Max. grid resistance

[stoo]

Hi!
Does anyone have a handle on "Maximum control grid resistance"?
The spec sheets don't always list this value and those that do don't match. This has to do with 6550 to el34 (and vise versa) conversions. I see 6550 amps with 100K bias feeds and El34s with 220K bias feeds. WHY?
Thanks all
Stew

[SY]

There are, sadly, a lot of designers (I use that word rather than "engineer") who build a few prototypes that violate datasheet restrictions, see that they don't blow up, then go into production. And everything might be ok, sometimes, for a while. And the reputation of tube amps as unreliable gets reinforced. Grid resistors are a perfect example- a particular batch of tubes might have a little harder vacuum than average, and one can get away with abusing the max grid resistance rating. But what happens when those tubes get older? Or when a more "typical" tube is plugged in at replacement time?

Bottom line: let others make mistakes, you don't have to. This is one rating I would follow carefully.

NB: There is often a difference between this max rating for cathode versus fixed (grid) bias. Don't tempt fate by using the cathode bias rating in a fixed bias amp!

[Bandersnatch]

hey-Hey!!!,
From the subjective and objective... The smaller this value the better I've liked the sound of the amp. That assumes that the preceeding stage is comfortably able to drive this load placed in parallel with its plate load. I do tend towards use of grid chokes instead of resistors.
cheers,
Douglas

[Miles Prower]

Originally posted by stoo
Hi!

Does anyone have a handle on "Maximum control grid resistance"?

There would be no limit if all VTs were truly filled with vacuum. However, this is not the case. The best vacuum pulled in a lab had a pressure of 0.00000000013 atmospheres. As impressive as that looks at first, that still represents 100 million atoms and molecules per cubic centimeter. 100 million is one helluvalot. And let's not forget that a successful pump-down of a VT is many orders of magnitude greater than that. So what you have is a not-so-vacuum tube where there are about a billion ionization events every second that thing is running.

It's these unwanted ions that can, and do, create control grid currents where there should be none. Granted, these currents are not very large, but can create some excessive voltages through high resistances. This can cause plate current run-away (just like an overheated BJT heading south). To make sure this doesn't happen, max values for grid DC return resistors are stated. These usually being around 1.0M, sometimes less (50K for the 50 for example).

[stoo]

Originally posted by Miles Prower


There would be no limit if all VTs were truly filled with vacuum. However, this is not the case. The best vacuum pulled in a lab had a pressure of 0.00000000013 atmospheres. As impressive as that looks at first, that still represents 100 million atoms and molecules per cubic centimeter. 100 million is one helluvalot. And let's not forget that a successful pump-down of a VT is many orders of magnitude greater than that. So what you have is a not-so-vacuum tube where there are about a billion ionization events every second that thing is running.

It's these unwanted ions that can, and do, create control grid currents where there should be none. Granted, these currents are not very large, but can create some excessive voltages through high resistances. This can cause plate current run-away (just like an overheated BJT heading south). To make sure this doesn't happen, max values for grid DC return resistors are stated. These usually being around 1.0M, sometimes less (50K for the 50 for example).

Thanks MP.
Ya I did read that 50K was the Max for 6550s but I routinely see 100Ks in 6550 amps. BTW I'm talking musical instrument amps here.
Stew

[GMuffley]

Originally posted by SY
There are, sadly, a lot of designers (I use that word rather than "engineer") who build a few prototypes that violate datasheet restrictions, see that they don't blow up, then go into production. And everything might be ok, sometimes, for a while. And the reputation of tube amps as unreliable gets reinforced. Grid resistors are a perfect example- a particular batch of tubes might have a little harder vacuum than average, and one can get away with abusing the max grid resistance rating. But what happens when those tubes get older? Or when a more "typical" tube is plugged in at replacement time?

Bottom line: let others make mistakes, you don't have to. This is one rating I would follow carefully.

NB: There is often a difference between this max rating for cathode versus fixed (grid) bias. Don't tempt fate by using the cathode bias rating in a fixed bias amp!

A good example is the original Quicksilver monoblocks that used the 8417 tube. Maximium recommended G1 resistance is 47K, but the designer used 164K. People complained about the 8417s being tough to match for those amps when the actual culprit was thermal runaway prompted by the high G1 resisitance.

[Richard Ellis]

Sy...
Really thought a production amp would have been "engineered" first, then prototypes would have been built, tested & measured et.al......as I am learning more and more, some schematics that exist in cyberspace are also suspect. I found an "error" in one , as it blew thru Vgk MAX............do your homework! This is the "improvement" we have heard about...that WE have to do....fixing lazily "engineered" gear.
________________________________________Rick...... .......
813s RULE

[Bandersnatch]

Originally posted by Miles Prower


These usually being around 1.0M, sometimes less (50K for the 50 for example).

ummm, Miles...the RCA data sheet shows a max of 10k for a 50's grid.

Now the curious thing about this grid current is that it is dependant on grid voltage, yet a large AC impedance doesn't do much to it. This isn't a linear relationship, let alone once you take the grid positive...
cheers,
Douglas