Kerchunk! ... or, rotary switch pop reduction

[CascoSieg]

I'm trying to implement the pre-input stage Depth Control we see in the hand-drawn OR100 https://el34world.com/charts/Schematics/files/Orange/Orange100.pdf
Here's my version attached. There's a major popping problem when I rotate through the six positions, whether or not there is a guitar plugged in. It seems similar to other posts about interstage coupling cap selector switches, but I'm not sure it's the same thing. To suppress it, I've tried a 2M2 resistor across the main poles of the switch. Sometimes it's quiet but mostly I get a pop/kerchunk sound. I'm at a loss for making educated guesses as to how to address it - looking for suggestions, but also wondering what exactly might be causing the sound... I'd think that with the grid-side of the caps always connected, they'd not have a chance to build up a charge, which is a cause I think I've read elsewhere.
Thanks!
~Sieg

[sluckey]

Connect a 1M resistor between grid and ground. Any joy?

[WimWalther]

There are two basic types of multi-pole switches.. break-before-make and make-before-break.

The difference is fairly straightforward: The latter makes the next contact before breaking the previous, the former breaks the current contact before making the next.

MBB is preferred for situations where contact loss (with a pop) can be an issue. BBM Is needed where two adjacent contacts cannot be shorted, even momentarily.

Which type of switch do you have? 

And sluckeys idea of using a 1M bridging resistor can be a good workaround, do try it.

[CascoSieg]

Will try the 1M grid resistor this evening and also check the switch type.
Thanks!

[PRR]

What they say. Then:

Change the first tube. Grid leakage which would not matter in a sensible design might deafen you in this one. Better OR worse is a clue. No change means this theory is bogus.

I'd also ask if you need to change mid-song. It seems to be Stage . . . Studio; you know which you are? Yes, you have to try them all at first but maybe not every gig?

Remember some Harley-Davidson riders do not complain if their chopper goes BANG!! when gears shift. (Not just the Clunk of the shifter, but a backfire.) Embrace the racket!

[CascoSieg]

So, Yes! installing the 1M resistor from grid to ground does eliminate the pop when transiting the first 3 positions (from no cap through the 3 largest ones) and reduces it significantly from the last 2 positions (smallest caps). The switch is a MBB, but I haven't worked out quite what's happening with the electrons to understand how that would impact them in this case.  Two other 12AX7s I swapped in worked bout the same. But to PRR's point, it really is only something that would occur a couple times in a set if at all - i.e. trimming off the bottom for a more overdriven amp setting and back to a cleanly comping a ballad, say.  And while I'm embracing what racket remains, I DO want to hear more about the sensibility or lack of in this design.    Is it just about the grid leak? No MV? Other?  It IS the nicest sounding amp to my ears, of the several I've made, so I'm keen to make reasonably glitch-free.
Thanks Sluckey, Wimwalther, PRR!

[WimWalther]

Remember some Harley-Davidson riders do not complain if their chopper goes BANG!! when gears shift. (Not just the Clunk of the shifter, but a backfire.)

They also think it's super-neato that they only hit about 3 out of 5 strokes at idle. It's their idea of "classic tone".

[PRR]

...The switch is a MBB,...

Then "in between" it contacts both terminals and tends to equalize charge. Even that may be a pop, especially with these very different cap values. (Filling a shot-glass from a wine bottle.) Left image.

The core sin is doing stuff IN the INPUT of a HIGH-gain amplifier. A few thousandths-Volt discharge is LOUD here. We are also messing with the self-hiss level but that's not such a thing.

Planned ahead, you can add 10Meg resistors across all switch cap-only terminals. Right image. Or across all the caps. But that may be a royal pain in an assembled amplifier.

[d95err]

I’d connect a high value (3-10M) resistor between the input and output side of the switch.

[CascoSieg]

3 out of 5 strokes

- love it!
Thanks for the explanation. I suspected, seeing Orange moved the feature to the next gain stage and never went back, there might be some flaw. I'll move it and see how that impacts tone shaping and kerchunking. It sure models better at the front, tho... AND demonstrates the limitations of circuit modeling.
Thanks again all,
~S