This isn't going to be one of your normal Instructables where I show you something that I've created on a breadboard, provide some source code, a schematic diagram, board layout, and finish with a picture of the finished project (usually crammed into a Altoids can or some other disposable plastic container).
This one is going to be different.
This instructable is going to focus on the most important part of any project, packaging. Your project might be the neatest gee-wizzer on the planet but it may not get a second look if it's poorly packaged for its intended purpose.
As the following pages demonstrate, inside your project may look like a rats nest of wires but outside it can look and function like a commercial product. When it comes to DIY projects, many times, beauty only needs to be skin deep!
Step 1: Start With the End in Mind
Every project starts life as an idea. If you're lucky, your idea came with a rough vision of how you (or others) will interface with it. If it didn't, STOP! Don't go any further until you have a basic understanding of what your project is going to do and how you're going to use it in it's final form.
In other words, Start with the End in mind.
In my case, I wanted a wireless DC voltmeter to transmit cell and bus voltage readings from a battery under test to a program designed to record the test results. A typical 24-cell 48 volt battery plant requires over 240 measurements during a discharge test to capture cell voltages at 10% intervals. That's a lot of readings.
The normal method for doing this is to measure each cell's voltage with a voltmeter then enter that data into a paper form, program, or spreadsheet. Trust me, this is not a lot of fun and prone to keystroke errors.
My idea came with a vision of how it should be implemented. First and above all, it had to be safe to use. The batteries to be tested range in capacity from 50 to over 2000 amp-hours. Short-circuits can be lethal. That means no loose wires, a rugged plastic enclosure and exposed metal only at the test lead tips were must-haves. The device needed to be battery powered, simple to use, allow field calibration, and accurate to within 2 millivolts. It also had to be wireless and plug-and-play drop-dead simple to setup (i.e. not Wifi!). Finally, it had to fit in two hands (no extra box hanging around your neck) so a pistol-grip housing would be desirable. Oh, and let's throw in a display to show you the number of the cell being measured, it's voltage, it's previously recorded voltage, RF link signal strength, and the internal battery's charge. Sounds like a pretty big order for a small package.
Now the idea has formed into a concept... but a concept is nothing without the hardware so let's start at the end and work our way back to the design.
I want this thing safe, small, light, plastic, and professional looking. I also want it in a pistol-grip housing with a trigger to send the readings to the remote computer. It needs a few extra buttons to navigate a few option menus. Sounds like a job for a 3D printer! Or not! I'm sure there are those out there with the skill to design what I needed and have access to the equipment to print it... that all sounds expensive and time consuming to me.
Step 2: Pistol-Grip Heaven
After looking around for ready-made pistol-grip cases for a while (OK, a few minutes) I noticed that Infrared thermometers were advertised all over the place, and some of them are even cheap! They are housed in well designed rugged pistol-grip plastic cases that include a trigger, mode buttons, a display, an internal 9 volt battery housing, and extra room for more "stuff". And did I mention they're cheap? Like USD $13 cheap? I chose the DT8380 thermometer from Amazon (you guessed it, it was one of the cheapest), but there's many others to choose from.
Being a nerd, the hardest part about using the IR thermometer for my project was destroying a perfectly good piece of electronic gear...
Step 3: We're Working Backwards, So Tear It Apart to Start
Of course the first thing you'll want to do with your shiny new IR thermometer is tear it apart. You need to know how much room you have to work with. Pay particular attention to the size and shape of the circuit board. If you don't have a micrometer, buy one. You'll need the most precise measurements you can get from the factory board because you need to duplicate the size, shape, and locations of the switches and display for your new board. You probably won't be able to reuse many parts, although I would have loved to reuse the LCD. If you choose the right display for your project you should be able to fit it under the LCD bezel. Just be extra careful handling the bezel because it's fragile and easily cracked (ask me how I know). I settled on the .96 inch blue/yellow OLED display from Banggood on my project. They're cheap (can you tell I like cheap?) and they interface to my chosen processor using I2C, so only a couple of pins are needed to drive it. Best of all, it fits under the original LCD bezel and actually looks better than the original LCD.
Step 4: Design Time
Now I have the housing, a circuit board size, and a display that fits my housing, it's time to design my project.
Knowing the form factor I needed to design to greatly streamlined (or should I say limited) the design and part selection. I knew I would need a power supply, processor, radio, input conditioning, buttons, and a display (oh, and a piezo buzzer too, I like key chirps and over-voltage sirens) so the challenge was finding the right combination of parts that would fit the housing and perform the functions I needed. The main circuit board had to be 1.25 x 2.85 inches with a taper down to .8 inches at the bottom.
