See the Radiotron Designer's Handbook 4th edition at the page 483 (Chapter 12 Sect. 2 (ii).
There this subject is dealt thoroughly.
http://www.tubebooks.org/Books/RDH4.pdf
There this subject is dealt thoroughly.
http://www.tubebooks.org/Books/RDH4.pdf
Last edited:
Assuming you have an ordinary circuit with nothing but grid-leak resistors following the cathodyne, then the foruma is simply:
C = 1/ (2 pi f R)
Where f is the low cut-off frequency, and R is the power tube grid leak resistor.
12AX7 would be an interesting experiment.
I tried changing from 6SN7 (-5v bias on the cathodyne stage) to E180CC (similar to 12AU7, with higher transconductance) and the bias dropped to -2v. Not sure what the optimal bias voltage is, but changing the cathodyne tubes definitely has a noticeable effect on the sound of the amp.
I tried changing from 6SN7 (-5v bias on the cathodyne stage) to E180CC (similar to 12AU7, with higher transconductance) and the bias dropped to -2v. Not sure what the optimal bias voltage is, but changing the cathodyne tubes definitely has a noticeable effect on the sound of the amp.
Bias voltage (Ug-c) is not essential measure.
Important is that the operating point of the cathodyne is such that the output voltage can swing sufficiently to both directions. When the clipping begins, it should take place symmetrically.
Typical advice that supply voltage should be divided by three; one third for anode resistor, tube and cathode resistor leads to bad performance.
It is easy to observe that about half of supply voltage should be across the tube and one quarter across both anode and cathode resistor.
Important is that the operating point of the cathodyne is such that the output voltage can swing sufficiently to both directions. When the clipping begins, it should take place symmetrically.
Typical advice that supply voltage should be divided by three; one third for anode resistor, tube and cathode resistor leads to bad performance.
It is easy to observe that about half of supply voltage should be across the tube and one quarter across both anode and cathode resistor.
since SY's suggestion appeared like a response to my post about the 2A3, but being too expencive, I took for granted 6S4 meant 6S4S, and refferred to the similar looking equivalent, 6C4C
now I see there are also 6S4 9 pin types
and since I also mentioned ecc99 in the same post, 6S4 might also be ... another 9 pin
so which is it, please 
It's the old American 6S4. Similar to a single section of 6BL7. 500V plate max means that you can run it on a high B+ and swing enough voltage as a cathodyne to drive normal output tubes.
http://tubedata.milbert.com/sheets/093/6/6S4A.pdf
http://tubedata.milbert.com/sheets/093/6/6S4A.pdf
Oh yes !! Any triode can do the job !! BUT 6S4 / 6S4A can do the job much better = linearity and able to handle higher +B ( 500 / 550 V ) thus more voltage swing and better linearity .
Yes it is , because the grid bias ( Vg - Vk ) Rp , Rk and +B , will put the tube in a very interesting operating point , regarding voltage swing and linearity.
Linearity is very important , it`s useless to have a big voltage swing
with lots of distortion
Last edited:
OK !! Now we agree !! The only issue is that I do not have a THD analyzer with very good sensitivity , BUT I think I can get very close with my DSOscilloscope with FFT function . Do you agree ??
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.