Embedded - 171: Perfectly Good Being Square and Green

Episode Date: October 12, 2016

Saar Drimer of Boldport (@boldport) spoke with us about the crossover of art to electronics and building a business around it.  Monthly, the Boldport Club ships aesthetically-pleasing electronics kits. We discussed past projects include The Lady and Touchy on the show. The seahorse board is on the blog. Micah Scott (@scanlime) has entrancing videos of putting together the first club project (Pease) and second one (Superhero). Saar uses PCBMode to create his circuits. He also wrote the tool. It is open source. Cratejoy is used for the sales and shipping logistics. 

Welcome to Embedded. I'm Eliseo White, alongside Christopher White. This week, we are going to talk about beautiful boards with Sar Drimmer of Boldport, a hardware hacker with entrepreneurial tendencies. Hi, Sar. Thanks for joining us today. Hi. Thanks for having me. I have heard a lot about you from Micah and I know you were on the Amp Hour in February, but things have changed since then and I have many different questions than what they ask. So can we start by asking you to tell us about yourself? Sure. Sure, my name is Saar and I usually go by engineer, it makes it simple. So I started my education at UC Santa Cruz, very close to where you are.

I did a bachelor's in computer engineering then I worked for Xilinx for a few years and then I came to Cambridge in England to get a PhD in the security research group. I was working on FPGA and volatile FPGA and banking security. Then I finished that in 2009 and I started Boltport to do online FPGA flow management for FPGA projects that lasted for a year until I learned a lot about the EDA business and decided to stop it.

And then I worked for a little bit and then I started consulting. Worked, I mean, full-time for a company. Then I did some consulting and then I started doing what I'm doing now. I started by writing PCB mode, which is the software that I use for doing all my circuits. And so that's what we're going to talk about. So I'll stop there. We can go into detail.

Okay. So before we do that, we usually do a lightning round so people get an idea just of your fast reaction time. However, it's not time. And while we ask for a short answer, and then we aren't supposed to ask you why or clarify or anything, that never happens. So let's try this.

All right. I'm doing this list? Yeah. All right. Science, technology, engineering, or math? All of them. I don't have to choose one specifically,

do I? That would be very difficult. Any answer is a fine answer. Yeah, I'd say all of them. Whatever works. Okay, I have an easy one for you. Hacking or making? All right. I don't like either.

That's a good answer. I don't like either. Engineering? Tinkering? Well, I like making, but in the dictionary definition, not in the commercial, now commercial definition. In fact, I would prefer hobbyist to maker. What makers used to be called 10 years ago was a hobbyist. And now it's a maker. You're a maker um it doesn't make sense to me because you you you

you're making things but now you you have all this kind of commercial baggage and uh marketing baggage associated with that word and it's similar to hacker um so i i don't i wouldn't i wouldn't call myself either of those um i'm just creating things that I find interesting. And in some ways, I want to disassociate myself with the marketing and hype that comes with these labels. It does feel like being a hobbyist has been branded from the outside. Let's call it something else so we can assign it to a movement and we can make money off of it yeah exactly and movement again this movement is is um is somebody's driving this um and i don't like to find interesting is that hobbyists used to be

uh didn't count right so if you did something for the heart for hobby hobbyists nobody if you wanted to raise money or you it's like oh hobbyists they don't pay they don't have any money they don't they're not it's not But now, if you're making stuff for makers, it's something that has an appeal. And I don't think, you know, not a lot have changed as to the mechanics of creating things. The tools might have changed, but I don't think the desires have changed. But now a lot more people come to it because of that hype machine. I don't know. Is that good? Is that bad? I'm not sure, but I don't.

I'm somewhat allergic to these labels. I like it. So I just create stuff. All right. We're going to stop lightning round because the stuff you create is awesome. And I want to start with seahorses. Seahorses, seahorses. Seahorses, yeah.

All right. Go ahead. Seahorses. You have a board in the shape of a seahorse. Yeah. Why? How? When You have a board in the shape of a seahorse. Yeah. Why? How? When? What? All of the questions. Okay. Yeah, this one has answers. Some of the stuff that I do just don't have very good stories.

But there's a company called Embicosum in the UK, and they're very supportive of open source and open source hardware. They do compiler optimization for embedded applications mostly. And I've done a board for them called the Cuttlefish, which is an Arduino, a bare-bones Arduino board that is based on something called the Shrimp, which is basically the ATmega328 with the minimum amount of components on a breadboard.

So I made a board for that. And called the cuttlefish a small board and then because I wanted another board with some more functionality so and wanted to keep the theme of

because the shrimp the cuttlefish so I looked for something that I could draw that had a good definition. And I came up with, in doing the research of different ocean animals, I came up with, I figured that the seahorse could fit well. Because you need to consider what you could do with the outline and what you could do with the silkscreen and make it work.

