Particle Electron – The Solution To Cellular Things

Just over a year ago, Particle (formerly Spark), makers of the very popular Core and Particle Photon WiFi development kits, released the first juicy tidbits for a very interesting piece of hardware. It was the Electron, a cheap, all-in-one cellular development kit with an even more interesting data plan. Particle would offer their own cellular service, allowing their tiny board to send or receive 1 Megabyte for $3.00 a month, without any contracts.

Thousands of people found this an interesting proposition and the Electron crowdfunding campaign took off like a rocket. Now, after a year of development and manufacturing, these tiny cellular boards are finally shipping out to backers and today the Electron officially launches.

Particle was kind enough to provide Hackaday with an Electron kit for a review. The short version of this review is the Electron is a great development platform, but Particle pulled off a small revolution in cellular communications and the Internet of Things

The Current State of Machine to Machine Cellular

Small microcontroller dev boards are nothing new. The technology to turn a hat, belt buckle, or bicycle into something that receives data from a cellphone tower has been around for more than a decade now, and a few very old blogs and forums preserved on archive.org reflect this fact. An old Siemens, Nokia, or Motorola phone, with the right serial cable,could be a microcontroller’s connection to the outside world.

SparkFun's Cellular Shield (above) and Adafruit's FONA 3G Cellular Breakout (above) Image source 1, 2
SparkFun’s Cellular Shield (above) and Adafruit’s FONA 3G Cellular Breakout (above) Image source 1, 2

More recently, Sparkfun and Adafruit have been putting together their own cellular modules for Arduinos and other similar boards. These modules, like SparkFun’s Cellular ShieldAdafruit’s FONA, and Seeed’s RePhone all provide easy-to-use cellular modules that plug into an Arduino. The problem is phone companies historically didn’t want to deal with a bunch of Arduinos blinking LEDs through their network.

While one would think more devices on a cellular network would be a good thing for a carrier, this is not what [Zach Supalla], CEO of Particle, found when creating the Electron. The success of a carrier in the eyes of shareholders depends on ARPU, average revenue per user, or the total revenue divided by the number of subscribers.

While the ARPU is great when AT&T, Sprint, and Verizon are selling multi-Gigabyte plans to smartphone users, it’s a terrible measure if a company is selling a lot of cheap one Megabyte plans. The economics of cellular carriers is why we don’t have a cellular Internet of Things; it doesn’t make sense for the carriers to do it.

Instead of going through the usual cellular carriers, Electron users will be getting their service directly through Particle. It’s a setup called a Mobile Virtual Network Operator, or MVNO, that allows carriers to work through Particle to provide plans without much data but at a very low price.

It’s a setup that works well for a cellular device that’s the size of a stick of gum. If you walk into your old Radio Shack, you will get blank stares (again!) if you ask about a low-cost plans that provide just a few Megabytes of data. Setting up a SIM card with Particle is as simple as entering a few digits into a webpage, entering your credit card info, and enjoying a Megabyte per month of data on a tiny microcontroller board.

The Hardware and Development Toolkit

UndersideParticle already has a few hardware products under their belt, such as the Particle Photon and Core, a very popular ARM and WiFi development board. It’s incredibly useful for people who don’t want to futz around with bare ESP8266 modules. To anyone who has ever used a Photon or Core, the Electron will quickly become very familiar.

Tucked away on the bottom of the Electron is the STM32F205 microcontroller, presenting 36 total pins to the outside world. The functions presented on these pins include UART, SPI, I2C, and CAN bus. In total, there are 12 ADC channels, 3 UARTs, 2 SPIs, 1 I2C, 2 CAN, 2 DACs, and 13 PWM channels. 1 MB of Flash is included, and 128k of RAM is available.  It’s an extremely capable board if you compare it to the Arduinos of a few years ago, and can go toe to toe with slightly more modern boards such as the Teensy 3.2.

The entire hardware design for the Electron is open source. Incidentally, Particle used Eagle to design the board.

ParticleSIM

Particle's web-based Arduino-ish IDE, allowing code to be flashed to the device over the cellular network.
Particle’s web-based Arduino-ish IDE, allowing code to be flashed to the device over the cellular network.

As for development for the Electron, choices abound. The easiest by far is a web-based Arduino-like development environment, Particle Build. Just like the Arduino IDE, Build will give you just enough to write some code and flash it to your device. Libraries for GPS modules, LCDs, switches, temperature sensors, and everything else that would normally be used in a one-off Internet connected project abound. While it’s not a full-blown IDE, it’s good enough and allows for over the air flashing of the Electron.

Particle’s Tinker app, a smartphone-based development environment is also available for the Electron. This app will allow you to read and write individual pins in a strange, ‘you know what’s cool? Visual programming’ way.

Javascript is possible on all of Particle’s devices, CLIs are easy, and if you’re coding for iDevices, your SDK is right here.

There are more than a few people who believe that a web-based IDE, especially one hosted on a server you don’t own, is a terrible idea. Lucky for us, Particle has made all their development tools open source, allowing anyone to roll their own personal cloud. For anyone concerned about the usefulness of Particle’s boards five or ten years down the line, this is a necessary feature.

