1st tube - triodes in parallel

[Bob M.]

Hi,

On my latest home brew - a tweed-ish thingy - I'm trying putting both triodes of V1 in parallel but I'd like to be able to switch it so I can use just 1/2 the tube (one triode) or have both sections of the 12AX7 or 12AY7 (whatever I end up using) in parallel. In switching this to parallel mode, I have to change the plate load resistor as well as the cathode resistor. In parallel mode, I think these resistors should be halved. Also, I have to connect the plates (pins 1 and 6) and the cathodes (pins 3 and 8) to one another in parallel mode. This calls for a 4PDT switch but I'd have to leave the grids (pins 2 and 7) permanently hardwired. Is there an easier way to do this? It's really only an experiment to see how it sounds. Any experience anyone has with this will be appreciated.

Bob M.

[loudthud]

A DPST would work. Each tube gets it's own plate and cathode resistor which stay connected all the time. The switch just connects the plates and cathodes together. The grids can stay connected all the time.

[Pedro Vecino]

The circuit of the Twin 57 can be interesting for you...
http://www.fender.com/support/amp_sc...Schematics.pdf

[cminor9]

I run the first triodes in parallel on a 5E3 and on a 6G3. Matchless did on their Lightning and Spitfire. I love it. I have no use for two separate channels, so rather than waste the triode...I like running them in parallel.

At one level, you don't have to do anything. Just run the plates into the respective coupling caps, then onto the volume control where they get tied together. I do this on my 6G3. Dirt simple, but it is a bit subtle. I guess I mainly like the idea of not wasting a triode.

Of course, you can tune it to your liking too and get a bit more sophisticated. I did this on my 5E3, and I detailed this mod here, complete with pictures: Tweed Deluxe Volume/Tone control mod - Telecaster Guitar Forum In my opinion, this is the best mod you can get for a 5E3 (along with lowering the coupling caps from .1 to .02). And I discovered it on my own! It's like giving the 5E3 a bass control that doesn't get farty -- it just fattens up the tone big time and is a real winner for single coils. I guess in a way it's the same thing as jumpering the two channels.

Before I found that mod, and what probably led me to it, I tried doing something like Bruce Collins did on his soulkicker amp with switchable cathodes (he detailed it here, thanks Bruce! Single channel 5E3). I liked this except 1) the switches popped when flipped, and 2) I wanted the ability to blend the tones. With the switches, I found I just ran it one way. With the triodes tied to the the two volume controls on a 5E3, I use different settings for diff guitars, which IMO is how it should be.

I read somewhere that running them in parallel has a canceling effect that will make things quieter. Not sure about that. It didn't make a difference for me, but then again, I don't have any real hum problems other than my guitars.

[Bob M.]

I guess my understanding was to halve the cathode resistor value when tying cathodes together, pins 3 and 8 on a 12AX7A. So instead of Fender's choice of a 1.5K cathode resistor, one should use a 750 ohm resistor for cathodes in parallel. Fender usually used an 820 ohm resistor when tying the cathodes together, for example, at V4 on a reverb amp (close enough, I guess).

Also, shouldn't the same principle apply for the plates (pins 1 and 6)? I guess you could just use the same selection of cathode and plate load resistors for both modes but the circuit wouldn't be optimized - it would favor one mode or the other or some sort of compromise in between. As it's an experimental idea, I'd kinda like to get the textbook part right.

Thanks,

Bob M.

[MWJB]

"I guess you could just use the same selection of cathode and plate load resistors for both modes but the circuit wouldn't be optimized" I'm not sure that optimised is the right word. Fender favoured the 100K/1.5K per triode combo, but occasionally strayed to 220K plate resistors and 820ohm & 2.7K cathode resistors.

Generally, they were aiming for a middle of the road sound.

As you go down from 1.5K per triode the tone will become fuzzier/hotter, as you go up the tone will have better fidelity.

Let's face the fact that you want some difference between the 2 settings anyway, so why not fit a 1K cathode resistor and 100K plate resistor, when you double up the triodes you get 2K/200K effectively. See how you go, if it's too dirty set to one triode, perhaps try 1.2K etc...

Limiting factor may be that with too small a cathode resistor value the amp may choke up under heavy chording.

Don't worry too much about the text book, your ears will tell you what the best compromise will be, as long as you don't go mad (<820ohms/>2K) you won't damage anything.

[Bob M.]

Good suggestion. I guess it's that 4PDT switch I've got lying around and I'm dying to use it. You're right, the ears are the final adjucator, no matter what the math says. Thanks, Bob M.

[cminor9]

FWIW, I tried adjusting the cathode resistors, and the plate resistors. I tried 220k, even a 270k plate resistor on my "hot" channel. I also tried 1.5k up to a 2.7k cathode resistor, maybe even stuck a 4.7k in there IIRC.

I put all the resistors back to stock values*. Why? I simply found I got more bang for the buck fiddling with the cathode bypass caps (larger values = more bass frequencies get through, smaller values focus the gain on the high frequencies) and the coupling caps (again, larger coupling caps for fatter but less cutting sound, smaller values for less bass but a nice lead tone. That, and I wanted the ability to twiddle a knob to get back to a stock 5E3 sound. That may or may not be a concern for you though.

There are all kinds of good threads on this forum that deal with biasing preamp tubes, I learned a lot from reading them.

* Stock value for v1 on a 5E3 being 100k plate resistors and an 820ohm cathode resistor with both cathodes tied together. Since I split the cathodes so that I could tune them via the cathode resistors, 1.5k on each cathode is roughly equivalent to an 820 on both. Maybe that answers your other question, too. Yes, if you have two cathodes tied together going thru an 820 ohm resistor, if you split them up then you should use 1.5k to get the same bias.