6V6 Screen Grid Dissipation - Is it an issue?

[bepone]

lets stay in class A for simplicity , then there is no off signal.

also to measure the power we need oscillograms , u, i, in the grid 2 circuit. after is easy voltage x current, because power dissipation is always , for anything U x I.

some advanced scopes have this math function integrated.

[martin manning]

...Allow me to illustrate.

100VDC across a 1K resistor is 10W. If the resistor is seeing 200V for half the time and zero for the other half, what is the dissipation ? The average Voltage is still 100V right ? When the 1K resistor sees 200V, the dissipation is 40W for that amount of time. Averaged out over the time when dissipation is zero, the average dissipation is 20W.

Instantaneous power is I V, which for the "on" part of the pulse is 40W, and for the "off" part 0W, which is then integrated over time to get average dissipation.

In this example there are just two levels of instantaneous power so it's easy. You can also use RMS values and calculate power as I^2 R or V^2/R, where V is Vrms and I is Vrms/R. For this positive-only pulse waveform, Vrms is Vp D^0.5, where D is the duty cycle = 0.5. Vrms is then 0.707 Vp. P = Vrms^2/R = (200 0.707)^2/1000 = 20W, or using I^2 R, ((200 0.707)/1000)^2 1000 = 20W. If the Vg2 waveform is near enough to a positive-only square wave, you could use this method.

For a random waveform, you don't know how to calculate the RMS value, or instantaneous power (I(t)V(t) vs. t). A digital scope with math functions could do it, but using a true RMS meter you could measure AC and DC voltages across the screen resistor and combine them as (Vac^2 + Vdc^2)^0.5, then divide by R screen to get the current. Vg2-k RMS could be determined in the same way (provided the meter has the bandwidth for the signal).

[LOUDthud]

Thanks @martin manning for the reinforcement. Average Voltage times Average current does not equal average power. I used rectangular pulses because you don't need to know calculus to come up with an answer.

lets stay in class A for simplicity , then there is no off signal.

also to measure the power we need oscillograms , u, i, in the grid 2 circuit. after is easy voltage x current, because power dissipation is always , for anything U x I.

In this example the Average Voltage across a 1 Ohm resistor is 1V, but the power is 1.25W.

Av_E_Av_P3.GIF

[bepone]

and? what this changes? where is the current in those sketch.. current is the most important thing.. you can make things even more complicate, current will be in impulse and voltage will be more steady, and you need to do the integral below the curve , but how you will calculate this integral below the curve? there is no solution of equation?

lets make life simpler..if no osc. , voltmeter, it is good enough.

[LOUDthud]

and? what this changes? where is the current in those sketch.. current is the most important thing.. you can make things even more complicate, current will be in impulse and voltage will be more steady, and you need to do the integral below the curve , but how you will calculate this integral below the curve? there is no solution of equation?

lets make life simpler..if no osc. , voltmeter, it is good enough.

In a 1 Ohm resistor, the Voltage waveform is the Current waveform, that's why I used 1 Ohm. If you look up AMP in a dictionary, it probably says something like "The current you get when you apply 1 Volt to a 1 Ohm resistor". Hint: When 1 Amp flows in a 1 Ohm resistor, the power is 1 Watt. But when the Average Voltage is 1 Volt, it doesn't mean the Power is 1 Watt. In the example below, the Average Voltage is 1 Volt, but the Power is 3 Watts.

Av_E_Av_P2.GIF

After thought: In all these examples I drew, the Average Voltage across the 1 Ohm resistor was 1 Volt and the Average current is 1 Amp, but the only time the power was 1 Watt was when the waveform was DC.

[martin manning]

After thought: In all these examples I drew, the Average Voltage across the 1 Ohm resistor was 1 Volt and the Average current is 1 Amp, but the only time the power was 1 Watt was when the waveform was DC.

Aside: This is what kills rectifiers when the reservoir cap is too large.

[roberto]

I do remember another amp of mine does overdissipate the screens. EL34s with 6.6K OT, 470V HT, Choke but 2.2K screen grid resistors. I think it runs the screens at 12W flat out and the limit for EL34s is 8W. Komets must be even worse given they use 1K screen grid resistors with a choke and 5.2K OT!

5k2 Raa (depending on screen voltage) can make screens' life easier compared to 6k6, as the plate reaches g1=0 at at higher voltage, so lower secondary emission, then lower screen current. that could be the reason of lower screen resistors.

I use russian 6V6GT in UL (that helps!) with 100 Ohm screen resistors without issues, but I don't have experience with recently produced 6V6GTs.

[Colossal]

I use russian 6V6GT in UL (that helps!) with 100 Ohm screen resistors without issues, but I don't have experience with recently produced 6V6GTs.

Roberto, what load are you running your 6V6s at in ultralinear?

Thanks

[roberto]

Hi Dave,

i run them at 340V B+ with 8k Raa and 23% UL, source followers DC coupled to the grids of the output tubes to reach AB2, plus a local feedback to triodize the curves (similar to the other thread about the 360k amp).
Here below the loadline and the curves with UL (but without the local feedback effect):

6V6GT 8kRaa 34VB+ AB2.png

[Littlewyan]

Thanks for the replies everyone, very informatative, great discussion! Sorry for being absent for most of it. I feel I understand average power vs instantanous power now, thanks for the images LOUDthud. Unfortunately I don't have an RMS meter nor can my scope cope with high voltages (It's a Picoscope 2204, I need to buy 100:1 probes). So for now I am trusting the loadlines in that it'll be ok. I've removed the extra dropping resistor I added to the power string so now I just have a 500Ohm choke with 1K Screen Grid resistors. The amp sounds much better this way. Given my loadline is straight through the knee of the 0V line I can probably get away with 470Ohm screen grid resistors, but there's nothing wrong with a bit of extra protection and it shouldn't affect the sound anyway.