Solder Sesh #31 Transcript: Dave Cox

Carrie and Dave talk about his home-made DMX light system, entering the STEM space, industrial knitting machines, unicycling across Canada, failure and typing again, and designing/using aluminum printed circuit boards!
Check out his Twitter @unicycledave
Follow his unicycle adventures on his website: http://www.unicycledave.com/

Full stream:

(0:56 – 6:44) DMX Lights
Carrie: I have Dave with me. He is @UnicycleDave on Twitter, and he has unicycled across Canada, which is pretty darn impressive. He also likes hardware hacking and he's got some lights, which I am hoping he'll tell us more about, because they're DMX lights. I don't know anything about that system; I know that there are protocols and systems involved for theater and stage lights. I would love to learn more about that!
Dave: The lighting for this part of my room sucks so I went on this over-engineering journey of building some lights and that started probably a year and a half ago. I bought these little smartphone size, bi-color, lighting things. They're fantastic! I love them a lot, but they die, you have to charge them, they don't run off of power, you can't control them remotely, and they're not as bright as I wanted them to be. I did a little bit of shopping on Digi-Key, and I found these 95 CRI modules in two colors. I did some research and learned about aluminum PCBs. Something else that tied into the project is I had sliders that came from one of those high-end film mixing boards that have a servo controller on them. I thought, what if I built a lighting board from my desk? It just led to vastly over-engineering this problem.
Carrie: What is the system and protocol because I hear that all the time, but I've never really worked with stage lighting or theater lighting.
Dave: So, DMX is RS485. Which is a serial protocol that’s also used for industrial control and stuff like that. It’s just RS485 at 250,000kBd and it’s called DMX 512 because you send out 512 different settings, one after the other. You just repeat that over and over again, then the lights just listen for the start of the packet, find where they are in the address space, and go those are my values. They translate that into whatever they're supposed to do. The light cannot communicate back to the board in most setups. The board is like, do this and the lights do it.
Carrie: Just dumb receivers basically. For anybody listening, who's not familiar with RS485, you might already be familiar with UARTS or maybe even RS232 protocol or serial communications via those old school DB9 dongles that used to plug into the back of your computer, which USB has pretty much made obsolete. RS485 is a serial protocol, there tends to be a transmit and receive. For RS232, there's a dedicated transmit line, a dedicated receive line, and usually have ground. You can have other signals for handshaking, but they're not absolutely necessary. It’s very similar to transmit and receive UART communications that come from a micro-processor, but just level shifted and inverted a little bit. In order to hook up a processor directly to a computer, assuming that it has an old-school DB9 port, you would usually still have to have a chip in between that did a little bit of level shifting and converting. RS485 is similar, but it's differential. Basically, you can run it for longer lengths, and it is less sensitive to noise. I imagine it would be very good for theater applications where you might be running hundreds and hundreds of feet between your lights.

