G sensor power consumption?
Interesting real world requirement here… We’re getting fed up with shipments being partly destroyed by couriers who think a box marked “fragile” is obviously a football, so we’ve decided to do some tests of the couriers…
To this end we need something that can record G against a time and date over a period of 24-48 hours.
My immediate thought was, Jeenode (minus RFM module), RTC, G-sensor board and an eeprom board for storage.
Now I just don’t know how much juice I can expect the G sensor to consume. I don’t currently have one, and I don’t have a JC style power consumption calculating uber-scope :-(
I see notes in the G sensor spec about interrupts, so hopefully I can put the ATmega into some sort of sleep state during low periods of abuse (half time?).
Any thoughts? Will I be able to accomplish this without an sealed lead acid battery hung underneath?
RE: G sensor power consumption? - Added by padvinder95 over 2 years ago
According to the sensor’s datasheet, it consumes 200µA when active and 1µA when in stand-by. I think that even without using the stand-by mode, you should be able to run this for 48 hours: a standard AA battery packs about 1000mAh, and 0.2mA at 3V means about 0.5mA at 1.5V, so theoretically the sensor should run for 2000 hours on a single AA.
I would advise first making a very dumb sketch that just reads the acceleration data from the plug every second or so, sleeping the JeeNode in-between. If that runs long enough on a battery, there’s no need to invest more time in improving energy-efficiency. If it depletes the battery too soon, you can always look at programming acceleration thresholds and subsequent interrupts into the chip, using that to wake up the JeeNode and recording, etc…