Low power beacon

RE: Low power beacon

agishii — Sun, 05 Apr 2020 07:01:32 GMT

Hello,

This has been resolved. After all, it was caused by DCDC switch noise. In the case of CR2032 + 100uF or AAA alkaline battery, no noise appeared. Thank you for your cooperation.

When using a LiPo battery or AC adapter, noise occurs when the power source voltage is significantly different from the target voltage.

Below is using AAA battery.

RE: Low power beacon

haakonsh — Mon, 24 Feb 2020 13:18:47 GMT

I think your measurements are off by a factor of 10x to high.

The TX current will be ~60mA with respect to the 40mA spikes, which fits well with ~6mA TX current of the radio with DCDC enable and transmitting at 0dBm.
See ITX,0dBM,DCDC in Radio current consumption (Transmitter).

If that's the case then those 40mA spikes would be ~4mA, which fits well with the CPU run current of ~4mA with DCDC enabled.
See IDDFLASHCACHEDCDC in CPU performance.

What's the resistance of the series resistor you're using for the measurements?

RE: Low power beacon

agishii — Sat, 22 Feb 2020 10:11:21 GMT

Hello Turbo,

Thank you for reply. You know a lot about battery-powered circuits.

Looking at your answer, I momentarily thought it was DCDC power. But, I am using DCDC via LiPo, but the current measurement is between DCDC and BLE. So, I think that this 200uS cycle current consumption is used in BLE.

RE: Low power beacon

Turbo J — Sat, 22 Feb 2020 09:55:21 GMT

What is your power source?

Ar CR2032 or similar coin cell with high internal resistance needs reservoir capacitors (around 100µF, more is better).

Otherwise the voltage drops significantly every time the NRF chip turns on CPU or radio peripherial, since the battery cannot follow these fast current changes.

Do you use the integrated DCDC switching mode regulator? If yes, these spikes might just be the DCDC startup shortly before the radio event. But that is just a wild guess at this point...