I need help troubleshooting my rooftop setup, where the RAK4630 module died repeatedly after 5 days under the sun (50C max).
My setup:
RAK4630, RAK 910421 solar-powered outdoor case with 1x 18650 battery and temp/pressure vent added. Latest stable firmware 2.4.0. for rak4631.
Picture here:
The problem:
My setup works fine outdoors for up to 5 days. Then suddenly it stops working.
The battery is fine at 4.0V. But the blinking LED is OFF.
I could not reach it via the Web serial console or BT app.
I could not see anything on the LCD screen after attaching the I2C LCD screen and rebooting it with a power reset. However, I could double-click on the reset button to bring it to bootloading mode, so the bootloader is still there.
After reflashing the firmware, it still would not communicate via USB serial or BT. On the web serial console, it just got stuck on “Messages: Loading…”.
This happened TWICE during my testing of the outdoor setup. The 1st time, I thought it was a defective chip inside my RAK4630 and I replaced the RAK4630 module. That has worked for the past 5 days. But today, it stopped working with the same symptoms.
Question:
Did anybody else experience the same problem with an outdoor setup? I had a BME280 temp sensor installed and the highest temp I recorded was about 50 deg C in the middle of the afternoon. However, at night the temp can drop to 12C here in SF Bay Area, California. So there is significant thermo cycling taking place.
Is it possible to revive this RAK4630 module? I am guessing the nRF52840 hardware is the issue?
If temperature is the issue, how can I deal with it without active cooling?
My other outdoor setup based on RAK4630 and a UL-rated solar power bank also failed today with the same symptoms after running perfectly for a few days. I feel very frustrated with the reliability of the RAKwireless module. Please help…
i have several raks running on solar with no problems.
almost same setup like yours, 19003 and using 18650 (but i use 4 pieces 18650 in parrallel !)
but i had several problems that made me climb the pole
on time it was my fault to disable bluetooth when leaving the spot… Rak no bluetooth–>-unstable, and crashed
the other time it was some problem with power source, that lead to the same symptons, that you describe… i had to reflash the node to get it back running. i dont know exactly, but i think it was some problems with voltage of battery and defective solar panel… that made the node loose its firmware… so i think i might be voltage problems…
since you only have 1 battery connected, maybe this one is causing the trouble, and sometimes the meshtastic voltage readings are just not correct !
so maybe try using an other , or better 2 batterys for a test run… and manually check voltage
i dont see a battery on this one ?!° if its running only on solar, you are turturing the RAK, when voltage is going down, or starting if i was that rak, i would go on strike
at the moment, before i put something out on a pole, i have it running in my garden for at least 3 weeks, fully configured, put it in basement, for discharge, and back in the garden to recharge again in sunlight
dont want to climb poles to retrieve nodes no more
also i test-drive the BLE updates on identical system before the ones on the poles are updated
Thanks for the note.
The second outdoor setup uses a Moultrie 10W solar battery pack, as a 12V power source. The 2 power wires come in from the top right via the white cable gland in my photo.
To host the RAK4630 module, I use RAK19016 5-24V Power Slot Module and RAK19009 baseboard.
My biggest challenge now is to revive the non-responding RAK4630 module if it is possible. When I tried to revive it, I already disconnected the battery and power it via USB cable. Thus the battery voltage is no longer an issue.
No. I did not experience a similar problem but I am not using that kit.
I do not like the direct powering of the RAK and so use a Waveshare D solar power manager and it allows you to make sure the power is always OK and reboot is ok if the battery is re-charged manually via USB. (The RAK runs for only about 5 days on one 18650 battery of 4500 mAh so you might be having this problem.) This was a topic in this thread Do RAK nodes always need manual reset after battery drain? - #10 by adingbatponder where I think what I was saying was misunderstood a for a while.
I do not know if temperature was the cause of a problem. I ran my node in a box in the sun and it went to 55 C a couple of times and did not suffer … I was worried about this so did what is shown here (RAK Solar enclosure on hot days? Too hot inside? - #18 by ScoJa ) which is to put a layer of rubber spacers around the box and cover the whole thing in aluminium tape. This resulted in the same node box (details here Solar node for top of tall building ) not getting hotter than a few degrees above ambient temperature when the box was pointing north and shielded from the south by a metal mounting plate. I would therefore expect that you can prevent it getting so hot with simple measures but of course if the panel in part of the box then this is harder.
I am not sure BTW that a 4.0 V voltage on the battery means that the RAK is necessarily getting the right voltage. (I am no expert on this) … I would swap to a new fully charged battery and see if that changes something e.g. it re-starts. You could also just run it via USB power from a fat power bank to see if the thing also gives up after 5 days or just keeps going for much longer. Then it would be a battery draining issue and a re-start after drain issue (see item 2).
