Bypass effect when PSU/battery fails?

[therecordingart]

I've been searching on "relay bypass switching" and variations of that, but I'm not sure that is what I should be looking for. I want to implement a true bypass (especially to non-bypassable pedals like buffers) that bypasses the effect when the battery or power supply fails.

Any thoughts on this?

I'm searching my butt off, but maybe I just suck?!? 

[R.G.]

It's an odd request, but a simple one to do.

Just wire up a non-latching DPDT relay (about $3.00) to bypass the circuit when the coil is not powered, then hook the coil up to the power supply.

Power good, relay enables the effect. Power fails, flips to bypass.

[therecordingart]

Thank you. I figured it would be easy, but thought I was overlooking something because that was my initial plan. 

[therecordingart]

Why is this an odd thing to do? I know it isn't the norm, but it makes sense that you wouldn't want power failure to ruin a gig. Lots of guys still swear by batteries and if a pedal dies in use that would suck. Then again...you don't hear of that happening very often. I'm probably just overthinking again.

I'm going to play with it and see how it pans out.

[PRR]

> see how it pans out

R.G. just set you on the path. There's details for you to work out.

The relay pulls power. On battery, failure comes sooner.

What is "fails"? Yeah, zero voltage is a failure, but what about 8V? 7V? 6V? 5V? Some chips need a full 7V internally plus a margin, crap-out at 8V. Some other chips and circuits keep going even down below 5V. A "9V" relay, once pulled-in, is sure to stay pulled to 5V, maybe to 4V. So if you could be faced with a drop (not to zero), you should co-ordinate relay and effect voltages. Since relays are not available in arbitrary dropout voltages, this may mean the added fun/complication of a voltage detector, adjusted to what you feel is the minimum useful voltage for your specific effect. Can be little more than an NPN and a Zener, or one of many common chips. (If you have a leftover opamp, and a low-current relay, you are nearly there.)

[therecordingart]

> see how it pans out

R.G. just set you on the path. There's details for you to work out.

The relay pulls power. On battery, failure comes sooner.

What is "fails"? Yeah, zero voltage is a failure, but what about 8V? 7V? 6V? 5V? Some chips need a full 7V internally plus a margin, crap-out at 8V. Some other chips and circuits keep going even down below 5V. A "9V" relay, once pulled-in, is sure to stay pulled to 5V, maybe to 4V. So if you could be faced with a drop (not to zero), you should co-ordinate relay and effect voltages. Since relays are not available in arbitrary dropout voltages, this may mean the added fun/complication of a voltage detector, adjusted to what you feel is the minimum useful voltage for your specific effect. Can be little more than an NPN and a Zener, or one of many common chips. (If you have a leftover opamp, and a low-current relay, you are nearly there.)

I had a similar thought to this when I was trying to go to bed last night. It's good to know that even though I'm very much an electronics noob that I'm thinking on the right track. Thank you very much for your help. I know this idea might be nuts, but DIY isn't for the sane.

[therecordingart]

> see how it pans out

R.G. just set you on the path. There's details for you to work out.

The relay pulls power. On battery, failure comes sooner.

What is "fails"? Yeah, zero voltage is a failure, but what about 8V? 7V? 6V? 5V? Some chips need a full 7V internally plus a margin, crap-out at 8V. Some other chips and circuits keep going even down below 5V. A "9V" relay, once pulled-in, is sure to stay pulled to 5V, maybe to 4V. So if you could be faced with a drop (not to zero), you should co-ordinate relay and effect voltages. Since relays are not available in arbitrary dropout voltages, this may mean the added fun/complication of a voltage detector, adjusted to what you feel is the minimum useful voltage for your specific effect. Can be little more than an NPN and a Zener, or one of many common chips. (If you have a leftover opamp, and a low-current relay, you are nearly there.)

So I could create my "voltage detector" from an IC setup as a comparator with the inverting input being help at my predetermined voltage with a zener diode, correct? Once the supply voltage drops below my zener voltage the output swings negative and allows the relay to switch states (bypass).

[R.G.]

Yes.

But as Paul pointed out, the circuit that does this, and the relay coil itself also does more battery drain, and makes the battery (where used) go down faster just by being there on the battery. The usual recourse is to latching relays, but now you have to save enough power to flip the relay to the not-enough-battery position when the battery declines.

I would use an 8-pin microcontroller with A-D inputs in common with a latching relay. I'd just tell it to watch the battery voltage and when it's less than X, whack the relay to the "low battery/no power" position, and then keep watching for the battery to come back up. The advantage is that it can do exactly what you want, and some microcontrollers are very, very low power. They don't eat much just watching, and can easily fire latching relays as needed. The latching relays don't eat any power unless they're switching.

But that takes the ability to program uCs, and that can be a big step all by itself. Low power comparators and analog circuitry can do much the same, but the design may be complicated.

[therecordingart]

Thank you, RG!

[Mark Hammer]

Um, apart from the various practical disadvantages it might have, vis-a-vis buffering, one of the advantages of a traditional bypass stompswitch is that it permits bypassing the circuit when the juice disappears.  It doesn't need reserve power, or microcontrol, and it works as intended, regardless of which side of the switch you decide to use for the bypass setting.

Sometimes low-tech is wonderful.

Having said that, i fully understand that sometimes relays are called for because they can do things that stompswitches can't (like remote switching).

[R.G.]

Yep. That was the original reasoning on using true bypass.

Of course, with a dead battery not erratically not working, then working, etc., swapping cables works too.