448: Little Squiggles All Around
Transcript from 448: Little Squiggles All Around with Carl Bugeja, Chris White, and Elecia White.
EW (00:00:06):
Welcome to Embedded. I am Elecia White, alongside Christopher White. Let us talk about PCB boards, making themselves move, and then soldering themselves. And origami. It is going to be a good week. Our guest this week is Carl Bugeja.
CW (00:00:24):
Hi Carl. How are you doing?
CB (00:00:27):
Hi. Thanks for having me.
EW (00:00:29):
Could you tell us about yourself as if we met, well, I want to say at an origami convention, but- Yeah, at an origami convention. Sure.
CB (00:00:39):
Sure. I am an engineer that lives on a tiny island called Malta, in the middle of Europe. My job is to design weird and novel electronic projects, and document all of them on YouTube.
EW (00:00:57):
And we will be talking about many weird projects.
CB (00:00:59):
<laugh>
EW (00:00:59):
But first lightning round, where we ask you short questions and we want short answers.
CB (00:01:04):
Okay.
EW (00:01:04):
Are you ready?
CB (00:01:07):
Yes.
CW (00:01:09):
What is your favorite fictional robot?
CB (00:01:11):
Okay. I guess it has to be WALL-E.
EW (00:01:14):
What is your favorite origami pattern?
CB (00:01:16):
Hmm. It has got to be the crane or the swan, I guess.
CW (00:01:22):
Flexible or rigid PCBs?
CB (00:01:25):
Flexible.
EW (00:01:27):
Can flex circuits be made into origami?
CB (00:01:30):
I think so, yes. <laugh>
CW (00:01:33):
Do you work on one project at a time, or do you have several in the queue?
CB (00:01:38):
Unfortunately, several <laugh>.
EW (00:01:41):
How far can your fold up rover go in one charge?
CB (00:01:47):
I think it can last around 15 minutes, but it heavily depends on how smooth the terrain is. On rough terrain it needs to have the battery fully charged to keep going.
EW (00:02:03):
Is that 15 or 50, like almost an hour?
CB (00:02:06):
15 minutes.
EW (00:02:07):
Okay.
CW (00:02:08):
If someone were to visit Malta, what restaurant should they go to?
CB (00:02:12):
Mm. It is hard to tell because there are so many restaurants here opened in the last five years. <laugh> It is a little bit crowded to be honest. There is a lot of Italian cuisine and stuff like that, so I think you have several options to choose from. <laugh>
EW (00:02:36):
Do you have a tip everyone should know?
CB (00:02:40):
Hmm. I guess, to never give up on anything that you are working on, because somehow, eventually, you can get it to work.
EW (00:02:56):
That is a great tip. So, I do want to talk about your origami actuators. DJ, a listener, sent in that video. But while we were talking about scheduling this, we then got another video, about a fold up rover. And then another one, about a self-soldering circuit. So where should we start?
CB (00:03:19):
<laugh> Let us start with the actuator stuff, I guess, because that inspired a lot of the foldable rover as well.
EW (00:03:33):
So you folded a bird and a dog and a couple of other things, and you had them flap, using a magnet and just a flex circuit. There was no traditional motor solenoid thing. How does that work?
CB (00:03:57):
The idea actually came about when I was- I had already developed the flexible actuator, and the PCB actuators. And then I had one project where I tried to turn a heater into a sticker, and see how and what applications it could be useful in. When I was testing it, I realized that putting an actuator on a sticker does not exist, and would be very useful in some cases, especially in origami, because flexible PCBs are like paper thin. So the idea was to develop a sticker actuator that actually connects to paper and you could bring life into origami.
EW (00:05:07):
Going back to the heater, I actually worked on something similar. It was interesting because we were making a thermometer that was non-invasive. You stuck the heater onto somebody's forehead, and it had some insulation, and then you heated it such that there was no temperature gradient between the heater and the human skin. That was a deep temperature measuring- I do not remember the acronym for it, but it was kind of cool. Have you done stuff like that?
CB (00:05:49):
What I have done is try to apply- One time I had received an email from a scientist in Argentina, and he told me that he was using my actuators as heaters. Basically, what he required is a heater that is paper thin. My flexible actuators were just 0.1 millimeter thick. I had this, "Let me try and actually put a heater on flexible PCBs, instead of using my actuators as heater." There were quite a few interesting results in that experiment, I think.
EW (00:06:46):
Did you melt any flex circuits?