 The Future of the Electron and Particle Cellular

Particle's certification matrix. They're working on getting the Electron certified for use in products. Image source
Particle’s certification matrix. They’re working on getting the Electron certified for use in products. Image source

More than any other Internet of Things company, Particle is seeing a lot of their modules become the basis of real products. There’s a reason for this: they’re one of the few hardware developers that are sinking money into certifications. If you wanted to build a device with a Particle Photon inside it, the certifications are already taken care of. The FCC and cellular certifications for the Electron module are in the works, but [Zach] says that will happen.

Of course, any developer building a piece of hardware with Particle boards probably isn’t using a stock Photon or Electron. There’s no need for USB ports or pin headers when a board is stuffed inside a black box, anyway. For the smarts that go into a product, Particle already has a WiFi module, the P0 and P1, and [Zach Supalla] says Particle is considering a cellular module that would also be FCC and CE certified.

The Electron is a great piece of hardware, but it’s not the biggest development in the world of cellular-enabled hardware. That would be Particle’s MVNO and the no-contract, one Megabyte cellular plan that costs only a few bucks a month. The burning question on everyone’s mind is, ‘will Particle sell SIMs to people who want to roll their own hardware?’. The answer to this, at least for now, is yes. Even if you’re not interested in the Electron itself, the cheap cellular plan is undeniably interesting. It’s cheaper than other MVNOs like Ting, is designed to be a purely machine to machine plan, and it will be coming soon.

The Requisite 2G Warning

Particle is offering three version of the Electron. Two are for 3G networks – the U260-based board is for North America, the U270 is for Europe. The third version of the Electron is for 2G networks. The 2G version is $20 cheaper, but this may be a false economy. In the US, AT&T will begin shutting down 2G networks “soon” – either starting January 1, 2017, or whenever it doesn’t make economic sense to keep the 2G networks running.

Conclusion

Particle already has a lot of experience building Internet of Things things. The Core and Photon are excellent WiFi development boards with a strong development platform and a very capable cloud backend. The Electron is the continuation of this, expanding the Internet of Things to devices that move more than a few hundred meters when being used, or operate where WiFi networks aren’t available.

While the hardware is good, the big story here is Particle becoming a cellular network. Smartphones, as we know them today, have been around for nearly a decade and until now, no one – at least at a large carrier – has realized the value of giving a few kilobytes of data a month to tiny battery-powered devices. Somehow or another, Particle solved this problem, and they’re not locking it down to only their devices.

Particle will be shipping rewards out to Kickstarter backers beginning today, and the Electron (and SIM) will be available on the Particle website in due time.

56 thoughts on “Particle Electron – The Solution To Cellular Things

      1. wait IED as in Improvised Explosives? I don’t really think that’s what DV82XL meant by “disruptive” I think he was using it in the tech world sense of shaking up the status quo, but even if he wasn’t I don’t understand how you jumped to IEDs. like honestly, I would be interested in the mental associations that you went through to end up there.

    1. I have a ting sim that i use for GSM phones. I just noticed that you can get another one for 6$(or as many as you want for 6$ each). plus whatever txt or data you use. cheaper than the SIM that sparkfun offers from t-mobile. if you usage is very low. theirs is 10$ for a single sim for 6 months. and it you device is remote you have to go and replace it. Ting would allow you to have it and just pay a monthly cost.

      1. Yessir – there is the Konekt Dash which offers similar functionality to the Electron

        You can also use our Global SIMs in any hardware including the Electron. It is $1.00 for the first MB and $0.60 for every MB after that.

    1. Well, if you have easily available WIFI, then the Electron is clearly not the right choice. If you don’t, like say monitoring weather conditions at a cabin in the deep woods, or in a field 60 miles out of town, then that’s when something like the Electron or various other cellular boards becomes useful.

      It’s all about using the right tool for the right job.

      1. oh! wow! like. LONG-RANGE? No-one will have to create a global COMMUNICATIONS network of terrestrial stations technology?

        you mean I don’t have to worry that no can invent and USE publicly available AMATEUR RADIO?

        *hint read only the caps-lock text.

        1. I know, right? It seems like HaD is neglecting HAM radio, which is basically free if you have a license, to appease some corporatesponsor that relies on (hackers like us) subsidizing their stupid broken network.

          I think the missing period after the second paragraph says it all.

        2. Didn’t have notifications on. I am a ham. I know you can use if for something like this, but it’s not always the right tool either.

          Amateur Radio prohibits encryption or encyphering of data passed over the air. If you need to maintain the privacy of your data, this doesn’t work. Or if you need to control the system either. Now you could do secure, unencrypted communications with key signing, using your signing key to verify authenticity, but again, if you need privacy of commands or data sent to the remote system, you are again SOL with ham radio.

          So get off your high horse!

