As a n tubey newbie I could use some advice before I conduct my next 'learning experiment'...and all comments and advice are welcome, especially if Enzo is out there...
I have this multi-section "ganged" pot...well really it's 4 seperate allen-bradley power pots with gear drive to one shaft. Each pot is now 100K linear and they are nicely spaced and reasonably insulated from each other, and look like they'd each handle several watts. I haven't opened one up or looked up the model, I guess I should...might even be wire-wound pots like my old Alembic preamp used; they're big, about the same size as were in the old Alembic FB2.
On my Peavey Classic 50 'rescue' amp that I'm using to experiment and learn on (but which has value to me):
Each of the 4 preamp sections (each 1/2 of a 12AX7, including the 2 sections inserted via relay for 'dirty' mode) has a 150K resistor (R2, R5, R1, R3) between the plate (anode) output and the appropriate VB+ or VB++. Being a relative newbie to tubes, I'm assuming the value of these resistors must affect the gain of the stage. Question 1: What is the common name for these resistors in this position in the preamp stage circuit? Plate supply resistor? The plate must get its power there, yet each also must vary from B+ or there would be no output signal. I'd imagine the plate's resistance to B+ and impedance to ground must represent much of the impedance it drives and affect the gain of the stage.
I've already added more volume controls so there is one between each stage. Now I want to use these new ganged pots to simulataneously vary the value of those four 150K resistor and listen / watch what that does and sounds like. Each pot section of these ganged pots actually lines up reasonably nicely with its relevant tube. I'm hoping it changes the gain of all 4 stages simultaneously.
I'm specifically not trying to just modify B+ and B++, I hope to add regulators for that later. I don't want the plate voltage to get dangerously low as a side effect of messing with these resistor values; can it fall below the grid and cause the grid to get too hot, like sometimes happens with power tubes?
I assume I'd want a resistor in series with these pots, as I don't want to ever connect the plate directly to its VB without one! So to limit the lowest resistance I'm thinking a 47K resistor in series with each 100 K pot would give me a range from Fenderish 47K like my Super Twin preamp to 100K like my old Alembic to the stock Peavey 147k (was 150k, close enough). Maybe I'd use slighlty larger (or smaller???) values in the later stages? Does that sound OK or would 75K to 175K be more 'dirty fun'?
Do I also need any resistors or caps across the pot in parallel with it from wiper to wired end for any reason? The stock Peavey fixed-value resistors are only 1/4 watt, which isn't much, so I assume the power handling of the pot won't be an issue. But will I want to further limit the range of adjustment for any reason or does this sound OK? I'm not changing the grid circuit... Also, I worry these pots might make a lot of noise if I turn them while it's running. Would a cap and/ or resistor across the pot (in parallel, from wiper to wired end) be advised for reducing the operation noise when the pots turn with signal present?
If I get squeals, I can add various caps later...or would that be advisable right away? Or perhaps should I limit the high-end or add stoppers to the EL84 power amp section tubes before I experiment, to at least contain any such problems? Even stock it would squeal occasionally with stupid 'dimed' settings of gain and presence, but seemed harmless if corrected quickly. This isn't a very powerful amp, and it's got 8 10" drivers connected and it never seemed to oscillate in the inaudible or RF range, so it doesn't seem prone to blowing up (yet). I was considering some HF-limiting caps in the preamp eventually. On the first stage I was considering a rather large value from grid to cathode, since miller effect won't mulitply its effectiveness. And if I'm changing gain with these new "gang of 4" pots that would change the miller effect and change the low-pass knee. So maybe I should put a cap from cathode to grid on each preamp stage? I would probalby do something like that to limit bandwidth on a transistor amp... The volume pots for the first 2 stages have a pull switch for a small 'bright' cap across them, which might also be dangerous without some kind of bandwidth limiting.
And one last issue. Like I mentioned, I now have 1M Audio taper volume pots between each stage. But I haven't turned it on since adding them. All these 4 stages are capacitor-coupled and have a 470K resistor between them. But I'm thinking that there's never any need to turn an inter-stage volume all the way down to 0, so I'm considering adding something like a 200 ohm resistor on the ground side of the pot, just so a little signal will leak thru when the inter-stage volume is set to 0. I can see some usefulness of turning the first pot or master volume pot all the way off, but not the ones in-between. What do you think?