CB (00:06:51):
Yeah, I tried to heat them to their limit. Flexible PCBs are rated to hundred degrees Celsius. What I try to do is to run a long duration test for around I guess it was two weeks or one month, something like that, because this was more than a year ago. The PCB that I was testing eventually turned into this brownish color. It was still functional and it was non-reusable, but it was still cool to see the limitations of these circuits.
(00:07:52):
But then eventually I also learned that the so flexible PCBs, there are different colors, and for some reason they are rated at different temperatures. So the black coverlay should sustain higher temperatures than the brownish one. That was the conclusion for that project.
EW (00:08:18):
You did something similar with self-soldering PCBs.
CB (00:08:24):
Yes.
EW (00:08:25):
Could you describe it?
CB (00:08:27):
Yeah, that project was actually stuck in my head for around five years, and after thinking about it a lot, I eventually decided to give it a try. Viewers from my channel probably know that I first tried to create a PCB hot plate. My plan was to first create a hot plate, then try to tackle self-soldiering circuits. I eventually found out that there are a lot of things to learn from my side about PCB heaters. So it was not that easy as I thought it was.
(00:09:13):
But eventually, from all the testing that I have done, I have found that PCB hot plates are actually a very bad idea, because PCBs are not rated to go to high temperatures for long duration of times. So basically, the conclusion was that a PCB hot plate would be only used for three times or four times maximum, until it start getting damaged.
(00:09:53):
But this still made the self-soldering circuit feasible, because you only have to solder the circuit one time, for the self-soldering concept. Once I eventually reached that point, I tried to make the video as geeky as possible, I guess.
EW (00:10:22):
You said not long duration and one time, and that makes sense, because the boards have to get hot enough to be soldered, normally.
CB (00:10:30):
Exactly.
EW (00:10:31):
And on the inside, if I remember correctly, you took a plane that might normally be a ground plane. Instead of having it all copper, you made little squiggles all round except for where there were vias.
CB (00:10:47):
Exactly.
EW (00:10:48):
How did that lead to heating? Did you have a ground? Do you have a short to ground? How does that work?
CB (00:10:57):
Imagine a resistor. When you pass current through it, it heats up. The same goes for a copper track. If the resistance is small enough, it will heat up. My idea was to use low temperature solder paste, to heat it up, so that it could follow the soldering reflowing profile.
EW (00:11:30):
Are the squiggles to increase the resistance so that it heats up?
CB (00:11:34):
The squiggles are just to make the track longer, because you have to achieve- The resistance have to be high enough, so that the current will not be too high. I would be able to just power it with nine volts or something like that, or 12 volts. That is why I try to increase the resistance as much as possible.
EW (00:12:04):
It was an interesting video. That you could make one of these from another one of these. The whole mother baking the daughter thing. But would you do this in real life? Are there production uses for this?
CB (00:12:22):
For hobby projects, I think it is perfect. Like for someone who does not own a reflow perhaps. Sorry, a hot plate. But for production, I am not totally sure, because one of the biggest criticism that the project received was that, in engineering school, they teach you that ground should have the shortest path as possible for high speed circuitry. So most probably this circuit does not pass an EMC test or something like that, because it actually has a very bad ground plane, even though I had a separate ground layer underneath it.
CW (00:13:20):
Huh.
EW (00:13:21):
But you could rectify that by- Right now you connect your squiggly layer to the ground layer, to have two ground layers.
CB (00:13:28):
Exactly.
EW (00:13:28):
But if you just, instead of connecting it, left it floating, then that would not matter, would it?
CB (00:13:35):
I do not think that that would matter, but I am planning to do some more testing on this, so that I can perhaps create even a better version, with less squiggly lines and the shorter path to ground. I am still sort of brainstorming at the moment, on how I can create a better version. But for me, it was just a cool concept that I had to try out.
CW (00:14:12):
I think it is one of those concepts that is so novel, to me anyway, that I cannot readily see the application. But I know there must be some really cool application for it, that I am missing. This is not a criticism, it is a "This is so outside what I would expect somebody to do, that I have to process it for a while <laugh>." It is kind of along the lines of all of the self-assembling or self-repairing kinds of things, like von Neumann machines and all that.
(00:14:44):
I feel like, yeah, okay, maybe something that needs to repair itself, could have a whole array of these layers. And, "Okay, well this is not working. We will try to reflow this one part of the circuit." Or something like that. It seems like there should be really important applications of this.
CB (00:15:01):
Yeah, for repair it would definitely- There were a lot of comments adjusting it to be used in repairs. Or even for something that has like critical data, where, for example, if stolen the PCB has to self destruct.
EW (00:15:21):
Mm, yeah.
