I'd agree with @wayneh then. A 555 timer is probably the easiest solution if you need the relay to be active longer than the LED.

I suppose the other possibility would be tapping into the buzzer circuit, but it might be tricky to find the right spot in the circuit to tap into to get a clean, sustained signal. The signal at the speaker itself is almost certainly some sort of AC signal which would be trickier to convert into a relay control.

Do you want the relay to be on the exact same length of time as the buzzer, or is it ok for the relay timing to be a little different?

The relay on time does not have to be exact. I have bought some Uxcell vibration sensors from Amazon , i was thinking glue it to the buzzer to trigger relay or transistor when buzzer on and vibrate. But this sensor is unreliable , uncontrollable, stay on and off at any time. If there is a device activate by sound , it would be easier.

 

Hmmm... I may have an idea. Measure voltage on J3, relative to ground, with the alarm on and with it off. I might be misreading things, but I think that might be a switched line that operates the buzzer. If so, it would read between 0 and 0.6V when alarm is off, and read at least a couple volts (probably 3.6?) when alarm is on.

If so, you could tap that for your control signal instead of the LED, and the rest of the external circuit you're making would go back to the original one transistor circuit (remove that extra inversion stage we added.)

 

Hmmm... I may have an idea. Measure voltage on J3, relative to ground, with the alarm on and with it off. I might be misreading things, but I think that might be a switched line that operates the buzzer. If so, it would read between 0 and 0.6V when alarm is off, and read at least a couple volts (probably 3.6?) when alarm is on.

If so, you could tap that for your control signal instead of the LED, and the rest of the external circuit you're making would go back to the original one transistor circuit (remove that extra inversion stage we added.)

Oh, wait. I forgot that if my other assumptions are right, the signal on J3 is going to be a pulsed square wave, not a steady DC signal. That may prevent some challenges of its own. You can certainly try what I said in my last post, and it might still work - if so it's a very simple solution. But the pulsed signal may not work on its own...

 

Hmmm... I may have an idea. Measure voltage on J3, relative to ground, with the alarm on and with it off. I might be misreading things, but I think that might be a switched line that operates the buzzer. If so, it would read between 0 and 0.6V when alarm is off, and read at least a couple volts (probably 3.6?) when alarm is on.

If so, you could tap that for your control signal instead of the LED, and the rest of the external circuit you're making would go back to the original one transistor circuit (remove that extra inversion stage we added.)

The j3 have 2.1 v when alarm off and 3v when alarm on

 

The j3 have 2.1 v when alarm off and 3v when alarm on

Oh, well I don't think that's going to help at all. I must've messed up when reverse engineering. Sorry about that!

Since the easy J3 signal didn't work out, and you say precision timing isn't critical, then it sounds like the 555 is probably your best bet.

 

Oh, well I don't think that's going to help at all. I must've messed up when reverse engineering. Sorry about that!

Since the easy J3 signal didn't work out, and you say precision timing isn't critical, then it sounds like the 555 is probably your best bet.

Once time I installed a electric water pump on 2009 ranger rover. I wired the relay for water to12v positive coil to ignition circuit capacitor. When engine off the water pump still run for about a minute then off every time. I think may be the capacitor continued to engage the relay until it power empty. If it correct, maybe i can use a capacitor to extend the relay on time. What you think? If it can be done , what kind capacitor should i use

 

Once time I installed a electric water pump on 2009 ranger rover. I wired the relay for water to12v positive coil to ignition circuit capacitor. When engine off the water pump still run for about a minute then off every time. I think may be the capacitor continued to engage the relay until it power empty. If it correct, maybe i can use a capacitor to extend the relay on time. What you think? If it can be done , what kind capacitor should i use

Not with the existing circuit. The NPN transistor draws current through the base in order to get its job done, and I think it would take an unreasonably large capacitor to hold enough charge to keep the NPN on for three minutes.

However, if we replace the NPN with an N channel MOSFET, there might be a simple solution with a cap, a resistor, and probably one diode too. Timing would be very imprecise, but if you're ok with that, it might work.

The 555 circuit is probably still the better solution. It has nice, discrete output states (on/off, nothing in between.) The capacitor to MOSFET idea goes through an awkward fading out phase during which the MOSFET may experience heating, and which adds dramatically to uncertainty in the timing. The 555 would be crisp and much more predictable. Have you looked at 555 one-shot circuits yet?

 

The 555 circuit is probably still the better solution. It has nice, discrete output states (on/off, nothing in between.) The capacitor to MOSFET idea goes through an awkward fading out phase during which the MOSFET may experience heating, and which adds dramatically to uncertainty in the timing.

