High-pass or high-cut caps

[EFK]

Looking at one old NAtional schematic, I see a 250 pf cap put across the pin5 output tube grids (yes it's one of those 50s rats nest amps). Also, I've had old Supros with paraphase inverters which use a 500 pf cap hanging off the grid of the first output tube, as well as varying values across both tube grid wires. So, my question is, is there a formula that can be used in this particular situation (as they're not bypassing resistors but rather bypassing the AC grid signal, I guess) to determine what frequencies are being cut?

My second question would be, is there any difference between a 250 pf cap across the output tube grids *or* two separate 250 caps off each individual grid. ???

[kg]

Looking at one old NAtional schematic, I see a 250 pf cap put across the pin5 output tube grids (yes it's one of those 50s rats nest amps). Also, I've had old Supros with paraphase inverters which use a 500 pf cap hanging off the grid of the first output tube, as well as varying values across both tube grid wires. So, my question is, is there a formula that can be used in this particular situation (as they're not bypassing resistors but rather bypassing the AC grid signal, I guess) to determine what frequencies are being cut?

as always it's 2piRC. the question is what the R is. it's going to be complicated.

My second question would be, is there any difference between a 250 pf cap across the output tube grids *or* two separate 250 caps off each individual grid. ???

at least that one is easy: the two grid lines are paraphase, meaning the voltage across them is always 2x the AC voltage on one side. that will make the capacitance between the two act like one twice as large. in your example, a 250pf across paraphase would be approximated by two 500pf from each phase to ground.

hth
ken

[kg]

i reread my last sentence and it's not very clear... not "two 500pf from each phase to ground" but "two 500pf, one from each phase to ground."

[EFK]

As far as the second part, I suspected as much with the paraphase. Thank you! As far as determining frequency, I'm a bit lost and for you indicate that it would be complicated means to me darn near impossible! I suppose in such a situation I'd be better off just using my ears - it's really more for the sake of curiosity and I wondered how those old Valco guys came up with some of the treble shunt cap values that they did. 500 pf for example seems quite high, seems to me as though it would be skirting with frequencies actually audible.

[kg]

well first, yes use your ears as the final judge.

second if you want to break it down, consider only one phase, with the "doubled" cap value, which will make conceptualizing what's going on a little easier.

the reason the R is hard to pin down is because it has many elements... the internal plate impedance of the tube, the load resistance, the coupling cap (which will have the effect of making the filter a bandpass), any series resistance, the grid return resistor, etc... they're all part of the function of determination of the break frequency.