And so I drew up that Seahorse. And there's some functionality that is from the Shrimp, the ATmega and the capacitors and reset circuitry and all that and the push button. And then I've added these two auxiliary boards. One is for LED and one is for an IR LED. And then there's a photo transistor on the on the board itself and the idea is that people could could saw to use conductive thread to connect those two auxiliary boards and put them on their clothes and play um play with with infrared and infrared is interesting because it's it's very simple to

to make it's only two components, simple to interface to a controller, but equally important, fairly simple to explain. So if somebody's running a workshop and Numbicosm is running a workshop with teenagers and people at hackerspaces, it's a really complete tutorial, I suppose. You build the board, you program,

and then you could also understand what's going on with a relatively simple diagram, and you could play. If you're not interested in the building, you might still be interested in the play part of it. So what I've learned is that there's a whole spectrum of interests of people. Some people like the building, some people like the result, some people like just the way it looks.

So I think that if a project can provide to those interests, one project can provide to those interests one project can provide to to several of those interests it's uh it's something i call it something interesting we're so used to seeing boards that are rectangles or if the rectangle doesn't fit in the enclosure for some reason it's a it's a rectangle with a cut off. And your boards are not like that. And it does draw more attention to them because they are so, I want to say unique, but of course they're not unique.

Do you think we should be doing more of this all over? Or is this sort of a special subset? No, I think that it makes sense in some cases, depending on what you want to achieve. If you want to draw attention to your board, then something that is visually appealing makes sense. Boards that are never going to be seen, they don't have to be visually appealing.

It's nice if they are. Some products, when the people take them apart and discover something nice, that's something that generates a little bit of a buzz around the product. No, I think, you know, I'm attracted to this kind of thing. But it's not something that I think necessarily can apply to every product out there. Some things are perfectly good, being square and green. I do have this conversation a lot when I talk to companies

and when I visit booths at trade shows and you see that all companies will have demonstration boards and they're all square and green or blue or some color. And I explained to them that if they want to attract people that are walking past, then something interesting could do that and could get a conversation going. So it really depends on what you want to achieve. Another example is education.

If you're trying to get people, young people or any age, interested in electronics, then something that is more pleasant and approachable is more effective. And in contrast to some things like an Arduino or Raspberry Pi, which are technically effective, but they're not, I don't think they're appealing. They're not effective at drawing, having that aspect of drawing people into in towards them if if you will to to to want to start engaging do you find that once you've started designing something or building something in a in a stranger shape or form factor that you're more free to make connections that you

may not have made before does this make any sense like oh Like, oh, if I make the PCB this shape, then I can do these traces in this way that aren't just artistic, but also solve a problem that maybe hadn't been considered? No, I don't have an example where I could say that the way I do things solved some problem, some serious problem. I do think about what I do is I try to find projects or do projects where the form contributes to the function in some way. So there are things like stretchable circuits that you need meandering paths, and that's the form contributes to the function.

And then things like education that I mentioned before, that if the function is to educate, then the for matters. But there is something in the way I do things that is different that gives me a slightly different approach than what you would get normally. And I'll explain. Normally, you would have an EDA tool or a PCB design tool. You'd have a schematic entry, a schematic capture that generates a netlist, and then that goes into the layout.

So this is like a KiCad or Eagle or Altium. Yeah, they all use the same flow. Schematic, net list, and layout. Now, the layout captures the physical. So the way I look at it is the layout contains all the information, and the schematic is a subset of the layout. But the problem is with the traditional flow is that you can't, but it never quite works to make changes in the layout that would then be what's called back annotated to the schematic. It usually goes in one direction more smoothly.

And traditionally, that's how it went so you have to you have to see a problem in the layout to fix it in the in the schematic and then regenerate the net list or let the tools uh so propagate that change but because i my tool pcb mode doesn't it doesn't have a schematic or or net list i i do everything by eye so there's no drc the design rule checking that to run it's it's just um it's just the shapes and i just connect them and um i don't know if we're going to get to pcb mode but just um i use i use inkscape which is a an open source uh sv editor, as my GUI. So I didn't want to reinvent a GUI.

So PCB mode is essentially a wrapper around source files, Inkscape. So I can use a proper drawing tool to draw all the shapes. Most of them come from those JSON input files, but the routing and so on, I do it in Inkscape. And then when I say when I need to rotate a component or I need to put it in some

sort of a weird arrangement, then I don't need to deal with that back annotation. I just draw whatever I want. And I don't need to get that back back annotation i just um i just i just draw whatever i want and i don't need to get that back into the schematic and i think that um and there's no there's no net list that i need to fix so if i all of a sudden want to connect the in in a resistor if i want to connect to one pad instead of the other of Of course, they're symmetric electrically,

but if you had a netlist, it would connect to one of the pins. But if I want to do to the other one, I just connect to that. So that gives me a bit more freedom. Well, it's a bit more freedom, but it's a lot more like a gaping chasm of errors for the rest of us. Chasm? Well, Saar knows how to do this because he's got lots of experience laying out boards. But if I used his tool to lay out a board, it would probably end up

not working because I need those design rules checking. I need the help it provides and I need to be able to turn my components around and have their connections stick because I don't know what I'm doing. And Saar really knows what he's doing and so he can make these beautiful traces that are nonlinear and beautiful. I guess I already said beautiful, but that's okay. We're just going to keep using that word over and over again as this podcast goes on.

