Point-and-Blend: fewer than 9 poles?

[stevewf]

Can it be done with fewer poles?

I've been toying with Bourns MN pots. Here's an attempt at a 3-coil setup that uses one:


In addition to the MN blender pot, it uses a 3-position switch (with, uh, 9 poles).

Here's what I'm trying to do:
A) The basic idea is that a "Main" coil is chosen via the 3-position switch (S1); the two remaining coils are selectively blended into the signal via the MN pot. That's the "Point" and the "Blend" in the title of this thread, respectively.

B) Note that there are two South coils and one North. The polarities come into play along with phasing to achieve some level of hum cancellation. When a South coil is selected as Main, then the other South coil has gotta be Half Out of Phase with it. I find that HOop usually sounds better than OoP.

C) When the Bridge coil is Main, then it's in series with the output of the blender; however when the Middle or Neck coil is Main, then the blender's output is in parallel with the Main coil's.

D) The blender is fixed to blend its two inputs in parallel with each other.

A table:
3-pos | Main   | Ser/Par | Blend      |
 1    | Bridge |   Ser   | N + [-M/2] |
 2    | Middle |   Par   | N + [-B/2] |
 3    | Neck   |   Par   | M + B      |
(position 1 is the bottom contacts in the diagram)


Example, with the 3-position in the middle position:
The Middle coil is the Main one. Its output goes to the jack, in parallel with the output of the blender. The blender gets its input from the Neck coil and from the Bridge coil - the latter having its phase swapped, and put in series with a capacitor to achieve HOoP. So in Position#2, when the blender is fully CW, we get M + N; fully CCW we get M + [-B/2]; centered we get M + (N + [-B/2]).

First, I'd be grateful for validation that the above diagram would provide that table. Then, I welcome suggestions for improvements, while respecting the requirements per the table.
-------------
Some notes:
- I'll probably be adding more controls in order to handle master vol and tone. Pots, of course. I hope the tone doesn't suffer too much. High-value and No-Load pots will be explored. Of course, the MN pot is available only in 250KΩ and 500KΩ versions.
- I might do this in conjunction with an "AND/NOR" option (mentioned in another thread, Bourns MN blender: Help "Flipping" the pot). It'd be the way to get a single coil (the Main one) without blending it with another coil (NOR).

I might be able to make a 9P3T rotary switch by bashing two of these together:
 (6P3T rotary on eBay)
but I'd rather avoid the work... which would have a high probability of failing anyway. I haven't yet ordered them.

Anybody able to reduce it down to 6 Poles while keeping to the table above? 6 Poles would allow me to use the above switch without modding it.
Or, anybody got a good source for a 9P3T rotary that fits in a 1/2" panel hole?

TIA -Steve

[MattB]

www.surplussales.com/index/Rotary-Switches.html

Might be something you can use in here. Check the sealed rotary switches section- they tend to be smaller.

[stevewf]

Dec 29, 2022 9:06:58 GMT -5 MattB said:

www.surplussales.com/index/Rotary-Switches.html

Might be something you can use in here. Check the sealed rotary switches section- they tend to be smaller.

Yup, that's a new supplier for me, and what a range! Thanks - S

[stevewf]

Maybe I can elicit some validation of the circuit via this diagram of what I want:

All three coils are always "active".
In each position, a different coil is selected as the Main coil (e.g. 1=Bridge)
In positions 1&2, a coil is to be Half Out of Phase (e.g. 1=Middle:HOoP). In position 1, there's to be a series flow; the other positions with parallel flow.


And here's a candidate scheme for getting it (same diagram as above):


Does it seem like it'll work? Can it be done with fewer poles on the selector switch? Thanks!