McCurdy AU300 Filament supply arrangement/ Hum question

[lassoharp]

From the schematic it appears the AU300 used either 12V or 36V on the filament supply.

Does anyone know why this arrangement was used here instead of a standard 6.3VAC ?

Or asked another way - Any reason why a 6.3VAC might create problems on this particular circuit?

I've combed through some  relevant threads but can't find one that addresses this question.

http://www.jlmaudio.com/au300mod.gif

 

[abbey road d enfer]

It looks like the arrangement allows powering from either 36V or 12V (by strapping 1-3 and 2-19). I don't think there's any significant difference between those two modes. The 36V mode may induce slightly more AC stray field, but I don't think it is is really of any serious consequence.
Practicality would prevail, 12V being much more common than 36.

 

[emrr]

I seem to remember the official psu's having either 12 or 36 DC avialable depending on the particular
model. Purely practical.  Some of the ampex cinema preamps are the same way.

 

[lassoharp]

Thanks abbey, Doug.

I've got this circuit on the breadboard now with a 6.3VAC and am having some issues with what appears to be AC hum.

I thought there may have been something in the circuit design(unusual vol control network?) I was missing that may have made it more susceptible to filament related hum issues. 

I'll have it running all day and hopefully get the problem isolated.

 

[abbey road d enfer]

Have you left the filament supply completely floating or grounded?
Usually, two 100R resistors from each leg of heater voltage to ground is enough to prevent any filament-induced hum problem.

 

[lassoharp]

The supply I'm currently has a grounded CT on the 6.3V winding.

Ok - I think I've found the cause.  The PS setup I was using was decoupling the output stage from V1 and V2 instead of having them all on the same leg as shown in the original.  I see that the Stalevel (3 stage PP) does this as well.  A similar generic circuit in the RDH decouples output from V1 and 2 as I had. The Pultec(2 stage PP) decouples each section. It doesn't seem as if it should make a difference one way or the other.

I connected all stages to a single leg and the hum was gone.

So I see that V1 and V3 are in phase - V2 out with 1 and 3.  This is raising more questions in the direction of decoupling practices.  I had been used to following the general rule of decoupling at least V3 from 1 & 2 when there are 3 stages - often decoupling all three in high gain/big coupling cap situations.  Even some 2 stage amps (RCA BA-2) have decoupling for each stage that doesn't appear to be for frequency compensation purposes (10K/10uF).  I'm really just thinking out loud here trying to make sense of each case relative to what I observed.  Any comments or more discussion is most welcome. 

 

[PRR]

If you are heating with DC and making DC with old-school rectifiers, 36V 0.3A is easier than 12V 0.9A.

The circuit is push-pull. In theory, B+ decoupling is not needed. But should not matter. That it matters for you suggests some problem, though I can't guess what.

 

[lassoharp]

If you are heating with DC and making DC with old-school rectifiers, 36V 0.3A is easier than 12V 0.9A.

The circuit is push-pull. In theory, B+ decoupling is not needed. But should not matter. That it matters for you suggests some problem, though I can't guess what.

The hum eventually came back with the altered decoupling set up

So far it's  been narrowed down to perhaps a grounding issue.  For some reason one of the input transformers I've tried works hum free, two others do not.  If it should make any difference, the two that do not buck the hum are UTC A-10 and A-18 - both designed for PP grids.  The one that does buck the hum is a Gates/Triad input - also designed for PP grids.  Tying the shield to ground has no effect on the two UTCs.  I've also tried tying the pri CT to ground with no attenuation.  I thought I would mention this in case it should point to something I'm not seeing.

 

[lassoharp]

Problem solved - I was somewhat surprised to find out the hum was simply being picked up from the PT which sat nearby on the work bench.  Surprised because nearby in this case was 12-15 inches away, which gets near the limits of a lot of chassis layouts.  This was the point where rotation would approach a null and hum had only slightly faded in the worst position.  Useful to note that the best null at this distance would still be unacceptable.  22in did the trick.

I've often worked with ITs in this distance range to the bench supply with no problems.  I was however using a different PS/PT than what I normally use - perhaps with a stronger field/less shielding.

Good to know which can has the higher immunity.  Chalk up another to testing the limits of how well your prospective PT gets along with your input transformer.

Other than that - 6.3VAC with CT is working just fine.

 

[emrr]

Cool.    Lots of 1930's literature talks about putting B+ and filament transformers at least 3 feet away from any audio transformers.    A strong argument for outboard PSU's. 

 

[lassoharp]

Officially it took a combination of a floating heater arrangement ala a Pultec EQP1-A and an artificial filament CT in lieu of the not so great one on the PT I'm using.  This reduced hum to an acceptable level = virtually unhearable with the output ran into a 50W power amp cranked wide open. 

A 100-200r pot worked best in this case vs two 100r resistors.  Elevation voltage was ~ 46VDC.

I suspect a better PT would have eliminated or drastically reduced both hum problems right from the start.