A little while ago I posted here about an organ amp conversion I'm just starting. My original post was a little vague admittedly hard to answer, so I thought I'd try to ask something more specific. I've since added in a tremolo circuit (slightly modified version of the Silvertone 1333 trem) using one of the other 6SJ7's available to me and refined the circuit quite a bit. Here is the updated schematic:
Now on to my issue. Throughout the process I have been adjusting the bias of the tubes and calculating the plate dissipation to make sure I won't blow this thing up once it finally comes in. I've been doing a lot of work in LTSpice to model and simulate my design. Measurements were taken from these simulations. These are the most recent numbers:
I've gotten the 6V6's to a place I'm happy with (~80% of max), but the numbers for preamp tube and the phase inverter seem a little low. I've been taking advantage of LTSpice's ability to input and output .wav files and from what I can tell, I like how the amp "sounds" (I get its not a perfect model, more a rule of thumb).
Here are my questions:
Is it normal for a preamp tube to be at 5% of max plate dissipation? Could it have something to do with it being a pentode/ something specific about the 6SJ7?
Should I bother trying to bias the preamp and PI hotter, if for no other reason than trying to maximize the potential of the tubes? Or should I leave the tubes cold because they are vintage tubes and I don't want to push them too hard?
If I were to go about getting these tubes biased hotter, how would I do that?The B+ heading into the tubes is pretty high, but the plate current is so low that its resulting in really low plate dissipation. How can I raise the plate current?
Sorry if any of these questions are ignorant. I'm just trying to learn. Comments and critiques are welcome.
Thanks,
Oli
Now on to my issue. Throughout the process I have been adjusting the bias of the tubes and calculating the plate dissipation to make sure I won't blow this thing up once it finally comes in. I've been doing a lot of work in LTSpice to model and simulate my design. Measurements were taken from these simulations. These are the most recent numbers:
I've gotten the 6V6's to a place I'm happy with (~80% of max), but the numbers for preamp tube and the phase inverter seem a little low. I've been taking advantage of LTSpice's ability to input and output .wav files and from what I can tell, I like how the amp "sounds" (I get its not a perfect model, more a rule of thumb).
Here are my questions:
Is it normal for a preamp tube to be at 5% of max plate dissipation? Could it have something to do with it being a pentode/ something specific about the 6SJ7?
Should I bother trying to bias the preamp and PI hotter, if for no other reason than trying to maximize the potential of the tubes? Or should I leave the tubes cold because they are vintage tubes and I don't want to push them too hard?
If I were to go about getting these tubes biased hotter, how would I do that?The B+ heading into the tubes is pretty high, but the plate current is so low that its resulting in really low plate dissipation. How can I raise the plate current?
Sorry if any of these questions are ignorant. I'm just trying to learn. Comments and critiques are welcome.
Thanks,
Oli
Last edited:
The 6SN7 uses a Pseudo LTP resistor.
Either tie the resistor to a very large negative supply voltage, and use much much more than 10k as the 'LTP' resistor, and set the neg voltage and 'LTP' resistor to get the current you want; or use a solid stage CCS and set the current at the value you want.
(In either case, a really large resistance 'LTP' resistance, or a solid state CCS, then use matched 6SN7 plate resistors).
Either tie the resistor to a very large negative supply voltage, and use much much more than 10k as the 'LTP' resistor, and set the neg voltage and 'LTP' resistor to get the current you want; or use a solid stage CCS and set the current at the value you want.
(In either case, a really large resistance 'LTP' resistance, or a solid state CCS, then use matched 6SN7 plate resistors).
Last edited:
For the input pentode, you are stuck with the following:
The pentode plate resistance, rp, is high.
That means its gain is dependent on the plate load resistor and all the network impedance to ground that follows the coupling cap that goes to the next stage's grid.
If the amp has plenty of of gain, you can triode wire the input pentode (take out the 2.2Meg screen resistor, and connect the screen to the plate using a 1k resistor), which lowers the effective plate resistance, rp (but will also lower the gain). Then you can reduce the 220k plate load resistance to get more plate current.
The pentode plate resistance, rp, is high.
That means its gain is dependent on the plate load resistor and all the network impedance to ground that follows the coupling cap that goes to the next stage's grid.
If the amp has plenty of of gain, you can triode wire the input pentode (take out the 2.2Meg screen resistor, and connect the screen to the plate using a 1k resistor), which lowers the effective plate resistance, rp (but will also lower the gain). Then you can reduce the 220k plate load resistance to get more plate current.
Last edited:
> calculating the plate dissipation to make sure I won't blow this thing up
The 6SJ7 has 220k to 354V. The worst-case would be half voltage on tube and resistor. 0.14 Watts. You CAN'T blow-up a tube with a large plate resistor.
Why so low? The gain of a resistor-loaded tube goes up as plate resistor is increased, until it gets larger than the audio load upon it. Since the audio load is around a half Meg, plate resistor of a quarter Meg gets gain about as high as it gets.