(17:34 – 22:55) Dave’s Work / Knitting Machines
Dave: I was going to say a little bit about where I work at the moment which deals with industrial circular knitting machines, which is really, really cool!
Carrie: Yes, I want to hear all about that. I’m a hand knitter so I think it’s really cool.
Dave: Okay, cool. So, this is a vastly different process, but it's still worth talking about. I work for Sheertex, a pantyhose manufacturer in Montreal that makes ultra-strong pantyhose. They have a special fiber that they use to make a knit that's stronger than most pantyhose. They work, they do what they say, they're a legit product! They can be quite expensive, so they're not for everyone, but I think the technology that goes into them is super cool. A lot of it's under NDA, but basically, the way the knitting machines work is they have 400 odd needles in them. It's a cylinder that spins around in a circle with a light vacuum pulling down the sock and it's kind of like a corking, I guess, is the name for it. There’re little fingers that are pneumatically actuated that actually dropped the yarn into the needles and pull it out based on the pattern that someone has programmed into the machine. We have so many of them and they speak RS485 so one of the things I had to do as part of the engineering work is we were trying to get them into our operations software. Turns out that you can talk serial to them if you say the right things in the right way. I got to build these little boards that actually do that, which was really cool. I don't have any photos of the machines, but it's worth looking up pantyhose knitting machines. You'll get videos that I don't even know, it's just one of these really cool processes. They're super intricate; it takes a long time to even learn how to program or use them because there’s a lot going on. You can do patterns, you can change the stitch height, so the taller your stitches, the more room you'll have in the leg for whatever thing you're doing.
Carrie: The looser the gauge, the more flexible it is.
Dave: Yeah. But the needles are a constant, they’re the heart of the machine. You can't change the number of needles and you can't change their size, they just are what they are.
Carrie: Right, so instead of changing the number of stitches in the circumference of a sock, they maybe change the yarn weight and the tension of each stitch.
Dave: Yeah, exactly!
Carrie: So, are these machines sort of specifically made for pantyhose or can they also do socks and hats and things like that?
Dave: Versions of the machines can do socks and I guess hats, I never even thought about that. Normally the yarn is quite thin.
Carrie: Yeah, that's got to be super tiny, tiny thread.
Dave: The yarns are wild, they float in the air, it's crazy!
Carrie: That must be nuts to try to work with yarn like that. That also is elastic, knitting yarn that’s fully elastic tension has got to be such an issue, consistency of tension as they're unwinding it from whatever spools or cones or whatever it comes on.
Dave: Yeah, they have a thing called a creel, which is all of the pipes and stuff that support the yarn packages that are above the machine. Then there's a bunch of feeders that sit in a ring above the needles that actually manage exactly what you're saying, the tension. When you get into stretchy Lycra and all this other stuff, the electronics are wild. I've taken apart some of these devices and it's so cool. They've got brushless DC motors and load cells and other stuff that they use on the fly to detect what the tension is and then adjust how much yarn is being fed into the machine. You also have to deal with temperature and humidity. I think a lot of industrial processes, which is one of the things I like about my job is kind of getting to learn more about that, like the intricacies of what makes those things work and what makes them not work and how to control them more accurately. It's a lot of fun!

(24:10 – 28:11) College/STEM
Carrie: How did you get into STEM and working on knitting machines?
Dave: One of the things I've always just had an interest in is electronics. Ever since I was a kid, I was dumpster diving for computers with my friends where I grew up it was just kind of my jam. I actually don't and never really have had much of a background in doing engineering on a more professional level. I absolutely have imposter syndrome about not being a “real engineer”. As I put LEDs on a circuit board that I designed myself – that works, got dammit!
Carrie: A college education is but one path and it is but four years of one's life.
Dave: This is super the thing and I'm slowly getting over it, but my background is actually in film. I worked on film sets for a number of years as an IAA certified data imaging technician and I used to do video backup. When the first digital cinema cameras came out, I was one of the first technicians that was backing up the files off of those cameras as my day job. It was great! The hours were terrible, but film sets were a good time.
Carrie: That’s the thing, right? You either have no project and nothing to do, or it's 24/7.
Dave: Precisely. My uncle introduced me to Linux at a young age and taught me how to use VI when I was 13.
Carrie: So, was he the one who mentored you with electronics and things in general?
Dave: It was more just the spark of interest, then it kind of ran from there with friend groups and exposure to other folks in my immediate community. There's a guy who ran an IRC (internet relay chat) server in the town where I grew up where you could get yelled at for writing C code if you wanted to. By that, I mean you would write C code, but it was the old school people were mean to you on IRC if didn’t know what you're doing. I don’t recommend it.
Carrie: I’ve never been on IRC, but I've heard about this.
Dave: Yeah, honestly, I'm glad that we're starting to move on from this kind of jam, but it was quite formative for me. I was always more drawn to hardware, but never had the resources for it.
Carrie: Yeah, it was really expensive, until recently, to make your own circuit boards or program your own circuit boards because the software was all very expensive initially.
Dave: Yeah, and now you can go grab a copy of KiCad and Fusion 360 for free (editor's note: sadly, Fusion 360 is not free anymore). I built my own 3D printer a couple of years ago, now I have a Prusa because I like myself and my time, but it was a very formative experience. I think the long story of that is my hobby experiences were very useful when they (Sheertex) were looking for someone to do prototype hardware engineering, machine interfacing, and data gathering stuff at Sheertex.

