I am looking for a wireless system that will have a transceiver that will have little power usage or be in a sleep mode until in range of the other transceiver. One of the units will be stationary and power won’t be as big of an issue but the portable one I am looking at making small. I am looking at a max of 2000mah worth of batteries and I would like this to last a few months if possible before needing to change batteries.
and… data volume? data rate? range requirement? unlicensed band? What regulatory agency? Proprietary wireless protocols or IEEE standards (e.g., IEEE 802.15.4)?
and so on.
Most all the low power wireless devices can do what you want, in terms of power management, if implemented correctly and IF the duty cycle is low enough, and IF there’s a good strategy for rendezvous in time.
there will only be a low amount of data and therefore the rate doesn’t need to be to high. The range is about 10-20 feet from the stationary transceiver. I would prefer an unlicensed band. The protocols aren’t a big deal as long as I can get it to do what I want. What are some things that I would have to do to get this done then. I was thinking about maybe bluetooth or using the nordic key fob from sparkfun and tweaking the code to have it always run. Thanks for the help.
I’ve been fooling around with the Nordic 24L01+ modules recently, and I’m starting to get to the optimizing-for-battery-life stage. I plugged a few numbers into a spreadsheet I’ve been working on to game out your scenario. If I’m doing this right, you can stretch a pair of batteries a lot farther than I expected.
Assumption: the mobile station will transmit one “ping” per second at 1Mbps at the -6dBm setting. It’ll use enhanced shockburst mode, with 3 address bytes, one TX payload byte, one CRC byte, no ACK payload. The mobile station will wait for an ACK, and if it does not get an ACK from the base station, it will go back to sleep (this is the more expensive case, if the base station ACKs, the mobile station will turn its receiver off faster, and will roughly halve the overall power consumption). So the budget for one such failed ping is:
oscillator startup = 1.5ms at 400uA = 1.98uJ
data upload at 1MHz SPI = 16us at 26uA standby-I = 1.37nJ
TX settling = 130us at 8mA = 3.43uJ
TX = 57us at 9mA = 1.69uJ
RX settling = 130us at 8.9mA = 3.82uJ
RX = 250us at 13.1mA = 10.81uJ
IRQ time = 8.2us at 26uA standby-I = 703.6pJ
---
Total: 21.732uJ
If we do that every second, it costs:
998ms in 900nA power-down: 2.96 uJ
~2ms tx/rx cycle: 21.732uJ
---
Total: 24.696uJ
If a pair of AA batteries has 18720 joules, they could power the nordic module doing this for 24 years (about 3 years from a CR2032 coin cell). This ignores a lot of things: the microcontroller (probably negligible), the battery’s self-discharge rate (not much you can do about it), the voltage regulator’s quiescent current (looks like this is going to eclipse everything else in my design)…
If you see any errors, please let me know.
Thanks for that it makes my project seem possible now and feel a little bit better about having to go buy the stuff. Now the fun can start as soon as I can put in an order.