Relay Toggle LED

[Buffalo Tom]

Here is a circuit Im using to toggle between two relays. It works, but the only thing that bothers me is that when power is applied to the circuit for the first time the led on the non active relay briefly flickers, its just a quick blink but its annoying. Why is it happening and how can I remove it?

[GibsonGM]

Add-on:  Is this due to the field building around the coil in the reverse direction as it energizes??   (for the gurus)  I know there is back EMF when the field collapses; what happens when it charges?  It seems that this might be part of the answer.

[bluelagoon]

Wouldnt be too concerned for a slight flash of LED on power up, a lot of effect pedals do that, doesn't mean they are inferior, For instance the VFE pedals do that at power on, at least the ones I have do. Its likely a cap discharge or something akin. But if the circuit works like you need it to, then not a real issue. Its actually a good indicator to let you know the pedal is powered on.

[Buffalo Tom]

Alright, I suppose I'll need to find a way to cope with this situation Perhaps relocating the LEDs to the switch instead of placing them above the relay coil would be a better solution.

[ElectricDruid]

Why is it happening and how can I remove it?

If we assume Relay A is the non-active one, here's my guess:

At power up, the caps are empty. Since R1/D1 is tied to ground, C1 isn't going to do anything and the C1 end of R2 is essentially tied to ground.
Since C2 is also empty, it's a short until it charges up. That means the bottom of the LED is tied to R2 and therefore connected to ground - effectively we have R3/LED/R2 from +V down to ground - so the LED lights up.

As C2 charges, the voltage across it increases and the voltage at the collector goes up. This reduces the voltage available to the LED, which then goes below its forward voltage and switches off.

Since C2 has to charge up via R3 *and* R2, the time constant is 5.3K/100n = about 2.7msecs. That's a brief blip, but I guess you might see it.

Does that sound about right?

[antonis]

I'd try deleting D1/D2 and wiring R2/R5 left leg to GND..

P.S.
@Tom: 2.7 msec (actually 2.12 msec due to LED + D1 forward drop) equals 4 X RC (capacitor 98% charged)
I'm not sure if the LED lights up during all the transient period..
But YES, it makes sense..!!

[GibsonGM]

"There is a very brief path to ground"

[ElectricDruid]

"There is a very brief path to ground"

Lol, nice summary!!

[antonis]

"There is a very brief path to ground"

Lol, nice summary!!

And scientifically rigorously documented..!!!

[GibsonGM]

I wasn't sure if the coil charging was involved - it appears not. I've seen this 'behavior' as caps charge/discharge, of course! 

[amptramp]

A marketing man that I once knew (who had a PhD in physics and an MBA) had one overriding principle that he lived by:

"If you can't fix it, feature it."

The momentary blip is like the warning lights on a car dashboard that all turn on briefly once you turn the key to "ON" to show you that all the lights are working.  You now have a lamp test function built in.

[MrStab]

A blue or white LED with a higher mcd and meatier resistor in series with it may have more difficulty becoming visible during power-up. Maybe even a diode in series, just so it doesn't have much time to get very far up that conduction curve.

Or use an infrared LED! Problem solved!