But if I do that, the chances are I'm going to hit another piece of copper and short power to ground and it's all going to go very badly. It's true and it happens to me. That's reassuring. more so in the in the past and but you know it's just i i have to i have to do a lot of uh manual checks and visual checks and hope and preferably have someone else look look it. The thing is that my circuits, the stuff that I do for the club, are not that big.

So they're only two layers and most of my projects are in two layers and it's manageable for me. But you're right, if I have a track that overlaps, nothing tells me that that's not going to work. If I fabricate it, it will be there. So yeah, it is not a process that can, at the moment, in the state that PCB mode is in,

that people without experience or confidence in what they're doing can easily use. On the other hand, the boards end up being worth it. I mean, it's much more... So it goes beautiful again well or handcrafted or you know aesthetically pleasing sure yeah so maybe that's a trade-off because the kikad and eagle and altium and all those tools, many of them were born decades ago. And they're sort of continuously patched up

to more recent advancements. So they weren't built to do this, to do these kind of things. But they should be. Should they be? They weren't built to do this, to do these kind of things. Even sort of curved. But they should be. Should they be? Yes, because I think about the BB-8's board, which is a circle. Right. And I think about the structuralness of the Pearson quadcopter board with its weird X shape that's curvy.

But all of its circuits are straight. You think about antenna for BLE and how they're usually on the board. If we could make these nicer, if we could do the Bezier curves instead of these straight lines, there is an opportunity for other things. So the way I see it is that there are, if Eagle, for example, has these ULPs and this kind of community scripts, and every tool will have some sort of a script that could take a graphic and put it on the board or to hack together something. So if you really, really, if you're dedicated to doing something, you know, not too big, you could do that. You could do what I do, some of what I do

with my tool in, say, KiCad. But the way I look at it is that you're trying to do something with the wrong tool. So those tools are arm twisted into into doing that and i when i first started thinking about doing this i looked at i knew that nothing really would get nothing out there would give me the kind of freedom that i wanted and it's not only about the shapes it's where you put them so I wanted to put shapes in copper and in silkscreen and on solder mask

and wanted to have control. And I also wanted to have accuracy. So part of what I, in the past when I was using Orcad, for example, I found myself zooming all the way in to get to the few decimal points to play something and zooming all the way out. And that was very tedious.

So when I designed the architecture for PCB mode, I decided that it's going to be text files and that I could type in coordinates to six, seven, eight significant digits if I wanted to. So a lot of the design is actually done in a text editor where I define the shapes for the footprints, for example, and the shape for the outline of the board, I cut and paste the SVG path, which most of the time comprises of Bezier curves, into that text file. So there's a lot of free hand in this, but there's also a lot of, there's a lot of accuracy because I can, I can move things in a very accurate way. When I do move things, mostly it's, it's through the text file. And the other aspect of it is, is that I feel confident in what I do because I know what I'm doing.

And I see sometimes when I come up with boards and they have these weird tracks, people say, you know, why isn't it... There's high-frequency issues here because it's not a straight track and it can cause a whole bunch of issues but um it's true i i do so it's not well it's it's it's true that under some circumstances at high frequency um with high frequency boards these can cause issues but when i do it i know that there's no high frequency components in there then i could do that i could do that. I could play with this and that's what I do.

If I can play with a track that is within the theme of the board, then I do that. Because it's important to me that every piece of the board is part of that concept that I'm going for. So I would use pads that look like fish, for example, the seahorse board, and tracks that might create shapes. And I know that that wouldn't affect the functionality of the board.

But if you're strictly looking at it from what to do and what not to do, the sort of the rules, rules of thumb, then yeah, what I'm doing seems wrong. Yes, it seems wrong. It seems advanced. I think it's what we, I mean, because sometimes when you watch me program, it doesn't look right the first time. And yet it's because I'm avoiding the first three mistakes that you'll need and I still want it to be nice. It's like optimizing, optimizing your, I mean, you're going in and hand editing the smallest possible things. Yeah. And that optimization leads to better craftsmanship. Right, yeah. So it's interesting because I'm an engineer and I came from a technical background and there's a sense that in engineering,

everything has to work the first time. Not for software engineers. No, of course the first time we're worried. That's very stressful. I must have loaded the code. Well, I came from sort of an environment in hardware. Because it's expensive, right? Everything needs to, you have to nail it the first time.

