@abhishek2101 no, it’s RTC1 of the nRF52840 set up to support a 64-bit system uptime clock at 32 KiHz. Measurements were on a bare Xenon, no external circuitry or peripherals. The duty cycle numbers are from a deployed device with the ALS-PT243-3C phototransistor and a HTU21D, waking at 1 s intervals to collect light intensity, temperature, and humidity, and transmitting observations and telemetry in beacons. There’s some filtering of beacons with insignificant changes but the average over the last 11.5 days was 28300 transmissions per 24 hours. Over that period battery voltage dropped from 4.045 to 4.030, with 99.92% of the time in sleep.
@pabigot, your work is great but since you are not using Particle firmware, the current measurements are “out of context” from the typical user. Nonetheless, they do illustrate that the Particle DeviceOS can be further optimized to reduce current consumption during sleep and that different methods of powering a Xenon will impact this as well. Thank you for contributing!
You’re welcome and that was my intent: identifying the hardware limits to power efficiency. If DeviceOS identifies other significant optimizations similar to powering down the flash to reduce energy consumption I’d like to take advantage of them.
Amazing topic @peekay123! thank you and thank you all
Now my question to Ryan:
What is the power consumption of an Electron in a similar test? You may have the values around, but if you don’t, I’m ok. Just curious to compare side by side.
Not really sure but I saw a graph @rickkas7 posed the other day which showed non LTE 3G power consumption at 2 to 3 times higher if I remember correctly.
What do you guys think about this?
Is it 4ma or 7.5ma used when the mesh is off?
I’m seeing 2.5-3mA when powering via USB with system mode manual and broadcasting every second via BLE.