1
\$\begingroup\$

I am a hobbyist in electronics and a HAM radio operator. I am currently designing a Colpitts oscillator using a 6J5 triode intended to oscillate on the 80 meter band (slightly above 3.5 MHz). I just started to learn vacuum tube circuits, and only have quite limited knowledge about oscillator designs, so I essentially designed the circuit as a LC tank circuit connected with a triode amplifier.

But when I simulated the circuit today, it shows an unusual waveform, as if the signal is modulated, and it seems like is oscillating around 10-15 kHz. I would like to know what is the issue with my circuit and your recommendations for improvements.

(Edit: I have not noticed previously that the sample time could be changed in the simulator, with 5ns being set, the result is finally sinusoidal. But another problem have arised, where the amplitude of the oscillation started to grow rapidly until the simulator stops functioning. I assume this has to do with the feedback ratio and the amplification factor of the tube. Could someone teach me how this calculation is done that ensures stable oscillation?)

The oscilloscope is connected across the bottom 200pF capacitor

Also I have a few general questions related to vacuum tube circuits:

  1. What factors affect the required feedback ratio to sustain oscillation, and how should I calcuate the required feedback ratio?
  2. If the oscillator is designed as electron-coupled using a pentode. How should I calculate the transconductance of the cathode-screen circuit, since the circuit is effectively a triode oscillator with the screen as the "plate"?

WARNING: High voltage is lethal if not handled properly, please always treat it with extreme caution!

I highly appreciate your answers and recommendations.

\$\endgroup\$
11
  • \$\begingroup\$ Possibly a simulation problem caused by a time-step too long. Try your simulation with a time step much shorter. Mind you, It'll take much longer to run. \$\endgroup\$
    – glen_geek
    Commented Dec 1 at 15:17
  • \$\begingroup\$ Thanks for your reply. But I have already used the maximum simulation speed and current speed. \$\endgroup\$
    – Morris
    Commented Dec 1 at 15:21
  • 2
    \$\begingroup\$ Resonator frequency is close to 3.5 MHz, which has a period of 0.28 microseconds. Your simulation should use a time step significantly shorter than that period. I'm unfamiliar with your simulator, so cannot suggest how to achieve that...perhaps allowing maximum simulator speed is a wrong choice. \$\endgroup\$
    – glen_geek
    Commented Dec 1 at 15:31
  • \$\begingroup\$ You need a slower simulation speed, I believe. To my knowledge, 'current speed' has no impact on the actual simulation in circuitjs, just how it's displayed. \$\endgroup\$
    – Hearth
    Commented Dec 1 at 15:32
  • \$\begingroup\$ I tried to change the applet's simulation speed, but the problem still persists. I think there is something wrong with the circuit, but I can't tell the exact issue. I did designed a similar circuit using BJT before and it worked well on the same simulator. \$\endgroup\$
    – Morris
    Commented Dec 1 at 15:42

1 Answer 1

Reset to default
0
\$\begingroup\$

Here is what I simulated wit microcap v12

Note that I changed the valve and L1 to 1mH.
The starting of the oscillation is "strange" but it start.

enter image description here

Here is the "steady" state.
The frequency is effectively 3.5 MHz.

enter image description here

\$\endgroup\$
5
  • \$\begingroup\$ Thank you so much for your efforts. Indeed there are more data available for 12AX7 tube. I will try to recreate this circuit and check its performance using a simulator. \$\endgroup\$
    – Morris
    Commented Dec 1 at 23:39
  • \$\begingroup\$ NB: I replaced 12AX7 with 6JB5. Same behavior. \$\endgroup\$
    – Antonio51
    Commented Dec 2 at 8:45
  • \$\begingroup\$ Thanks, but isn't 6JB5 a beam pentode, or you used the screen as the "plate" of the oscillator circuit? \$\endgroup\$
    – Morris
    Commented Dec 2 at 9:39
  • \$\begingroup\$ No, I am "sure" that it is a triode although there is also a pentode with quasi the same name ... the name 6BJ5 is not complete. \$\endgroup\$
    – Antonio51
    Commented Dec 2 at 13:21
  • \$\begingroup\$ It is a 6JB5T-Ayumi, triode. \$\endgroup\$
    – Antonio51
    Commented 2 days ago

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.