4 Tube Amp From Scratch

[TubeNoob]

Potentio-what?!?

The Kalamazoo has the cathode of the 2nd preamp stage tied to the cathode of the trem circuit; oscillating the bias. In this case, both cathodes feed a single cap/resistor to ground - putting a pot in between them means one of them won't be.

Given this situation, where would this potentio-thingy and cap go?

Kalamazoo Schematic:
Kalamazoo Amp Field Guide: Model 2 Schematic

[überfuzz]

Sry, I didn't knew you had an oscillator all fixed and neatly placed in the circuit. Seems you could simply replace the R11 resistor with a potentiometer + resistor.

[TubeNoob]

Interesting, ok. 'Cause R11 on the Kalamazoo schematic is carrying over 200v; didn't realize you would put that much juice thru a pot.

[überfuzz]

Eh hmm, lets put it this way. Imagen a system of lakes and waterfalls. Resistance in our schemes are like water falls in our analogy. The pot we're talking about is going to be like a little sweet water fall sitting between two beautiful lakes high up in the mountains.

See? The voltage drop over the potentiometer is going to be dependent on the resistance over the potentiometer. You're just going to use the potentiometer to slightly change the amplitude of the oscillating signal imposed in the bias circuitry. Try to find a pot + resistor combination that is: pot at half + resistor = R11.

Edit, oh about the to much juice. If you mean juice = voltage*ampere you're right. In this example you're going to have 20 to 40 volts tops. So, it's hunky dory.

[TubeNoob]

That was a loverly description, thank you!

So, in fact, the voltage is 'high' there but the tube is drawing very little current thru a resistor only allowing a very small amount of wattage to hit the pot. Correct?

Updated schemo attached.

Added a presence control to NFB (at the top); ripped it off the Fender 5E7.

Inserted a potentio-mabob in series with the trem's plate resistor for 'intensity'.

Anything jumping out at anyone or dare i say that soldering time is nigh?!

[überfuzz]

I woke up thinking, +200VDC on a control-pot to chassi ground, is that something we want? I've never done it, and honestly don't know... :-/ It might be a good idea to add a couple of coupling caps and a big resistor to ground. So that you have the oscillation traversed over the pot at a low potential.

+200VDC ---||--- 0VDC ---||--- +200VDC

Your potentiometer and resistor goes in series with the 0VDC line. Add a BIG resistor, a couple of MOhm, from the 0 VDC line to ground. Highpass filters, the capacitors and the big resistor is going to work as a highpass filter. Google how cutoff frequencies are calculated in a highpass filter to see how to match the capacitors and the big resistor.

[TubeNoob]

Whoa, ok. Out of curiosity, what about controlling the grid of the trem tube, instead?

[überfuzz]

Whoa, ok. Out of curiosity, what about controlling the grid of the trem tube, instead?

The oscillator is stabile, don't mess with it.

[Chuck H]

I think the two of you may be on different pages here. I think uber is suggesting using the oscillator plate modulation to move the cathode bias and I think Toob is thinking to control the oscillator with a pot in place of the plate resistor while still coupling the cathodes.

[TubeNoob]

Hi Chuck H. Thanks for the input!

So, ok, how about the attached schematic / idea...i know my kung fu is unorthodox...but...

What if there is a resistor and bypass cap on both tubes' cathodes - but the pot (R18) switches between which one the preamp's 2nd stage tube 'sees'?

So presumably, at no time is a cathode without a resistor/cap to ground, but on one side you have the trem and on the other you don't.

[Chuck H]

It's good to see a piece of an actual schematic! But... It's pretty easy to spot two problems that would hamper functionality. It's alright. That's why we do this, right? So...

I see you have your NFB introduced at the same cathode where the trem is intended to modulate the bias. It won't work like that. With the cap fully bypassed there won't be any signal to interact. Introducing NFB at a preamp cathode only works with an unbypassed or only partially bypassed cathode resistor.

And if you re examine your schem you'll see that with the proposed intensity control the cathode resistance of V2A would quickly adjust out of any useful bias and cutoff the tube.

The idea of a variable plate resistor intrigued me for a moment. I've had too much trouble with reducing gain on working oscillators. They tend to stop oscillating rather than reduce in intensity. Though uber didn't describe it perfectly his later discussion of coupling the plate via a cap was a clue that he may be proposing a modulation of the V2A cathode voltage with the oscillator plate rather than coupling the oscillator cathode to the V2A cathode. What I'm considering is that if this oscillator has an action you like it's important to keep it the same, right down to duplicating the stock amps voltages, otherwise it will sound very different or may not work at all.

[überfuzz]

The idea of a variable plate resistor intrigued me for a moment. I've had too much trouble with reducing gain on working oscillators. They tend to stop oscillating rather than reduce in intensity. Though uber didn't describe it perfectly his later discussion of coupling the plate via a cap was a clue that he may be proposing a modulation of the V2A cathode voltage with the oscillator plate rather than coupling the oscillator cathode to the V2A cathode. What I'm considering is that if this oscillator has an action you like it's important to keep it the same, right down to duplicating the stock amps voltages, otherwise it will sound very different or may not work at all.

Yeah, this is kind of the reason I told him not to mess with the oscillator it self as it seem stable. My idea was to rather tamper with the affect the oscillator has on the circuit. The coupling capacitors was just a trick proposed so that he doesn't need to have have the +200VDC to ground in the intensity pot.

[überfuzz]

tubenoob - An oscillator is kind of tedious to set up. You need to analyse the frequency behaviour an the feedback necessary to maintain the oscillation. Your oscillator looks like a phase shift oscillator, which would suggest that the signal is attenuated 29 times (it's been a while since I vent to school, but I'm pretty sure its 29 times). Now you have to set up the 3 stage oscillating stage so that is has an amplification factor of 29, and a frequency range you please with. In your case this would be a little bit like inventing... the wheel. The thing with the alteration of the amplification of the oscillator tube is that you run the risk of altering the amplification needed to maintain the oscillation. Well, this kind of boils down to, tamper with the oscillator and it's probably going to stop oscillating. :-)

[Chuck H]

I just changed this post because I had a eureka idea. There's no need to change the nature of the trem at all. You could use a dual ganged pot (I checked, it's available) to variably couple the cathodes or allow them to operate independently.

[TubeNoob]

Gents, thanks very much; you've pointed out some things i didn't know and are clearly really important.

Chuck, super cool of you to think on it some more and post a schematic for it. I was going to ask if these two pots should be 'reversed' for it to work, then i saw the little black dots to indicate the pots' directions.

Very elegant solution!

Did this Eureka moment occur whilst bathing?! :-)

I have some dual-gangs - will they have to be 5k or do i have some room on that? I'm gonna dig thru my box...


Uberfuzz, thanks for pointing out the need to keep the voltages the same as on the original design for the trem, i'm going to have to work on that.


So, any options for the presence / NFB at this point? On my Princeton I do have the NFB going to V2a's bypassed cathode and it does seem to work fine.


Revised schematic: