The electron is great for building remote, configuration less devices. At present the operating temperature of the electron is -20 to +60°C. -20 is fantastic, however it would be nice to have the upper maximum temperature in the future or a new version of it to +70 or +85°C. Since the device will be placed regular outside in a box, temperatures can rise pretty fast in these enclosures. So I was wondering if there are any plans on releasing in the future an electron with +70 or +85°C operating temperature. That would make the use of the electron even much wider.
@zach, thanks for the reply. @BDub, what do you think about an industrial grade Electron with an operating temperature range of -40 --> +85C (or al least a higher range than now)? What are the limiting components for the operating temperature? I suppose the lips battery could be an issue, or maybe the sim card, however IoT sim cards in Belgium are available in the range of -40 --> +105C. Thank you for your opinion / future vision on this matter.
The ublox modem on the Electron can be had in -40 to +85 C ratings, so I think that would be maximum range for the current electron hardware design. Obviously a new hardware design could have a higher temperature range.
You do have to be careful with automotive rated ICs in my experience since often the trade-off is extended temperature range vs. long life. In many automotive designs a 10-year lifetime is considered the norm and so in many cases the ICs you are buying with 105 C ratings are identical to the 85 C rated devices, but with a shorter lifespan due to metal migration etc. What you are often paying extra for is more manufacturing oversight, tracability and defect analysis, not a “better” part.
@bko, I totally agree with you on the long life, and thank you for your reply as well. However if we consider 2G/3G operation, a lot of the applications will reside outdoors and therefor requiring a greater operating range as -20 --> +60C. The -20C temperature is quite OK for most places, however, the +60C is easily reached in a lot of places. If the electron is placed in a box and the sun shines on the box with an outside temperature of +30C, the temperature in the box will easily rise to +60C. So to be able to use it for outdoor applications I would need a +85C or at least +70C version. If the battery and sim card would be the only limiting factors, than we are all right with the current design. However if there are other components that are limiting the operating range then indeed a new electron design will be needed. I started this topic since I think it really makes sense for the electron to have an industrial temperature grade version for outdoor applications.
I am also pretty interested in this. To the point where I dug into the BOM of the Electron to see what the limiting factor is on the device for temperature. I was more focused on the lower end -20 temperature, as I will be putting my devices in the cold, but I am sure it is applicable to you.
In summary, I found the following components to possibly be a problem:
LIPO battery (-20 – ??)
Maxim battery fuel gauge (-20 – 70)
NR3015T2R2M Inductor (-25 – 120)
TS-1185A-C Buttons (-20 – 70)
Here’s a link. Its the BOM from the GitHub repository with a couple extra columns added.
Not sure where the 60C limit comes from. My best guess, though, would be the LIPO battery.
I would be interested to know what happens at -20C and below. Especially if putting the device in deep sleep mode will allow it to achieve its -30C storage rating. What do you think @BDub?
@hwestbrook, thank you for the good work on the BOM for adding the operating temperature! Good to see support of you for this topic. I thunk it will apply to a lot of use cases (either cold or warm). The fact is that outdoor applications will require a larger operating temperature. As the Electron is a product that is build for use in remote (and most remote places and applications are outdoors) places, an industrial grade temperature range would be such a big benefit. @BDub can you please give your comments / ideas on this please?
One other thing to consider here is that the LiPo battery is /not/ actually rated for -20C full operation. It can discharge down to -20C yes but it cannot be charged at all below 0C, effectively making the battery useless for real usage below freezing since it can’t charge up at all until the battery warms up (which may conceivably not happen for weeks on end depending on where the device is). This rules out automotive or outdoor industrial / stationary applications in any climate where it freezes. Yes, it may work for some time if you do allow charging below freezing but it does do long term irreversible damage to the cell (specifically plating where lithium gets caked on to the innards of the cell) which can lead to spontaneous combustion of the cell (obviously extremely undesirable).
For this and several other reasons (including the whole shipping / exporting nightmare for lithium batteries, the high volatility of Li-Polymer chemistries and that the battery connector is not vibration resistant at all) we are not able to use the LiPo battery in our applications and instead are forced to rely on a beefed up power supply for the Electron (which adds considerable cost).
So we too would be very interested in a more ruggedized version of the hardware that might allow a higher degree of independence from some of these concerns.
I am looking at using the Electron in a commercial product with the battery connector (but a different Li-ion battery pack). Your comments raised some concern on my part. I have no prior experience with the JST connector and would appreciate if you could please expand on the concerns.
a) Is the vibration concern related to fretting? Could this be addressed with a thermal grease/lubricant?
b) Is the concern related to the solder joint quality of the JST connector or the crimping quality of the wires?
The connector aspect of the battery was something of a secondary concern to us (the larger issues were the hazards that the lithium battery itself presented), but it’s an issue for certain applications none-the-less.
Our chief concern with the connector was that it doesn’t have any locking aspect to it (no tab or latch that has to be depressed to allow it to be removed). A lot of our products get integrated into applications where there may be various degrees of vibration (eg: mobile applications) so obviously such a critical part would need to have the ability to latch in place and not wiggle loose in such circumstances (the battery itself would also need to be physically supported inside the case such that it can’t vibrate loose or get damaged under excessive vibration). Technically the SIM card really isn’t designed to stay inserted with vibration either but that at least could be largely mitigated by inserting a post or screw in front of the card once it’s inserted to hold it in.
Applying grease to the battery connector is probably not a good idea as this could impede the contacts in the connector, it’s possible some sort of RTV (which likely wouldn’t penetrate down to the contacts) might be able to be used to help secure the connector though. Obviously if you RTV it in place then you’d have to ship your product with the battery permanently connected which may be a problem if it then sits on a shelf for a long period of time though.
If our applications were exclusively stationary then the vibration aspect would be a little less concerning to us (though transportation of the product would still introduce vibration to some degree).
Again our primary concern is that the lithium battery in general (unless you are using one specifically designed for extreme weather operation and charging) adds significant liability and complexities both logistically (shipping restrictions) and environmentally (usage in sub-freezing temperatures negates the ability to charge the battery safely and usage in extreme hot conditions such as direct sunlight or in an enclosed vehicle where temps could easily reach 45C+ which can have severe longevity and definitely safety implications for the battery–simply storing the battery at elevated temperatures can irreversibly accelerate aging). There are very few lithium cells designed to operate fully in extreme temperatures and they unfortunately are typically pretty expensive.
To be clear, we are overall pleased with the Electron platform. We do feel that there are places where the platform could be improved down the road though to make it more viable for commercial use.
@dmfische@solarplug@BDub, we too had a problem with the connector of the Lipo battery. Our biggest concern was not vibration (nor temperature) related in this case, but in size. When the Lipo battery was attached to the Electron, it was too big to fit in to our housing, luckily there was a way to fix this, which might actually be interesting for you guys as well.
The Electron’s first pin is Li+ (which is the positive terminal of the Lipo battery). Unfortunately, this pin has no pin header soldered to it (really do not know why Particle did not solder this one). So we soldered a 1 row pin header to this terminal and connected an 90 degree angled connector to our PCB of which one pin was connected to the Li+ pin and the other pin as connected to the ground. So we overcame the issue in this was. Maybe you can use the Li+ pin as well for choosing a more suitable connector for your applications.
@solarplug - thank you for the comments, as they were most helpful. We too are most pleased with the Electron platform and one of our main focus is to harden the platform for environmental/EMI concerns.
@BDub - in looking at the github repository, I am not sure if the data sheets for all of the connectors are from Kaweei. It would be useful to have the Kaweei specs available (SIM, JST - male and female, 1x18 header, uFL) so that things like contact resistance, material, plating thickness, etc would be known. Could you please advise?