(48:20 – 56:04) Unicycling Across Canada
Carrie: Tell me about how you planned your trip across Canada. What was the remotest area that you went through and how long did it take?
Dave: I planned it just by… I don't know how to describe it. Sometimes there's stuff that you just have to do. I'm not sure that I ever really had a choice on the whole, you’re just like I'm going to do this. I started to obsess about it, and I really am a planner at heart. I think in general I really like knowing the details of how I'm going to do something. I had a pretty chill day job; at the time, I was working at Technicolor. I was just researching what the routes I could take were and how I would carry the stuff I would need to carry.
Carrie: Yeah, how do you carry stuff that you need to carry?
Dave: Carrying stuff is really interesting and everyone does it differently! One of the things I love about touring cycling is there's no real one size fits all for how to store your gear on your bike or whatever you're doing. I think it's so cool because when you're touring, you find other people and they set it all up differently. The way that I did it was I bought off-the-shelf bike racks, drilled holes in my frame, and put one facing forwards and one facing backwards. I bought 20L dry bags; big, blue dry bags from Mountain Equipment co-op.
Carrie: Because it actually rains in Canada, not like California where it does not rain, ever.
Dave: The weather was interesting at times, for sure. So, I want them to be waterproof and I bought replacement equipment for attaching panniers to the racks. I put a piece of Coroplast inside and then screwed that to the side because the problem with panniers for unicycling is that you get heel strikes because your cranks are so short.
Carrie: How is there even room for a rack on a unicycle?
Dave: I might have to explain a little here. I have a hole for a rack that goes here. The actual panniers would only hang down just a little bit. You get the clearance, but they have to be custom, and they have to be modified.
Carrie: Okay. That makes sense.
Dave: I was using a 36” tire, so there's a little more room for stuff, but I did carry a tent, drone, GH5 camera, tripod, sleeping bag – all this stuff. I had way too much crap with me. It was a lot of fun! It took me 111 days to do the trip and it was 9,250km; I have no idea what that is in miles. (5747.7mi)
Carrie: It’s a long way, it’s across Canada, right? It's across north America and not at the skinniest point.
Dave: Yeah, and I took a lot of side trips because I was like, well, I'm here and I'm in good shape. I would go to see friends; I would go to see stuff that was worth seeing that was on the way. It was really fun.
Carrie: What was the typical daily mileage for you?
Dave: So, my brain works in kilometers, but-
Carrie: We'll try to translate here.
Dave: You’ll just have to translate for me, or I can get out an app. But I would do a 100km a day roughly. About an hour’s worth of driving, I would do in one day. I don’t know how many miles that is. (62mi)
Carrie: Well, it depends upon how fast you're going. Are we talking freeway driving or?
Dave: Absolutely freeway driving. The reason why I didn't do it on a bike, which makes way more sense and is a lot easier, is you meet so many more people when you're that weirdo out on the road. People are like, what are you doing? Why are you here? How did you get here? Where are you going? On a daily basis, several people would pull over and they'd offer me snacks and places to stay and just kind of concerned people. I've received two bibles while riding my unicycle, I guess I looked lost a little bit.
Carrie: Right, this guy definitely needs God.
Dave:  Yeah, if anyone needs a bible it’s definitely that guy. So, it's really cool because you kind of have that gimmick that opens up a couple more doors than you might, if you're more normal biker tourist type.