And then I started working with product designers in London. Very talented and I've learned a lot. And coming from Cambridge, which has this kind of culture of very technical culture to London, has more of a design, creative culture. It really opened my eyes to the iterative process that I wasn't quite appreciative of. In design, you prototype constantly and you iterate constantly and you try things constantly. And I realized that things don't have to work the first time. You create prototypes and that's part of the process.

You don't feel bad about it. And as long as you get something reasonable out at the end, that's worth that process. And I think a lot of people, that part is invisible to them. So I would say that there's probably, for every project, there's a few months at least of things just happening in my head, thinking about, randomly about, you know, this is kind of where I want to get to. How can I do it? And so it'll sort of ferment there.

And then there's a fairly long process of just drawing on a piece of paper in my notebook, concepts. And then at a relatively late stage, I go to actually put things in a digital form. So when I go to PCB mode to design, I already know more or less what it's going to look like. And I've already tentatively identified the components that I'm going to use and research them. Because everything feeds into one another. And again, in engineering, we have this thing where the functionality is more important than the form. So the plastic around the electronics is secondary, right?

It has to function. And there's these points in time where there's a clash between the enclosure and the industrial designers and the electronics, and this doesn't match, and oh, we need to move this knob here, and so on. But I look at it that they're equally important, and when they feed off each other, you get something that is nice. I can't really claim that is better, but at least I find more effective. So the form and the function are equally important in this sense. but at least I find more effective.

So the form and the function are equally important in this sense. I feel like we should eventually ask you about the Bold Park Club because this is how people get to own one of these projects and build and work with and just generally get to see the artistry you put into the boards. Yeah. So the Boltwork Club is, I started that earlier. Well, January, I started advertising it and the first project went out in march but uh you know before that i was working for about two and a half years or so or three years on just exploration i was creating things that never really sold these are projects that i i put online and i got feedback and i at some point i understood that

there is audience for that sort of thing. People appreciate it. Particularly, I got two kinds of responses from practicing engineers. Either they appreciated the novelty of it and they saw things that they haven't seen before done with a medium that they are working with, or people that got terribly offended.

What am I doing to their precious methodologies and way of doing things? And I was sort of rocking the boat. There's comments on Reddit and things. Why the hell would you do something like that? Really? Of course, because square and green makes sense. It's how we've always done it. It's proven.

It's optimal. It's all these things that we may not need to focus on so much. It's because when you look at the aesthetics, that's not always the most efficient path. The cheapest path. Well, yeah, sure. But goals are different. Yes. But I could see how if your mindset is such that you're so used to making things cheap, to look at some of these boards and say, well, a cuttlefish is not going to fit on a PCB efficiently.

You could do so much more with that space. And yet, I totally understand Sara's point of view that beauty draws other people. It makes these more accessible. I actually agree with both. I think the worst thing that somebody could react with is meh. Right. So as long as i get a reaction whether it's uh what what the hell are you doing or wow this is nice um i'm good if if if i do

if i do something and people are going you know it's like it's like I've seen hundreds of these before. Then I didn't do my job, I feel. So I agree and I accept the criticism. To me, it's good news because it means that I am getting a reaction from people. And I'm perfectly fine with that yeah so so the the response has generally been been good I thoroughly enjoy doing it doing doing this kind of work so it's been it's been encouraging I've had this portfolio of projects that that I've done over the years.

And I had an online shop, and I'm not very good at marketing, so they were available, but no one was really buying it. And I think I've had ideas along the way, and then I had a conversation with Star Simpson in I think it was November of last year and I think it was the first when she was starting to work on the forest mims boards she's been working on I think that was a bit later or maybe I heard she was working on them in November but I don't think it was public then. Maybe. I don't know. But I think that was our first voice chat. We've only done emails in the past.

And we had a conversation, and and that that that brought back some ideas and and she was suggesting um i can't remember the details but it it it sort of um i i got into gear then about doing the club and i was looking at patreon at the time and that didn't really quite fit what I wanted to do. And then it turns out that monthly boxes are very popular these days for various things. And so there's a platform called CrateJoy that I use. CrateJoy? Is that what it's called? CrateJoy. CrateJoy. Like shipping crate. Yeah. Okay. And they manage all the logistics in terms of subscription and so on. So it's very convenient. So I went with that and I knew that I had at least three, four, five projects that I can borrow from what I've done previously so I could

concentrate on building up the mechanics and the flow of creating a monthly project. And I expected about 25 to 50 people to sign up in the first couple of months or three months or so. But before I sent out the first project, I had 170 people sign up. Wow. Yeah, which was great. And at the moment, so I'm about to ship Project 8 and I have 340 people active, so active members. So there's been, of course, there's cancellations and so on, but the churn is incredibly low. I'm at a retention rate of 97%, which means...

