Cathode resistor and IM distortion

[mike-p]

I have some unwanted IM distortion on my hodge podge mess (lite 11b with double power section running 4x 6p1p ) especially when running in pseudo SE mode with an mv on one half of the power section.

I currently have 16v across 220 ohm 5w cathode resistor bypassed with 2200uf cap, shared between each pair of power tubes.

I have 180 ohm 5w resistors handy but only 100ohm 1/4w to try the tail resistor/ unbypassed portion of the cathode resistor. Is 3 of those in parallel after the bypassed 180 ohm a hodge podge bodge too far? Just to try out it out.

[Stevem]

That's a unbelievable huge bypass cap if it's not a misprint?

[mike-p]

Apparently they are common on 18 watt marshall style circuits.

[maxkracht]

The 18w guys' theory is giant cathode bypass cap = closer feel to fixed bias. Haven't poked around the 18w sites in a while, but 1000uf used to be pretty common, I guess the value has inflated. I have played around with 470-1000uf, sometimes I think I notice a difference in feel with giant caps, and sometimes I can barely tell the difference between no cap and a 47uf. No idea why.

[WhopperPlate]

The 18w guys' theory is giant cathode bypass cap = closer feel to fixed bias. Haven't poked around the 18w sites in a while, but 1000uf used to be pretty common, I guess the value has inflated. I have played around with 470-1000uf, sometimes I think I notice a difference in feel with giant caps, and sometimes I can barely tell the difference between no cap and a 47uf. No idea why.

Some days it’s hard to finger a barre chord . Other days it’s hard to make any mistake. We are human .

[maxkracht]

maxkracht wrote: ↑Mon Feb 12, 2024 11:23 am
The 18w guys' theory is giant cathode bypass cap = closer feel to fixed bias. Haven't poked around the 18w sites in a while, but 1000uf used to be pretty common, I guess the value has inflated. I have played around with 470-1000uf, sometimes I think I notice a difference in feel with giant caps, and sometimes I can barely tell the difference between no cap and a 47uf. No idea why.

Some days it’s hard to finger a barre chord . Other days it’s hard to make any mistake. We are human .

I am the first to assume my own incompetence or tiredness above anything else, but I have noticed this on multiple occasions where a bypass cap just doesn't do much of anything or a very big bypass cap isn't different from a normal-sized one. Triple checked wiring, and confirmed by client asking what the hell does that switch do, i can't hear any difference. Has happened most when rebuilding old projector amps, so maybe it's due to poor low frequency response in the OT.

[maxkracht]

I have 180 ohm 5w resistors handy but only 100ohm 1/4w to try the tail resistor/ unbypassed portion of the cathode resistor. Is 3 of those in parallel after the bypassed 180 ohm a hodge podge bodge too far? Just to try out it out.

I think those resistors would let you know before anything bad happened. Could also run resistors for each tube, before the shared 180r, to get some extra feedback going.

[mike-p]

Yes sorry I was going to run the extra resistors from each tube to the shared resistor on the board. Do I need to double the value of the non shared part of the resistance? Ie going from a shared 220 ohm to a shared 150 ohm I need to make up 70 ohms equivalent shared resistance. Am I looking for 70 or 35 or 140 ohm for the individual cathode resistors?

[maxkracht]

I think you would want 140R if using one for each pair of tubes. Total is in parallel. So 280R if using one per tube?

[mike-p]

I'm really confused now. Currently each pair has 220ohm shared resistor bypassed with 2200uf as stated.

Case 1.Say I change that to 180 ohm bypassed plus 40ohm still shared then I've changed nothing except the bypassed portion of the cathode resistor.

Case 2. If I make the non bypassed part individual to each tube then I need to double it so; 180 ohm bypassed and shared plus 80 ohm individual to each tube. Am I correct?

Is there an advantage to case 2, ie extra localised feedback or is it just extra faff?

[maxkracht]

Sorry, I misread. Thought you had one single, shared, cathode resistor and were changing to 2 or 4.

Benefits for option 2: potentially allows a bit more mismatch between each tube and you are spreading the resistors' heat out a bit more if you're using something underrated for the job. Probably not a significant difference and no idea if there are any other benefits, haven't tried.

[maxkracht]

I have some unwanted IM distortion on my hodge podge mess (lite 11b with double power section running 4x 6p1p ) especially when running in pseudo SE mode with an mv on one half of the power section.

I've never heard of a pseudo SE mode. How exactly are you doing this? I would imagine some weirdness if half of the output isn't working.

[mike-p]

I called it pseudo SE, it's in a KOC book apparently, but its basically a mv for one side out of the phase inverter grounding the signal to 2 out of 4 tubes with a pot. I got it from the vajra schematic (pp-se mod) on 18watt.com there are various versions of it. I like it a lot except for excessive IM distortion. But I'd like less IM distortion when the amp is running all 4 tube normally too...

[maxkracht]

Weird, I would have assumed the OT would do some weird stuff with that setup...

Have you tried this?

https://www.18watt.com/storage/18-watte ... fo_311.pdf

[mike-p]

Yeah I thought so to but couldn't find any negative reports. I have the Paul ruby mod installed.