This circuit works on the bench with a few more parts including Q1.
View attachment 281363

Hi sghioto,

There's a problem with the quiescent current of 2.6ma when the switch is connected. As I want to run this on 9v battery, so 2.6ma is quite big. Do you have any suggestion to remove this?

 

There's a problem with the quiescent current of 2.6ma when the switch is connected

The circuit I posted using a CD4093 in post #9 uses essentially no quiescent current (just the leakage current of the gate).

 

There's a problem with the quiescent current of 2.6ma when the switch is connected. As I want to run this on 9v battery, so 2.6ma is quite big. Do you have any suggestion to remove this?

You can increase R1 and R3 to 1meg and R2 to 100K or even larger possibly, that will reduce the current to appx 110 ua. Use a 9 volt lithium battery?

 

This version will reduce the standby current to appx 5ua and was good to below 5 volts.

 

This version will reduce the standby current to appx 5ua and was good to below 5 volts.
View attachment 281630

Thanks, it now consumes less than 20ua including the rf433 modules.
I have a question: as far as I know, the capacitor self discharge over time, so if the circuit connected for several days or weeks, when it's disconnected, can the capacitor still output its pulse?

 

Thanks, it now consumes less than 20ua including the rf433 modules.
I have a question: as far as I know, the capacitor self discharge over time, so if the circuit connected for several days or weeks, when it's disconnected, can the capacitor still output its pulse?

Not sure what you mean by when it's disconnected.
As long as the 9 volts is connected to the switch the circuit will pulse every time the switch opens and closes no matter how long the time is between the open and close . If you disconnect the 9 volts you will get 1 pulse to K2.
Does that answer your question?

 

Not sure what you mean by when it's disconnected.
As long as the 9 volts is connected to the switch the circuit will pulse every time the switch opens and closes no matter how long the time is between the open and close . If you disconnect the 9 volts you will get 1 pulse to K2.
Does that answer your question?

I mean the switch is on for several days and when it's off, will the pulse still sent to K2.

Yesterday, I kept the switch on for around 12 hours, when I switch off, there's no pulse sending out to K2. When I switch on again, also nothing happened to K1 and also no to K2. However, when I keep it off for around 2 minutes, then it's sending out pulse again to both K1 and K2, respectively.

 

Yesterday, I kept the switch on for around 12 hours, when I switch off, there's no pulse sending out to K2.

I don't see any reason for K2 not to get a pulse when the switch is OFF.
Capacitor C1 does require at least 5 seconds to discharge so the switch would need to be OFF for that amount of time before switching back ON.
Should add diode D3 back in the circuit for the K1 pulse, it was left out of the previous schematic by mistake but should not have any effect on K2.
I'll repeat your test and leave the circuit connected as you described. Will check in the morning after 12 hours.

 

I don't see any reason for K2 not to get a pulse when the switch is OFF.
Capacitor C1 does require at least 5 seconds to discharge so the switch would need to be OFF for that amount of time before switching back ON.
Should add diode D3 back in the circuit for the K1 pulse, it was left out of the previous schematic by mistake but should not have any effect on K2.
I'll repeat your test and leave the circuit connected as you described. Will check in the morning after 12 hours.
View attachment 281801

My input is 9v, I try many different capacitors from 10uf(below this, it can't trigger the rf433 module) to 200uf, I use same uf capacitors for both side.

It stopped working after several on/off within 1 or 2 seconds apart, esp K1 will stop working first. And if I off them for a few minutes, it'll start working again.

However, the original design(the one that consumes 2.6ma) is working fine.

Thanks;

 

After 12 hours the circuit is still working fine.

It stopped working after several on/off within 1 or 2 seconds apart, esp K1 will stop working first.

That is normal as I explained in my previous post, you need a minimum of 5 seconds between ON and OFF because of the large value resistor R1 (10M). That value was chosen to reduce the standby current to a minimum.
As far as K2 is concerned what part # transistor are you using and diodes?
You can try changing R1 and R2 to 1M but should not need to change the cap values.

 

After 12 hours the circuit is still working fine.

That is normal as I explained in my previous post, you need a minimum of 5 seconds between ON and OFF because of the large value resistor R1 (10M). That value was chosen to reduce the standby current to a minimum.
As far as K2 is concerned what part # transistor are you using and diodes?
You can try changing R1 and R2 to 1M but should not need to change the cap values.
View attachment 281848

My diode is 1N4007, Transistor: 2N3906

 

My diode is 1N4007, Transistor: 2N3906

Have you tried the circuit in my previous post?
Do you have any 1N4148 diodes you can use?

 

Have you tried the circuit in my previous post?
Do you have any 1N4148 diodes you can use?

Yes, I have the zener diode. I try it but it looks the same. when using 1M Resistor, it is around 70ua(including my rf433 module)

I think the problem of K1 not working properly is due to there's some current from K2 port, which cause conflict to K1. When I connect only K1 pin to the RF433 module, it works fine.

5 seconds delay is a bit too long, anyway to reduce this to 2 seconds? I change to lower uf capacitor but this can't help much. The RF433 module has a feedback pin when it's sending data out(the red circle in my attached photo), can this be useful?

 

Yes, I have the zener diode.

The 1N4148 are general purpose diodes not zeners.
Try the original circuit again as modified below.
In reality what is the minimum time the water pressure switch will stay open, more then 5 seconds?

The RF433 module has a feedback pin when it's sending data out(the red circle in my attached photo), can this be useful?

I don't think so.

 

Another version using a CD40106
Current draw 2ua with switch closed.

 

Another version using a CD4069.
Current draw 2ua with switch closed.
View attachment 282558

Sorry, due to COVID situation, I can't test your design yet. I will be back next week. Thank you very much for your help.

 

Another version using a CD4069

You should use the CD40106B Schmitt-trigger version to avoid the possibility of oscillations for the slow input voltage fall-time from the RC circuits.

 

You should use the CD40106B Schmitt-trigger version to avoid the possibility of oscillations for the slow input voltage fall-time from the RC circuits.

Agree, the 40106 does produce a sharper pulse when the switch opens and closes.
EDIT: see post #36.