RG Keen latching relay bypass circuit - max voltage?

[intripped]

I've built this circuit and it's working very well, with both Panasonic TQ2-L2-5V and NEC EA2-5TNJ relays.




Since many pedals have the option of being powered with different voltages, 18V also, I don't know if this circuit would work or something would be fried, the relay expecially.

...if that's the case, what could be a way for utilizing this circuit in such pedals?
Thanks

[R.G.]

There are two power supplies to worry about here. One is the CMOS inverter package. The CD4xxx family is good to at least 15V, and some of them (check the datasheet!) are OK to 18V. The second power supply is the one which runs the relays. The two relays you mention are specified at a nominal 5V. Many relays are good to a good overvoltage, sometimes as much as 200% of nominal. But those relays will overheat if they're on very long with a supply over 9-10V. 18V is probably a death sentence.

The good news is that the logic chip and relay don't have to use the same DC power supply. The relay supply voltage can be anything that the relay and transistor will survive.

If you simply must use 18V, change to the 9V rated versions of these relays. That at least has the chance of surviving 200% of specified coil voltage.

I personally don't understand the trendiness for powering pedals from higher power supply voltages. I guess maybe if you have a powerful urge to tinker and mod and supply voltage is all you can really tinker with, I can see turning the knobs you have access to. But the likelihood of coming out with something good instead of reduced pedal live is quite low.

[intripped]

Thank you very much RG for the answer!
I will check data sheets for sure; I was uncertain about the relays because it looked already strange to me that a 5V relay is used in a 9V-powered circuit.

What if I use a voltage regulator for the relay and transistors?
Would a 78L06 or a 78L08 do the job? Even at 18V?

...or what about a simple voltage divider?

Regarding the overvoltage trend: i for example prefer to run my CE2 clone at 12V because it sounds much better IMO. Same for PT80 delay.
And many commercial pedals, mainly transistor-based, would give a different ...flavor if supplied with more than 9V; marvel drive from rambleFX for example, Fetto, ... and others that now I can't remember.
To my eyes, noob eyes, it looks like another available option for shaping your tone

[PRR]

For LONG-term power applied, you can be sorta-sure of 125%, and 140%-150% if it is not a life-and-death situation (elevator controls). 200% is seen rarely.

BUT this is a PULSE application. Power is applied only about 1/100th of a second. That is far too short to over-heat a coil. You could go way higher. With the fine wires, at some point the wire will glow and explode like an incandescent lamp, but I think that point is like 10X over-voltage.



I think this mod will be OK to 22V (a hot "18V"). The CMOS current is nearly negligible. The LED is a pig but with modern bright LEDs we can up the resistance. Insert a 1K resistor, and a 12V Zener. The CMOS will never feel over 12V, safely short of the 15V serving suggestion. We could probably add a fat cap here and power the relay pulses, but I think they will be fine on raw 22V. Transistors and diodes must be rated >23V and this is not a problem.

For 24V+ be sure the 1K is a 1/4W part (above I assumed 1/8W). Over 30V use 1K 1W and start to worry about relay coil failures. Since R.G. says these are nominal 5V relays, at some point you should instead switch to a 7805 regulator, which will take 7.5V to 35V, deliver 5.0V, and shut-down if shorted.

[R.G.]

What if I use a voltage regulator for the relay and transistors?
Would a 78L06 or a 78L08 do the job? Even at 18V?

Those would work, but as Paul says, as long as the relay is >>pulsed<< you're probably OK. My history with circuits has led me to appreciate "Murphy's Law", the idea that if anything can go wrong, it will. And sometimes things go wrong even when they can't.    If one of the US$0.05 transistors shorts and the power supply is too high, it kills a US$3.00 relay. The regulators work, and may be a solution, but toss in another US$0.50 to solve the issue.

...or what about a simple voltage divider?

It is possible to do this, but it takes a long time to get the divider right for all cases. It's not a simple task.

Regarding the overvoltage trend: i for example prefer to run my CE2 clone at 12V because it sounds much better IMO. Same for PT80 delay. And many commercial pedals, mainly transistor-based, would give a different ...flavor if supplied with more than 9V; marvel drive from rambleFX for example, Fetto, ... and others that now I can't remember.
To my eyes, noob eyes, it looks like another available option for shaping your tone

That's a fair statement, and it's pretty much in line with my not liking intentional overvoltage application. For someone who thinks that the world of effects and tone modding is ever so attractive and who does not have the circuits training to go muck with the circuits inside, it's almost irresistable to tinker with the "flavor" by messing with the voltage knob.

That's a perfectly fine technique if you're willing to bet your pedal that nothing will die, either immediately, or long term in terms of higher noise level or future failures. For instance: many commercial pedals intended to run on 9Vdc batteries or a 9V power supply have internal capacitors rated at 10V. Caps like that may survive a while at 12V, but over the long term, they either have shorter life or drifted capacitance value. Zener diodes may be overheated by a much larger supply voltage and also have future failures (shorter life). Bipolar transistors may have increased noise because if the conditions inside the pedal >> at power off time << cause the base-emitter to be reverse biased and broken by a higher emitter voltage, even for microseconds, the noise performance of the transistor is permanently degraded. There are other issues that can happen, as well. Those are just off the top of my head.

In my "day job" of designing pedals for a commercial pedal firm, I have to anticipate this abuse and use 25V rated capacitors, use extra robust zener circuits, and otherwise in general expect this kind of abuse and at least try to make it not kill the pedal as much as I can - and do that without running the cost of preventive design up too much.

I suppose I should just keep quiet about this, as it can only increase sales, but I still think it's a pity, and shortsighted. But that's just my opinion, and so it's definitely not the mainstream opinion. 

[intripped]

I will modify the circuit adding the resistor and zener diode and report back
Thank you very much guys for your time and the detailed explanations

[intripped]

Ok, I've done the mod and it works perfectly.

I've used two zener diodes, 5.6V and 9.1V in series, to get 14.7V.
Both diodes are 0.5W, so max allowed current is around 34mA.

I've calculated that the resistor, in the worst scenario when supply voltage is 22V (7.3V voltage drop) and current at max (34mA) could be 215 ohm, with power diss. of 248mW
I've used a 560 ohm res 1/4W.

Tried at 9V, 12V and 18V - no issues.

Now that this problem is solved I've noticed another one though:
If I unplug the power supply jack, this circuit doesn't return to bypass state, so that signal is actually muted.
I know that this is a very rare situation, but anyway, I know Murphy's law as well ...and maybe there's another easy solution that my noob eyes can't see...