I have this multi-section "ganged" pot...well really it's 4 seperate allen-bradley power pots with gear drive to one shaft. Each pot is now 100K linear and they are nicely spaced and reasonably insulated from each other, and look like they'd each handle several watts. I haven't opened one up or looked up the model, I guess I should...might even be wire-wound pots like my old Alembic preamp used; they're big, about the same size as were in the old Alembic FB2.
On my Peavey Classic 50 'rescue' amp that I'm using to experiment and learn on (but which has value to me):
Each of the 4 preamp sections (each 1/2 of a 12AX7, including the 2 sections inserted via relay for 'dirty' mode) has a 150K resistor (R2, R5, R1, R3) between the plate (anode) output and the appropriate VB+ or VB++. Being a relative newbie to tubes, I'm assuming the value of these resistors must affect the gain of the stage. Question 1: What is the common name for these resistors in this position in the preamp stage circuit? Plate supply resistor? The plate must get its power there, yet each also must vary from B+ or there would be no output signal. I'd imagine the plate's resistance to B+ and impedance to ground must represent much of the impedance it drives and affect the gain of the stage.
I've already added more volume controls so there is one between each stage. Now I want to use these new ganged pots to simulataneously vary the value of those four 150K resistor and listen / watch what that does and sounds like. Each pot section of these ganged pots actually lines up reasonably nicely with its relevant tube. I'm hoping it changes the gain of all 4 stages simultaneously.
I'm specifically not trying to just modify B+ and B++, I hope to add regulators for that later. I don't want the plate voltage to get dangerously low as a side effect of messing with these resistor values; can it fall below the grid and cause the grid to get too hot, like sometimes happens with power tubes?
I assume I'd want a resistor in series with these pots, as I don't want to ever connect the plate directly to its VB without one! So to limit the lowest resistance I'm thinking a 47K resistor in series with each 100 K pot would give me a range from Fenderish 47K like my Super Twin preamp to 100K like my old Alembic to the stock Peavey 147k (was 150k, close enough). Maybe I'd use slighlty larger (or smaller???) values in the later stages? Does that sound OK or would 75K to 175K be more 'dirty fun'?
Do I also need any resistors or caps across the pot in parallel with it from wiper to wired end for any reason? The stock Peavey fixed-value resistors are only 1/4 watt, which isn't much, so I assume the power handling of the pot won't be an issue. But will I want to further limit the range of adjustment for any reason or does this sound OK? I'm not changing the grid circuit... Also, I worry these pots might make a lot of noise if I turn them while it's running. Would a cap and/ or resistor across the pot (in parallel, from wiper to wired end) be advised for reducing the operation noise when the pots turn with signal present?
If I get squeals, I can add various caps later...or would that be advisable right away? Or perhaps should I limit the high-end or add stoppers to the EL84 power amp section tubes before I experiment, to at least contain any such problems? Even stock it would squeal occasionally with stupid 'dimed' settings of gain and presence, but seemed harmless if corrected quickly. This isn't a very powerful amp, and it's got 8 10" drivers connected and it never seemed to oscillate in the inaudible or RF range, so it doesn't seem prone to blowing up (yet). I was considering some HF-limiting caps in the preamp eventually. On the first stage I was considering a rather large value from grid to cathode, since miller effect won't mulitply its effectiveness. And if I'm changing gain with these new "gang of 4" pots that would change the miller effect and change the low-pass knee. So maybe I should put a cap from cathode to grid on each preamp stage? I would probalby do something like that to limit bandwidth on a transistor amp... The volume pots for the first 2 stages have a pull switch for a small 'bright' cap across them, which might also be dangerous without some kind of bandwidth limiting.
And one last issue. Like I mentioned, I now have 1M Audio taper volume pots between each stage. But I haven't turned it on since adding them. All these 4 stages are capacitor-coupled and have a 470K resistor between them. But I'm thinking that there's never any need to turn an inter-stage volume all the way down to 0, so I'm considering adding something like a 200 ohm resistor on the ground side of the pot, just so a little signal will leak thru when the inter-stage volume is set to 0. I can see some usefulness of turning the first pot or master volume pot all the way off, but not the ones in-between. What do you think?
Oh my. Yes, these ought to make just tons of noise when you turn them.