That's fantastic. Yeah, that's very good that means that people are happy with what they're getting did you have to rethink your logistics planning to do X

and suddenly it's you know 10 times as much in terms of time or in terms of money or both or just kidding

yeah or even just mentally oh now there are more people watching uh no no uh no that i think i i try to do things that are interesting and push boundaries a little bit. So, for example, the last project that I sent is an SMD surface mount project, the first one that I've done, and it has a QSOP on it, which has a 0.65 millimeter pitch component on it. And that would be fairly challenging to people who are not used to it. But I decided to do that

in order to... Well, there's a few reasons. One, the first is I wanted to push people to accept the challenge. And I wanted them to

fail if they needed to. So not everything has to be canned and working. That's just what engineering is. And so if they don't succeed, then that's okay. And if they tell me, then I'll send them a new kit if they really messed up. And I created a practice board for it so they could have ordered a practice board from the shop if they wanted to practice before committing to the chip on the project. This is Touchy, Project 7, right?

Yeah, Touchy. Okay. Yeah. Yeah, touchy. And the other reason is that I didn't want to limit myself to things that are easy to solder. So if there's a very nice chip and that Silicon Labs chip is very nice, but the largest package is the QSOP package, and I didn't want to not create a project around it just because it's hard to solder. And lastly, I wanted people to know that the it isn't geared towards beginners. There are plenty of other projects for learning how to solder and learning how to start with Arduino or Raspberry Pi, but the club isn't about that. Right. It's about getting something interesting that you haven't seen before and changing the experience that you have with electronics. And that starts with the packaging and the unboxing experience. And it's the community we have on Slack that is quite active.

And people are helping each other and posting the progress. And when they need help, they get it there for me and from other members and they're they're very knowledgeable people and very experienced people um and in the club so um if you look at the portfolio of the projects you kind of you now as opposed to the beginning you get an idea and i wanted to have that those challenging projects there so I don't send projects with a lot of explanations and it's not a it's not a something that you get and there's instructions I won't and there's very detailed instructions there's some but I want people to look at something and say well I don't really know how to connect this I'm'm going to look at the data sheet. And to me, that is what I wanted people to experience also,

that part of engineering that is a bit less glamorous, that sourcing components, understanding the physical properties and looking at the data sheets for the footprints and what is pin one. that sourcing components, understanding the physical properties, and looking at the data sheets for the footprints and what is pin one. And if you can encourage them to do that for fun, if you can encourage that struggle and get something lovely out, I don't know, that sort of learning is hard to force other people to do. It's one of the reasons that when I have a personal project, I get to that struggle part, and sometimes that's where I give up.

And if I had, I'm looking at the ladybug, or the lady project, it's a ladybug with these resistors that come out and act as the legs, and you have them shaped so, curved so nice as legs, and I'm thinking, if I did that, they would be crinkly and not beautiful. But I would want to build this because it's pretty. That's a very good point. I didn't think about it like that. You push yourself more when you want to complete something, right?

So if you got some project in a baggie from Radio Shack or Maplin here, then you might not push yourself to finish it. And you would because it's something that you do want to put on your desk. Right. Yeah, I think there's an element of that, definitely. I want to go back a little bit to what you were saying about having the people who are assembling it kind of need to explore a little further. You know, I used to build a lot of kits when I was a kid, you know, radio kits and stuff,

and it was mostly kind of a paint-by-numbers approach. It's interesting to hear you say, well, I purposely make it so you have to think a little harder or, you know, figure out what this means or what this component, its orientation means. I'll give an example. An example is, say, a polarity of an LED. The marks are there, so there will be a square hole for the anode and a round hole for the cathode. So if I needed to point somebody out, point that to somebody and say, okay, you put the long leg in the square hole, then I could do that.

But I wouldn't necessarily put it in the documentation or in the page. I somewhat expect people to have that curiosity of looking at, okay, there's a long leg and this is a short leg, and I'm not sure where to put that, uh, put that in. So how to put that in, because it can go both ways. So then they would either ask or, uh, they will look up led or they'll look at the schematic or blow it up. And all of those are better ways of learning than following somebody else's directions.

Yes. And what will happen if it doesn't work? What will happen? It might trash the LED or it might just not turn on. Most likely it would just not turn on. So, and the LED might not work later. But then you wouldn't, like you say, you wouldn't forget the long leg and what it means. And next time when you have an LED, you would perhaps know where to look.

