Making an Altec 1592a Quieter

[sircletus]

I'm playing around with an Altec 1592a I bought awhile ago as a project to futz around with.  It's in pretty good shape and came with 3x 1588B, 1x 1588C, and a 15095A line transformer.  Everything works fairly well, the noise floor overall is good, but there's a lot of 120Hz.  Noise floor overall is hovering around -90 dBu (Wow!), but 120Hz is sitting at -51dBu.  And that's with the master volume all the way down!  Modules in, modules out, doesn't matter.

Every electrolytic capacitor in this thing has been replaced fairly recently (I did all of them on the power amp board), with Panasonic FCs in all power supply positions (photo attached).  Original filter caps are in place, but out-of-circuit. I did notice, however, that the Panasonics' negative terminals are all connected to the original filter caps' negative terminals, as they look like they make a convenient binding post for multiple leads, and are all connected to chassis.  Not sure if this would be potentially problematic or not.

Anyone have any ideas where I should look and/or what I could change or eliminate to get this 120Hz under control?

Thanks!

 

[gridcurrent]

not seeing the capacitor to chassis connection.
one should localise power supply ripple before connecting circuit common (0V) to chassis.
did you read  the grounding note in the manual ?

 

[sircletus]

not seeing the capacitor to chassis connection.
one should localise power supply ripple before connecting circuit common (0V) to chassis.
did you read  the grounding note in the manual ?

Yes, I saw that note.  There's a circuit ground connection to chassis that's clearly original near the output transformer.

The negative lead of EACH of the Panasonics is tied to a different old capacitor negative lead. Then all of the old capacitor negative leads are tied together, which is how it was wired from the factory.  For shits and giggles I did the "right" thing and tied all of the Panasonic negative leads together to circuit ground, which then connects to the chassis at the aforementioned point.  The result?  A 40dB increase in noise.  Go figure.

 

[ruffrecords]

If the interference is at 120Hz this implies it is insufficient B+ smoothing. Have you changed C25 and C26?

Cheers

Ian

 

[gridcurrent]

There's a circuit ground connection to chassis that's clearly original near the output transformer.

The negative lead of EACH of the Panasonics is tied to a different old capacitor negative lead. Then all of the old capacitor negative leads are tied together, which is how it was wired from the factory.

I've seen serious mistakes on original, unmodified Altec gear, notably 1594b amplifiers, so not all that surprising.
Your opportunity to rewire and do it right.

Are you confident about the new filter caps ?  Can you test the old ones ?

While not the problem, easy to delete the zener and install a 7812 regulator for the input devices.

 

[sircletus]

I've seen serious mistakes on original, unmodified Altec gear, notably 1594b amplifiers, so not all that surprising.
Your opportunity to rewire and do it right.

Are you confident about the new filter caps ?  Can you test the old ones ?

While not the problem, easy to delete the zener and install a 7812 regulator for the input devices.

Interesting to know!  I'm confident enough in the new filter caps, FWIW, but I can test the old ones.  Not sure I care enough about the zener, if I'm being completely honest, though I could probably fit one on the little turret board where all but the rectifier diodes are mounted.

I have one more idea I'm going to try for circuit ground layout for those filter caps before I just pull 'em and hook up the old ones again (which I'll test, of course).

Thanks for the responses, gridcurrent!

 

[sircletus]

I've seen serious mistakes on original, unmodified Altec gear, notably 1594b amplifiers, so not all that surprising.
Your opportunity to rewire and do it right.

Are you confident about the new filter caps ?  Can you test the old ones ?

While not the problem, easy to delete the zener and install a 7812 regulator for the input devices.

Plot twist!  Pulled all new filter caps, put originals back in place.  Overall noise floor is up just a hair, but 120Hz is down 10dB.

Looking around in this thing, grounding seems weird.  Pin 1 on all input XLRs are soldered to a bus bar with a direct chassis connection via a standoff.  Could have just jumpered each pin 1 to the XLR chassis tab, but whatever.  Fine. Same bus bar is also connected to all octal sockets' pins 2 and 6 and...it's also got a lead soldered to audio ground: this is where the ground terminal for gain pots for each of the inputs are bussed and connected to the chassis.  There's ANOTHER audio ground-to-chassis connection at the output transformer.  That appears to be the circuit ground for the power amp.  Messed around a bit with lifting each and connecting them to the same point on chassis with no noticeable improvement in hum.

For shits and giggles I lifted both of those chassis connections to see if that isolated circuit ground from chassis and...nope.  So there's at least one circuit ground-to-chassis connection somewhere else in there that I can't find.  Naturally.  :

 

[emrr]

These problems can be infuriating and tough to find....as you're seeing. 

I've been working on something lately that has several shield plates and cases, and it has to be completely reassembled to assess any changes, makes for very slow going. 

 

[lassoharp]

You're likely dealing with proximity problems. The bottom of the PT sits just inches away from the amp card as I recall. I went through the same struggles you did some years ago and wasn't able to improve the hum level very much. Try opening and closing the front panel and see if the hum follows as the amp card moves away from the PT bottom. You might be able to fashion a metal cover for the card which may help some.