I have deployed about 50 electrons so far and counting. I noticed that their heat specs are much lower than what an electron in a box in the sun is receiving. (Some sim cards have even melted!) I am adding small heatsinks and fans to try and keep the temps low. Has anyone else had a similar experience? If so, what was your solution?
Do you know how hot it’s getting in the box under the sun?
It’s melting sim cards?
Field tech says when he was able to measure he was getting an avg of 20F over ambient temps. So during the day were sitting at around 120-130F for around 6 hours. We are under the specified maximum heat ratings, so I’m wondering if the operation of the electron is adding extra heat that we aren’t able to catch yet.
The Electron will generate it’s own heat also, so that’s contributing to the heat problem if it’s in a small enclosure.
If your case is not white, then that’s not helping either. Shade the case from direct sunlight if possible.
We are using a sand colored case that is standing in the middle of a field. Oil and gas locations in the dessert are not kind!
@calebatch, have you considered creating artificial shade by adding a small canopy above the case? Do you have any venting holes or does the case need to be sealed?
I am currently looking into a way to mount a solar panel above the enclosure to take off some of the heat. I think that plus a heat sink might help. I will see if I can add vents as well, but these are in a hazardous environment and might have to be sealed.
Hey there! Even with a PCB design that elevated and robust temperature specifications, the SIM card will be a material limitation in integrated products. This has prompted a good conversation internally about how to improve temperature robustness from a hardware perspective, so more thoughts forthcoming.
1 Like
That is good to hear! I’m not sure what to do about the sim card, but it sure is interesting. Maybe the material of the sim can be changed to handle the higher temps?
There are industrial temp rated SIM cards; see Gemalto for example:
http://www.gemalto.com/m2m/solutions/mim
The “trick” with sealed enclosures and heat generating components is to make sure the surface area of the enclosure can dissipate the required heat load of the electronics into the ambient environment. In practice this usually means that a heat sink, which increases the surface area (all those fins usually) is required, but it could come from a larger enclosure.
Do you have power available?
You may be able to use something like this TEC1-12705 12V 65W Heatsink Thermoelectric Cooling Peltier Plate from amazon. It will pull about 5 amps, and you will need to add heat sinks for both the hot and cold sides.
I have had them in coolers that would keep a 6 pack cold or freeze a small ice tray.
They come in various sizes and wattage and price is low. You could also have the electron act as a thermostat to control the temperature say < 100° F. or <37.8° C. for the non-Fahrenheit group.
Chuck
1 Like
I wonder what the internal temperature would be if your casing were brilliantly chrome instead of sand color.
For cheap testing you could add a DS18B20 to two of your boxes and cover one of them with tinfoil, on the 180° or so which get the most intensive photon bombardment.
I’ve had a similar problem here in coastal North Carolina where my devices are often mounted on top of a commercial building and are exposed to direct sunlight. Temperatures this summer have regularly approached 100˚ and then you have to add the effect of direct sunlight on the medium gray enclosures. We had cellular hotspots overheating and going offline. Tools left on the roof for 15 minutes could not be picked up due to their temperature!
I went to Lowes Home Improvements and purchased some Reflectix reflective roll insulation and wrapped the enclosure in that material and haven’t had a problem since. Just passing this along for your consideration. It’s a bit of a Rube Goldberg, bit it works! In my case, I’m using Photons at this location, but I’ve noticed they generate more heat than Electrons. The enclosures I’m using are about 12"x10"x5".
1 Like