Chris Gordon : Alright, so I work for Intel Corporation. And what you are looking at here is our making for manufacturing table. These are two of more than 150 Galileo based proof of concepts that we have developed in our Intel factory network. What I am going to show you here is our executive dashboard. I work as part of a Data Visualization Engineering team. And so we are always looking for new and novel ways to be able to represent data. One of the personal frustrations that I have is that I spend sometimes hours a day just bouncing from one web report to another trying to pull all of these different indicators together to really understand what’s happening in my factory.
And what I’d really like is not to have to go chasing after my data, but for my data to really come to me on my own terms. So I am personally a little bit of a levied and I spend way too much time staring at a computer screen anyway. So what I wanted was really to be able to bring that data out of the PC and into something that was a little more palatable for me. So what we did is use our Intel Galileo board here to essentially act as a data aggregator. So that it goes out and pulls data from a wide variety of sources, in the case of air quality index or temperature from some public APIs or in case of the factory indicators, you see here from our internal company data systems.
So the Galileo goes out, collects all of that data real time and basically translates it into a pulse width modulation signal, so that we can take that digital output. And I will put it as an analog voltage into our meter interface circuit. So our meter interface circuit basically uses a latching op amp, so that the micro can just pulls the output voltage to the interface here and it will latch its output accordingly to actually do the heavy lifting of driving and holding those meters, so your micro is freed up to go off and do other tasks for you, it doesn’t have to burn cycles just holding the meters in place.
Now today our monkey here is tied to the Intel stock price, so that every time the stock has gone up since the previous refresh cycle, he will clap for you, but if the stock price has gone down since the previous refresh, then his eyes bug out, he bares his teeth and starts screeching. So you know things are headed in the wrong direction.
T i m: What is the good catch percentage?
Chris Good catch percent is actually a factory system that we use to record potential safety issues. And so one of the things we are interested in is what percentage of our population is actually participating in this good catch system and logging these safety good catches so that we can proactively address safety considerations in our factory.
Tim: Talk about the details of looking out to old analog meters with digital wave, so where did you get these meters and how do you actually interface with it?
Chris So it’s actually getting harder to find analog meters here since – as you know the world has gone digital on us. But a lot of these were mail ordered off of different sites, mostly from China. And as I mentioned earlier, what we are basically doing is using pulse width modulation to kind of bridge that digital to analog gap there, so that what we do is use a scale from 0 to 255 that represents the percentage of full scale deflection that you are interested in on your meter. And then the meter interface circuit just reads that as an analog voltage coming in and matches that accordingly to hold your meter in position. As you can see, all of these were hand scaled, we lovingly scaled each one of these meters by hand with pencil and paper...
Tim: And you frame them as well?
Chris Absolutely. And absolutely no expense was spared, everything was about attention to detail with this project. So everything you see here is antique oak and real brass, the monkey itself is an over 50-year-old antique symbol monkey from the early 1960s, no reproductions, even the wiring here is done in real cloth wrapped electrical tape for authenticity.
Tim: Of a toaster – reminds me of a toaster.
Chris Hopefully a safe toaster.
Tim: And I think I gave it away on that basis. So you work for Intel?
Chris I do.
Tim: And you also have accessed a lot of factory sensors and data and work bench time I am sure.
Chris Oh yes.
Tim: If someone wants to do this with a little bit less experimentation, is it packet accessible thing to do, or at this point are we still looking at monkeys and toaster wire and frames to put them in?
Chris This is super easy for anyone even with limited experience to be able to do. If you have ever done any kind of Arduino development, you are already eight tenths of the way there. This whole project from start to finish was completed in less than six weeks and probably four of those six weeks were actually spent on building boxes, getting brass hardware and actually doing the physical assembly putting this together the way we envisioned it.
Tim: Do you use any of this kind of technology of sensing at your house, at your home?
Chris I do. So maybe it’s a bit of a misnomer to say I am a luddite. I am really fascinated with the technology, but I don’t like to interface with it. I don’t like looking at wires all over my computer. I don’t want a big ugly computer block sitting in my living room. And so everything I do is to try to hide the technology so that I get the advantages of it without having to look at plastic boxes sitting all over my living room.
Tim: Like a nice tangible output?
Chris Absolutely. So this is something that we have actually developed to be flexible, so that I can just as easily take this home and repurpose this for whatever metrics I might want to track around my house. I can keep a chores remaining meter or paternal displeasure index or whatever I might like to track at home and set this whole thing up in about 15 minutes to pull from whatever data source I am interested in.
Tim: PDI does sound like an interesting index, but I don’t know what sensors will be the best for that?
Chris Child volume is at least one input.