PPK2 stops power input for short periods of time in ampere mode causing device to reboot.

[Arttu]

Hello.

Wiring 1:

wiring 2:

Background to the case: We tested this with 2 different ppk2s and always got the same results. Our custom board works fine if we wire it directly to the battery or use the ppk2 in source mode as a power supply. The problems start when we try to connect the battery to the ppk2 in amp mode to measure how much current the PCB draws. It seems that the PPK stops supplying power to the PCB for short periods of time, causing the PCB to restart on a regular basis. We have tested wiring 1 and wiring 2 several times. The results are the same.

What could cause this behaviour?

General info:
battery 3V 950mAh
Power Profiler v4.1.1

Best regards,
Arttu

[Håkon Alseth]

Hi,

 

Which device is on your PCB? And what peak currents are expected in your application?

 

Kind regards,

Håkon

[Arttu]

Hi.
Our custom board is using nrf9160. Peak currents are up to 500mA when modem is connecting.
peak currents shouldnt matter cause  ppk caused many reboots during sleep current (below 4µA).

Regards,
Arttu

[Håkon Alseth]

Hi Arttu,

 

Thanks for providing these details.

What type of battery are you using? You mention 3V, do you mean 3.7V? 

Arttu said:

Our custom board is using nrf9160. Peak currents are up to 500mA when modem is connecting.
peak currents shouldnt matter cause  ppk caused many reboots during sleep current (below 4µA).

Is the issue when going from PSM to LTE_ACTIVE state? ie. from 4 uA to >100 mA? Do you have a trace of such a reboot-issue?

 

Have you tried monitoring VBAT while this occurs to see if it goes below 3.0V and triggers a brown-out reset?

 

Kind regards,

Håkon

[Arttu]

Hi Håkon.
our battery is 3v 950mAh from minamoto, type CR2.

The reboot issue appears randomly , some times during peak current and some times just middle of sleep current. Unfortonately we dont have trace. We havent monitored Vbat because the device operates completely normally if we take PPK2 out of the circuit.

Best regards,
Arttu.

[Håkon Alseth]

Your battery's nominal voltage is on the very edge of what the hardware accepts:

https://docs.nordicsemi.com/bundle/ps_nrf9160/page/recommended_op_conditions.html

 

Please monitor the VDD voltage while testing to ensure that it does not trigger a brown-out/power on reset scenario.

 

Kind regards,

Håkon

[Arttu]

I forgot to mention that,

We use boost converter so it should be 3.3V at all times. boost converter accepts 400mV to 3.3V as input and should give 3.3V output. Also the Device operates completely normally without PPK. Wouldnt the brown-out power reset scenarios occur with just the battery as well if the battery would be the reason? Our tests point to ppk dropping power, not the battery.

Regards,

Arttu

[Håkon Alseth]

Hi Arttu,

 

Thanks for explaining the scenario. If you have a boost converter, the voltage level should not be problematic.

Does this happen if you run the PPK2 in source meter mode and provide 3.0V to your PCB?

If this issue happens even at low uA current consumption, small periods of switching in resistors should not have an impact, as the total capacitance in your design should be able to handle this. Could it be that there is a bad connection between to/from the PPK2?

 

Monitoring the voltage after the PPK2 will help here to determine what happens.

 

Kind regards,

Håkon

[Arttu]

Hi Håkon.

Håkon Alseth said:

Does this happen if you run the PPK2 in source meter mode and provide 3.0V to your PCB?

The device runs normally in source mode. The problems only occur in ampere mode.

Håkon Alseth said:

Could it be that there is a bad connection between to/from the PPK2?

We did run the test with 2 ppk few times, so its unlikely.

Håkon Alseth said:

Monitoring the voltage after the PPK2 will help here to determine what happens.

I will get back to you after we are done testing this.

Thank you.

Best regards,
Arttu

[Arttu]

After updating to  Power Profiler 4.1.2 our probleblems were fixed, now we cant re-create this reboot behaviour and everything seem to work as intended.

Regards,
Arttu