You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for making it available. I built the v0.1 edition and used an 18650 battery, rated @ 2600mAh, (could be wrong / fake). I made the following changes:
Battery lasted 5 months with the sensor attached to the front door, the circuit stopped working with battery @ 2.3V, which exceeds the voltage regulators specifications.
I paired it with a DS90 sensor and monitored both to see how reliable each of them were, and I think for 99% of the time, both were identical. The ESP would occasionally drop a "CLOSED" signal, if open/closed happened quickly enough.
For this reason I think a watchdog timer waking up the device every x minutes, and reading the current state, would probably be a good idea.
While I had hoped for 12 months of running time, I'll need to see if the battery holds the promised capacity, and see if its possible to trim the timings to lower power usage.
Do you have any data for real-life running of your own?
The text was updated successfully, but these errors were encountered:
I trimmed the timings a little and added period wake-ups. Here is a graph of one of the sensors. Ignore the reported voltage numbers, its off due to poor calculations compared to resistor values used.
Thanks for making it available. I built the v0.1 edition and used an 18650 battery, rated @ 2600mAh, (could be wrong / fake). I made the following changes:
Battery lasted 5 months with the sensor attached to the front door, the circuit stopped working with battery @ 2.3V, which exceeds the voltage regulators specifications.
I paired it with a DS90 sensor and monitored both to see how reliable each of them were, and I think for 99% of the time, both were identical. The ESP would occasionally drop a "CLOSED" signal, if open/closed happened quickly enough.
For this reason I think a watchdog timer waking up the device every x minutes, and reading the current state, would probably be a good idea.
While I had hoped for 12 months of running time, I'll need to see if the battery holds the promised capacity, and see if its possible to trim the timings to lower power usage.
Do you have any data for real-life running of your own?
The text was updated successfully, but these errors were encountered: