Investigate power options

[0x736E]

Battery:

• LiPo
• CR123A

Other:

• Light Fixture
• Wall Plug

[estuans]

+1 for Light Fixture, we want as low maintenance as possible ideally.

I'll spend a little more time seeing if I can get some kind of trickle charge from a solar cell working.
My thinking is if we can get more than 7 or so mA out of it then there'll be a slow charge whilst the device is in deep sleep.

[0x736E]

I suppose we should first decide how often we want the devices to 'wake up' so we can determine how much power they use?

I would like as much resolution as possible, but any more than 1 per minute seems OTT. somewhere in the range once per 1-10min seems suitable

[estuans]

Right - I guess it depends on the amount of current drawn by the sensor too.

Maybe use an LDR to drive an interrupt on the more "heavy" sensors. we don't need to airquality / gas in rooms that the lights aren't currently on for? - just a thought.

[0x736E]

I'm not so sure.

Air quality is not a binary state. We may want to graph how air quality changes over time.

In the case of gases, we may wish to discover a leak etc. which would need regular monitoring.

I would rather we 'poll' the environment for state at regular intervals, and report the sensor data regardless. We can do smart things with that data on other threads / projects.

Speaking of, we may indeed use such sensor information to detect human presence, regardless of lighting.

[estuans]

I guess so - I don't think it needs to be any more frequent than 10 minutes or so. it's all a merry-go-round, we'll use more power logging frequently, and driving more thirsty sensors.

This also raises a question - if we were to use a PIR to detect movement within a room, do we log each interrupt? or timeslot it somehow?

[0x736E]

That's a good question.

I don't know the answer. We should respect privacy though.