SparkFun Forums 

Where electronics enthusiasts find answers.

Have questions about a SparkFun product or board? This is the place to be.
By andywatson
#132994
I have about a dozen Arduino Pro boards that I've bought from Sparkfun over the past 8 months, usually in orders of 3 or 4 at a time. In the past few weeks, 4 of the boards have started getting flaky. I have them mated with Adafruit datalogger boards, and they've been working great for months but recently they've stopped recording data after being deployed anywhere from a few hours to a few days after I put new batteries in and everything. When I go to retrieve the memory card, batteries are still good, the data card is empty or it stopped recording data shortly after being deployed, and the pin 13 LED on the Pro board is blinking rapidly. Pressing the reset button only makes the LED stop blinking if I hold down the reset button. When I let up, it goes back to blinking rapidly. Cycling the power off and back on seems to make it stable again, but it'll start blinking again a few hours later or the next day. Did I get a bad batch of Pro boards, or is there something else that is indicated by the blinking pin 13 LED?
By TomasReabe
#133140
Would need more info on what you are doing and what your code is. It could be an error in the code a bad flash, or any number of things. But more info is needed on your hardware and software to help. I know I had a data logger that had problems and what it was is I had too big of CAP and the voltages were still nuts when the SD card was being setup. It was a pain to find out what was going wrong but was fixed by a simple wait comand in the code.
By drienkm
#193781
Years later... I'm having this exact issue and I'm not finding many documented cases out there. Did you ever find resolution?
I'm using 5V Pro Mini boards from Sparkfun. They run on a battery for days outdoors (in a sealed enclosure). I checked two of them the other day which were in the same location. It appears they had both run for several days (performing every part of the sketch many times), but were both stopped and fast blinking the pin 13 LED (maybe 10 or 20 Hz). My sketch doesn't reference pin 13. The batteries were both fine, but one was significantly higher than the other, which tells me power issues are not a likely culprit. The only common factors I can think of are temperature and runtime. The temperature wasn't changing much... it was around 3 or 4 C.
My sketch is pretty basic. It watches a sensor over I2C, a couple analog sensors, and operates two tiny relays (10mAh coils) and a few LEDs.
If you ever got anywhere in diagnosing your issue, I'm all ears!
Thanks.
By drienkm
#195099
I figured out my issue. I had 100k (to conserve power) pull-up resistors on my serial and clock pins for the I2C sensor. I found some guidelines on proper sizing of the pull-up resistors (linked here: http://www.ti.com/lit/an/slva689/slva689.pdf) and after switching to 15k resistors, I have never seen the glitch again in many thousands of hours logged. Totally worth 117 uA!

The processor was occasionally getting in a jam in the I2C code whenever the data or clock signals would get a little too sloppy due to slight changes in node capacitance... perhaps due to small temperature or humidity changes, or the proximity of... maybe an ant or something :roll:

Hope this helps someone.