0
\$\begingroup\$

Does this exist?: A mechanical rotary switch that - as it's rotated from position 1 to position N - connects the wiper to the contacts cumulatively rather than only one at a time.

For example, imagine a 1-pole 3-way switch with the following twist:

  • In position 1, the wiper is connected to the first contact only (so far so normal);
  • In position 2, the wiper is connected to the second AND the first contacts at once;
  • In position 3, the wiper is connected to the third, second AND first contacts at once.

Reason for asking: I want to create a variable bass-cut tone control in an audio project. A cumulative rotary switch as described above would allow a number of capacitors to be progressively connected in parallel. (Please don't point me at alternative ways of making tone controls - I'm aware of circuits such as Baxandall, etc.)

Since this is an AC application I don't think I can rely on diodes.

Since composing this question I stumbled upon a web page that says "Rotary switches can also be designed to have many contacts happening at once on a single switch position." So it sounds like what I've describing is an example of that. But are such switches available to hobbyists?

I also read here that a multi-pole switch can be wired to achieve this. Sounds extravagant and, I confess, the way it would need to be wired is not leaping out at me.

\$\endgroup\$
2
  • \$\begingroup\$ The premise to have the capacitors added in parallel of each other cumulatively makes the requirement on the rotary switch unnecessarily complicated. As pointed out by several answers below, you can achieve exactly the same thing by using capacitors of different values, and then using a simple single-pole ratory switch, which would be much easier to find, and cheaper. And the "Please don't point me at alternative ways of making tone controls - I'm aware of circuits such as Baxandall, etc." justification puzzles me. \$\endgroup\$
    – dim
    Nov 4, 2021 at 13:10
  • \$\begingroup\$ @dim, unfortunately I don't have a large collection of miscellaneous capacitors. Although the example above was for three positions, in reality it's more like six. And the actual requirement is to handle two stereo channels. So, since I'll need to purchase the capacitors, I was thinking it might be cheaper to buy a few multi-packs of say 3 different sizes rather than two each of six different sizes. But yes, if a switch to do what I want can't be procured at a suitable price then the 'simple' approach might win after all. \$\endgroup\$ Nov 5, 2021 at 16:14

6 Answers 6

3
\$\begingroup\$

You're looking for a rotary wafer switch.

enter image description here

Image source: India Mart.

The wafers on these are typically available in 1p12w, 2p6w, 3p4w, 4p3w and 6p2w variants. You probably want 12-way wafers to give you ten or twelve settings.

Posn A B C D E F G H I J
   1 
   2 =
   3 = =
   4 = = =
   5 = = = =
   ...
  10 = = = = = = = = =
  11 = = = = = = = = = =

... where = is a connection through that layer.

If you're trying this I suggest a big knob so that you have enough leverage to turn it.

\$\endgroup\$
3
  • \$\begingroup\$ In principle a modified single wafer could do it without stacking 12 wafers. The question is whether such a thing is produced. \$\endgroup\$
    – user253751
    Nov 5, 2021 at 9:41
  • \$\begingroup\$ @user253751, yes, if you're thinking of a 180° wiper segment that engages with more and more contacts as rotated. All the contacts would have to be on one semi-circle of the wafer resulting in 15° steps rather than the more normal 30° for a 12-way switch. \$\endgroup\$
    – Transistor
    Nov 5, 2021 at 14:24
  • \$\begingroup\$ (15 degrees is not really strange at all, though - it's pretty normal for a 12-way 2-pole switch) \$\endgroup\$
    – user253751
    Nov 5, 2021 at 14:26
3
\$\begingroup\$

All you need is a two pole rotary switch with three positions.

The first capacitor is always connected, so the rotary does not have to do anything with it.

So rotary only ever needs to connect zero, one or two additional capacitors.

First position, both extra caps disconnected.

Second position, only pole 1 connects extra capacitor 1. Pole 2 keeps extra capacitor 2 disconnected.

