Trip Report: Maker Faire Detroit 2013

This marks our third time exhibiting at this Faire, and fourth time attending overall. Our project this time was the DIYPinball system, which we’ll get to in a bit. As with years previous, it was a great time but with some lessons to bear in mind in future.

Leaving Ontario

Unlike last year - when all the projects to be exhibited were ready before we left - we started Friday without a pinball machine ready to go. We would up spending nine hours on Friday getting the machine wired and done, only to turn on the power supply for the solenoids and have them all fire at once. This was a bad thing to have happen, considering that the solenoids aren’t rated for continuous use, and that we hadn’t issued the command to turn them on. We shut it down - quickly and with much yelling - and unplugged the solenoids.

Once they were out, we fired it up again, and noticed that:

• All but five of the solenoid LEDs came on, and
• The boards were smoking.

More yelling, more frenzied power cutting. We figured that the 2N3904 transistors that drive the solenoid MOSFETs had fried, as they’re only good for 40V, and we were putting closer to 50V across them. Given that it was already 9:00 PM and we had yet to leave Waterloo, we packed up parts for replacement, soldering gear and the pinball machine, and made our way to the Adoba Hotel in Deaborn. We arrived around 2am.

Lessons so far: Have the project ready to test before the day before the Faire, if only so that you know if it’ll blow up the first time you plug it in.

Upon arrival at the hotel, we discovered that I’d screwed up the hotel reservation, booking it for Saturday and Sunday nights, not Friday and Saturday.

Lesson: be more careful with hotel reservations.

With all that sorted, we got about four hours of sleep before we had to wake up, and be at the Faire so we could try and get the pinball machine working.

The next morning, at the Faire

We arrived for check in at around 8:30 AM, sans breakfast or working project. We were inside the museum this year instead of in a tent, which is better for comfort but not as good for exposure - a lot of people don’t seem to venture indoors unless there’s rain (which we got later that day).

Some frantic hot air rework in the booth to remove the 2N3904s and some resoldering led us to find… that we hadn’t fixed the problem (I should point out that all of this is happening while the Faire is in progress). I was about to replace the MOSFETs when I noticed that the transistors we bought for this were rated to 20V, when we’d specced 100V originally. Remember that the solenoid supply is 50V - running that through the 20V transistors lets out the magic smoke REAL quick. We gave up on getting the solenoids working, installed the boards, and tried to get some sort of demo going.

Further testing revealed that the switch inputs worked fine, but that none of the lamps were working. Our first thought was that the matrix scanning speed for the lamps - which was tested with LEDs - was far too fast for the lamps, and it was switching away before they lamp had a chance to turn on.

Slowing it down, however, didn’t fix it.

Even more testing revealed the true culprit: we’d wired the lamps wrong. The lamps had two obvious terminals, with a diode joining them. We’d assumed one was positive and the other negative; in fact, one was floating, basically just there to wire in the diode in place, for the old circuitry’s matrix scanning. The other terminal of the lamp was the bracket itself.

More soldering, and we had the lamps working. With that, we wrote some quick demo code to turn on a bunch of lamps from the left and right flipper buttons, and had ourselves a demo. This was about noon on the Saturday, at which point we donned lab coats and began the exhibition.

Lesson learned: basically, bring every tool you could possibly need to fix the thing.

I was surprised at the interest we got, in a way - I wasn’t expecting people to be that interested in pinball. I must have talked with at least a dozen people who were either rebuilding their own machine, or knew someone who was. The fact that we were using CAN for our system - an automotive networking technology - drew a lot of surprise and interest, which makes sense given the Detroit audience.

Other exhibits

I once again didn’t get a chance to see a lot of the other exhibits. The guys across from us, who were doing EMG stuff, had some VERY cool demos, and a very cool project. They were using electrodes to measure motor neuron activity, and in one demo were opening and closing a robotic gripper by opening and closing their own hand. The other demo was a text-to-speech system designed to let a family member who’d been in a major accident speak again. Sad that they needed it, but well-done stuff.

We had trouble missing the 501st Squadron this year, since we were right across from them. It was an unending parade of stormtroopers, rebel pilots, an amazing Lando Calrissian, Jedi and R2 units. I did not realize, though, that the members of the Great Lakes Garrison were also into other costumes, though it makes sense in hindsight. The Muppet Stormtroopers and the person in the Iron Man costume were particularly good.

The R2 builders were also good. One in particular, who’d been driving his R2D2 around by remote all weekend, did a particularly fine job with a few kids. As the Faire was closing on Sunday, a bunch of kids came across his R2, but didn’t see him. “I think it’s real!” the kids said, as the builder hid behind a pillar with his remote, playing along. The kids posed for pictures and were generally over the moon. Very cute.

Some of the other exhibits I saw were pretty cool as well - HackPGH had their system that turned their glass brick wall into an exterior-facing display, and someone brought a pumpkin carving bot that was neat - but as I said, I didn’t see tons of other stuff. Lesson for next year: more people, and schedule shifts in the booth so that people can see other stuff. I don’t think Mike saw much either, and Alan went a couple times to shoot photos but I don’t think he got all that far.

Entertainment and Refreshments

We’ve got this trip down to either a science or a bunch of closely-guarded traditions at this point. At some point - hopefully Friday but Saturday this year - we hit up the Walmart at Ford Road and Southfield and spend far less on beer than any Ontarian is used to. This year I’m happy to note that we didn’t make fools of ourselves laughing in the aisle at the price of beer in the US.

On Saturday night, we stop at the Five Guys in Southfield en route to the i3 Detroit party and enjoy some burgers. On Sunday, after cleaning up the booth we grab Jet’s Pizza and devour it in the car, taking a photo with the box to make the folks at home jealous.

We stayed later at the i3 party this year than I think we’ve ever stayed before, and had a blast in the process. Lots of very cool people - some Canadian, most not - building cool things with neat stuff to talk about it. I’d had a few at that point, so you’ll forgive me for being light on specifics. I do recall a tricopter with a POV camera transmitting to the LCD TV in the hackerspace that I was sure was going to get tangled in the extension cord hanging from the ceiling. I may have also called Alan names for perceived creeping on other people we were talking with, and I’m pretty sure I started singing the theme from The Polka Dot Door at one point.

I will say that staying out even later the second night was a lousy idea after getting in so late on the first night - we didn’t get back until about 3:00 AM, with two guys from HackPGH along for the ride. Their compatriots who drove were evidently unable to handle partying past 10:30 PM. At the time, I was dismissive of that idea; afterwards, I saw a bit of wisdom in that approach.

Overall verdict: the stuff that happened outside of Maker Faire was awesome.

Logistics

Mike’s Ford Explorer, when loaded with an AF-TOR pinball machine, is cramped. Alan was in the back seat and was definitely starved for space. We need to either figure out a different vehicle or make a smaller pinball machine for future exhibitions.

I’ll also note that we need a better solution for in-car audio in future. In order to play music from our phones, I brought along my Bluetooth speaker, which I’d purchased for listening to podcasts in the shower. It says on the box, “Surprisingly good audio quality”. It was $20 on eBay, shipped. It’s not sufficient for in car use. Also, if you stick its suction cup to your forehead, dance around a bit, and then try to pull it off quickly, you’ll get quite the bruise. Trust me on this one.

So, yes, the vehicle situation needs more consideration next time. It didn’t suck, but I think improvements could be made.

Conclusion

It was looking pretty grim for a bit there, especially after we blew up the solenoid MOSFETs. Overall, though, it was a great Faire with a lot of interest in our project. Our next step is to get the thing working for Maker Faire Ottawa, and hopefully have something even better to show next year at Detroit.

Photos