The Power Profiler Kit

RE: The Power Profiler Kit

Mathias — Tue, 19 Sep 2017 13:28:50 GMT

Thanks for the clarification!! :)

RE: The Power Profiler Kit

wlgrd — Mon, 18 Sep 2017 09:03:13 GMT

That's not stupid :) BLE advertises on three channel, hence you can always see three radio distinct radio pulses per event. If you do a submit here devzone.nordicsemi.com/.../, you'll se that with a regular advertisement, it looks for connection requests (also on all three channels), and you have an extra radio pulse right after each TX pulse which is slightly lower on average.

So, since we see three distinct radio pulses, it is only TX. If we have seen two on each event, then it is TX and RX. If scan response was enabled on this beacon advertisement, we'd also see that the events current profile will change, showing you that it is doing RX in case it needs to adjust the packet for a scan response.

RE: The Power Profiler Kit

Mathias — Tue, 08 Aug 2017 06:48:07 GMT

Hi Wilhelmsen!

Great blogpost! I have a (stupid) question though, what do you mean by "We can also look at the trigger window, where we clearly see that we only have TX events."? How can you see that you only have TX events?

Thanks!

Mathias

RE: The Power Profiler Kit

wlgrd — Thu, 08 Dec 2016 09:24:37 GMT

Hi Steven,

Yes, this component was not included on the original pcb, but we discovered that in some cases we might need it. Thus, we strapped it in just in case.

RE: The Power Profiler Kit

steven — Thu, 24 Nov 2016 14:05:16 GMT

hi nordic,

Soon after we received this part from Mouser Electronic, when i got it check the board which some question,

we examined its outward appearance and found that this part had been revise . (at C10 connect one part to MCU) we need check if this is a Refurbishment product then feedback to Mouser.

see follw attachment file nordic_pf_kit.png.

I hope you will be able to provide the information.

RE: The Power Profiler Kit

wlgrd — Tue, 15 Nov 2016 12:01:49 GMT

The kit is not created as a development kit, and the firmware is not publicly available so I wouldn't advice you to create any generic BLE application on it. So even though it is possible, this kit should be used in colaboration with an nRF5x DK or a board you have created.

RE: The Power Profiler Kit

steven — Fri, 11 Nov 2016 14:44:19 GMT

hi ,it is very interesting kit.

Can use this kit to develop general BLE application on it?

how to do that ? to open some circuit

RE: The Power Profiler Kit

wlgrd — Mon, 07 Nov 2016 17:07:39 GMT

Yes, but the amount of improvement is not measured since the ppk should always be compatible with the dk's.

RE: The Power Profiler Kit

Adrian Eggenberger — Mon, 07 Nov 2016 10:24:12 GMT

You were right. It was the trigger that was enabled by default with a level of 2.5mA. If I disable the trigger the lag mostly disappears. Will the behavior change if a faster J-Link is used?

RE: The Power Profiler Kit

wlgrd — Sat, 05 Nov 2016 19:41:24 GMT

@moritz, that is great to hear! Yes, the data interface is unfortunately not fast enough to handle both windows at full speed. Hence the user guide advise you to stop the trigger when you don't need it or have high accuracy needs for the average window.

RE: The Power Profiler Kit

Moritz — Fri, 04 Nov 2016 16:22:53 GMT

First and foremost: Thanks for this really nice tool! Already helped me to resolve a long standing power issue.

I think I can reproduce the problem that Adrian describes when the trigger level is set to a value that triggers many times per second. J-Link driver version V6.10i

RE: The Power Profiler Kit

wlgrd — Fri, 04 Nov 2016 14:24:32 GMT

What is your general setup? We have experienced some lag with very slow computers, but many seconds should not happen. Have you tried stopping the trigger window while running it? Also, make sure you have the latest SEGGER drivers.

RE: The Power Profiler Kit

Adrian Eggenberger — Fri, 04 Nov 2016 09:42:47 GMT

I started playing around with my new Power Profiler Kit. The graph with the advertising looks really nice. One thing I noticed is that the displayed graph is many seconds behind realtime. If I establish a connection the graph changes to the expected shape many seconds after the connection has been established. Any way to improve this? What causes this lagged output?

Regards Adrian