constant momentary need info

I need to create a momentary pulse from a constant switch on as a one time event (Power comes on, pulse is generated, power stays on but no more pulse). I found a circuit here that looks close, but I'm not sure I understand it or that it does what I need. I also think I need an enhancement, but first things first.

Here's the circuit I'm looking at using: https://www.the12volt.com/relays/relaydiagram20.html

I think I understand how it works... switch on the power, voltage flows through the coil to the capacitor and the relay closes while the cap charges. At some level the charge is high enough to overcome the resistance of the resistor and it discharges and the relay opens.

But, won't this circuit just re-cycle after the cap discharges if the switch remains closed? That is, after it discharges, it will charge again and run through the same cycle?

The resistance is always there. There is no "overcoming" of it.
The effect of the resistor is negligible. It does form a voltage divider with the relay's coil, but since the coil resistance is much smaller (ie, 50R-500R compared to 10kR) and - ignoring the effect of the cap - the voltage at 85 will equal 86.
In fact that 10kR is so high that the relay won't even energise.

But then add the cap. Now, there are various ways to explain what happens. Hmmm, let's see...

The cap is initially discharged.
When +12V is first applied, the cap's voltage is 0V so the (relay) coil has 12V across it and the relay energises. (As textbooks say, when voltage is first applied, a discharged cap behaves like a short circuit.)
As the cap charges thru the coil, the cap voltage increases and hence the coil voltage and current decreases until the coil de-energises (drops below the relay's drop-out voltage).

The 10k is there only to ensure the cap discharges when voltage/power is removed.
In practice the 10k may be unnecessary if the input is GND when off as in the case with most IGN & ACC circuits (ie, other loads act like a ground).

Ah... I see... so the cap never discharges in operation. Makes much more sense!

I believeI have a followup question, but I need to puzzle over the expansion I need to make to this circuit and see if it already answers my question.

Thanks for the explanation.

Yes. You summed it up nicely.

It's tricky understanding it doesn't "discharge" in normal operation. But ignoring the effect of inductance (not relevant here), there would have to be some threshold or switching to cause any cycling - like the a resistor "overcome & self-healing" that you alluded to.


It reminds me a bit of a audio dude that started a thread herein about audio caps and referred to the fast charging & discharging of a cap under "constant" DC (or in comparison to a battery).
He was new to theory and was confusing several issues.    
Like I often write, some things are real tricky to understand! (Then seem so easy in retrospect, until...)

OK... I've done my pondering. I need some additional professional comment.

In addition to sending a momentary pulse when the main switch closes in this circuit, I also need to send a pulse when the main switch opens.

Any ideas on how I could add that facility?

Can the switch be changed to a Single Pole Double Throw On-On type switch?

Would it bother things if the switched wire were connected to ground when the switch was turned off?

Nevermind, I figured out how to do it.

It will take an additional relay. Switched wire to whatever it is switching, as well as the additional relay terminal 86. Ground terminal 85 of the new relay. Terminal 30 of the new relay to terminal 86 of the relay in the link you posted. 12v to terminal 87 of the new relay. Ground to terminal 87A of the new relay.

You already understand how the initial pulse is provided. For the second pulse, When the switch is turned off, the new relay will switch it's output from positive to ground. With Ground on 86 and the charged capacitor on 85, there will be 12v across the coil of the relay. Ditch the resistor. You may need a larger value capacitor if the turn off pulse is not long enough.

I did not build or test this circuit. Logically it should work. Let us know how it works and if you had to modify the capacitor size.

Thanks. I am diagramming this out so I can understand it and I will get back with any questions... or move on to component purchase and let you know the results!

OK... drew it out... studied it a bit. One point I might not understand: Do I really need the 12v to 87 on the second (new) relay? It looks to me like the only relay contacts that need to be used on R2 are the NC 30 & 87a - so that I get a ground switched to R1's coil when power is removed(then the cap can discharge back through the R1's coil and send another pulse). R2 gets power from the switched 12v - does it need constant 12v for anything on 87?

Also, thinking about something my digital electronics buddy tried to teach me about state change... when power is initially applied, there's a ground path from 86 on R1 through 30/87a on R2. Will that create a short at the moment power is switched on before R2 pulls 30/87a open?



Thanks,

Switched power from the switch goes to only 86 of the new relay. 30 of the new relay goes to 86 of the original relay. It has to be this way so the new relay can provide ground to 86 when the switch is turned off.

It would work the way you were trying to do it. But from a safety standpoint, much safer the way I suggested. If the relay failed to energize you would put power onto ground provided from the new relay. If something fails to energize my way, it just fails to provide a pulse, but we still drive away.

As always... allow for me not understanding something... but this is my confusion:

...in the original circuit, switched power goes to 86 on R1. So if 30 from R2 connects to 86 on R1, 30 is getting switched power. Contacts 30/87a on R2 are normally closed, so the grounded 87a on R2 is connected to R2-30 when the switch is thrown just before the relay energizes. So, the way I see it, switched power is connecting to ground for a split second when the switch is first thrown before R2 can pull the 30/87a contacts open.

Am I still missing a key concept? Should I post what I have drawn to see if I've miss-connected something?

Yes post the picture.

Thanks.. got it figured out. I was leaving a line in place that you had removed. I'll post a diagram for future reference for anyone that wants to do the same once I draw it up a little more neatly.


Thanks for all the help.