nRF52840 stops running after a while (>48 hrs) while powered by 2 AAA batteries

[catfish]

I'm running a code based on the ble_app_uart example using the PCA10059 dongle. I have the dongle connecting to a raspberry pi 0 W via BLE and when this runs using a power supply ( 3.3 volts connected to VDD out and jumpers SB1 shorted and SB2 cut) and the connection runs fine. I found that when I use 2 new AAA alkalines the code will maintain connection for > 48 hours and then suddenly start disconnecting to the point that I have to reset the dongle and then I can reconnect for a couple more hours before it disconnects again. The voltage of the batteries is still > nominal (3.59 volts). The USB port is not connected during testing. Any thoughts what I should be looking at?

Thanks   

[ketiljo]

Are you sure about the nominal voltage of the batteries? 2xAAA shouldn't have more than 3.0 V

When you cut SB2 and connect SB1, you will operate the device in normal mode. So unless you have solder issues or do anything wrong with the code, I can't see why would drain the batteries this fast. Are any of the LEDs lit?

[catfish]

I know, the battery voltage is a bit of a surprise being that high. I normally do see new batteries exceed nominal (1.5) but not usually THAT high. When I started out it was even higher and I wondered if the DC-DC was influencing the value due to the low current draw. None of the LEDs are on and the batteries aren't draining much since they are still > than the expected 1.5V/cell. I restarted the code, with the same batteries, and it's running a day later so I'm not sure what could be causing a hangup but as of today it's running as expected. I'll keep testing and report back if there are any further issues. To note, I don't have the HW uart connected while running. The only connection to the dongle is the BLE connection.

As a side note, I'm really impressed with the API functionality provided. It's a steep learning curve but after getting through the basics the API is very thorough and a great guide (examples) to get started.

Thanks

[ketiljo]

You should check the instrument against a known voltage source or another instrument. It's definitively not normal to measure this kind of voltages off a AAA cell. 

[catfish]

i verified that my portable DMM is reading incorrectly while the batteries are in circuit. Not sure why, but the voltages are as expected when read with a benchtop HP DMM. So far the DUT has been functioning all day without a hangup so I'm not sure what caused the issues yesterday. Anyhow, seems to be working correctly so far. Thanks