It's an example. And again, I'm exploring. I'm not necessarily trying to force this kind of way. If people get back to it, the first project, for example, when you design something, you become so familiar with it that it's hard to see sort of glaring possible faults. So people have assembled the components on the side that I didn't intend them to. It can be done wrong. It will be done wrong. I don't say wrong side because it's the way it was

presented i wouldn't call it wrong it was just not the not the side that i i i thought that they would put it on uh put the components on and so after that i thought well how can i how can i resolve it in an elegant manner so so i came up with a symbol that says this is the solder side and this is the component side without being too verbose and something that would take 10 words to describe and take attention from the other visual aspects of the board. But from every project I learn, and you ask about think it's a it's it's an excellent question about about what what I've changed and I and I change all the time every project I learn new things particularly about the assembly process so last project had an issue with the packaging that I designed it was the first packaging that I designed it Because it was an SMD project, it was fairly flat

because the components are flat and they don't take much height. And I designed this foldable cardboard package that's supposed to stick with glue, double-sided glue that I can apply with the supplier applicator. But that didn't work. It wasn't quite big enough, and then the glue in some of the packages came undone. And it was too late to fix, and I just sent an email to people and said,

look, there's a design flaw and flawed testing methodology. I'm sorry. And some of you would not have a perfect experience. If something, let me know what you think. And that's definitely something I would consider for future projects. There is availability now that I'm doing more stuff with microcontrollers and integrated circuits that are more than just the LEDs and diodes. The issues of availability are more of an issue

because my timelines are so short and I can't wait that three, four weeks even for things to become available. And also, I don't quite know that I'm going to use them up to, say, two months or a month and a half before I need them. About a month before I ship because then there's the assembly. There are other issues like you just learn, even if you want to add a label to the project, you realize how much time it takes for the 600. So I'm building, I'm shipping about now 360, 370 projects a month because there are people that are on three projects every month and people buy extras.

So I'm kitting for 600. So even the step of taking a label and putting it in a bag adds that time. And every component is, first you need to lay it out on the table and then you need to put that component in the bag. That adds, say, 15 seconds to your process. And now you have to multiply it.

So you start thinking about that. You think about the height of the overall packaging because if it's under two and a half centimeters i can ship it as a large what's called here called large letter and that's much cheaper than what come the step above which is a parcel so if i um so i start thinking about these sort of things when i when i when i design the circuit uh and but that that is that these are things that come through through these experience of of actually sitting there and and putting all these components in bags and then the lady for example the the the ladybird project had a I try to be efficient and and use the box as the

shipping box so the box of the kit is also the shipping box but just and and there are flaps inside and other things to glue. And that took a long time, just gluing things and folding them together. It just takes a long time. So boxes, for example, that come flat, but you pay for it later in assembly time. Because your time has value. Although, actually, now I want to get into the impertinent question portion. Okay. Is this your full-time job?

Yeah, yeah. It has been for almost four years. I have a non-impertinent question. Do you try to keep the assembly time for the the customer to kind of a certain amount do you find yourself okay this project is a little bit too complicated let me pull stuff back because it's way you know twice as long assembly as the previous one or do you just kind of let it go as it goes you mean the when people receive the kit if that's the assembly time for them. Yeah. No, I don't.

That's not really something I considered. I actually think that, you know, the longer it is, the more they get out of it. Right, right. I mostly think about... So the way it is, is that people are paying a fixed amount of money per project. And because every project is different,

very different, they're distinct in their packaging and the looks and the components. It is my challenge to create an interesting project that I still make money from because it's a fixed amount, again, that I get paid for it.

I need to make it below a certain amount, but yet still have the properties that I'm after. So it's very challenging to cheap, but are a novelty within that context. So I think project three was the cordwood puzzle, and that comes in a coffee bag, which is a few cents. Right. a few cents. But in the context of an electronics project, that's something people haven't seen.

Or the tin, a small mint tin can that a small project comes in. So yeah, it does drive quite a lot of

discovery on my part. But I try to be flexible. Some projects cost me less, some projects cost me more, and that's okay. What I want is for people, so we haven't mentioned the club you pay three months in advance. So you pay for three projects at a minimum. And that's 49 pounds. 49 pounds, yeah. That includes everything.

So shipping, taxes, whatever. And I want people to feel they're getting value over those three projects. And like I said before, some people appreciate different things. And some people, what one person sees as something that isn't great value, other people would see that it is for various reasons. They might notice things, they might touch them a different way. So overall, for the three projects, I try to give people a good value.

And for those of you who are wondering, that's about $61 US. Yeah, right now. Right now, as of two minutes ago on Google. Thanks to Brexit. The people from around the world are getting a discount. Yeah, the pound is pretty weak. Are you still doing consulting to make up any wage differences?

Yes. So the club is great. The margins are not that high, factoring labor and all that. And so I have to, and I want to continue consulting. I do a few projects for fairly big companies to demonstrate their products. And it's pretty exciting. These things take time to come out. I hope to come out with them, that they would come out in the next few months.