Third position, pole 1 connects extra capacitor 1 also in this position. In addition, pole 2 now connects extra capacitor 2.

\$\endgroup\$
6
  • 1
    \$\begingroup\$ OP wants an N-way switch. I presume N > 3. \$\endgroup\$
    – Transistor
    Nov 3, 2021 at 21:19
  • \$\begingroup\$ N poles are needed for capacitors count M minus 1. However It would make more sense to just not add capacitors cumulatively, but simply use a 1-pole switch and always connect a single capacitor, so a single say 1-pole 12-way rotary switch would suffice. \$\endgroup\$
    – Justme
    Nov 3, 2021 at 21:41
  • \$\begingroup\$ @Transistor, yes, N was 3 in my example but is 6 for the actual application. Also (and perhaps I should have mentioned this from the outset) my actual requirement is to handle two stereo channels (so I guess that doubles the number of poles needed). \$\endgroup\$ Nov 5, 2021 at 16:15
  • \$\begingroup\$ @DavidEasley Yes, you should have mentioned N=6, and even doubling thay for two stereo channels. You can still add them to the question but with the extra restrictions it invalidates a lot of answers made to the initial specs. Also can you add why you want to rather use your existing caps and buy an unobtainable special switch which will be hugely more expensive than say buying a bog-standard two-pole six-position rotary switch and twelve new caps? \$\endgroup\$
    – Justme
    Nov 5, 2021 at 16:26
  • \$\begingroup\$ @Justme, really appreciate your input. <br /> "why you want to rather use your existing caps" - what existing caps? <br /> "... and buy an unobtainable special switch which will be hugely more expensive" - well, that's what I was seeking to find out by posting the question. <br /> "[rather] than say buying a bog-standard two-pole six-position rotary switch and twelve new caps?" Yes, the premise may seem moot but I've tried to explain in my comment to @ dim, above. \$\endgroup\$ Nov 5, 2021 at 16:38
3
\$\begingroup\$

This is how to wire up the standard rotary switch in case Transistor's wafer switch suggestion is unattainable for whatever reason. You could also try wiring up the capacitors in series and having the switch reduce the capacitance as you add more or use a single capacitor and have your (I assume) resistors ganged on the rotary switch. You could probably get more consistent steps that way since resistor tolerances are generally tighter than capacitor ones. Switch wiring

\$\endgroup\$
3
\$\begingroup\$

Yes they exist. For example in an old audio amp I repaired recently there was a switch like that, but with a wider contact (arrow) so it hits several contacts at the same time. This was used to enable one or both speaker output terminals.

enter image description here

Same principle for a wafer switch:

enter image description here

However... a wider wiper that spans several contacts will probably be special order.

You could make a PCB rotary switch, like in multimeters:

enter image description here

This looks like a good option for DIY because you can just print the pattern you want. Then you can mount the spring contact wiper on another PCB that rotates relative to the other.

Wiper suggestions: one two three four

I don't know if the contact will be reliable enough with vibration though. If it is intermittent, it will produce unpleasant sound.

Failing that, you could just use a row of toggle switches, or a simple rotary switch driving CMOS analog switch chips.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ The way you've worded it, it sounds like the first photo should be an example of a rotary switch with a wider contact, I was confused for a while. \$\endgroup\$
    – DamienD
    Nov 4, 2021 at 9:44
  • \$\begingroup\$ The wider contact is often called "make before break", to distinguish it from the usual "break before make". \$\endgroup\$
    – Simon B
    Nov 4, 2021 at 16:48
2
\$\begingroup\$

You could use a gray coded encoder for that. Once upon in the time (perhaps I could find a document) I have already computed standard resistors and excel table, to be used like a potentiometer. So it is doable.

\$\endgroup\$
1
\$\begingroup\$

Here's the schematic for the multi-pole, multi-way switch option.

enter image description here

Using a single-pole, multi-way switch would simplify the wiring.

enter image description here

Various combinations e.g. 4-pole, 3-way or 1-pole, 12-way are available in the same package.

enter image description here

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.