|
|
Post by radioholic on Jun 15, 2019 22:35:34 GMT 12
The valve data sheet for the 42 / 6F6 says the maximum value for the grid resistor should be 0.5meg with cathode bias and 0.1 meg with fixed bias. Yet, many radios with fixed bias used values much greater than 0.1 meg. Examples : Philco 37-620 used 1 meg, Pilot 403B used 500K, RCA T6-1 used 470K, and Wells-Gardner 7D used 600K. Can anyone resolve this discrepancy?
|
|
|
|
Post by Richard on Jun 17, 2019 20:31:23 GMT 12
There are more complicated explanations on the effect of the value of the Grid leak resistor, but for a basic answer as far as I remember, the 1M ohm grid leak resistor will give you a slightly higher gain, a lower value grid resistor,slightly lower gain, but better frequency response.
(adding to your list above,not forgetting the Ekco SW86 "42" output tube had a 250K grid leak resistor).
|
|
|
|
Post by Philconut on Jun 18, 2019 16:51:48 GMT 12
I had a look at my various circuits & the value varies from 300k to 1.5meg. As well, the plate resistor of the audio driver ranges from 150k to 350k. These 2 resistors are effectively in parallel across the audio signal appearing at the grid of the output stage and with the plate resistance of the driver, form a voltage divider for the audio signal that appears at the grid of the output stage. So, some theory but no definitive answer.
John
|
|
|
|
Post by OwenYoung on Jun 19, 2019 11:04:20 GMT 12
Hi Chaps - I'm a bit of a radio novice & I don't have an exact answer to radioholic 's Q but my understanding is.... From the venerable Radio Designer's Handbook & from good ol' Morgan Jones... the grid R exists & is sized as large as poss in order (i) to minimise the minor gain loss due to the potential divider presented to the preceding stage (as Philconut describes) & (ii) to allow the preceding coupling cap to be as small as poss. However, this so-called 'grid leak' R has a specified maximum value in order to minimise the risk of grid current leading to valve 'runaway'. The grid R value is specified much higher for fixed bias because there is no cathode degeneration 'stabilisation' as in cathode bias mode. But of course, you know all this  |
|
|
|
Post by OwenYoung on Jun 19, 2019 11:35:59 GMT 12
RDH says:
"... in general, 1 megohm with self bias & 0.5 megohm with fixed bias are suitable maxima & that with receiving valves having an anode dissipation exceeding 10 watts, the grid-circuit resistance should not exceed 0.5 megohm when self bias is used & 0.1 megohm with fixed bias. "
Further, for RC-coupled pentodes & with a series screen resistor, RDH gives calculations for maximum grid resistor vs anode load R & concludes....
" However, it is generally advisable to limit the grid resistor to 2.2 megohms maximum for amplifiers having a reduced frequency range & 1 megohm or less for amplifiers having a maximum frequency of 10,000 c/s or more. "
|
|
