Grid leak resistor vs. MV pot before power tube

[Yoda]

So I’m building a couple of 5F1 style amps (not exact clones) and in looking at how others have implemented the master volume pot into this single ended amp circuit I’ve seen a couple people say to replace the 6V6 power tube’s 220K grid leak resistor with the master volume pot. One such source of this is Rob Robinette’s webpage on his Six Shooter build.

I already have another 5F1 clone I built a couple years ago with a master volume and I left the 220K grid leak intact, between the MV pot and the 6V6 grid and the amp seems to be working fine this way.

I’m wondering if someone could explain to me what the difference is between these two setups? If I understand correctly the grid leak resistor provides a ground reference for the cathode to grid circuit correct? Why is it not needed if a master volume pot is installed just before the power tube? The way I see it the master volume just controls how much signal is present before it gets to the grid leak and power tube. With the MV at zero the grid leak is there to perform it’s usual function.

Thanks in advance.

[sluckey]

The MV pot also functions as a grid leak resistor, just like the volume pot functions as a grid leak resistor in the 12AX7 grid just prior to the 6V6.

[Yoda]

Simple enough answer.

So let’s say the volume pot is 250k, when the master volume pot is at zero does the ground leg and middle wiper of the pot provide a 250k resistance to the 6V6 grid similar to the 220K grid resistor? Looking at a schematic it seems like this is how it can provide the grid leak function. But it also seems like if the resistance is all on the signal legs of the pot to kill the volume then the other two legs would be providing zero resistance. That would be like having zero resistance from grid to ground when the volume is turned all the way down. Is this correct?

[R.G.]

You're getting into second order effect now.

1) to a first approximation, the MV does provide a path to ground for any charge building up on the grid, which is what grid leak resistors are for
2) tubes are in general fine with grid leak resistors below some maximum, which is usually specified on the tube data sheet; if the original grid leak was 220k, then a 250k master volume should be fine at any setting, including zero
3) pots are imperfect; yes, a pot's wiper would work to provide a grid leak, but wipers on pots always have some series resistance of a small fraction of the pot value; the second order Bad Thing that can happen is that pot wipers can become completely open with age and dirt

If the pot wiper is the only grid leak resistance, and it goes open, charge builds up on the grid and the tube can go into runaway and kill itself. This will not happen often, but it's unpleasant if it ever does. Like a bias pot in a fixed bias amp, it's a reasonable practice to put a safety resistor so the grid has some resistance to ground or bias if the pot wiper opens.

[martin manning]

If the pot wiper is the only grid leak resistance, and it goes open, charge builds up on the grid and the tube can go into runaway and kill itself. This will not happen often, but it's unpleasant if it ever does. Like a bias pot in a fixed bias amp, it's a reasonable practice to put a safety resistor so the grid has some resistance to ground or bias if the pot wiper opens.

This safety resistor is a feature of the Lar-Mar variation on the KF Type 2 post-PI MV, and it is typically a 2M2 from the wiper to the bottom (CCW) lug of the MV pot. With a 250k pot, at max volume the resistance is very close to the usual fixed 220k.

Another way to go about it would be to connect the safety resistor from the wiper to the top of the pot, and make its value something lower, say equal to the pot value. This would work in essentially the same way at max volume, where the resistance to ground (or bias voltage) would be the pot element resistance (250k), but the resistance in the event of a wiper failure will be at most 2x the pot value instead of a rather high 2M2. The other consequence of this is that the resistance to ground at low volume is reduced to a value approaching half the pot value, which will result in a one-octave increase in the corner frequency of the LP filter formed by the coupling cap and grid leak resistance. That might be beneficial at low volume settings. It will also affect the PI's AC load line. It would be an interesting experiment to see which sounds better as the volume is reduced.

[Yoda]

Lots of info to wade through here, thanks guys, I appreciate it.

It sounds like it doesn’t hurt anything then if I were to leave the 220K grid leak though it may act as a voltage divider when the MV is not all the way up correct?

[martin manning]

Not sure what you mean. Post your proposed schematic.

[R.G.]

This safety resistor is a feature of the Lar-Mar variation on the KF Type 2 post-PI MV, and it is typically a 2M2 from the wiper to the bottom (CCW) lug of the MV pot. With a 250k pot, at max volume the resistance is very close to the usual fixed 220k.

Didn't know the specifics of [someone's special type of MV]; didn't know if the OP did either from the post.

Another way to go about it would be [...] a one-octave increase in the corner frequency [...] It will also affect the PI's AC load line. It would be an interesting experiment

Yes, many ways to ensure DC connection in the face of flaky pots, each with its own side effects. Thought and circuit analysis is needed to see which one is better or worse for the intended application. But it's really fundamental to realize that a sliding connection can simply become open. Many newbies don't know that.

[martin manning]

Didn't know the specifics of [someone's special type of MV]; didn't know if the OP did either from the post.

That was the reason for adding the background ;^)

[Yoda]

I don’t have a schematic for this at the moment. When I have some free time I will draw one up and post it just to see what kind of answers I get to my last question from my previous post (regarding having the MV and leaving the 220k grid leak between it and the 6V6).

[Yoda]

Here’s a quick schematic showing what I mean by 220K grid leak resistor between the MV pot and the 6V6 grid. Ignore the other broken lines on the schematic as I don’t have anything proper for drawing schematics so I had to piece this together from some 5F1 schematics I printed:

[sluckey]

That's fine.