I built a tube (UK: valve) amplifier but it has one problem. When I turn it on, it starts buzzing at 150 Hz but only if I connect the audio output transformer.

If I directly connect the speaker to the plate it works pretty well, but it's not powerful.

Changing the transformer or the capacitor's value changes the frequency of the buzz.

How can I keep the transformer in circuit but get rid of the buzz?

Here is the schematic:

enter image description here

  • 1
    \$\begingroup\$ Sounds like you’ve made an audio oscillator. Is it only yhe output stage oscillating? What tube are you using with a plate voltage of only 24V? And the output transformer impedance? \$\endgroup\$
    – Kartman
    Commented Jan 27, 2022 at 12:11
  • 3
    \$\begingroup\$ You didn't say which capacitor value changes the buzz, but the cathode bias bypass capacitor is orders of magnitude too small. (And I see no supply decoupling) \$\endgroup\$
    – user16324
    Commented Jan 27, 2022 at 13:51
  • \$\begingroup\$ As the comments say, we need a bit more information. What tubes are you using? What is the plate voltage? Which capacitor have you changed? \$\endgroup\$ Commented Jan 27, 2022 at 14:17
  • 1
    \$\begingroup\$ 47uF is enough to stabilise the cathode for all the audible frequencies that the human ear can detect in a typical EL84 or 6BQ5 single pentode output stage. 470nF is crazy small. \$\endgroup\$ Commented Jan 27, 2022 at 14:31
  • 1
    \$\begingroup\$ @GeorgeChita - the ECC84 has a standard 6.3 volt heater, and the PL84 is a series-heater chain 0.3A TV type valve/tube with a heater voltage of 15 volts. How are you heating them? Both usually have an anode voltage over 150 volts. Also I can't help thinking that getting HT ('B+') supply from a stack of dry cells is going to be rather high impedance and make for instability. Where did you get this circuit? Off some guy on Youtube? Note the PL84 is not a 'low-voltage EL84'. It is more like an EL86. \$\endgroup\$ Commented Jan 27, 2022 at 14:57

1 Answer 1


Neither the ECC82 nor the PL84 you say you're using are rated to operate at such a low B+ as the 24V you show in the diagram. While it's not impossible to coax the ECC82 into almost functioning at a B+ of tens of volts, it does require deliberate and non-trivial work - e.g., just reducing the anode resistors as shown isn't near enough to get an ECC82 to bias properly at 24V B+. And after you do all that non-trivial work, you'll likely be disappointed in the results. You really have only 3 reasonable options here:

  1. Increase B+ to the 100's of Volts that the tubes you're using need.
  2. Use low-voltage tubes (and suffer from their limitations and non-availability).
  3. Use BJTs.

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.