Measuring Screen Grid Dissipation

[JazzGuitarGimp]

Did you subtract the 70V from the screen node HT voltage when calculating that screen dissipation?

I am pretty sure I did. However, memory NOT being what it used to be, I will go back and measure it again.

[JazzGuitarGimp]

I had another look at my measurements today, and it indeed appears I made my calculations based on the actual screen grid voltage.

Here's what I've got:

___________________________A_____B_____C_____C-B_____Power
At Idle_____________________372V__362V__362V__0.36V___0.28W
At Full Clean Output__________345V__301V__297V__1.7V____1.08W
At full Square Wave Output_____325V__244V__239V__4.4V___2.26W

A = Voltage at original screen grid node (output of choke)
B = Voltage at new screen grid node (after 2.7K resistor)
C = Actual screen grid voltage wrt cathode
C-B = Voltage across 470R (actual is 466R) screen grid resistor

-----------------------------------------------------------------------------------

Also, I think I figured out where the additional power is coming from. Remember that faulty octal socket I had? I have a terminal strip at each socket where I anchor the control grid stopper and the screen grid stopper. I had to resolder the screen grid connection on the terminal strip in order to add the 2.7K resistor to the circuit. It just so happens that this was at the same socket that gave me noise when I bumped into the tube. I suspect the 470R was not making connection at this tie point, which would have left the screen grid floating. I don't know if floating screen grids behave the same way as floating control grids (seeking 0V). But if they do, that would have put that tube in cutoff. I was able to verify this by removing the tube, jumpering out the 2.7K resistor, and remeasuring the output power. Bingo! - right back to 34W. I feel better because this was supposed to be a 40W amp by design.

[Littlewyan]

Well if you did have a floating screen before then there would have been less current across the choke which means less voltage drop. So now that you've sorted it maybe your screens will be fine without the 2.7K resistor? If you jumper across the 2.7k resistor now what is the screen grid dissipation?

I think when the screen grid is disconnected the valve shuts off. I'm sure I recall someone having an issue on here and it was due to the screen grids not being connected. Then again if the valve was shut off then you wouldn't have measured 34W as surely that would be way too high for one valve?

[JazzGuitarGimp]

Then again if the valve was shut off then you wouldn't have measured 34W as surely that would be way too high for one valve?

This amp has a four cylinder engine. 4 x 6V6

[Littlewyan]

Oh yeah duh . Try bypassing the 2.7K resistor. The extra anode current might pull the HT down enough to run the screens cooler at full tilt.

[JazzGuitarGimp]

Oh yeah duh . Try bypassing the 2.7K resistor. The extra anode current might pull the HT down enough to run the screens cooler at full tilt.

I will try, but I'm not hopeful. If you notice in the "full square wave output" measurement above, there is an 81V drop across the 2.7K resistor (325V - 244V). I'm pretty sure that resistor has found its new home. 😀