Does anybody out there know how to create a battery powered system that will last for about a year before needing battery replacement? I know I could achieve this by running a large stack of batteries in parallel but that seems excessive for such a simple little device. I want it to run as a stand alone system where I don't have to replace batteries every few months or run a power cord to every pot plant. The problem that I found was that the hygrometer is drawing 8.3mA when powered by 3 AA batteries. This is too much current as AA batteries have, at best, a 3000mAh lifespan which will give me approximately 2 weeks before the batteries run flat. I was able to reduce the current draw down to approximately 1mA on average by using only 2 AA batteries instead of 3, replacing the DO LED with a 1k resistor, removing the power LED, and powering the board intermittently through an astable multivibrator (a hack from a flashing LED circuit). The hygrometer is now powered for less than a second and then turns off for about 2.5 seconds. This is setup also gives a good frequency of the signal LED flashing when the soil runs dry. Does anyone know if there is something else I could do?
How often do you typically need to water those plants? I'd think that a sample frequency of, say, ten times the watering frequency should be adequate. To get a visible indication, you might to capture the state of the monitor's output at the end of the measurement cycle - maybe in a continuously-powered CMOS memory element - and use that, along with the 0.4 Hz astable multivibrator that you already have, to blink an LED. Or, if it suits you, you can enable the indication with a button, so it'll only flash when you ask it.
I have similar questions for a project I am working on that will send data twice a day via a GSM shield. I have been told that the gsm shield can use a lot of power. I have read a lot about putting the arduino to sleep. Am I better off using an alarm on a DS3231 RTC which I assume still uses some power, or use an 8 second sleep cycle and lowpower library? I want to try to keep it simple as it is my first project. I also know I could save power with a barebones arduino but would still like to use an Uno.
I am looking at running it on a 3600mAh Lithium battery, how long could I make the arduino run for sending data twice a day?
Gopster:
I have read a lot about putting the arduino to sleep. Am I better off using an alarm on a DS3231 RTC which I assume still uses some power, or use an 8 second sleep cycle and lowpower library?
8 second sleep sounds good to me.
Gopster:
I am looking at running it on a 3600mAh Lithium battery, how long could I make the arduino run for sending data twice a day?
What do you mean by "the Arduino"?
Putting an Arduino Uno into sleep mode is pointless because of all the other junk on the board.
Putting an Arduino Uno into sleep mode is pointless because of all the other junk on the board.
Sorry I am using an Uno.
I only need the uno to be awake twice a day to take a reading, so everything can sleep the rest of the time.
I do understand that a bare bones set up would be a lot better but I have an Uno and would like to make it work. I think trickle charging via solar will be the way to go.
cjdelphi:
The atmega processor can simply be lifted off the board, simply buy a replacement cpu and pop it back in...
(Gently prize it up both ends equally until it pops out, careful not to bend the legs)
You could do that but why bother? A Pro Mini does the same thing for the same price and with much less work than building a board with 16MHz crystal, etc.
If you are only doing soil moisture, I would think the easiest way to go would be to use a solar walk light, the cheap units they sell to push into the ground and has a solar charger. the 2 battery versions are harder to find, but would be a good base.
then use an ATTINY85. internal resonator, so it does not need all the extra chips. the AA batteries can power the LED for a couple hours when fully charged. but if you control the LED with the '85 you can make it last for weeks by flashing the LED.
the '85 can be programmed from the UNO and you do not need to buy anything special.
Hi guys, I am interested in this too, are you saying that the power savings by sleeping the configuration using an Uno or other "typical" Arduino (for the best part of 24 hours) are still not worth it compared to just running a bare bones setup 100% of the time?
I.e. Load cell-ADC/amplifier-Arduino Uno-GSM waking twice per day to perform task vs load cell-ADC/amplifier-Seeduino/barebones-GSM
cjdelphi:
The atmega processor can simply be lifted off the board, simply buy a replacement cpu and pop it back in...
(Gently prize it up both ends equally until it pops out, careful not to bend the legs)
You could do that but why bother? A Pro Mini does the same thing for the same price and with much less work than building a board with 16MHz crystal, etc.
Fair enough, just throwing out options... but does the pro mini have a dedicated usb to tty chip? Powering that will suck up juice, so you buy a cheap clone with just the micro controller for $3.50 on ebay, sure now buy a programmer $5 more, at least pulling out the mc from the board is a little cheaper but yeah the mini is a better option than soldering your own
NZbeek:
Hi guys, I am interested in this too, are you saying that the power savings by sleeping the configuration using an Uno or other "typical" Arduino (for the best part of 24 hours) are still not worth it compared to just running a bare bones setup 100% of the time?
I.e. Load cell-ADC/amplifier-Arduino Uno-GSM waking twice per day to perform task vs load cell-ADC/amplifier-Seeduino/barebones-GSM
Clarification here would be appreciated.
if you can put the chip into a deep-deep-deep sleep, you basically eliminate using any power at the chip.
however, every other device will hum along wasting watts.
20 years on a coin cell ? not quite, but in that direction.
we are talking about sleep mode to save the most energy. just getting rid of the extra bits can save some, but going that little bit further can save much more.
NZbeek:
are you saying that the power savings by sleeping the configuration using an Uno or other "typical" Arduino (for the best part of 24 hours) are still not worth it
It could save about 25% of the base power consumption.