On the other hand, if this is just an experimental procedure, they don;t have to be quiet, they are not panel controls.
I guess it must be exciting to envision all these things turning at once. But really, why do that? Each stage is doing something different, I'd be wanting to tweak them individually. I'd wager you are more likely to find your perfect spot for each one at a different setting. I see no reason to expect that the ideal thing will be for ALL of them to go to 86k instead of 150k. You are limiting yourself. What if you find that a series like 150k 90k 220k 86k sounds best? Of course it might be interesting to just crank the whole bank around, and i am sure it will be interesting. But TO ME it sounds more like taking something to an extreme just for fun rather than a practical experiment.
But...
Yes, I'd add a 47k minimum resistor in each spot. A 100k pot would then allow up to 147k, close enough to stock. But why limit yourself? Plenty of circuits using 220k plate loads.
Those would be the plate load resistors or simply the plate resistors. They are PART of the impedance calculation. I leave that discussion to others who like math.
Once I modified an old SIlvertone for a guy, and he just couldn't settle on resistor values inb the tone stack and plate loads and a couple other places. We'd change something, he'd like that but then that made something else not right.
After a while, I installed four little trim pots, pointed them out to him and where they were in the circuit, and sent him of with "So here you go, tweak them until you find a combination that sounds best." And my involvement stopped.
You can swap out plate resistors, but the cathode resistors and their bypass caps are also improtant factors in your gain formula. Don;t forget them. Improtant? Important too.
The plate of the tube is the point of largest signal amplitude, and running wires out to pots and back offers a whole new world of possibilities for noise, instability and hum. Implementing this will have headaches, I'll bet.
Here is a thought experiment. How large a resistor would it take to make the plate voltage lower than the grid voltage? Hint, grids are at pretty much zero volts. Or some other way to get the plate to be negative?
I don't think that is a concern.
On the other hand, if this is just an experimental procedure, they don;t have to be quiet, they are not panel controls.
I guess it must be exciting to envision all these things turning at once. But really, why do that? Each stage is doing something different, I'd be wanting to tweak them individually. I'd wager you are more likely to find your perfect spot for each one at a different setting. I see no reason to expect that the ideal thing will be for ALL of them to go to 86k instead of 150k. You are limiting yourself. What if you find that a series like 150k 90k 220k 86k sounds best? Of course it might be interesting to just crank the whole bank around, and i am sure it will be interesting. But TO ME it sounds more like taking something to an extreme just for fun rather than a practical experiment.
But...
Yes, I'd add a 47k minimum resistor in each spot. A 100k pot would then allow up to 147k, close enough to stock. But why limit yourself? Plenty of circuits using 220k plate loads.
Those would be the plate load resistors or simply the plate resistors. They are PART of the impedance calculation. I leave that discussion to others who like math.
Once I modified an old SIlvertone for a guy, and he just couldn't settle on resistor values inb the tone stack and plate loads and a couple other places. We'd change something, he'd like that but then that made something else not right.
After a while, I installed four little trim pots, pointed them out to him and where they were in the circuit, and sent him of with "So here you go, tweak them until you find a combination that sounds best." And my involvement stopped.
You can swap out plate resistors, but the cathode resistors and their bypass caps are also improtant factors in your gain formula. Don;t forget them. Improtant? Important too.
The plate of the tube is the point of largest signal amplitude, and running wires out to pots and back offers a whole new world of possibilities for noise, instability and hum. Implementing this will have headaches, I'll bet.
Here is a thought experiment. How large a resistor would it take to make the plate voltage lower than the grid voltage? Hint, grids are at pretty much zero volts. Or some other way to get the plate to be negative?
I don't think that is a concern.
Thanks! More to think about. So would a cap across the pot reduce the noise when operated?
220K plate loads on a 12AX7? Guess the reason I'm thinking lower is that I've mostly only been inside Fenders as far as tube amps, and the Alembic preamp.
The wires might be really short (maybe about an inch) putting the gang where the little reverb tank used to be. The shaft would stick out the left side of the head.
220K plate loads on a 12AX7? Guess the reason I'm thinking lower is that I've mostly only been inside Fenders as far as tube amps, and the Alembic preamp.