Do we care that we are not using the 2.5 Watts Pdiss we paid for? Is a smaller tube cheaper? Not really. The 6J7 comes from the earliest good indirect-heat tubes. The cathode structure worked well and was about as small as mass production handi-work allowed. A smaller cathode would cost more. (As seen near 1939 when the acorns and sub-mini tubes first came out.) Also a pentode has a lot of grids so inevitably has a large plate to contain them all. The plate is just iron (maybe Nickel, but iron was popular) so a little more didn't cost much.
Look at another situation. You build a house. The main beams you want to design to 80+% of allowable stress, to save cost, same as you work those 6V6es at near full rating. OTOH you build a nik-nak shelf, to support a half a pound. But the smallest board in the lumberyard would support 20 pounds. Do you care? Do you ask for a 0.75" board to be shaved down to 0.123" so it is "just barely safe"? No, that would be unthrifty. You use the "oversize" board and be done with it.
The 6SJ7 has 220k to 354V. The worst-case would be half voltage on tube and resistor. 0.14 Watts. You CAN'T blow-up a tube with a large plate resistor.
Why so low? The gain of a resistor-loaded tube goes up as plate resistor is increased, until it gets larger than the audio load upon it. Since the audio load is around a half Meg, plate resistor of a quarter Meg gets gain about as high as it gets.
Do we care that we are not using the 2.5 Watts Pdiss we paid for? Is a smaller tube cheaper? Not really. The 6J7 comes from the earliest good indirect-heat tubes. The cathode structure worked well and was about as small as mass production handi-work allowed. A smaller cathode would cost more. (As seen near 1939 when the acorns and sub-mini tubes first came out.) Also a pentode has a lot of grids so inevitably has a large plate to contain them all. The plate is just iron (maybe Nickel, but iron was popular) so a little more didn't cost much.
Look at another situation. You build a house. The main beams you want to design to 80+% of allowable stress, to save cost, same as you work those 6V6es at near full rating. OTOH you build a nik-nak shelf, to support a half a pound. But the smallest board in the lumberyard would support 20 pounds. Do you care? Do you ask for a 0.75" board to be shaved down to 0.123" so it is "just barely safe"? No, that would be unthrifty. You use the "oversize" board and be done with it.
Thank you both for the thoughtful responses!
6A3sUMMER:
I would like to avoid triode strapping the 6SJ7. Perhaps its silly, but I would like to have a true pentode preamp. If I feel strongly enough about it once I assemble the amp, maybe I'll look into it more.
As for the phase inverter, I've been toying around with negative supply voltages on the tail resistor and I like the idea, but I can't quite figure out how to get a negative supply out of the power transformer. I've been trying to sort of bias the center tap of the 340-0-340 center tap, but its just not working. The negative supply is diving way down to like -400V and the B+ ends up at like 50V. I'll probably try to sleep on it, but if I can't figure it out I might have to just leave it as is.
PRR:
I may have given the wrong impression when I said "maximize the potential of the tubes." I'm not necessarily looking to get more gain out of the preamp. I've read, and correct me if I'm wrong, that a sufficiently cold-biased tube can cause the amp's output to be thin and weak. As the amp has yet to arrive and I have yet to make these changes, I suppose I have no real way to verify this. Perhaps the lesson is that I should just wait to hear what I've got before rushing to "improve" the circuit.
Thanks again,
Oli
6A3sUMMER:
I would like to avoid triode strapping the 6SJ7. Perhaps its silly, but I would like to have a true pentode preamp. If I feel strongly enough about it once I assemble the amp, maybe I'll look into it more.
As for the phase inverter, I've been toying around with negative supply voltages on the tail resistor and I like the idea, but I can't quite figure out how to get a negative supply out of the power transformer. I've been trying to sort of bias the center tap of the 340-0-340 center tap, but its just not working. The negative supply is diving way down to like -400V and the B+ ends up at like 50V. I'll probably try to sleep on it, but if I can't figure it out I might have to just leave it as is.
PRR:
I may have given the wrong impression when I said "maximize the potential of the tubes." I'm not necessarily looking to get more gain out of the preamp. I've read, and correct me if I'm wrong, that a sufficiently cold-biased tube can cause the amp's output to be thin and weak. As the amp has yet to arrive and I have yet to make these changes, I suppose I have no real way to verify this. Perhaps the lesson is that I should just wait to hear what I've got before rushing to "improve" the circuit.
Thanks again,
Oli
You can retain the Pentode wiring of the input tube, and increase its current.
Try plate resistors of 180k, 150k, 120k, 100k, etc.
But the gain will be reduced, and . . .
at some point the distortion will go up, and worse than that, eventually you will not be able to drive the next stage to full output before the Pentode clips.
You may be able to optimize the pentode stage by reducing both the plate resistor and the cathode resistor.
Try plate resistors of 180k, 150k, 120k, 100k, etc.
But the gain will be reduced, and . . .
at some point the distortion will go up, and worse than that, eventually you will not be able to drive the next stage to full output before the Pentode clips.
You may be able to optimize the pentode stage by reducing both the plate resistor and the cathode resistor.
- Status
- Not open for further replies.