newbie here with basic questions

[dbw]

Hi, I've been trying to learn about tube amps from books and stuff but I'm still pretty new at this. Can you guys answer some questions for me?

Is there any reason not to share the cathode RC network between gain stages, if they both want the same values anyway?

Here's a similar question: Say I have 2 channels that I want to mix together, each ending in a common-cathode stage. I could use an anode resistor and coupling cap on each of these, then mix them with resistors. Or I could short the anodes together and use a single anode resistor and coupling cap. What are the advantages/disadvantages of this?

Thanks!

[pdf64]

See thread
http://music-electronics-forum.com/t14638/
for some viewpoints on this.
If I've understood the arrangement you've put forward correctly, I would link anode to anode and cathode to cathode if the cathodes were to be fully bypassed, but keep them seperate if the cathodes were to be only partially bypassed. Peter.

[dbw]

See thread
http://music-electronics-forum.com/t14638/
for some viewpoints on this.
If I've understood the arrangement you've put forward correctly, I would link anode to anode and cathode to cathode if the cathodes were to be fully bypassed, but keep them seperate if the cathodes were to be only partially bypassed. Peter.

Thanks Peter, that link answered my first question. I'm not sure I understand your answer to the other question... surely I can link the anodes together regardless of whether the cathode is bypassed?

[loudthud]

On the common cathode question, there will always be a little bit of crosstalk. It's up to you to determine if this is tolerable.

On the common anode question, you can just connect the anodes together. There will be some signal loss. This is know as the "tweed mixer" used on some early Fender designs. You could use a pair of "mixing" resistors and then a single coupling cap. There will be some signal loss but not as much as the single anode resistor. You'll see this on some Blackface Fender amps. If you use a pair of coupling caps and then the mixing resistors the result is about the same but you have the option of using different caps for different amounts of bass. Anytime you have two mixing resistors, you can connect a small cap (250pF or so) across one resistor that will brighten one channel and darken the other.

[R.G.]

Is there any reason not to share the cathode RC network between gain stages, if they both want the same values anyway?

As with all electronics issues, it depends.

The whole point of a cathode resistor is to limit the current that flows through the tube it serves. It does this by converting the current which flows through the plate to the cathode and then to ground into a voltage. The grid is tied to ground by a resistor and pulls no current to speak of, so the cathode is forced higher in voltage than the grid. This has the effect of decreasing the current which flows in the cathode to create the voltage.

It's a negative feedback setup. It stabilizes the DC current in the tube. Now if you put a capacitor across the cathode resistor, or across part of the resistance by splitting the cathode resistor into two parts, the capacitor shunts all the AC parts of the signal above a certain frequency to ground, and this increases the gain of that tube section. So there is a frequency-gain effect if you use a capacitor.

That all being said, back to your question. Can you use a common cathode resistor and cap? Sure. But there are some things to be sure of. If all the previously-separate tube sections have the same value cathode resistor, and equal plate resistors, then it's fine. This is done a lot in power amp setups where power tubes share a cathode resistor.

To share a cathode resistor over several tube sections, you need to ensure that afterwards they have the same DC current as they had before. So for N tubes, you want a shared cathode resistor of 1/N times the resistance of the previously separate cathode resistors. With N times the current and 1/N times the resistance, the net DC voltage is the same as it was, and all of the tubes will be similarly biased to their previous condition. This is very much trickier if the plate resistors and cathode resistors of the tube sections was not identical before you started this.

If there are cathode caps, there are other issues. The cathode is a feedback point for AC signals. If you common all the cathodes, you had BETTER bypass all of the signal to ground with a BFC (Big Freakin' Capacitor) or you will have a condition where later tubes are sending signal back to earlier tubes. 50% of the time, this will be positive signal feedback and the thing will oscillate like crazy. Or you could get lucky and just have peaky, squarky sounding frequency response. Bad juju.

Here's a similar question: Say I have 2 channels that I want to mix together, each ending in a common-cathode stage. I could use an anode resistor and coupling cap on each of these, then mix them with resistors. Or I could short the anodes together and use a single anode resistor and coupling cap. What are the advantages/disadvantages of this?

A mixing stage is one place where this can work very well. Anodes are a high impedance place, so tying two of them to one plate resistor makes sense if mixing is what you were trying to do.

[Bruce / Mission Amps]

On the common anode question, you can just connect the anodes together. There will be some signal loss. This is know as the "tweed mixer" used on some early Fender designs.

I've used it and also heard that called the tweed mixer too... but without spending a lot of time... is there a Fender schematic out there that shows it?

[Enzo]

A number of them, here is the Twin 5C8

http://www.schematicheaven.com/fende..._5c8_schem.pdf

Also:
Bassman 5B6
Super 5C4, 5D4
Pro 5A5 thru 5E5

Those are all also at SChemtic Heaven.

Maybe others...

[dbw]

The early Hiwatt preamp (cloned by Ceriatone) has 2 channels which are mixed with resistors before going into another gain stage, which is 1/2 a dual triode. The other half of that tube is unused.

The later version (http://schematicheaven.com/newamps/hiwatt_50w_dr504.pdf) got rid of the mixing resistors by using the other half of the tube for the second channel and mixing them at a common anode resistor, ...the "tweed" mixer, or so I'm told

I was sort of wondering why Hiwatt made this change.... does it sound better?

On the common anode question, you can just connect the anodes together. There will be some signal loss. This is know as the "tweed mixer" used on some early Fender designs. You could use a pair of "mixing" resistors and then a single coupling cap. There will be some signal loss but not as much as the single anode resistor.

Why is the signal loss worse with the common anode? I would have guessed the other way around