Building The Ultimate Raspberry Pi Automation Controller

At this point, we’ve lost count of how many automation projects we’ve seen with some variant of a Raspberry Pi at the helm. Which is hardly surprising, as the boards are cheap, powerful, and well documented. The list of reasons not to use one has never been very long, but with the PiCon One that [Frank] has been working on, it’s about to get even shorter.

The project takes the form of an IP65 industrial enclosure and support electronics that the Raspberry Pi Zero W plugs into. While expandable in nature, [Frank] has a core set of features he’s aiming for as a baseline such as additional serial ports, integrated uninterruptible power supply, a battery-backed Real Time Clock (RTC), an array of programmable status LEDs, and support for XBee and GPS plug-in modules. Feedback is provided through a pair of four digit seven-segment displays and a color 320×480 TFT screen running a custom user interface.

[Frank] envisions the PiCon One for use as a rugged solar power controller, eventually able to measure array output, energy consumption, and even operate motorized mounts to keep the panels pointed at the sun. To that end, he’s recently been experimenting with running JPL’s Horizon software on the Pi to determine the sun’s position in real-time. But the device is capable of so much more, and would make an ideal controller for many home and potentially even industrial applications.

34 thoughts on “Building The Ultimate Raspberry Pi Automation Controller

        1. I would assume you wouldn’t. From my reading, the GPS module itself has it’s own battery backed RTC. Not sure why you’d want another RTC on board when the GPS clock is going to be self-correcting. Someone let me know if I’m wrong.

        2. As this is an “industrial” design, the possibility of the end product being located inside a metal enclosure in a metal building with out windows, is very likely. GPS would be (almost) useless in that scenario.

          1. If the purpose of the GPS is to provide date/time or its own position to determine the elevation of the sun for solar panels. Does it even matter if it’s inaccurate by a mile or two?

            Unless the user is located smack bang on a date line I don’t believe even in extreme degree of inaccuracy its going to cause any concern.

        1. I got on the bus, sat down and noticed a beautiful Asian woman crying in the seat across from me. I moved over and asked her why she was crying. “I don’t usually bare my soul to strangers,” she said. I replied that sometimes it was perfectly fine to tell your story to a perfect stranger. She nodded and said, “I just came out of my therapist session and he says there is no way to cure me.” I asked what exactly was her problem. She said, “I’m a nymphomaniac, but I only get turned on by Jewish cowboys. You know, I do feel better. By the way, my name is Kim.” “Glad to meet you,” I said. “My name is Bucky Goldstein.” – Steven Wright

      1. “I put a new engine in my car…
        but I didn’t take out the old one.
        Now, I can drive 200 miles per hour!”

        “I replaced my car’s headlights with strobe lights.
        At night, it looks like I’m the only one moving.”

    1. For a mass production device, a purpose designed board is going to be more cost effective because you only implement what you need. The advantage here is prototyping where a general purpose system is easier to program and you have access to tools and widgets already developed. Once refined to a product you want something more energy efficient with no extra interfaces or hardware.

      1. You do realize PLC’s have modules that allow the customer to customize them to their needs. Also there is the issue of support and documentation, Not to mention handling hash environments and voltage spikes? I know I wouldn’t trust what amounts to a throw away board meant for kids and hobbyists running part of a Lime Kiln.

        Are you going to have a paid staff support the product on a 7×24 basis, and I don’t mean so,me scrpt reading Hindu in Bombay getting paid $7 a hour.

        Then there is the software. s it easy enough for a electrician to use and modify the code like it is for a PLC.

  1. First Nice clean refreshing build. Took you a bit of time to package it nicely with the features for a variety of applications.
    Those questioning why might go actually read the build to discover the answers to your comments.

  2. Not a bad hobbyist central point / simple dashboard. I’m just craving a small PoE board with BT and GPIO to use as satellite points on the same network, to report back to a dashboard like this.

  3. I have worked in Industrial Automation all my life. The new micro PLC’s on the market cost under a 100 bucks, are very reliable and a breeze to program. Doubt an RPI based controller can beat that.

      1. The Micro800 series from Allen-Bradley isnt too bad but the programming environment is a bit funky (VB-based but it is free), not enough to complain for <$100USD though. I've also played with one of the newer Siemens micro units (S7-1200 maybe?) a few years back and it seemed like a great product.

        My shop is entirely AB but my exposure to Siemens has always left a better taste in my mouth. Could just be a "grass is greener" thing though.

      2. We have used the Micro800 in several applications…so far so good. But the easiest to use and program would be the Automation Direct line of PLC’s. We have several of the Click! based ones controlling several pieces of our equipment. Their HMI’s are really great too.

  4. I like it, clean design and a good place to start. Don’t know why everyone’s commenting on the GPS part. If you can’t think of a reason why to have GPS you shouldn’t complain or make negative comments. Many industrial vendors have similar design options that and also have one variant that includes GPS.

  5. Wow, the lack of imagination in those posting here is incredible

    This is an incredible build. IP67, modular, less expensive than any PLC and it includes GSM and GPS… best yet it’s built for an engineer that actually understands the entire stack. I commend your build. We could specifically use this both in mobile implementations and at thousands of locations in our gathering and pipeline operations.

    Real world usecases include asset tracking with real coordinate tracking instead of guessing based on addresses, survey information, or rarely accurate manually input lat/lon. If the asset moves. Pending it will have an external antenna, this out of the box solves all my problems day 1. I can have an entire system operating tick by tick with the ability to have accurate time for pipeline modeling and because the GPS can give me altitude, I don’t have to guess based on aerial land survey data or even pay for that data.

    With GSM and a directional antenna, it simplifies installation and eliminates the need for expensive and often complicated back haul systems in areas that are already lacking. Add a VSAT or BGAN IDU to the install and you could have communication resiliency.

    Best idea out of the IOT world I have seen yet. Keep up the good work!

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.