did you check those resistors for tolerance ?    All the Gates limiters I have serviced needed some of them replaced.

. I only wish it could catch LF transients faster

I don't use this sort of limiter for material where I want the bass transients caught.  In fact I use them where I don't want them caught which is one of the characteristics I like about them.  If you want to nail the bass attack, then that is the job of a different compressor/limiter imho.

 

[buildafriend]

There is quite a bit of R drift happening in more than one of these old carbon composition resistors..  :-[ Someone beat me with replacing the more common ones but they mostly seem to have gone under either slightly or too much. The large 20W 1500 Ohm resistor looks a little bit fried too, but is still in spec.

 

[emrr]

Oh, you haven't done the DC heater mod?  It's a must.  I never heard a usable hum free unit without it.  These things are designed for 3-5dB gain reduction, so it's special effect past that. 

 

[buildafriend]

I'm curious if people usually do the DC heater mod on only the first one or two pairs of tubes or if they do just about everything that's not a rectifier

 

[emrr]

SA-39B schematic.  It's on there.  3 parts.