Last time I wrote, we talked about Beardon’s first foray into the world of hardware development. I went over the way the idea started and the early development phases it went through, as well as the decisions that were made regarding where to take it going forward. Now I’m going to walk through the later stages, and talk about the device itself a little more.
We decided it would be suitable to approach the project like we were designing a simple monitoring device, built with features that a more casual user might want. Perhaps the user would want to know if someone entered their room or house while they were away? A parent may want to be alerted of when their child gets home from school. Keeping those ideas in mind, we narrowed down the list of things someone might want a sensing unit like ours to accomplish.
We eventually narrowed it down to four sensing modules. We wanted the user to be able to check the temperature of the space around the device. That would allow for pet owners to check on the habitats of their lizards, birds, or other pets that need a heating source. So, I whipped up a thermometer-like sensor using a TMP-36, that records temperature in degrees C.
Retaining the light sensor was a good idea,we reasoned. It’d be easy to implement and a user might want to know if they left the lights on, or if someone entered a room in the house. Elderly parents getting out of bed and the device alerting their concerned son or daughter with a message? It might help put them at ease without having to go through the trouble of a phone call every day.
A microphone to detect when noise reaches a certain level, such as a party getting too loud or a baby crying, could also serve a use. I worked up a small subsystem comprised of a few capacitors and resistors, a dual op-amp, and an electret mic. The hand-drawn schematic, as well as some simulation runs of it, are below– the objective was to amp up the smaller sounds made by a footstep or speech and make them recognizable on a 5V scale, as opposed to a 100mV scale.
The last thing we decided to implement was a motion sensor. The standard most commercial devices use is a passive infrared sensor; it takes an IR ‘snapshot’ of the area it’s pointed at (usually a wide cone), and then takes another and compares the two for differences. If it detects that infrared radiation in its field of view has changed (such as the heat from a person walking by a wall, changing the reading from wall-temperature to body heat-temperature), it outputs high voltage.
Companies like Adafruit have prepackaged PIR sensors, like this one, that are plug-and-play, turning on when they detect motion. With some basic testing, it seemed to do what was required. So, we had all four modules at least up and running– sound, motion, light, and temperature.
From there, the question of packaging came under inspection. We wanted the device to be compact enough to fit discretely in a room corner, and be simple to set up and forget. That meant no ethernet cables, just a lightweight power adapter. We set a target goal of 3.5”x3.5”x2” in volume, with a standard 5V cell phone charger for an adapter.
So I sat down and started laying out a printed circuit board. Initial design concepts were on paper. We wanted it to be organized in a certain way; the temperature sensor should be as far as possible from the microcontroller, because the latter generates heat which can impact the temperature readings. The micro needed to be near a wall so it could be plugged in, since it was providing the power for the rest of the devices… Things like that impacted our layout. So, I whipped up a quick paper model of what we wanted the final product to look like.
After that I used the model as a base to visualize how much board space I would have, and then drew a rough concept sketch of how the components would fit. It was really basic and rough, but it definitely helped me transition to a digital layout later.
Then I chose a board design software to start laying things out. EAGLE is a fairly popular one among electrical engineers, so I decided to give it a try. It has a free version that allows for boards to be produced with dimensions 3”x3” or less, which was fine for the purposes of the project. It also has a lot of libraries pre-made for it, including one for the Particle Core. These libraries provide a symbol to represent a part, and more importantly, a physical footprint for when the schematic is ready to move on to board design.
First came laying out all the components in the schematic editor.
It worked similarly to most circuit CAD software I’d grown used to from college, like PSPice. Circuit elements represented by fairly standard icons, placed around and wired together and labeled. EAGLE, like most board designing software, provides an interface between the schematic and layout modes. Components placed in the schematic and linked to one another maintain those links when the mode is changed.
Once the board’s physical dimensions were laid out (in this case, a 3”x3” square), the system let me drag components around to where they needed to be and place them. After putting them all where they needed to be, I used the program’s built in auto-routing feature to draw traces between all of the connections. These traces are represented by lines that would be laid out in copper on a fiberglass board, forming connections between the circuit elements just like wires on a breadboard.
Once that was all set up, we went through a few options to get it fabricated, and eventually settled on OSH Park. They were pretty helpful with getting the files from EAGLE they needed and taking care of the rest for us. It was fairly priced and we got them faster than expected– I do recommend! If you plan to design a circuit that you’d keep around for a while, there’s no substitute for having a real PCB hold the components.
Once it came in, I was pretty eager to assemble it. That’s where most of the components from the original breadboarded layout went, and why we don’t have a breadboard model laying around anymore; they were cannibalized to put onto the new, professionally printed board. Assembly just took a couple hours, plus some time to fix bugs; the colored light in the center was a challenge to solder on without any metal unintentionally touching.
Then we had a prototype! All of the pieces were in order, it came together and looked nice, and fit into the little paper model I’d made neatly. I worked with Tyson to get a little html file going, something I could put up on a computer screen that would display the readings from all of the sensors. Then I started testing and improving on the design, and working to make it something people might want to buy in the future.
It’s actually grown into a pretty promising little project! We took it to Oklahoma State University’s Pitch and Poster competition, hosted by the Spears School of Business. RoomPing went head to head with thirty competitors, where we presented a posted and the prototype to a passing crowd. Judges picked the top ten out of the thirty one and offered them a chance to make a ninety second pitch, and we placed.
After that, we made our pitch. We talked about all the things we wanted our product to do and what it could do already, and we discussed its market potential, potential production costs, and possible price points.
They loved the idea. RoomPing placed third out of thirty one in the contest, and gained a little bit of traction. Some of the judges and fellow contestants asked for more information on the project going forward.
Right now, we’re in the midst of looking for places to take the project going forward. We’re in the market for manufacturing, for testing, for app development. I’m definitely keeping my ear to the ground for more options, so if you have any suggestions, please feel free to comment about them! I promise I’ll read it and give it some thought.
That about wraps it up for RoomPing’s development for the moment. Maybe we’ll have another post about it in the future. Right now, the project is on the backburner until we can find some more information about production. I hope the journey was engaging! If there’s anything you’re curious about, please feel free to ask. If I can answer, I will. And if there’s anything you’d like to hear more about in general, our work environment, our projects, I’d be happy to elaborate. Until next time, take care!