Estimating power use of NRF52840 dongles in the Online Power Profiler

[DookDook]

Hi,

I am using the NRF52840 dongles for BLE communication with S140 and the NRF5 17.1.0 SDK. I have plugged each dongle in a Raspberry Pi 3B+, and have set the DCDC regulator to 1.

I have 7 questions:

1. What value should I use in the Online Profiler for the voltage? I have no other way of measuring it. The dongle's operations include frequent BLE communications with multiple connections and communicating with the Raspberry Pi with usb_cdc_acm.
2. I haven't made any hardware changes to the dongle. So I should choose the Internal RC as the LF clock option and not external crystal correct?
3. When two dongles are connected as master-slave, is the sleep clock of both 500 ppm because of the Internal RC? Is there a way to lower the ppm?
4. If the slave latency is higher than 0, do the central dongle's time phases in the power profiler change when the slave did not receive the message because of the latency? I was thinking maybe the Radio RX phase would become 0us, or would just be lower, not sure if i'm correct.
5. Are Pre-processing, Crystal ramp-up, Standby and Post-processing times and current always the same for the communications of a dongle and not related to their other application processes? Especially the Pre and Post processing, are they shown in the graph just for the sake of showing that something is processed during those times, or do they always occur in the same way?
6. In the sniffed packets I see that when a dongle sends a packet, the other sends an ACK packet, so that the sender knows if it needs to retransmit or not. I do not notice this in the Power Profiler. To make the estimation as close to reality as possible, should I add a Radio switch and Radio TX/RX (depending on role) to estimate the sent ACK packet.
7. The communicating dongle sends full payload packets with the Data Packet Length Extension. But sometimes in the sniffed packets, I see that such a packet is split in two and sent separately. Do you know why this could happen? It seems to happen randomly and also occurred when I made the maximum payload size 5 bytes shorter.

Kind regards.

[Kenneth]

Hello,

If you haven't already I suggest running through the https://academy.nordicsemi.com/ 

Be aware that the nRF52840 USB dongle is not well suited for development, since there is no on-board programmer or debugger, you should get yourself an nRF52840-DK instead. The nRF52840 USB dongle is better suited for demo purposes down the line.

If you are making an BLE application you can find decent power numbers here:
https://devzone.nordicsemi.com/power/w/opp/2/online-power-profiler-for-bluetooth-le 

Though be aware that the power numbers do not include anything beside the BLE link, so any other application activity such as serial interface and leds are not included. To find what is actually drawing current having an Power Profiler II is very useful:
https://www.nordicsemi.com/Products/Development-hardware/Power-Profiler-Kit-2 

Also, I recommend to check out:
https://docs.nordicsemi.com/bundle/ncs-latest/page/nrf/test_and_optimize/optimizing/power_general.html 

Finally I would like to mention that I can see ask a bit about some low level details of the radio, be aware that we do provide various radio protocols (e.g. BLE or ESB) with a set of api for sending and receiving data, so all these low level timing is not something you need to consider.

Kenneth