Separate Cathode Bias Resisters for Dual 6V6 Amp -> Garter Biased 6V6 Songwriter

[maxkracht]

From the article, he mentions that the bypass caps need to handle twice the wattage with this method, but I assume the resisters probably only need half the power handling since there are twice as many.

Voltage between cathode and ground doubles, so you need bypass caps rated for double the voltage. I'm guessing the voltage across the cathode resister is around 20v, double that, so a 50v cap is probably fine, 63 or 100v is better. (it's hot over there so added margin is nice)

As far as the tube is concerned, there is no difference between the single cathode resistor per tube option and and the garter circuit, aside from the extra voltage dropped on the added resistor. Pretend where the grid leak resistor connects is ground for all bias calculations. You are still measuring plate to cathode and across a single cathode resistor. Use the same wattage resistor you would use normally. You could probably get away with 2w resistors but 5w are cheap. Hopefully one of the smart kids will step in if i'm getting any of this wrong.

[bcmatt]

From the article, he mentions that the bypass caps need to handle twice the wattage with this method, but I assume the resisters probably only need half the power handling since there are twice as many.

Voltage between cathode and ground doubles, so you need bypass caps rated for double the voltage. I'm guessing the voltage across the cathode resister is around 20v, double that, so a 50v cap is probably fine, 63 or 100v is better. (it's hot over there so added margin is nice)

As far as the tube is concerned, there is no difference between the single cathode resistor per tube option and and the garter circuit, aside from the extra voltage dropped on the added resistor. Pretend where the grid leak resistor connects is ground for all bias calculations. You are still measuring plate to cathode and across a single cathode resistor. Use the same wattage resistor you would use normally. You could probably get away with 2w resistors but 5w are cheap. Hopefully one of the smart kids will step in if i'm getting any of this wrong.

Ok, thanks.
Well I do have 5x 470/5w so I guess that's what I'll use to start for my 4 resisters. I also happen to have a couple 47uF/100V caps, so I guess that is deciding that value for me too. Actually a just found a couple little radial 22uF/100V too. I guess this will more depend how I'm trying to lay this out.
I thought I was clearing tonnes of space, but as soon as I made a stack of filter caps internally, things are a little tighter.

[bcmatt]

While waiting to get close to the filter caps and power tubes, I got the VVR installed. I'll wait for the silicone to dry over night.

20241019_203539.jpg

Maybe I can get back to it tomorrow sometime and hopefully get this garter wired up.

[maxkracht]

I thought I was clearing tonnes of space, but as soon as I made a stack of filter caps internally, things are a little tighter.

It definitely makes things a bit cramped around the power tubes. I thought my layout had plenty of space on paper vs what it was in 3d. Not helpful for your current setup and current stash of parts, but modern radial caps are great for fitting into small spaces around the preamp. More room for other stuff and potential for an improved ground scheme. Nichicon makes some surprisingly small 10-15uf 450v caps with good specs.

[bcmatt]

Well, I thought I had finished wiring it up today with the Garter circuit. I put some known new well matched tubes and fired it up...
but it's super quiet.
I'm thinking I will skip the VVR from the circuit for now to make sure it's working first... I just couldn't resist.

I feel like I must have missed grounding something maybe. I'll pour over my wiring to see what I might have screwed up.

The garter circuit I put on a terminal strip to the right of the filter caps, and the 220 resistors I left on the tube sockets.

I wonder if installing a speaker jack could have done something since the ground(-) would touch the chassis... not that I really know which would be the negative from that OT.

20241020_213238.jpg

[pdf64]

The OT galvanically isolated the speaker output from the output valves, there's no DC interaction between them.

[bcmatt]

The OT galvanically isolated the speaker output from the output valves, there's no DC interaction between them.

Good call. I switched the jack type. I had obviously messed with something I didn't understand. (But isn't that my entire life?)

That brought the volume back.

The VVR is doing nothing. I wonder if I fried either the MOSFET or zener with my other indiscretion.
I also think the treble pot is too far gone and needs to be replaced.

I'd like to check some voltages now.

I'd also like to figure out an appropriate value bleeder resister to put on the first filter cap.

[maxkracht]

I'd also like to figure out an appropriate value bleeder resister to put on the first filter cap.

Bleeder resisor value doesn't matter too much, you can go arbitrarily large and wait longer for it to bleed, just don't go too small. 220k-1M is probably a good range, 1W+ is probably fine.

Is that red (screen?) resistor actually that dark in the center or is it just lighting?

[bcmatt]

I'd also like to figure out an appropriate value bleeder resister to put on the first filter cap.

Bleeder resisor value doesn't matter too much, you can go arbitrarily large and wait longer for it to bleed, just don't go too small. 220k-1M is probably a good range, 1W+ is probably fine.