          Also, ham radio is only LONG-RANGE in the right conditions. With VHF/UHF, you’ve got line of site considerations. APRS can help with this, but that’s not guaranteed. Perhaps you meant HF? Well, then you need a larger antenna. And that may not work for every project. If you’re trying to make a small remote temp-sensor, you might not want to attach a big mag-loop antenna or big ass dipole to your project.

          I don’t know why my comment about using the right tool for the right job brought up your HATE and soap-box RANT

          *hint: my caps text is purely mocking. Please disregard.

    2. Well, yeah, if you’re in range of a known wifi connection this would make very little sense, but that’s not the intended use. Think data coming from solar-powered sensors outside, or GPS coordinates from your bike if it’s stolen, or any number of other uses that don’t involve being near a friendly wireless access point.

          1. Reminds me of my wife telling her friend about the justification for getting a carphone back in the day: “It’s like having a remote control for your husband.” The look on her husband’s face was priceless.

      1. Gosh, I’ve never wanted to send someone back in time to WWII to work a Radio in a trench until now. I very much want to elect you to travel back in time now. Do you speak Navajo?

  1. I can get one of these https://www.google.com/search?q=ZGPAX+S6&tbm=shop and a prepaid SIM with 1GB data that lasts a year for another USD $80

    Look at the specs on that thing! 3G WiFi GPS….

    I just need to get GP I/O lines via the USB interface, i.e. Add a Arduino Nano, which I could also run off the included battery.

    That is a lot of options and functionality for under $200. And I bet if you ask around in Shenzhen somebody can get you just the main board for the watch for $20. So if people are going into the business of competing in that market they had better do a brilliant job in the areas of support, tutorials, project ideas, add-on and community.

  2. Pretty ambitious in scope for the cell plan– Only additional comment to piece would be, yes the 2G model could/would be a ‘white horse’ in the US, but I don’t believe there is a mention it is ‘global’. In my mind, this fact is a ‘really big deal’ as all of a sudden you *really* open up IoT in many much more potentially rural or remote areas for basic tasks/monitoring.

    1. You only need one node with a long range link, the rest can use $2 WiFi modules as part of a mesh network to get data back to the main node. Even WiFi can be set up to create directional beams that operate over hundreds of kilometres.

      If you do have GSM back to a phone tower the tower is going to have a microwave link back to more populated areas, in many cases.

  3. Does this thing have coverage anywhere in the world or just when you’re connected to their tower? In Europe you can get gigabyte for around $6, depends on country and carrier, so $3 for a meg doesn’t seem like a good deal.

    1. Depends on country. Cheapest italian carrier is 3 which offers 3Gb/5€ but it’s inconvenient to use as IoT because it stops you unles you load a page with a browser, then it redirects to internet.tre.it, after that it unlocks a 20Mb packet for 5-6h, after that you must unlock again. Other carriers often have silly rules which prevent robot uses like you must do some calls or risk sim blocking or telemarketers calling you and you get billed to be notified by SMS of the missed call.

  4. You guys forgot about Konekt (disclosure: I’m an engineer there)
    We have much better data pricing, have been doing IoT cellular for way longer, we have a board similar to theirs but it’s already shipping and we have an open cloud. Granted, we’re a little smaller and still building out features, but I think we have some cool stuff :)
    Check out konekt.io for more info

    1. Is Gilgamesh like Sonny Bono or Ringo Star? Yeah, yeah, Gilgamesh that guy that collected all the the cursed Pre-Sumerian weapons including weapons wielded Abraxis blessed avatars before *cough*. Yeah. No. Sun meeting the Moon scares people for some dumb reason. Let’s agree to drop it because most Gen Y, Z? Won’t get it.

    1. I think it’s just AT&T and a handful of regional carriers. I have yet to see a CDMA or LTE based kit that would work on a Verizon compatible networks. My home area network was all built by Cellcom, which is now a Verizon partner network. GSM connectivity is somewhat sparse outside of larger towns.

  5. This tech is utter and complete landfill, crap and GARBAGE. Do you want “cleanskin” domestic cells? Because this is how you make cleanskin domestic cells.

    In all sincerity you have zero clue about channel 0, the channel that the cell operators use to send YOU texts. That said, F’ AT&T and F Carrier IQ.

    Gimme a CB or HAM operator with some cred and honor and I’ll show you a peace keeper.

  6. I went to our local supermarket the other day and got phone for $10 that included a Telstra (Australia) SIM card with $10 credit – so with this I also get a battery/charger, case, keypad, camera and a display, plus the internal radio can rx FM radio. I am thinking it would be better to hack the cheap phone, it is not much bigger.

    1. Yup, so long as you can tell people how to unlock and root the device and point to the correct set of build tools so they can strip all the bloatware off it and have room for the code they need. Once a phone gets that sort of treatment they get very popular :-) Or you can just treat it as a set of parts, but then you have the problem of interfacing them to a new MCU. I have a growing collection of old gear like that, lots of useful parts, if I can get the specs on them etc..

  7. I just got mine and when setting it up I realized you can only flash via the cloud or command line. Seems a little weird since they have their own version of Atom, but it only has upload via cloud options…..

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.