NRF52832 - Power spikes in eddystone example

RE: NRF52832 - Power spikes in eddystone example

Jared — Wed, 27 Jul 2022 13:05:42 GMT

Hi,

Great,

Keep me updated.

regards

Jared

RE: NRF52832 - Power spikes in eddystone example

Martin — Wed, 27 Jul 2022 12:37:54 GMT

Hi Jared

thanks for verifying this.

Then I will move to a lab and try to measure the current with an osci again.

Cheers

Martin

RE: NRF52832 - Power spikes in eddystone example

Jared — Wed, 27 Jul 2022 12:35:24 GMT

Hi there Martin,

I was able to reproduce your results with the PPK2 and after discussing it with other HW engineers I pretty sure that the spikes are introduced by the PPK2 itself. I wasn't aware of it, but this is actually a known issue. The spikes are caused by the PPK when it switches to a smaller measurement resistor because of the sudden current increase. Ideally the power profiler firmware should have filtered it out. Hopefully we'll get a Power Profiler fw fix for this in the future. In short, the 14-15 mA spikes are caused by the PPK2 and the nRF52 consumes less current, much more in line of what is estimated by the online power profiler.

I hope this explanation helps.

regards

Jared

RE: NRF52832 - Power spikes in eddystone example

Martin — Thu, 21 Jul 2022 12:28:59 GMT

Hi Jared

it is very puzzling, as I see no reason why this is happening...

Checkout the screens.

Cheers

Martin

RE: NRF52832 - Power spikes in eddystone example

Jared — Thu, 21 Jul 2022 12:25:34 GMT

Hi,

Well, it's difficult to say as the plot doesn't look like advertisement events. It would be good if you could update this case when with a new plot, and see if the pattern is repeating accordingly to the advertising interval.

best regards

Jared

RE: NRF52832 - Power spikes in eddystone example

Martin — Wed, 20 Jul 2022 14:02:35 GMT

I'm currently not in office to produce a new plot.

It always happens at the advertisement interval. it is really puzzling me, because I can't really pin the exact point down where it happens...

RE: NRF52832 - Power spikes in eddystone example

Jared — Wed, 20 Jul 2022 14:00:14 GMT

Hi,

Can you share a bit more zoomed out plot? Is there any other peaks before the "peak pattern" in your plot repeats.

regards

Jared

RE: NRF52832 - Power spikes in eddystone example

Martin — Wed, 20 Jul 2022 13:43:28 GMT

I have the device supplied over PPK1, which is powered from a hub with external power supply.
Usually I found it is not very noisy.

Reworking it to the DevKit would be a big hassle...

Any ideas?

RE: NRF52832 - Power spikes in eddystone example

Jared — Wed, 20 Jul 2022 13:30:04 GMT

Hi there,

That doesn't quite look like an advertising event.

Can you describe your measurement setup?
Are you using the PPK2, in source mode or in ampere mode?
How is the board supplied, by USB? A laptop USB port is usually a bit noisy, it's better to use for example a power bank to supply if you're using USB.
Have you been able to replicate this on a development kit?

regards

Jared