Just wondering, is it worthwhile to worry about the power dissipation of the "top" grid in a cascoded arrangement of two triodes? Last night I was playing around with a curve tracer, and hooked up a 12BH7 in a cascoded configuration. If I set the voltage to the top grid to around 200v (quite high, I know), the grid would start to draw a pretty decent current as the input signal reached saturation. At a point just a bit below the knee, the plate current was approximately 30mA, and the grid current was approximately 14mA.
Initially, I thought that meant the grid of the top tube was dissipating 200 * 0.014 = 2.8 watts (!), but on reflection, it seems to me that since the cathode of the top tube is itself elevated to almost the same voltage as the grid, the dissipation is probably a few milliwatts at most, despite the current. Is that right?
Initially, I thought that meant the grid of the top tube was dissipating 200 * 0.014 = 2.8 watts (!), but on reflection, it seems to me that since the cathode of the top tube is itself elevated to almost the same voltage as the grid, the dissipation is probably a few milliwatts at most, despite the current. Is that right?
Hi Guys
Grid current must come from somewhere and that place is the cathode, along with the plate current. If it is a cascode then both currents flow through the lower tube, so the 2.8W is likely the case. There is no need to guess at the dissipations and the tubes in question actually have significant ratings in this regard.
Depending on how much output signal swing you are targeting and what the supply voltage is, you might want to readjust the proportion of voltage for the two tubes.
Have fun
Grid current must come from somewhere and that place is the cathode, along with the plate current. If it is a cascode then both currents flow through the lower tube, so the 2.8W is likely the case. There is no need to guess at the dissipations and the tubes in question actually have significant ratings in this regard.
Depending on how much output signal swing you are targeting and what the supply voltage is, you might want to readjust the proportion of voltage for the two tubes.
Have fun
The 'screen' dissipation will depend on Vg1-k for the upper triode, so will normally be small. This is fortunate, as most small signal triodes are not designed to get heat away from the grid - 12AU7 may be an exception as it was based on the 6C4/EC90 power triode.
The total dissipation (most of which will be in the anodes) will be greater. So most of the heat developed by the upper 'screen' current will appear at the anode of the lower triode.
The total dissipation (most of which will be in the anodes) will be greater. So most of the heat developed by the upper 'screen' current will appear at the anode of the lower triode.
Thanks, that's exactly what I wanted to know. So, in essence, the current flowing through g1 of the upper triode should be added to the anode current of the lower triode? For example, if in the upper triode Ia = 30 mA and Ig1 = 14 mA, Ia in the lower triode is 44 mA, with a corresponding increase in plate dissipation?
Hi Guys
I believe the original point has not addressed: the heating of the upper grid. I think it is safe to assume this equals the grid current multiplied by the Vgk of the upper tube. This is easy to plot in LTspice.
A grid-stop resistor should be used between the upper grid and its reference voltage. This limits grid current and usually improves the stability margin, too.
Have fun
I believe the original point has not addressed: the heating of the upper grid. I think it is safe to assume this equals the grid current multiplied by the Vgk of the upper tube. This is easy to plot in LTspice.
A grid-stop resistor should be used between the upper grid and its reference voltage. This limits grid current and usually improves the stability margin, too.
Have fun
See post 3. I didn't actually spell out that it is Vgk times grid current; I assumed that was obvious from what I said.
- Status
- Not open for further replies.