Agreed. The behavior at the switching point could be a problem. The MOSFET might get hot, the relay might chatter and fail prematurely, the load may not behave well.

It might be worth testing, to see how it goes. I'd choose a MOSFET with a much higher rating than you would choose otherwise. Put a finger on it while it switches and listen to the relay.

 

Not with the existing circuit. The NPN transistor draws current through the base in order to get its job done, and I think it would take an unreasonably large capacitor to hold enough charge to keep the NPN on for three minutes.

However, if we replace the NPN with an N channel MOSFET, there might be a simple solution with a cap, a resistor, and probably one diode too. Timing would be very imprecise, but if you're ok with that, it might work.

The 555 circuit is probably still the better solution. It has nice, discrete output states (on/off, nothing in between.) The capacitor to MOSFET idea goes through an awkward fading out phase during which the MOSFET may experience heating, and which adds dramatically to uncertainty in the timing. The 555 would be crisp and much more predictable. Have you looked at 555 one-shot circuits yet?

I am looking to purchase that now. As soon as i got timer. I will try it.

 

I am looking to purchase that now. As soon as i got timer. I will try it.

Your lights are 12V DC? For what it's worth, once you have good switching from the 555, you may not need the relay. A MOSFET to switch even a large (20A?) current at 12V is inexpensive. If your lights are AC, stick with the relay.

 

Your lights are 12V DC? For what it's worth, once you have good switching from the 555, you may not need the relay. A MOSFET to switch even a large (20A?) current at 12V is inexpensive. If your lights are AC, stick with the relay.

Thank you

 

Thank you

All my lights are 12v

 

All my lights are 12v

AC or DC?

 

12v dc

 

Here's a different concept for getting the extended relay activation. With this circuit, we're taking the signal from J3 (varies between 2.1 and 3V depending on alarm state) and using a pair of op amps (not as bad as it looks - it's two op amps in one small package, the same size as the 555 timer chip discussed previously.) The op amps are used to get a clean on/off signal out of the small change in voltage at J3. This circuit will keep the relay active as long as J3 is above ~2.55V, which presumably should be any time the alarm is on.

One thing worth noting, both for this and the 555 circuit, is the need to use decoupling caps (common values are a 10uF and 0.1uF in parallel with each other) as close as possible to the power supply pins of the IC. Decoupling caps are a standard requirement for nearly any IC, but they're often omitted in simulations and rough-draft schematics, the assumption being that the builder will know to include them later.

There are probably lots of op amps that could do this job - I chose the LMC6482 (https://www.digikey.com/product-detail/en/texas-instruments/LMC6482IN-NOPB/LMC6482IN-NOPB-ND/364330) because I knew it would work and I'm familiar with it, not necessarily because it's the best or only option.

***EDIT: I've just realized R1 should probably be a higher value, maybe 2-5k. I reused part of the earlier circuit there, but forgot that it was previously fed from just a few volts, and here it would be getting 12V. I doubt the exact number is critical, but we do want to limit the current we're drawing from the LMC6482 outputs to a reasonable number.

 

Here's a different concept for getting the extended relay activation. With this circuit, we're taking the signal from J3 (varies between 2.1 and 3V depending on alarm state) and using a pair of op amps (not as bad as it looks - it's two op amps in one small package, the same size as the 555 timer chip discussed previously.) The op amps are used to get a clean on/off signal out of the small change in voltage at J3. This circuit will keep the relay active as long as J3 is above ~2.55V, which presumably should be any time the alarm is on.

One thing worth noting, both for this and the 555 circuit, is the need to use decoupling caps (common values are a 10uF and 0.1uF in parallel with each other) as close as possible to the power supply pins of the IC. Decoupling caps are a standard requirement for nearly any IC, but they're often omitted in simulations and rough-draft schematics, the assumption being that the builder will know to include them later.

There are probably lots of op amps that could do this job - I chose the LMC6482 (https://www.digikey.com/product-detail/en/texas-instruments/LMC6482IN-NOPB/LMC6482IN-NOPB-ND/364330) because I knew it would work and I'm familiar with it, not necessarily because it's the best or only option.

***EDIT: I've just realized R1 should probably be a higher value, maybe 2-5k. I reused part of the earlier circuit there, but forgot that it was previously fed from just a few volts, and here it would be getting 12V. I doubt the exact number is critical, but we do want to limit the current we're drawing from the LMC6482 outputs to a reasonable number.

View attachment 203716

I would like to say thank you to all, with the help i got for this project; complete success. Thank you very much

 

I would like to say thank you to all, with the help i got for this project; complete success. Thank you very much

Great to hear. Which arrangement did you end up with?

 

Great to hear. Which arrangement did you end up with?

i look through my garage and found timer relay board previously purchased on eBay