Page 1 of 1

Graphing barometer - storm Doris

Posted: Sat Feb 25, 2017 12:11 pm
by pythoncoder
I built this about a year ago. Note the display showing Storm Doris passing over. Horizontal scale one week, divisions each day. The circle adjacent to the Y axis shows rate of change of pressure, a forward indicator of weather.
Image
Image
I've not posted an image of the slaves as they are featureless white boxes ;)

It gathers data from three remote locations using NRF24L01+ radios, logging it to an SD card. The remote units (with one exception) have run continuously on three AA cells. All four units wake up hourly from deep sleep. The master has a calibrated RTC and each slave updates their RTC from the master every hour, so that precise synchronisation is maintained. Each slave has a one minute time slot in which to communicate with the master. Typically each slave wakes for a very few seconds only - the one which has needed new batteries has tended to have communication problems owing to having a chip antenna rather than a duck.

The pushbutton wakes the Pyboard and displays an alternate screen which tabulates data from the remotes. Data returned includes the battery voltage which is displayed if below threshold so dying cells can be replaced.

The master uses a rechargeable LiPoly cell as it wakes for four minutes in every hour and has to refresh the ePpaper display. Originally it too was powered from AA cells but I was having to change them every three months. It's demonstrated the ability of Pyboards to run with very low power consumption, and to survive under conditions of changing temperature and humidity.

[EDIT]Code references for subsystems:
ePaper display driver https://github.com/peterhinch/micropython-epaper.git
NRF24L01+ radio protocols https://github.com/peterhinch/micropython-radio
RTC calibration https://github.com/peterhinch/micropython-samples.git see DS3231.
Micropower operation of Pyboards https://github.com/peterhinch/micropyth ... opower.git

Re: Graphing barometer - storm Doris

Posted: Sat Feb 25, 2017 1:50 pm
by devnull
Nice, what is the battery life using lipoly and what is the battery capacity.

I am also trying to decide on high capacity AA cels, or lipo / liion foi my project which will be deployed on the other side of the world !

How do you calibrate the RTC ??

Re: Graphing barometer - storm Doris

Posted: Sun Feb 26, 2017 7:11 am
by pythoncoder
I use a 2AH LiPoly. In principle this should provide about two months operation but I haven't tested this. You'll need to measure your current consumption which is dependent on hardware and software. With low power systems which spend most of their time in deep sleep you'll need to measure what happens when they wake up: I used a low value series resistor and a scope.

I calibrate the RTC using a DS3231 which is a very cheap, highly accurate RTC. A calibration to clock quality can be achieved in five minutes using some code here https://github.com/peterhinch/micropython-samples.git. Once you've acquired the cal factor you can remove the DS3231. Put the factor in your code so that it persists after reset or power cycling.

Absolute accuracy is not a requirement for the comms to work as the slaves sync to the master, but I wanted it for the data logging.

Re: Graphing barometer - storm Doris

Posted: Sun Feb 26, 2017 8:02 am
by Roberthh
That is truly a very well made-up set, in all aspects. Using the ePaper for that purpose is very appropriate, since it requires no Energy once set.

Re: Graphing barometer - storm Doris

Posted: Sun Feb 26, 2017 8:28 pm
by marfis
consider selling this as a product :)

A hackable ultralow power uPy weather station, excellent!