The wires might be really short (maybe about an inch) putting the gang where the little reverb tank used to be. The shaft would stick out the left side of the head.
very valuable experiments. My suggestion is to do them all, but do them one at a time!! If you do a bunch together the effects might not be fully appreciated and if you end up with some kind of trouble at some point in your tweeking you will know what caused it. I find tweeking cathode values produce bigger changes.
Last edited:
I agree with Jeep, don't focus only on plate resistors. Play with catho0de resistor values, and to me even more, the bypass cap value.
I don't see them much now, but when I was a kid learning, and I am talking 50 years ago, substitution boxes were popular. Imagine a rotary switch that selected from a range of caps, put in a box with a couple banana posts on top. You could clip this into a circuit, then dial various cap values into the circuit to see which worked best.
Easy enough to make one. For coupling caps you could switch instantly between .1, .047, .001, .0047, etc. For cathode bypass experiments you could make the film caps more larger and add a few lytics. They did the same with resistors, but you are already using pots for that.
I have a cap box next to me, and several resistor decade boxes. The decade box had knobs to select 0-9 in the ohms, 10 ohms, 100 ohms, 1000 ohms, etc. You could dial in ANY resistance from zero ohms up to 999,999 ohms. A pot is more convenient, but this was more precise.
OH wow, seek and ye shall find. I had no idea. I went looking for a photo, and I find Ted Weber sells them on his web site, check it out. Inductors even.
https://taweber.powweb.com/amptechtools/subs.htm
I don't see them much now, but when I was a kid learning, and I am talking 50 years ago, substitution boxes were popular. Imagine a rotary switch that selected from a range of caps, put in a box with a couple banana posts on top. You could clip this into a circuit, then dial various cap values into the circuit to see which worked best.
Easy enough to make one. For coupling caps you could switch instantly between .1, .047, .001, .0047, etc. For cathode bypass experiments you could make the film caps more larger and add a few lytics. They did the same with resistors, but you are already using pots for that.
I have a cap box next to me, and several resistor decade boxes. The decade box had knobs to select 0-9 in the ohms, 10 ohms, 100 ohms, 1000 ohms, etc. You could dial in ANY resistance from zero ohms up to 999,999 ohms. A pot is more convenient, but this was more precise.
OH wow, seek and ye shall find. I had no idea. I went looking for a photo, and I find Ted Weber sells them on his web site, check it out. Inductors even.
https://taweber.powweb.com/amptechtools/subs.htm
An externally hosted image should be here but it was not working when we last tested it.
It won't be a really big cap across the active part of the 100K pot, as it will reduce the gain at high frequencies, which could be a good thing if I'm careful. Anyone want to do the math to make a recommendation that won't affect the usable frequency range? I don't want to limit treble in the first tube stages.
I'm leaning toward a larger resistor in series with each pot, as Enzo recommended, almost 100K resistor in series with each 100K pot.
My next question is whether the fixed resistor should go at the B+ end and the resistor at the plate, or vice-versa with the pot at the B+ and the resistor at the plate? The drop will be the same across each either way, the only difference might be which end is more sensitive to picking up noise on the wires. So I'm inclined to put the pot at the stiff B+ and the fixed resistor right at the tube socket. I'll put a molex socket on each, so I can remove the ganged pots and plug in resistor plugs later if I choose.
Yes, I was looking at those just today myself. I used to see decade boxes at shopgoodwill.com but not much anymore.
The reason I decided to start with plate resistors was because that was where I saw the greatest variation between my more familiar Alembic/fender circuit and higher gain stuff from Peavey. So that's a continuum whose sounds I'm somewhat familiar with. I'd like a single-channel preamp that could be adjusted anywhere across that spectrum, though most people would just get both and switch channels.
The reason I decided to start with plate resistors was because that was where I saw the greatest variation between my more familiar Alembic/fender circuit and higher gain stuff from Peavey. So that's a continuum whose sounds I'm somewhat familiar with. I'd like a single-channel preamp that could be adjusted anywhere across that spectrum, though most people would just get both and switch channels.
I'm doing a bunch of changes at once, but they're mostly on switches or capable of the original factory settings, except the transformers are on plugs so I can swap even them quickly until I bolt them down.
I just miss having an oscilloscope and suspect I'll get another as soon as I can afford one. I'm lusting for one of the thin 'hybrid' scopes that are a bit bigger than the handhelds and will work standalone or plug into your PC, plug into the wall for power, and have knobs instead of just buttons.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.