Cap switching methods

[MikeH]

Are there any "tone sucking issues" (or any others for that matter) with switching caps using an SPDT, like in method "A" below?  I've only done this using a DPDT like in method "B", which works great, unfortunately I only have a bunch of SPDTs on hand right now and no DPDTs.

[R.G.]

The issues that matter are ones of the residual voltage on the caps.

The SPDT version works just as well as the DPDT version for switching the cap values. However, if the caps are being used to block DC as well as conduct AC, the cap that is NOT connected will tend to leak down from whatever DC voltage was across it the last time it was connected. Then when you switch back to it, the sudden charging of DC into the cap causes a pop.

The way you get around this is with one high-value resistor per capacitor to pull the being-switched end to the same DC voltage it will be at when the switch connects. 1M is usually good unless you're switching input caps in a FET or high impedance opamp circuit. The high value resistor keeps the DC level correct on the cap so there is no pop when you switch between caps.

If both sides of the caps are at the same DC level, you can just switch away and not worry about the popping.

[MikeH]

Well, I guess I should be more specific.  I'm using this to switch input caps; so I guess it's blocking DC?

[Processaurus]

Recently I tried putting two caps right on a spdt toggle switch, with each cap connected to an outside lug, and then to the common.  The in and out were connected to the outside lugs (the two throws).  That way it was like the two caps were in series, but the toggle switch would either short out one or the other.  No pop.

Yes, you don't need to "true bypass" caps like in figure B.

[cab42]

For the switching if input cap in my rangemaster I used a method suggested by Mark Hammer (as far as I remember). I have searched but I can't find  the thread.

It uses a spdt on-off-on switch for switching between three values. Input is connected to common. Capaciator A is connected to common, capacitor B is connected to one outside lug and capacitor C  to the other outside lug. The other ends of the capacitors are connected together and to output.

This way you can switch between capacitor A (switch in off position) or A+B in parallel or A+C in parallel.

It should pop as B or C are allways unconnected, but I don't remember. I only use one posistion.

I used 5nF for A, 50nF for B and 10nF for C. This way I can switch between 5, 15 and 55nF.

Regards

Carsten

[Pushtone]

I once read here, I think, but could not find again 

a method using rotary switch to "tap" at point in a series of caps.

[Processaurus]

I once read here, I think, but could not find again 

a method using rotary switch to "tap" at point in a series of caps.

neat idea, I have a bunch of 2-12pst rotaries, that could possibly make for an interesting, custom taper feeling one knob tone control.  So would you have the caps in order of largest to smallest, and short between the out of the smallest and the different tap points between the caps?  Though you would get the series capacitance, so you'd have to dust off some formulae.

[Mark Hammer]

The original idea for this was given to me by John Hollis with his "mode" switch on the Zombie Chorus.  He placed two caps in series and used a SPST toggle to shunt one of them, providing a shift in BBD clock frequency without popping.  The cleverness of this approach is that at no time does either/any cap lack a path to ground, so popping from switching should either not occur, or at least be drastically reduced.

I'm fond of the 3-position center-off toggles because they can be easily used to shunt caps for 3 choices.  Of course the thing you tend to rely in in these instances is that the math will work out for you.  That is, between the standard values of caps in the general range of interest, that series switching will yield a meaningful and practical change in sound.

There is nothing at all wrong in principal with switching parallel caps.  In many respects, it provides a greater likelihood of achieving "ideal" capacitance values.  The question that needs to be grappled with is whether it is the sort of function you might ever switch in a gigging situation, such that audible switch-popping would be problematic.  Ultimately, if the effect is bypassed, then it doesn't matter what cap-switching approach you use, because neither will be audible.  For some folks, though, having to bypass the pedal to be able to switch settings is a bit of a nuisance.

[Bucksears]

Couldn the voltage surges released from caps when they 'pop' degrade/damage components within the circuit?

- Buck

[Bucksears]

bump

[idlechatterbox]

5-position, rotary switch, used to select cap in a version of the Bazz Fuss. Double-pole switch, with one "bank" of poles having one leg of each cap. There are 4 caps. You just can't see the small black one. The other legs of each cap are tied together and connected to the red wire. Orange wire is the "center" wire, not connected to either leg of the cap. Orange and red then go on the stripboard where the original cap was supposed to be. (Ignore the yellow wires; they are on the other "bank" of poles and are used to turn the pedal on/off, since  don't like stompswitches)   

[alderbody]

my version of the rangemaster input cap switching



6 pos rotary.

values start from low (super treble) and climb up to full boost with a .15uF as mentioned in the Rangemaster "bible"...(Technology of...)

works fine, anyway.

[remmelt]

I'm fond of the 3-position center-off toggles because they can be easily used to shunt caps for 3 choices.  Of course the thing you tend to rely in in these instances is that the math will work out for you.  That is, between the standard values of caps in the general range of interest, that series switching will yield a meaningful and practical change in sound.

I've posted a follow up to this in another thread, is this what you meant?
http://www.diystompboxes.com/smfforum/index.php?topic=56671.msg438331#msg438331