And yeah, I am looking for more work because practically when I get paid my daily rate, that's when big chunks of money come in so I

can continue going. For the first three years, I, I wasn't quite making any money. So it's good. It's good the club actually started having some surplus of money. And it's exciting because that means that I can do more stuff that I want to do in this area. That is exciting, that the club is paying for itself and a little beyond. Yes, yes. Yes, it's very nice.

Given how much time I imagine you can put into these boards, I mean, you're doing not only the labor of kidding, but the making each line be what you want it to be. How do you know when you're done? I mean, it has to be- It's true of all art. It's an art thing. You have to be done enough and you have to be willing to let it go. But do you have a way to figure out, okay, this is'm i'm ready i'm done i'm gonna go spin the boards no so uh the the it's an external event right so uh done is a moving target.

Done is weird because you say I'm done, but you never really are because if you continued working on it, you'd, well, in some cases, you'd introduce problems. You know how it is. You add things and then you uh you introduce more more issues i i think for for the club projects there there's there's a there's there are deadlines right i have to send a prototype two months before i want to ship and um or a month and a half if i'm if it get it tested and and then you know make any any changes I try to get the prototypes out with as much as much as I can as much of the detail as I can and not leave it for later because I like to see how things turn out physically. And then I don't want to introduce new problems

when I commit to ordering those 700 boards. At some point, it feels like enough all right that's it you you kind of and and you know i i look at it and i say well this is the this is the ballpark club there is the project which means that it doesn't necessarily need to be 100% finished. I think that the club members appreciate that and if I had to spin, if I had to create another

version of every project, I will have things that I want to improve on. And that's okay. We always do. And there is no doubt in my mind that one project, I will order boards

and there will be an electrical problem on there. And I think that with the community that the club has attracted, if I tell people, look, I messed up, there's a problem, here's instructions how to fix it. Let's have an experience of what an engineer feels like almost every time a board is back. You have to cut this trace and connect a wire from here to there.

And I hope that that's going to be acceptable. So I would say that the projects themselves are a bit of kind of like a 90% done or 95% done. I could always add more and after the fact either I come up with something or people say hey why didn't you add this and I thought that would have been such a such such a good idea. Like adding a little, you know, those to the, to the ladybird, somebody suggested adding a little,

these micro vibrators thingies. Yeah. So if you add that and then it can move. Yeah. Yeah. So, well, that,

that is something if I do, if I do another project like that, that was definitely something I would add. There are many other insects. What do you suggest? A wonderful grasshopper visited us yesterday and I was looking at it thinking, that would be kind of cool. It's also green already. It's true, it's a green board.

The Lady Bird, I always intended for it to be a series. So to create more sort of personified insects that have some sort of a, you know, circuit on them that makes sense. And I have an idea for... Bees, bees, buzz, bees. Sorry, go ahead.

For a chameleon. There's a name I was going to call the gent and it's going to have sort of LED lights, sort of these 3mm LEDs as eyes and this kind of coiled piece of solder as a monocle. Yeah.

I like it. But that's only in my notebook at the moment. Yeah. So you know, I think you may have alluded to before about changing schedules and things. quite I had at the beginning I had a lot of spare time because I I had the previous projects that I've already

done, I've proven and I just needed to create a new version of them and now I'm done with that now I have to come up with a new interesting project for every month and that's started to become quite...

That's a bit of pressure. It's a bit of pressure, yeah. It's a good kind of pressure. I think it makes me more creative, but it means that there's not a lot of time to play. And that somewhat concerns me. I fear that eventually it might affect the results. I don't think it's not there yet,

but I am thinking ahead of how to deal with that and maybe getting some help and planning a few more projects ahead of time. So there is more time to have that creative process because it's what I try to explain to people who hire me to do this kind of work, because I tell them, look, this is a creative process. It's not something I can cram into an overnight design on Altium. I have to think about it and get some concepts to you, and then there's iterations, and it has to rest for a little bit before I get back to it. And with some, some new ideas and creating prototypes, mock-ups of the, of how it's going to look like.

So yeah, I don't, I don't want to lose that because of the pressure. So I think I only have one more question. Do you have any art training or background? I know you have a PhD in FPGA awesomeness, but... happened because I've been doodling my whole life and I've been painting here and there just just for fun kind of I get these these periods where I decide that I want to paint or draw or whatever and so but I've been doing my notebooks from high school to to now are full of doodling. And if you didn't tell me that you heard me drawing on a piece of paper, I would be, the paper in front of me would have been filled with nonsense doodling.

See, we've stopped genius ideas from forming. I should have asked him to move the mic. But I never paid attention to it. I never thought that this was something, this was a skill or this was something that is worth pursuing. It was just something I do. So the genesis of the PCB mode was that I had this period when I started painting, and I ended up painting circuits.