(1:00:42 – 1:08:55) Failure and Trying Again
Dave: It was definitely a long time coming. I had tried to Unicycle across Canada once 10 years previous, but that is a long and torturous tale. I didn't get super far; I made it to Alberta and then I went home.
Carrie: But that is also good for people to know. People tend to hear only about the successes and sometimes, especially when you're trying to do a big thing that you've never done before, it might take you a couple of tries to be successful and that's okay. That's part of the whole thing.
Dave: Yeah, that's a very good point. I don't often think about that, but failure is a super important part of success. I will also say through the wisdom of having failed a number of times in my life, sometimes it's important to fail and just walk away like you're never going to pick it back up again. Maybe in a year or two or whatever, depending on the project, the time will come back around where it's right to complete the thing that you wanted to do. Time is such a useful tool for overcoming problems.
Carrie: In knitting, we call that hibernating; projects are hibernating. When they’re giving you too much trouble or you don't like them, they're being a pain in the ass for some reason. Okay, putting this aside, hibernating now.
Dave: Yeah, and then when you come back to it with a fresh brain, after your subconscious has had some time to chew on it. There's something really cool about that. I like that on the larger scale, but also on the day-to-day scale where you're like, “Screw this, I'm out of here.” Then the next morning you're like, “That was easy, I should've gone to bed first.” I give up so easily now and go to bed. It's move number one for me; I’m like, “Nope, I'm going to go write some emails and go to bed.”
Carrie: I like it! The wisdom we've learned along the way, just give up and go to bed. It'll be fine tomorrow.
Dave: It's so true.
Carrie: Yes, Bob (from the livestream chat) is saying that very few people follow a strictly linear path through life, and many go through multiple iterations. I feel like that is so true and yet, it is so weird that sometimes we still feel like failures about not having this direct linear path or it still feels like the expectation somehow or the thing that we should be doing or the thing that we should be achieving. Why is there that narrative?
Dave: That's so true. One of the things that I aim for in my life overall, and I feel like I'm achieving it by and large is, setting my life up to have as many interesting opportunities come my way as possible by just saying yes to the weird stuff and going with it. I also recognize that there's a lot of privilege in that as well. I don't say that without qualification, but it's really cool. Sometimes it's worth it to take the thing that is interesting and just go for it because it leads to other opportunities that you might never have been able to see coming or anticipate. I don't know how true that is for everybody, but it's worked for me.
Carrie: I absolutely agree. Also, what you said about privilege is really true because if you're in a position where it's a daily marathon or a struggle just to meet basic minimum needs then oftentimes you don't have the luxury of doing the weird things. Or you're not in the head space where it sounds like a good idea or that you have the energy to take advantage of it.
Dave: Exactly, I just need to be really clear about that. Because I know not everyone has the ability to quit their job and take this other weird job that sounds like it'll be cool.
Carrie: But we can work towards making a place where that is possible, where everybody does have those opportunities.
Dave: Yeah, exactly, wouldn't that be great. It's what landed me learning about industrial controls and stuff like that, because I had gone and explored a lot of this stuff on my own. I don't have a traditional engineer background, like I was saying earlier, so there’s a lot of imposter syndrome. Being able to actually do stuff professionally really, really helps.
Carrie: I think it's funny too. The whole having a traditional engineering background because even though, yes, I have an engineering degree and you could say that I have a traditional engineering background, I did not take classes and what I did in college is not what I'm doing now. Yes, it's set me up for some things in certain ways, it definitely opened doors and gave me opportunities for sure, but the classes that I took in college were not what I'm doing now, at all. Engineering is such a huge, huge topic, right? I mean, you don't learn everything you need to know in college. Nobody could. You're always going to be continuing to learn and you're going to be learning on the job and learning from other people and maybe learning through your hobbies, too. I think that there's this real fallacy of, “I learned this in school and therefore, it is somehow more important.”
Dave: Yeah, totally. I'm getting over it, but it was a long journey thinking I can't do electrical engineering. What if there's stuff that I didn't learn in the magic sauce? Like the magic sauce class, they don't tell you on the internet.
Carrie: Most of the magic sauce stuff was stuff that the profs had learned by experience. Right? The really interesting gems of information that I picked up in college was when professors were telling you about these weird things that there was no way of learning except by actually just getting out there and doing them.

(1:22:05 – 1:28:00) Aluminum PCBs
Dave: This is really interesting trying to solder to an aluminum PCB!
Carrie: Yeah, tell me about that. I'm curious – Where did you have it made? How much did it cost? Was it really expensive compared to a normal PCB?
Dave: I learned so much making these. This was done with a credit with SeeedStudio and one of the options they had was for an aluminum PCB. For the least expensive type of these, you can't do vias. The original version of this board had these KiCad 2-pin connector layouts. Obviously, that doesn't work. I got a message back and they were like, you can't do that. So that's why I'm doing this really silly wiring. It's really interesting because the heat sinking is quite good on the board, soldering is really difficult because the minute that your iron leaves, it just hardens right away.
Carrie: Do you have an idea of how much more than a normal circuit board the cost would have been if you hadn't had the voucher or anything?
Dave: Yeah, I think it was between $50 and $75 for 10 boards of this size. The quoting is pretty good on their website; JLPCB may not do it, but Seeed definitely does it. There's one other place that will also do aluminum boards that I forget the name of, but they will give you a quote if you upload your Gerbers. It's worth checking, but the big thing is making sure that it's a single-sided board.
Carrie: No electroplating aluminum.
Dave: Yeah and I think if you pay a fair bit extra you can do it, but the process is a lot more complicated.