0
\$\begingroup\$

I am building a 18W tube amplifier based on this layout. I found a schematic for a similar amp.

my problem is that there is low plate voltage on all of the tubes (by plate I mean pin 3): when I measure the pins without the tubes it's around 120V and with the tubes it drops to ~17V. there is clearly something wrong. After rectification I can measure ~300V on the (+) of the bridge rectifier.

I am concerned about that 220K resistor, which is (in the layout) next to the rectifying diodes. would it make a difference if I remove it? because i cannot see it on the schematic.

note: I have a power transformer without center tap, which is rectified using a bridge rectifier as discussed in tube amp power transformer

\$\endgroup\$
0

2 Answers 2

1
\$\begingroup\$

Posted as a comment on the related question : tube amp power transformer

That "220K 1W" resistor looks like a problem. Was it meant to be 220R?

In a little more detail : the original 1950s circuit that inspired this design would have used a vacuum tube rectifier with a significant internal impedance. I haven't been through the details but something like a 5Z4 or GZ32 if you want to look up the datasheets.

That would impose some natural current limiting, partly to soften the impact on high voltage electrolytics, but also to save the circuit in the event of a flashover in the output valves (spectacular!).

Now is 220R about right? Anode current could reach 50-70mA for a pair of EL84s. (Look at the cathode bias resistance, and the grid voltage curves in the EL84 datasheets : at what cathode current does this resistor drop the grid voltage?)

In practice, measure the voltage across that resistor : it tells you the anode current.

Take an upper limit of 100mA : voltage drop of 22V, power = V^2/R = 2.2W. So 220R 2W is very plausible for HT currents up to say 90mA.

\$\endgroup\$
1
\$\begingroup\$

The plate voltage is way to low, even with the tubes unplugged.

If you unplug all the tubes no current should flow and you should measure a voltage close to 300V on all plate connections.

First, re-check all the wires. A photo of your amp-build may be helpful for us to pin-point errors. Also: Check your measuring instrument. Is the battery fresh? Is it a cheap device or something proper?

For further troubleshooting I suggest that you measure the voltage before and after each resistor in the B+ supply path (e.g. in your schematic that would be R28, R22 and R21. That way you can find out where the voltage is dropping and we can make a good guess why that is happening.

Please leave the 220K resistor in place. It's job is to slowly discharge the capacitors if you turn the amp off. They store a lethal amount of energy and having a resistor that discharges the tubes can be a life-saver.

UPDATE:

@crasic pointed out the strange connection of the 220k resistor and that it is not connected as a bleeder resistor.

The resistor should be in parallel to the first capacitor. There is clearly an error in the layout here that explains your low plate voltage.

Here is an update of the layout:

enter image description here

Cut the connection from the lower side of the 220k resistor to the + terminal of the capacitor (big cross in magenta). Connect the resistor on this side to ground instead (cyan connection). Connect the upper side of the 220k resistor to the + terminal of the capacitor (magenta connection).

The resistor will now be in parallel to the capacitor and do it's bleeding job and will not be in the B+ supply line anymore.

\$\endgroup\$
5
  • \$\begingroup\$ I'm not so sure about the bleed resistor. It is connected in series with the standby switch so how can it bleed if standby is normally open? I assumed it was a soft start resistor to avoid slamming the anodes with full B+ in an instant but the value also seemed high for this purpose \$\endgroup\$
    – crasic
    Jul 13, 2015 at 10:27
  • \$\begingroup\$ @crasic, good point! The resistor should be parallel to the first capacitor. In the way it is connected in the layout the resistor makes little sense. Regarding soft-start: There are already multiple Resistor/Capacitor filters in the B+ supply chain that somewhat limit the slew rate if the tubes are pre-heated. \$\endgroup\$ Jul 13, 2015 at 10:33
  • \$\begingroup\$ @crasic Hey! After looking at the layout again the 220K resistor is indeed the culpit! There is clearly an error in the layout. I'll edit my answer. \$\endgroup\$ Jul 13, 2015 at 10:35
  • \$\begingroup\$ Agreed, it seems incorrectly drawn. Maybe 220R x 2W for extra soft start and supply filtering for the primary cap. A 10hz low pass filter makes some sense \$\endgroup\$
    – crasic
    Jul 13, 2015 at 10:42
  • \$\begingroup\$ @crasic To be honest, after such a huge error in the layout I'd double-check the entire layout for more errors of this kind. \$\endgroup\$ Jul 13, 2015 at 10:58

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