So, you know, on canvas, and sort of drawing lines and these squares and things between them and i said okay i'm drawing circuits uh this okay that that is that it's not very special that's what i do so it's kind of uh understandable and then i thought well why can't i design um uh design paintings right so um why can't I have a circuit board that is also something that I can hang on the wall? And that's when I, together with my frustration with traditional EDA,

and I thought, and I still think, that there's a lot of, we need to innovate there more. I wanted to, I created PCB Mode to do that, to demonstrate some concepts in how we could do EDA differently, or the process of design differently, but also to design paintings. And so I've done a few wall-mounted stuff, and there's that the pieces that appeared

in Mary Claire magazine so yes back to the original question so late in my life I discovered I have this skill that I've been using my whole life, but I never really paid attention to it. It's this, this kind of aesthetic sense of, uh, my aesthetic sense of what I find pleasing. I, uh, I have a particular style and, uh, that's what I applied to my work and so it was quite neat that discovery to say well

now I've had this skill and now it's part of what I do and so I don't know to me the things that kind of clicked together, the circuit design, programming for PCB mode, and the artistic or visually pleasing things or beautiful things, whatever, all coming together was very neat for me to realize that that's what I do now. I think you just answered what was going to be my last question, which is, do you think this has made you a better engineer? Yeah, yes, just because I've been exposed to more components, materials. There's a lot of interesting materials these days to do circuits with. Interesting people, definitely. It's really nice to see,

once you start doing things in the creative world, how many other people are doing that and and are um and and that starts the conversation and you have these these interesting conversations and and you learn from them like somebody i talked to introduced me to z-axis tape um do you know what that is no it's um it's a tape a double-sided tape that only conducts in the z-axis once you press on it so yeah no that doesn't seem possible i'm sorry that's breaking the laws of physics so they use that they use that for uh connecting um lcd screen uh ribbon cables to PCB. So you'd have the exposed pads and then you just glue the ribbon cable of the LCD onto the PCB. And it conducts only in the Z-axis. And I've had I had a client project with it now, and it works.

And so that I learned from someone that I got through this work. I've worked with a company that does stretchable circuits, and that I didn't know existed. But it turns out that the way I do things is really useful for maintaining stretchability. Ooh, touchy would be neat if you could make it stretchable. Yeah. A lot of these would be really, yeah. Okay, yeah, lots of ideas there.

Well, and I did see Micah Scott put together one of your boards and the video was so beautiful. I mean, the board was lovely and the way she did the video was awesome. I loved it. I think I wrote a tweet about it and I said, you know, whether you like the style

or the music or whatever, or the board or not, just presenting projects in in that way is interesting and we should encourage that a different a different a different way of of um presenting our work and it's it's essentially what i do right it's a different way of presenting electronics and Mike has been doing a great job I mean her videos are just the technical detail

is impressive and also the creative part on top of it is great and I support her on Patreon, and I'm happy that she's able to continue that work. All right.

Before we get into a conversation with Micah, we should probably go back about our business. Yes. We've kept you quite late. No, that's all right. I'm not a morning person, so I do most of my

work about this time of the day. So, it's all right. Do you have any last thoughts you'd

like to leave us with? You warned me about this. I do. Before recording it, I warned

all of the guests in case they haven't read the outline. And I thought I'd come up with something. But I didn't let you draw. i think um we covered it was good conversation we covered a lot of the things that um i i want to write about sometimes but i just don't and uh so i think that's that it's good that this that we're having this conversation um I, hmm, no, nothing profound comes to mind, really. It's because your boards are so beautiful. Okay. you know, speak for itself kind of thing. And people take what they can or want out of the work. Not necessarily what I say about it.

Well, thank you so much for speaking with us. Thank you. It was very, very pleasant. What was that? Oh, yeah. I was surprised we didn't have more of that. That's just an ambulance or something. They're not coming for me.

No, they're just best. Our guest has been Sar Drimmer, engineer and founder of Boldport. I do want to mention that I

messed up the time with Sar twice, and then we lost internet during recording and he waited. He's been so

patient, so thank you for that as well as the good conversation. Thank you also to Christopher for producing and co-hosting. And of course, thank you for listening. Hit the contact link on embedded.fm if you'd like to say hello, sign up for the newsletter, or you can subscribe to the YouTube channel around there and the blog. You should just go to Embedded FM. You know, it'll be fun. We'll meet

there, I'm sure. Also, apparently the word chasm is pronounced chasm, which is not how it is spelled, so I don't understand. A final thought from me, as long as this is how it's going, Ursula Le Guin in The Left Hand of Darkness wrote, I talk about the gods. I am an atheist, but I am an artist too, and therefore a liar. Distrust everything I say. I'm telling the truth. The only truth I can understand or express is logically defined a lie, psychologically a symbol, aesthetically, a symbol. Aesthetically defined, a metaphor.