Don't even think about a Raspberry Pi or full-sized Arduino for something like this, they're HUGE! Same for a Zigbee radio, there's no room for it. I needed the processing power and number of inputs of an Arduino but in the form factor of an over-sized postage stamp. Luckily, two companies provide just that kind of device. LowPowerLab has their Moteino and Anarduino.com has their MiniWireless Arduino boards. Both are available with RFM69xx radios already piggybacked on them and are the ideal size for this project. Just pick the frequency and power you need. I chose the MiniWireless with HW915 radios (915 MHz high-power version) for my project and I get just over 700 feet on the RF link.
Once the parts were chosen the actual design revolved around the number if I/O points I had on the processor and what I needed them to do. With the basic design in mind and the components selected everything was put on a breadboard and debugged. From there the design was moved to a perf board. Remember the part about beauty only needing to be skin deep?
The perf board was cut to match the original circuit board. This allows you to fit the major modules together to make sure everything fits.
Step 5: Putting It Together
These photos show how the project is assembled. The first two photo shows the wired up perf board next to the original board for reference. As I said earlier, you need to match the board size and locations of the display and buttons if you expect it to fit your housing.
My design ended up being a four-board stacked configuration. I used point-to-point wiring on the prototype to get it running and ring out the bugs. The Anarduino with it's piggyback RFM69-HW915 radio attaches to the bottom of the perf board. The "glue electronics" and input conditioning are done on the perf board and the OLED display attaches to the top of that board. I used a Pololu DC/DC converter to take the 9 volt battery voltage down to 5 volts for the Anarduio which provides 3.3 volts for the OLED.
I mentioned earlier that safety is a primary consideration in this design. The detail of the voltage probes merits some discussion. The (+) probe was built from a 3 inch #12 machine screw. It was bolted into the port where the laser was mounted on the IR thermometer. The shaft was covered with heat shrink and a red tip protector was used to cover the pointed tip when not in use. The exposed nut at the base of the probe will be covered with rubber sealant. It would take quite an effort to break the probe off it's mounting plate. The (-) probe attaches to a safety (shielded) banana jack, similar to the ones used on all multi-meters today. It fit nicely in the hole used for the IR temperature detector. This allows me to use a standard multi-meter probe with the safety jacket (i.e. no exposed metal on the plug). This protects the operator in case the negative probe pulls out of the jack while working around the battery. The jack allows me to easily use a longer (-) lead or an alligator clip if I'm measuring voltages referenced to a common ground.
Step 6: RF to PC
Now I have a working remote DC meter I need to find a way to get the data into a PC. Since the RFM69xx radio is not one of the built-in options in PCs, I needed to create an external "base station", or gateway.
One of my original objectives was to have a simple plug&play interface on the PC side. Luckily, Low Power Lab has just what I needed. It's called a Moteino-USB and comes with an attached RFM69xx radio and a built-in USB port. This device fits nicely inside a small Hammond 1591 project case.
Step 7: The End Results
The final results are shown here.
The first photo shows the original IR thermometer and the new remote DC voltmeter. The next photo shows the original LCD display on the thermometer and the OLED display on the voltmeter. Finally, the voltmeter is shown measuring a 9 volt battery and sending it's value to the program running on the PC. The display on the PC shows the cell number being measured along with the battery voltage and received signal strength indication (RSSI) from the remote voltmeter. This same information is displayed on the remote voltmeter.
The three push buttons under the display allow you to increment or decrement the cell number and save the sample run in the program.
Step 8: Epilogue
The title of this instructable is Beauty is only skin deep - Package your Project like a Pro, implying that you can stuff your ugly-duckling project into suitable housing for a professional look. And you can, as evidenced by the preceding slides.
But do you really want to leave it that way? I don't. For one thing, point-to-point wiring (aka "rats nest") is pretty fragile. It's easy to break a wire and very difficult to troubleshoot. For me, part of the challenge is getting my original idea from a concept to a finished product, something I'd be proud to "take the top off" and show off.
To do this right I needed a circuit board. Luckily there are a lot of options and a lot of help on the web to walk you through the process. I chose to work with Eagle from CadSoft because their demo is free and there are a lot of part libraries for it. It has a somewhat steep learning curve but once you understand it you'll have a lot of fun placing parts, moving them, and ripping up traces. It also has integrated schematic capture so it's easy to document your design.
Once you have your board designed there are a lot of board houses that cater to students and hobbyists. I chose to go with Advanced Circuits "Bare Bones" option because the boards are cheap and it's easy upgrade to a "real" board with silk screen and solder masks if I need to.
The photos on this slide show the new bare board next to the original IR board, followed by photos of the finished voltmeter ready to be installed in the IR housing.
I hope this instructable inspires you to take your idea from the breadboard all the way to a final product, even if you're the only end user of your creation. You'll learn valuable skills along the way that you'll be able to apply to your next inspiration.
So, is beauty only skin deep? Yes, it can be, but it doesn't have to be!