Cathodyne Phase Inverter Resistor Ratings with Grid Leak Bias

[Diablo]

I started to draw out the schematic of my old Capitol guitar amp and noticed that it has a Cathodyne PI. The PI triode is half of a 6EU7 tube (12AX7 equivalent) and uses a 39K resistor for the anode and a matching 39K resistor for the cathode. It has a 2.7M grid leak resistor from the grid to the cathode for grid leak bias. I'm wondering why the designer used 3-Watt carbon comp resistors for the 39K plates. Can current draw be that high under certain conditions with grid leak bias? I notice the identical 39K-3Watt resistors are used in four other places in the amp for voltage dropping between filter caps. In those spots, I would expect to see large wattage resistors. Do you think the designer just used the same sized resistors in the PI for convenience? The resistors are original to the 51 year old amp.

[mooreamps]

When I go to source parts, I don't always find "exactly" what I am looking for...

-g

[Rhodesplyr]

Do you think the designer just used the same sized resistors in the PI for convenience? The resistors are original to the 51 year old amp.

Yes. It would be nice to think that parts choices in vintage amps were always completely intentional, but the fact is that, sometimes, they just used what they had on hand to get them out the door. In this case, if they had a box of 3W resistors, there's no reason not to use them as plate/cathode resistors, even if they're overkill. Maybe it was more economical simply to order one type of resistor.

[J M Fahey]

Agree and add.
How do you know they are "3W" ?
Because of the size?
Remember technology advanced (51 years) and resistor sizes shrunk a lot.
They are probably 1/2W .
Their design, a carbon/ceramic paste (that alone takes up a lot of space), compressed inside a Phenolic tube (very poor/inefficient "heatsinking") is big out of necessity.
When I started this (late 60´s) there were still many of those around, specially in repairs , not any more over the counter.
Carbon film resistors (the ones most used today) have a carbon film on the outside of a ceramic tube: much cooler; thus, much smaller.
Metal film ones also coat the outside of a ceramic tube, but can safely work quite hotter, so they are even smaller.
I think a 50´s 3W carbon comp one should be as large as half your pinkie. No kidding.

[MWJB]

Building amps for profit, as a commercial enterprise, means keeping parts costs low. Factories make big benefit from bulk ordering, so if they are buying 39K resistors and they already are using 39Ks in the B+ rail for instance, then it makes sense for them to use the same resistors wherever they can, rather than order another parts line & pay more for each of the 2 different versions of the 39K.

As JM says, I'd be surprised if it was a 3W carbon comp...not impossible, just unlikely.

[Diablo]

Thanks all for confirming that the large resistors on the PI plates aren't really needed. The reason I suspect they are 3 watts is just based on the size. I replaced most of the original 1/2 watt carbon comp resistors and the replacements were the same physical size. Check out the orange / white / orange / gold resistors - one is sitting over a yellow Sozo cap. It's 3/4 inch long by 1/4 inch diameter.

[Tonewood]

Anyone know if you can put grid stoppers on these grid leak biased preamps and splitters without messing up the bias? Seems like changing the RC time constant may mess up the bias. I really don't understand this type of bias.

Thanks

[Bruce / Mission Amps]

Thanks all for confirming that the large resistors on the PI plates aren't really needed. The reason I suspect they are 3 watts is just based on the size. I replaced most of the original 1/2 watt carbon comp resistors and the replacements were the same physical size. Check out the orange / white / orange / gold resistors - one is sitting over a yellow Sozo cap. It's 3/4 inch long by 1/4 inch diameter.

Those big resistors you are pointing to look more like old early 1940's-1950's one watt carbon comp to me.

[Diablo]

OK, thanks Bruce. I'm reworking the grounds on the phase inverters (yes there are two, it's stereo) to see if I can lower the noise floor. I figured I might as well replace those large plate resistors since I have to desolder them anyway. I'm running the PI grounds directly to the PI filter caps.

[Diablo]

I finished the mods, installed the 1/2 watt PI plate resistors and moved the PI grounds from the local chassis ground to the PI filter caps, and that really did the trick on quieting the hum. That's been bugging me for 6 months and I was starting to think it was characteristic of the design with the split chassis.

[Gtr_tech]

What Bruce said. Those are carbon comp 1w...

[trobbins]

I doubt you would move the bias significantly if you added a moderate amount of grid stopper (eg. up to 100k), given such large grid leaks.