How to achieve 0.7 muA current draw as stated in the datasheet?

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

niclasBume — Mon, 26 Jul 2021 17:24:14 GMT

That's refreshing news!

Thanks a lot for the info 🙂

Best,

Nic

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

Simonr — Mon, 26 Jul 2021 08:49:14 GMT

I discussed this with one of our current consumption experts, and these spikes are due to the refresh mode being enabled for your regulator. In system OFF mode the refresh interval is at 100-150ms. In refresh mode the current draw between spikes will be ~70nA. Whether the device is set to refresh mode or normal mode is controlled by the MCU itself automatically, and is based on the load current from the chip. So there is nothing to worry about here, and this is normal behavior.

Best regards,

Simon

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

niclasBume — Fri, 23 Jul 2021 08:00:54 GMT

Hey!

Thanks for the response! The components should be within the specs and we calculated the values to fit the application scenario.

On the measuring setup:

Keysight N6705C Power Analyzer [Meter view]: Source the MCU directly with 1.8V at VuC visible in the schematic with current lowest current sensing range of 10uA (nothing else attached). It measured ~70nA across multiple channels. Obviously the current spikes are not detected in meter view as the bursts are too short.
Keysight SMU: Source the MCU directly with 1.8V at VuC visible in the schematic with current lowest current sensing range of 10uA. It measured ~70nA across multiple channels. So the same result with the slower reaction of the SMU.
Keysight N6705C Power Analyzer [Power Profiler]: Source the MCU directly with 1.8V at VuC visible in the schematic with current sensing range in auto. This returns the previously posted plot:

Best,

Nic

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

Simonr — Thu, 22 Jul 2021 07:53:40 GMT

I did a quick review of your schematics and in general it looks good, just a few comments:

L3 doesn't seem to have a value here, but I assume this is 10µH and okay.
The crystals you are using have rather large capacitor values it seems, but I assume this is intended, and that these are the correct values according to the crystal datasheets. The load capacitance we recommend is following this formula: Ccap = CL * 2 - 4pF, where CL is the load capacitance in the crystal's datasheet.

Can you explain in detail how exactly the device is set up for current measurement, because the average current consumption sounds very strange, and can't be SW related. I'm guessing there is a connection either between the measurement tool and the device or a bad soldering somewhere.

Best regards,

Simon

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

niclasBume — Wed, 21 Jul 2021 09:42:29 GMT

Hi Simon

No worries, thanks for taking over!

Unfortunately we can't share the entire schematic/PCB, but basically we just soldered the minimal components (to isolate the measurements as much as possible) as in this image (sorry for butchering it):

Our main concern, is that the mcu shows a deep sleep current averaging at ~70nA which is much lower than in the datasheet, but has spikes of multiple 100uA as can bee seen in the previous image.

Cheers,

Nic

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

Simonr — Wed, 21 Jul 2021 08:01:33 GMT

Due to the summer vacation period here in Norway we are short on staff at the moment and delayed replies must be expected. I'm sorry for any inconvenience this might cause. Håkon is out of office for the time being, so I have taken this case in his place.

Can you upload the schematic and PCB design files so we can do a HW review of your design?

Best regards,

Simon

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

niclasBume — Tue, 13 Jul 2021 18:03:15 GMT

I might have to take back my statement. At closer inspection and soley powering the MCU we measure an average of 70nA with periodic current spikes (Keysight N6705C Power Analyzer). We double and triple checked that the MCU is in fact operational and in system off! The PCB only includes the MCU necessary components as per the datasheet. Could it be that there is decoupling going on that is missing from the datasheet?

Cheers,

Nic

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

Håkon Alseth — Thu, 08 Jul 2021 08:09:01 GMT

Glad to hear that!

Hope you have a wonderful day!

Kind regards,

Håkon

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

niclasBume — Wed, 07 Jul 2021 16:49:24 GMT

I can confirm that it is leakage on the DK. On our custom board we were able to measure the the stated current 🙂

Thanks for the help!

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

Håkon Alseth — Tue, 06 Jul 2021 12:43:15 GMT

Hi,

Yes, It sounds like you have an offset, highly likely not caused by the nRF; but rather the surrounding logic on the board.

Kind regards,

Håkon

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

niclasBume — Tue, 06 Jul 2021 12:36:10 GMT

So the current consumption overhead could be due to the leakage?

So the posted code actually should put it in the deepest possible power consumption state?

[quote userid="2115" url="~/f/nordic-q-a/77066/how-to-achieve-0-7-mua-current-draw-as-stated-in-the-datasheet/318758#318758"]f you enter systemoff with no ram retention, what do you measure at this point?[/quote]

Still get the same readings...

Best,

Nic

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

Håkon Alseth — Tue, 06 Jul 2021 11:07:14 GMT

Hi,

[quote user="niclasBume"]What range are we talking about, regarding the DK?[/quote]

The analog switches and so-forth can add some micro-amps.

If you enter systemoff with no ram retention, what do you measure at this point?

Kind regards,

Håkon

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

niclasBume — Mon, 05 Jul 2021 11:52:38 GMT

Hey

Thanks for the reply! Yes we are testing on the DK and it is USB connected & powered. However we also tried with it powered by the battery of the DK and got similar results. Would you advice to power it with an external PSU?

What range are we talking about, regarding the DK?

Best,

Nic

RE: How to achieve 0.7 muA current draw as stated in the datasheet?

Håkon Alseth — Mon, 05 Jul 2021 11:48:07 GMT

Hi,

Are you testing on the nRF52-DK? Is the USB cable connected, or are you powering it externally with a lab power supply?

Note: Small leakages can occur on our DK's.

Kind regards,

Håkon