Which DPDT foot switch do you prefer?

[JFace]

I'd like to use a DPDT switch for Millennium 2 bypassing, but I'd like to know about the feel and the longevity of what's on the market without having to buy everything. Do you have any comments on the alpha, x-wing, or mammoth regular and industrial switches?

[R.G.]

My congratulations on the way you stated that. Nicely done.

I'm being serious.

[JFace]

I'm also interested in which switches tend to debounce more/less with all other considerations taken into account (e.g. pulldown resistors, DC offset, high gain circuit).

My congratulations on the way you stated that. Nicely done.

I'm being serious.

Thanks! 

[deadastronaut]

alpha...+1. nice short travel. 

[R.G.]

I'm also interested in which switches tend to debounce more/less with all other considerations taken into account (e.g. pulldown resistors, DC offset, high gain circuit).

That's kind of an imponderable. It's dependent almost entirely on the internal mechanical design and any lubrication/damping grease used. The electrical parts can only suppress electronically whatever the mechanical contacts do, so the other electronic considerations don't make much difference to the actual bouncing, only what you see of it. It's an interesting question, but the answer is probably to put them on a scope and look. Manufacturers sometimes change the innards without changing the outside, so it may change at random (as seen by us). It's also likely that very few people here have done any debounce looking.

[Chugs]

I predominantly use Mil bypass and prefer Alpha to X wings.

[JFace]

It's an interesting question, but the answer is probably to put them on a scope and look.

I was thinking the same thing. What inspired me to ask this question was that I used a 3PDT PCB mounted switch on a well designed circuit with low gain, pull down resistors, and even input and output series resistors. The pop was so loud I set out to find a better solution. It is a persistent problem, why hasn't anyone done a test on the various switches and schemes? Just use a simple buffer circuit as your control, and change the switching scheme, and measure the dBU of the click.

[R.G.]

I think the answer is that the answer would be specific to that switch, and would vary with any signal through it.

Hard contact mechanical switches can never be completely pop-free, if only because they are hard-contacts, and they send the output line from open circuit while the contacts are in between states and moving to as much as the peak signal amplitude when they "make". And they may do this several times as they bounce. That's the *best* case.

However, this is typically limited to the size of the signal. If you're seeing "the pop was so loud", then I would suspect that something else is causing a problem. Is this on only one circuit, or all the circuits you've ever tried?

I personally left hard switches behind years ago. Hard switches persist for two reasons: (1) they're easy to understand compared to soft switching and (2) they've become intertwined in the mythos of pedals, and therefore hyped to high heaven by pedal makers without real technical chops.  True bypass switching is ingrained as something aking to home, hearth, motherhood and applie pie, and users simply demand it, not fully undertstanding the issues, but having been conditioned to say " um, it's true bypass isn't it?", with the implication being that if it's not, it can't be good.

In many ways, Pavlov should have used people. The dogs took longer to train.

[Mark Hammer]

In many ways, Pavlov should have used people. The dogs took longer to train.

And maybe drummers; dogs don't drool quite that much, and Pavlov measured saliva as his dependent variable. 

The dab of grease inside mechanical switches is both a boon and a curse.  As I've blathered on about for years, the damping grease inside stompswitches plays a role in debouncing the contacts.  At the same time, many of the switch-failures that people here report stem from overheating that damping grease during soldering, such that it liquefies, flows along the contact, and serves as an insulator, impairing switch continuity.

In an effort to keep the good and lose the bad, it seems like the optimum strategy is to disassemble any new switches one gets, pull out the rocker contacts with fine tweezers, grease and all, do whatever soldering needs to be done, place the rocker contacts back in (with their grease), and reassemble the switch.

OR one could migrate to momentaries and electronic switches. 

[JFace]

However, this is typically limited to the size of the signal. If you're seeing "the pop was so loud", then I would suspect that something else is causing a problem. Is this on only one circuit, or all the circuits you've ever tried?

I've made perhaps a dozen circuits with 3PDT, mostly the Taiwan blue type, and more recently the Alpha. I pondered if I could quicken the assembly process like some of the boutique builders by getting a PCB mount 3PDT, and that was the loudest of any of the previous arrangements.

Electronic switching is also on my mind for comparison of clicks. I've breadboarded the boss style flip flop circuit, and it was silent (to my ears). I've tried CMOS switching and noticed a slight bit of switching sound. Everything is a trade-off. I'm hoping to explore all avenues and find the best balance of:

PCB real estate
Number of switch poles
Assembly time
Cost

And this consideration will probably be different according to each project. For instance, if a project has a few components, and the cost is low, then a high parts count electronic switch would make sense. But if I'm cramming a bunch of components into a 1590A twinkie box, CMOS switching is not possible for me.

[Mark Hammer]

Just about every Boss pedal you will see that uses FET switching, will feed the gate of the JFET via a diode.  That diode is preceded by a 1M/47nf pair intended to provide juuuuuussssttt enough lag to turning on the JFET that it won't click, while not providing so much lag that the changeover from effect to bypass is perceptible.

[PRR]

> It is a persistent problem, why hasn't anyone done a test on the various switches and schemes?

As Mark says-- Roland/BOSS/etc settled this a long time ago. Timed FET switches get pop way-way down, and audible side-effects are nearly zero. Given a choice 0.02% THD or POP, I'll take the trace THD all day long (especially in distorty effects).