Hi everyone,
I'm just beginning the process of planning an Aikido line stage build utilising the new Rev. E Noval board and PS-21. After a bit of reading, I'm a bit confused as to why most of the circuit examples I've seen use a larger value Rk for the input tubes than for the output tubes, even when they are of the same type. Like this diagram from tubecad.com:
John Broskie also notes the same in the manual when providing typical value suggestions:
Yet, on the previous pages that hold many example values, the R3,4 and R6,7 resistors don't differ for the same B+ setting:
I feel like I'm missing something simple, but fundamental. I do recall reading on TubeCAD that JB likes to run one of the stages hotter than the other. Could it just be this preference that's showing through?
I'd appreciate any guidance you could give. I am a bit embarrassed that I have not been able to work this out for myself.
I'm just beginning the process of planning an Aikido line stage build utilising the new Rev. E Noval board and PS-21. After a bit of reading, I'm a bit confused as to why most of the circuit examples I've seen use a larger value Rk for the input tubes than for the output tubes, even when they are of the same type. Like this diagram from tubecad.com:
John Broskie also notes the same in the manual when providing typical value suggestions:
Yet, on the previous pages that hold many example values, the R3,4 and R6,7 resistors don't differ for the same B+ setting:
I feel like I'm missing something simple, but fundamental. I do recall reading on TubeCAD that JB likes to run one of the stages hotter than the other. Could it just be this preference that's showing through?
I'd appreciate any guidance you could give. I am a bit embarrassed that I have not been able to work this out for myself.
Last edited:
FYI Fully.
I built an Aikido LS from scratch using all 6CG7s @ 300 volts on tag boards and point to point. It sounded nice but too tubey and slow for me. Cut the B+ back to about 160 and changed the output tubes over to 6DJ8s. Better but still a bit too warm and slow. Once I went to 6DJ8s for all tubes did it sound the way I like, not tubey and slow. Quick and detailed. Quite a few tube rolling choices with the 6DJ8 family too.
Steve
I built an Aikido LS from scratch using all 6CG7s @ 300 volts on tag boards and point to point. It sounded nice but too tubey and slow for me. Cut the B+ back to about 160 and changed the output tubes over to 6DJ8s. Better but still a bit too warm and slow. Once I went to 6DJ8s for all tubes did it sound the way I like, not tubey and slow. Quick and detailed. Quite a few tube rolling choices with the 6DJ8 family too.
Steve
Thank you kodabmx. I think I understand.
I see that in the page 12&13 charts a 6CG7 at 250V B+ has both the input and output Rk listed as 626ohm, but Broskie recommends 820 on the input and 470 on the output the following page.
Can I conclude then that 626ohm would work fine as a baseline, but the recommended values are different due to them being optimised for the bias requirements of each stage?
Thanks Steve. That's quite interesting, and I appreciate the info.
I already have a nice stock of 6DJ8 tubes, so am considering going that way. On the other hand, the 6CG7 would seem to have the more 'correct' gain for a LS.
Did you notice much difference in gain between the two types?
Also, I don't suppose you could share your Rk resistor values ?
Yes. FWIW I use 240R for 6CG7 (or any other lowish mu tube like 6N1P, 6N6P, 6DJ8, 6N5P, etc.) in my point to point Aikido with a B+ of 260V ish after the filter...
Thanks, that's also helpful. I think I've become too fixated on getting the 'right' values, whereas the Aikido circuit appears quite pliable.
The other section of the manual that was a bit baffling to me was this:
I suspect that the highlighted example formula is a typo? I can't see how it works out.
With your 260V B+, 240Rk, and estimating an rp of around 8000, your idle current would be 260/2(8000+[20+1]240)=0.0099 or almost 10mA? Might that be true?
Last edited:
Closer to 8ma, depending on the tube, but the B+ might be a little lower than 260V since even though I built the damned thing, I can't remember what value resistor I used in the filter... It's anywhere from 560R to 2K. Still... Close enough for government work.
Very nice
I've been drawing dots on these charts for the last couple of days, without knowing if I was doing it correctly. This makes it very clear, thanks.
It is an *amplifier*. It takes a small signal and makes a large signal.
It is reasonable, and thrifty, to run the early stages at lower current than the final stage.
Even when they are the same type.
There is no "right" current for a tube, but a wide range. Easily 10:1 of current, which may imply 100:1 range of bias resistors. In "good audio" we tend not to go way-low, because bandwidth (treble extension) falls off. Also as long as early stages are "somewhat" cooler than final stage, there is no economic advantage to starving them brutally (does not reduce total power consumption enough to care).
Thanks, I appreciate your time. I think that between yours and kodabmx's answers this novice may have come to understand much more what the Aikido circuit is doing. And also that while concentrating on the plate and cathode, I had completely ignored the grid.
At the risk of looking a bit foolish, here's my summation:
The first stage tube is the voltage amplifier, amplifying the source signal that's being presented to the grid. This stage doesn't necessarily warrant a large amount of current across it (mostly, I assume, because the signal input itself is low voltage and being a line-stage amplifier there isn't a large amount of gain required).
The second stage takes the amplified signal from first (and doesn't itself amplify?), and must drive a load which warrants 'more juice'. In this case the load is the power amplifier input(?).
I'll work on choosing some sane operating values, stop worrying so much - and enjoy the experimentation.
At the risk of looking a bit foolish, here's my summation:
The first stage tube is the voltage amplifier, amplifying the source signal that's being presented to the grid. This stage doesn't necessarily warrant a large amount of current across it (mostly, I assume, because the signal input itself is low voltage and being a line-stage amplifier there isn't a large amount of gain required).
The second stage takes the amplified signal from first (and doesn't itself amplify?), and must drive a load which warrants 'more juice'. In this case the load is the power amplifier input(?).
I'll work on choosing some sane operating values, stop worrying so much - and enjoy the experimentation.
The first stage drives the second stage. The input of the second stage is an "infinite" impedance. An open grid is like hundreds of MegaOhms parallel with dozens of pFd of capacitance. (If the 2nd stage were a Voltage Amp it would have more parts and more input loading.)
The second stage drives..... what? We don't generally know. It could be 1 foot cable into a 1Meg input. It could be 100 feet of cable into a 10K input. I have seen lower. Up to a point we design a line-level output like a passenger car: It can carry 1 person, it can carry 5 people, makes little difference.
I just built an Aikido using 6N14P (ECC84) as first tube.
There is virtually no noise at all, a very clean choice for a low noise application like first tube in a phono amp. (I know I'm going to build one using it!) Get a few as the gain can vary from tube to tube, but still within about 5%.
As far as cathode resistor values, I settled with a B+ of 280V, and using a 510 Rk I get about 8mA, 140V, 4V across Rk per section. It looks exactly as I thought it would when I eyeballed the curves.
There is virtually no noise at all, a very clean choice for a low noise application like first tube in a phono amp. (I know I'm going to build one using it!) Get a few as the gain can vary from tube to tube, but still within about 5%.
As far as cathode resistor values, I settled with a B+ of 280V, and using a 510 Rk I get about 8mA, 140V, 4V across Rk per section. It looks exactly as I thought it would when I eyeballed the curves.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.