Craftybytes
Senior Member
Re: Battery life for remote sensor circuit - extra info
Hi Jeremy Leach -
You might want to consider instead to actually switch off the battery power to the Picaxe and sensor circuits UNTIL you need to take the measurements THEN turn the power on via a low Rds-on N-Channel FET. I'm using a similar circuit in one of my set-ups but as it includes an LDR light sensor & a voltage level sensor as the 'trigger' switches, it is fairly easy to 'turn on' the battery power to my circuit. In your case, the types of 'sensors' you are using don't give much in the way of suitable methods to 'trigger' the battery power on NOT if you also remove power to the sensors between each set of measurements to reduce battery drain. So I suggest you will have to go for a low power (micro amps current drain) 'time delay' circuit to turn on the battery power & on completion of sensor measurements & processing - use one of the Picaxe output pins to turn off the power. So if you set the 'delay' for say 15 minutes between sensor readings & then 'power up' say for 30 seconds - take & store the readings, etc, then power off via the Picaxe output - the battery drain could be considerably reduced making them last much longer before you need to replace them.
I have a circuit that may be suitable for the 'time-delay' function - uses 1 x 4093 IC, 1 x BC547 transistor, 1 x N-Channel FET (<= 500ma), 4 x low value capacitors, 8 x resistors, 7 x IN4148 diodes. Sounds slightly elaborate but is actually quite simple in implementing the 'wait for a set period then turn-on the battery power' (to the main circuit) function you need.
The technique used is to dump small discrete amounts of current into a charge pump circuit several times a second. Because it is no longer a continuous current flow, it takes a lot longer to charge up the 'time-delay' capacitor. This allows for a low leakage smaller value (6.8uF) type tantalum capacitor to be used instead of the extra large value (470uF or even larger) electrolytic capacitor if one was to use say an 555 timer chip.
AND because the 4093 is a CMOS chip, it has very low current drain - very attractive for the proposed 'time-delay' circuit. Helps to keep the battery drain lower than otherwise possible.
Just changing the value of ONE resistor allows for 'delay' periods from seconds up to at least 18 minutes (achievable).
If you're interested I'll Email you the circuit diagram - just Email me first.
Best of luck with your set up.
Hi Jeremy Leach -
You might want to consider instead to actually switch off the battery power to the Picaxe and sensor circuits UNTIL you need to take the measurements THEN turn the power on via a low Rds-on N-Channel FET. I'm using a similar circuit in one of my set-ups but as it includes an LDR light sensor & a voltage level sensor as the 'trigger' switches, it is fairly easy to 'turn on' the battery power to my circuit. In your case, the types of 'sensors' you are using don't give much in the way of suitable methods to 'trigger' the battery power on NOT if you also remove power to the sensors between each set of measurements to reduce battery drain. So I suggest you will have to go for a low power (micro amps current drain) 'time delay' circuit to turn on the battery power & on completion of sensor measurements & processing - use one of the Picaxe output pins to turn off the power. So if you set the 'delay' for say 15 minutes between sensor readings & then 'power up' say for 30 seconds - take & store the readings, etc, then power off via the Picaxe output - the battery drain could be considerably reduced making them last much longer before you need to replace them.
I have a circuit that may be suitable for the 'time-delay' function - uses 1 x 4093 IC, 1 x BC547 transistor, 1 x N-Channel FET (<= 500ma), 4 x low value capacitors, 8 x resistors, 7 x IN4148 diodes. Sounds slightly elaborate but is actually quite simple in implementing the 'wait for a set period then turn-on the battery power' (to the main circuit) function you need.
The technique used is to dump small discrete amounts of current into a charge pump circuit several times a second. Because it is no longer a continuous current flow, it takes a lot longer to charge up the 'time-delay' capacitor. This allows for a low leakage smaller value (6.8uF) type tantalum capacitor to be used instead of the extra large value (470uF or even larger) electrolytic capacitor if one was to use say an 555 timer chip.
AND because the 4093 is a CMOS chip, it has very low current drain - very attractive for the proposed 'time-delay' circuit. Helps to keep the battery drain lower than otherwise possible.
Just changing the value of ONE resistor allows for 'delay' periods from seconds up to at least 18 minutes (achievable).
If you're interested I'll Email you the circuit diagram - just Email me first.
Best of luck with your set up.