Experimenting around with a tweed era circuit

[eschertron]

My guess is that the designers did not expect the amp to be dimed.

And I should add that maxing the vol control limits the tone control's effectiveness; but you already know that from first-hand experience.

The first stage of the preamp generates the most noise (any noise here is amplified throughout the rest of the signal path). That's the classical criterion for 1st stage design. Max gain, min noise. That doesn't mean that you shouldn't experiment with the circuit.

edit: before it's all over, I just know that you'll try cascading the two input valves

[mort]

One thing about the tone control, and it's much like the tone control on the 5e9-a tremolux circuit (my favorite), it also acts a sort of drive or sizzle control. It does change the treble but the treble is pleasing to my ear at any position so I many use it as a "hot" pot.

And yes I'll try just about anything. I'm getting to that phase, no pun intended. This will be my ever changing amp. Gonna put post PI mv in when my next order gets here. Will probably also try a VVR in it at some point too. But for right now I'll swim around in the preamp for a while.

[Mike Sulzer]

And I should add that maxing the vol control limits the tone control's effectiveness; but you already know that from first-hand experience.

The first stage of the preamp generates the most noise (any noise here is amplified throughout the rest of the signal path). That's the classical criterion for 1st stage design. Max gain, min noise. That doesn't mean that you shouldn't experiment with the circuit.

edit: before it's all over, I just know that you'll try cascading the two input valves

But noise apparently was not a concern in most guitar amp designs because the grid stopper resistors are so large that their noise exceeds that of the tube by 10 db or more. And that is assuming that the resistors have theoretical noise. In fact the carbon comp resistors used for grid stoppers back then had significant excess noise.

[Chuck H]

Which seems to support escher's assessment that the designers never expected the amps to be used as distortion boxes Indeed it's more common today to see smaller grid stops at the input. Usually 10k to 33k. That 33k value still has significant thermal noise, but new designs don't use the old CC resistors and usually dump hiss related frequencies through a "bleeder" cap somewhere in the preamp. It might also be worth noting that if one is using an AC filament supply, it's a good idea to fully bypass the first stage cathode to well below 60Hz because it helps keep hum from that grid. In this case all frequency control is with the coupling cap/s. Smaller cathode bypass caps do more than just alter gain @ frequency. There's also some NFB and an increase in impedance at the unbypassed frequencies. This can be a good thing in the right circumstance.

Even though I do it sometimes, it's bad practice from a design standpoint to use less than a full bypass cap value on the first triode with AC heaters. But if other hum conscience design practices have been employed it can be done with only a small (not objectionable) hum addition.

[mort]

Ok so now I've taken off both bypass caps and changed the 470r to a 1k on V1 and left the 2.2k on V2. Now we're back into the goods. That little bit of head room is back and it gets a real nice vibrant, present sound. Still moreso on V2 than V1. I'm gonna try a different x7 in V1 and see what happens. Also gonna try a .22uF bypass cap on V2 for kicks. I'm recalling Enzo saying that it's not necessary to use polarized caps. I've got Mallory 150 .22 cap that I would normally think of as a signal cap and that should be safe to use, correct?

Also, do I have it right that no matter which input I'm plugged into, that the guitar signal is going through 1 triode before hitting the phase inverter? And if that's true, using a tube that has different triodes in it would just give two flavors split between 2 inputs? I've had a pair of early 60s/ late 50's NOS RCA 12bh7a's sitting in a box and if memeory serves those have differing triodes, don't remember what off hand and haven't looked them up yet but that might be cool to play with a little later, no?


Edit: V1 had a dull sounding tube

[eschertron]

I've got Mallory 150 .22 cap that I would normally think of as a signal cap and that should be safe to use, correct?

150s are good to 630 volts. So fine for this application. I'd hate to try to source a film cap that was bigger than 1uF or so, though!

Also, do I have it right that no matter which input I'm plugged into, that the guitar signal is going through 1 triode before hitting the phase inverter? And if that's true, using a tube that has different triodes in it would just give two flavors split between 2 inputs?

If I'm looking at the right schem, each jack can go to 1 triode (the other grounded) or to both triodes in parallel. you may need to play with how the jacks are wired if you want to select the desired half for the 'single dip' mode. Also, jumping ch1 and ch2 can net you up four triodes in parallel at the input. Talk about flavors!

edit: scratch what I said about jumpering channels. Doesn't look like that can be done.

Edit: V1 had a dull sounding tube

If all our conundrums could be fixed with a tube swap...

