Calling Mr. Keen and others...Switching comparison

[modsquad]

I am having some problems reconciling the best way to do remote switching based on my research.  It seems that every method I have come across has its merits and its pitfalls.

I think it would be interesting if someone could help with explaining the pros and cons of the following methods for switching:

Relays
Opto/LDR like H11Fx
CMOS
   - 4013, 4016, 4066, SL74HC4053
JFET

It seems they all do what I want them to do as far as signal switching but I get different opinions on their effectiveness.

[Jobet]

IMHO, the best remote switcher will involve something that directly hard-wires the signals together rather than causing the signal to take different paths through semiconductors.

This may involve switching circuits such as bistable multivibrators (flipflops), but these will involve energizing reed switches or magnetic relays instead of passing the signal through FET's. They might exhibit some nasty behaviour like pops and what not, but the upside is that you don't get any coloration caused by having to pass the signal through semiconductors. This is what I intended to do with my PIC-controlled pedalboard, with the PIC energizing reed relays which pass either the bypassed or processed signal through.

[widdly]

Don't forget analog cross pointing chips.  One of those babies would allow you to change the order of effects in addition to switching them on and off.  They are pretty expensive though.

[Branimir]

I have tried CMOS switching, it works but I wouldn't recommend it.

Built one to make a switch between two GT2 tonepad units and I think I used 4066, used only one chip to make the "true bypass matrix" and I had problems. While one of the pedals was on, the other one was also audible. Then I used two separate IC's to do the switching work, actually I used one IC for input switch and an another IC for output switch, I also got "true bypass" but there was no problem with leaking one signal to another since outputs and inputs were physically separated (different IC's). Lots of fuss and time, and it could have done far easier with a simple DPDT switch.

Although I can see some advantages of the CMOS switching, I remember I did led indication for both units also at the same time...

If you can, stay clear of the semiconductors heh...

[modsquad]

The only problem with a mechanical i.e. 3pdt is that you have to run the signal to the switcher.  I should have clarified that I am refering to remote switching where you are basically sending a control signal to the main unit that switches/routes the signal.

[Jobet]

The only problem with a mechanical i.e. 3pdt is that you have to run the signal to the switcher.  I should have clarified that I am refering to remote switching where you are basically sending a control signal to the main unit that switches/routes the signal.

Yeah I did understand what you meant.

What I meant though, is you can take the basic bistable multivibrator from just about any boss circuit (those 2 BJT's back-to-back), buffer the output(s) with switching transistors and let that actuate a reed switch or magnetic relay. It is through the magnetic reed or relay that the actual signal moves through and hence doesn't go out of the device.

It's the spst switch that actuates the bistable that you can drag out from the device you're controlling. And it doesn't contain any of the audio signal, just a path to ground.

[modsquad]

Gotcha.   Anybody have any idea why you get signal "bleed" when the control current and signal go through the IC?   I noticed doing a search that IC SPDT type circuits are used all over the place for audio switching.