Is that red (screen?) resistor actually that dark in the center or is it just lighting?

Thanks.
Ya that is a dark red. Those are the original B+ dropping resisters for the filter caps. They seemed to all still measure about 10K so I didn't change them at this point.
Here are a couple more that I pulled out:

20241021_093431.jpg

By the way, thus Blumlein circuit is fascinating and the tubeCAD experiments. Have you considered trying any of the transistor-based designs that JRB comes up with near the end of this article you shared?
https://www.tubecad.com/2005/May/blog0046.htm

Or are they just more complicated methods to lose less B+ voltage? (Which I dump anyways with VVR)

[bcmatt]

The VVR was fine; I just had the pot wired backwards... noticed while I was checking my B+ voltages.
I still want to measure all the voltages but side-tracked by this treble pot that was measuring 3.5m across, but also cutting in and out throughout the wipe. I put a 1M pot in and also noticed mine doesn't have this cap so I just added it since I don't love the distortion sound at all. I'm also going to try it connected to the my creamback cab to see how much is the speaker's fault.

Garnet 45B.jpg

[maxkracht]

Ya that is a dark red. Those are the original B+ dropping resisters for the filter caps. They seemed to all still measure about 10K so I didn't change them at this point.

I'm familiar, you find those in British/German stuff from the era. Just making sure the center hadn't darkened indicating overheating.

By the way, thus Blumlein circuit is fascinating and the tubeCAD experiments. Have you considered trying any of the transistor-based designs that JRB comes up with near the end of this article you shared?

Haven't tried any of the transistor versions, I think they are just more efficient like you suggest. He warns on the more recent blog post not to use the older transistor version.

[bcmatt]

I'm familiar, you find those in British/German stuff from the era. Just making sure the center hadn't darkened indicating overheating.

You know, now that I look at it, I think it is a little darkened in the middle and perhaps the coating bubbled a bit. I think I will try to replace it with something higher wattage... But I may end up changing the value since I may be messing a bunch with this circuit. For one, my heater voltage is up at about 7.2VAC. I'll try a thermister on the main power input....

I took some DC voltage readings at Full Power and then just under 1/3rd power using a good pair of 6V6s. I'm curious how this will compare when I try some other random 6V6s.
That first triode runs some pretty high voltages.

Garnet 45B Cathode Garter TAD6V6s.jpg

[martin manning]

This is an interesting experiment. Looks like the power tube idle current stays very close at either extreme. Any idea how close the current would be without the garter circuit?

[bcmatt]

This is an interesting experiment. Looks like the power tube idle current stays very close at either extreme. Any idea how close the current would be without the garter circuit?

That is a great question. I was actually trying to think of a way to switch between regular Isolated circuit using a DPDT switch that I installed on the back panel... but got sidetracked by trying to get things working first. Let me know if you think it is possible to put on a switch.

Currently, I was re-measuring everything after installing a couple CL90 Thermisters to try and get the Heater voltage under control.
I remeasured with the good 6V6s at full voltage, then installed some old unmatched 6V6s to remeasure. I started with a couple old RCAs out of my old 74 deluxe reverb that had drifted. One of them was flashing a little and then started to make a loud hum, so I yanked the suspect tube and replaced it with the single mystery brown-based 6v6 that came with this amp. I was scared to turn it to full voltage right away so I recorded some voltages at half power on the VVR.
Here is what I found so far:

Garnet 45B Cathode Garter TADs and mismatche at half.jpg

Now I'm going to get brave and record these mismatched tubes at full voltage...

[bcmatt]

Ok, I tried out these mismatched tubes at full voltage and it seemed to go OK... so I played through them a bit. I actually think I liked the sound better than the healthy tubes. It's probably because the mismatch adds a bit of harmonic distortion... and since I don't like the sound of turning up the gain very much on this amp, adding a bit of harmonics to the clean tone sweetens it up a bit.
Here are the Voltages in Orange. Please let me know how I ought to interpret this and if I ought to measure something else for this experiment:

Garnet 45B Cathode Garter Mismatched Voltages.jpg

Do these seem like reasonable voltages to run on this thing? Or is it destined to explode, ruining my $10 investment?
So I think I do need to tweak this circuit a bit in the preamp at least. It may have worked fine as a bass amp, but for guitar, I think it could use a much smoother sounding overdrive, and maybe a bit more drive in general. I'm wondering about trying to adapt a bit of a Rocket style amp since the front panel has Volume, Bass and Treble. I'd be happy to give up one of the inputs for another control..

There are a couple holes in the chassis for other preamp tubes. I'm not sure whether the PT can handle more, but maybe that explains the higher heater voltage? There is also an unused triode since it utilizes a Cathodyne Phase Inverter.
Thoughts?

10dollar guts.jpg