[mort]

There's actually pretty good surplus store in town that has lots of random, low voltage rated oddball caps for cheap. I could spend 5 bucks and have a whole pile of caps at almost any imaginable uF in the desired range for this application. They don't have most of the typical stuff I would use but they have everything else in between

[mort]

Ok nice with the .22 on the V2 cathode it basically sounds like it's got a light OD pedal on it or a simple boost. The characteristics are the same but sounds a little hotter and would cut through a mix a bit stronger. Still all the vibrancy to make it sound all goose bumpy. This is a nice part of the playground

[g1]

using a tube that has different triodes in it would just give two flavors split between 2 inputs? I've had a pair of early 60s/ late 50's NOS RCA 12bh7a's sitting in a box and if memeory serves those have differing triodes, don't remember what off hand and haven't looked them up yet but that might be cool to play with a little later, no?

Both halves of the 12BH7 are the same. It's a higher current tube sometimes used as a driver in things like SVT. I think you may be thinking of 12DW7, also a fairly common Ampeg tube, which is like half 12AX7, half 12AU7.

[mort]

V1 got a .1uF on it's cathode and gonna play around with that some more. Probably not worth mentioning much more on that...


... but what about the idea to use a resistor to set a minimum bias at say 800r and then a 2k pot to vary the cathode bias as a panel option. Any reason not to do that? It's very low voltage and not alot of current so I would think it's ok?

[Chuck H]

I've seen that done. I wouldn't trouble with it. The first gain stage only amplifies the guitar, which has an output of about half a volt IF you're using humbuckers and IF you slam a power cord. That's not enough to clip even a fairly well skewed bias so a variable cathode resistance would act sort of like a volume control that doesn't go all the way down.

Apply the same thing to a later stage and you might actually get the ability to change the clipping character from saturation, to centered, to cutoff and any blend in between, BUT, you also change the output of that stage and would need to simultaneously compensate for that with another stage to make use of the feature intuitive. I wouldn't trouble with it. You'd likely find a sweet spot you like and then rarely use that knob again. Which brings up another possibility...

Put pots on the cathodes, find the sweet spots and then measure the pots to find the fixed resistance you like best. To do this you really have to keep in mind that each stage affects the next, so differences are compounded. You also need to do it in a way that doesn't allow setting to impractical operating parameters. AND you probably should look at what's happening on a scope to see that your gain structure doesn't end up all cattywhompus and ultimately limit drive to your power amp.

[mort]

When you say each stage effects the next, is that to say that one triode effects the other? Because if I'm understanding correctly each input is a single triode into the phase inverter so no subsequent gain stages. (Even though I've removed the signal caps and changed the cathodes, the skeleton of the drawing is still the same)

[eschertron]

When you say each stage effects the next, is that to say that one triode effects the other? Because if I'm understanding correctly each input is a single triode into the phase inverter so no subsequent gain stages. (Even though I've removed the signal caps and changed the cathodes, the skeleton of the drawing is still the same)

There are 4 input stages mixed and feeding a gain stage right before the PI, unless you've changed the PI from what it is in the schem to an LTP?. This gain stage can amplify the input stage signal up to clipping (or nearly so) which can audibly color the sound. Hot, cold, or in between are distinctive when copious amounts of harmonic distortion are generated. I've played with a couple champs - and while it doesn't get into Neil Young territory until the output tube is slammed - intentional voicing can be effective on the preamp tubes, especially if there's a MV that allows the preamp voicing to be heard.
I agree with Chuck that the goal of the input stage is to get gain. Not much else can be expected at small-signal levels. Hit the input with an OD pedal and it's a different story though. Then the input stage is no longer an input stage, innit?

[Enzo]

It is indeed low voltage and low current, but it still is DC on a pot, so it WILL be scratchy every time you adjust it.

Like so many tiny tweaks, you will probably find one setting you like and then never bother it again. Why waste space for that on the panel? Even larger changes act that way. I converted a guy's amp to cathode bias or switched bias at the flip of a switch. After playing both ways a little, he set it one way and never flipped it again. Some of the Fender Pro Shop amps had switchable fixed/cathode bias plus tube/SS rectifier. Again most of us found the one we liked and then didn;t touch it. You may indeed find it useful, of course, but experience says it is more a tweak than a feature.

[The Dude]

What he said^^^^^^. It sort of reminds me of something a bit unrelated that I run into often at my shop, particularly with kids who play modern metal and like the tight SS rectifier sound. They brag about having a dual or triple rectifier amp (obviously not knowing what that even means), but never use anything but the SS rectifier because that's the sound they like.