Boron - One Year Battery Life Project

Social Project Share
21

Exactly, I wanted to be able to run it off the lipo for testing (why burn primaries when testing) as the electron's ADC top end is 3.3V. I can't find it explicitly in the boron datasheet but I think it's a safe assumption that it's the same for the boron. Although assumptions can make an ass out of me and you :slight_smile:

Ya it is a tiny little guy to solder. A stencil and hot air seemed to help a bit.
But definitely let us know if you find a different boost!

I'm a small fish, with a few prototypes deployed, I doubt I count in the "existing product" category. But now that the boron works in Canada it is looking very enticing.

22

My project have the same problems with possible theft and therefore we decided to keep both the battery (2000mA) and the solar panel as small as possible to make theft not worthwhile. I read sensors every hour and send the data to the server once a day and my battery life is 4 months without an solar panel. Afterwards we added a 25 mA / 5 Volt solar panel (25 mm X 45mm) that is about 10 times stronger that we needed but we could not find an smaller one. We have a few units installed and some of them is even installed under tree canopies and the batteries stays full 100%. We theoretically therefore never have to visit an installation again. We also decided to keep the solar panel stronger than needed cause solar panels degrade over time and this will help a lot to counter it. I use Electron’s for my project.

3 Likes
23

@jamesza,

Thanks for sharing your approach. I like your approach to collect hourly and send daily, this could be a very significant savings in power.

I have recently looked at the process to collect data frequently and report in-frequently and wrote a sample sketch for the folks at Colorado State University here.

In my approach, I stored the data payload as an object in EEPROM each 15 minutes and then sent the data each hour. As this was my fist attempt at this, I figure there may well be a better way. Any advice or tips you might be able to give on how to manage this process?

I am especially concerned that the daily send would be a big payload and I would hate to loose a day’s data if the payload is corrupted, does not go through or there is a system outage that prevents a connection for the backend.

Thanks,

Chip

1 Like
24

I seem to be working in a similar area as this, here’s a few bits and pieces that might come in handy:

  • Boost converter: I found these guys to be very efficient even when I have a sensor that uses a lot of current for a short period of time. Datasheet lists it as 6.5uA quiescent.
  • Might think about using alatching circuit for your secondary microcontroller/counter system if it can fit in to your design requirements. This would allow you to be drawing next to nothing, trigger off your (PIR?) and only keep the (ATTiny85?) powered for a very short time while it adjusts your counts/writes to memory/de-powers itself. There are some ICs for this as well if you don’t need the flexibility and have limited board space.
  • This is a decentRTC with alarm function, biggest downside is ~130uA current draw in standby. But if you can use it to remove power from the boron between checks…
  • These are great for a ‘oops I drained the battery, start cutting power to anything I don’t absolutely need’ option. Several of my sensors are active with useful sleep modes, and if anything goes wrong or the batteries are seeing their end of life it’s more helpful to have the boron start sending help messages than to just have it die.
1 Like