Sudden Change in Moteino Low Power Performance

[DonpK]

I have been running a Moteino USB on 4 AAA batteries for several years. Typically, the Moteino, which has several sensors attached, runs on the AAAs for well over a year before the batteries need to be replaced. Several months ago, the Moteino stopped running, the batteries were depleted and I replaced them with four fresh AAAs of the same type. However, instead of running over a year, the unit needed new batteries after about a month. Again, I put in fresh batteries and again they needed replacement after a little over a month.

The Moteino sensor module still performed normally with fresh batteries, but the life time was much shorter. I measured the current draw during the wake and sleep periods. The wake current was about 7.9mA and about 0.700mA in the sleep mode - much higher then when the Moteino module was first put in service and explained by the battery life had become much shorter. I measured the current on a different Moteino USB board wired the same way as the first unit and loaded with the same software. The current measurements were about 7.9mA waking and only 0.014mA in the sleep mode, in other words much lower than the original Moteino and explaining why the battery life had suddenly before much shorter.

This is the sleep section of my sketch:

Code: [Select]

      radio.sleep ();
      LowPower.powerDown(SLEEP_8S, ADC_OFF, BOD_OFF); 

 
My question: Could something have happened to the original Moteino after the recent battery changes to increase the sleep mode power consumption? I've read about problems resulting from letting the supply voltage drop too much with the brownout detector turned off, but I have no idea if this is related to my problem.

[Felix]

Could be various things.
If the node is in good condition and factors such as corrosion are not involved, you have to determine where that big power draw is coming from.
I would first ensure that you can obtain the low power draw on a known good board with no such issues.
Then load it on the questionable board. Then isolate the problem, this is not always easy.

[TomWS]

Could something have happened to the original Moteino after the recent battery changes to increase the sleep mode power consumption?

I'd put my money on corrosion.  It's hard to see but you get these little fuzzy bridges of leakage between the leads, the pads, and under the processor and it bleeds the batteries in a hurry.  Once it starts you can pretty much trash the mote.  I've found the best preventative is, as you first deploy a mote, liberally 'paint' it with CorrosionX.  It's messy and it stinks, but, man, is it effective.  I had one outdoor enclosure leaking like a sieve, water was actually accumulating around the processor board (ESP8266 in this case), and it was still working.  Unbelievable.

[DonpK]

Thanks for the answers. It sounds like neither of you buy the warnings in this Adafruit article about the problems of code and bootloader corruption by "blindly" disabling the BOD "or you'll regret it later" Although they add there's a 1% chance of it happening plus I wouldn't have any idea how to troubleshoot or repair such a problem

Corrosion is a possibility, although this is an indoor unit and the problem seemed to start after a recent lost of power and battery replacement. Still, the same code works fine on another Moteino, so replacement of the processor and board seems like the solution

[alexelite]

@DonpK, thank you for posting the adafruit link. Good to know about this.
I have only one battery 8 MHz mote in a remote location, powered from an old phone lipo, and after almost 2 years the voltage just dropped below 3.9V, but I will now be careful to recharge it when it drops below 3V as to avoid this type of corruption.