CB (00:15:23):
Someone even suggested using the heater to keep satellites warm from the cool environment of space. So there are a lot of areas where this could actually be useful.
EW (00:15:44):
I was starting to think of things like places where you do not necessarily have all the tools you want. But it is difficult because you also do have to solder paste it and put all the components on. So it is not like you can do it in a field test, unless you have a very quiet place to be. Could you do it in space? Why would it be better to do this in space than to have it heater? Or could you do it in Antarctica, where you do not necessarily have everything and you want to be able to configure circuits on the fly?
CB (00:16:22):
Yeah. It can be portable, I guess, by just connecting a battery. But the space idea was just related to keeping the actual circuit warm-
EW (00:16:35):
Yeah.
CB (00:16:36):
Not soldering it. But it can also be soldered in space, but it would be messy, I guess.
EW (00:16:45):
Well, it was more, if you went to Mars, what would you need to take with you? At some level, you do not know what you need to take with you, so you need to take flexible things. So it was same perspective as Antarctica, for me.
CW (00:17:02):
You could also selectively blow parts of- Like, if you had a whole array of these, you could address parts of the circuit with a heater. You could like blow fuses that way, I guess, or-
EW (00:17:10):
Well, he did say-
CW (00:17:11):
Blow up an EPROM. Just one though. Not the whole thing.
EW (00:17:14):
Trashing your- Yeah, I mean-
CW (00:17:15):
I want to selectively modify the circuit <laugh>.
CB (00:17:17):
<laugh>
EW (00:17:23):
In an effort to scientifically determine the maximal limit, you also set one of these on fire, which I found highly amusing.
CB (00:17:32):
<laugh>
EW (00:17:32):
How far did it go? What was the power where it just said, "No, I am not ever going to work again"?
CB (00:17:42):
PCBs are rated, their material are rated, on a parameter called Tg rating. Usually, the highest rating, for example for the PCB that I used, was 170 degrees Celsius. I am not familiar with Fahrenheit, I am sorry. You can translate it if you would like.
CW (00:18:12):
<laugh> Hot. Very hot.
EW (00:18:13):
Above 300.
CB (00:18:14):
Yeah, very hot.
EW (00:18:14):
Way above boiling.
CB (00:18:17):
So trying to heat up the thermal track higher than that value will damage the PCB, especially for long durations. In that test, I think I powered the track with around 30 volts, if I am not mistaken. So that brought up the temperature way high. The interesting thing is that since that squiggly track, once the PCB starts getting damaged and getting melted, the tracks themselves start to get shorted. So by that process, they are essentially lowering the resistance even more, creating a bigger damage.
EW (00:19:20):
A cascade effect. Yeah.
CB (00:19:22):
Exactly.
CW (00:19:24):
You see more current after that.
CB (00:19:25):
Exactly.
EW (00:19:27):
<laugh> Okay. So how do heaters relate to flap actuators?
CB (00:19:34):
They actually relate a bit, because one of the limitations of PCB actuators are the thermal limits of the PCB. So like I said, for flexible PCBs, for example, the maximum temperature it can withstand for it to be okay is hundred degrees Celsius. So when passing current through the copper tracks, it will obviously heat up, even though it is a coil, not a PCB heater. That is actually the limitations of the coils themselves, is that they have to operate within the maximum temperature ratings of the PCB.
EW (00:20:30):
Oh, okay. So it is basically, if you take a flap actuator and use it too much or power it too much, it will burn itself out.
CB (00:20:44):
Not- At a much higher voltage it will, but even at say- I usually try to keep their voltage ratings below five volts. But even say if you power it at eight volts for like one month, it will eventually damage itself.
EW (00:21:12):
We have not really described what the flap actuator-
CW (00:21:15):
<laugh> That is very hard to say.
EW (00:21:17):
Sorry, that is hard to say. Looks like. Since we are podcasting, that is always a challenge, but I think the way to start is with the circuit pen that you used to make one of these on paper. Can you describe that?
CB (00:21:36):
Yes. This idea came about when I traced- I just bought some conductive ink, and I was playing around [with] it trying to make a coil that was similar to my PCB actuators. The downside of this project was that conductive ink has a low conductivity when compared to copper. So the effects are not as- The coil's magnetic field is not as efficient as the PCB one. But it worked and I did manage to attract paper with magnet.
CW (00:22:34):
These are basically deconstructed solenoids in a sense, right? You have got a coil, and then it is acting on a permanent magnet.
EW (00:22:41):
And the coil just looks like a spiral.
CW (00:22:43):
Yeah. So you apply...
CB (00:22:46):
It is like a low, very inefficient electromagnet. That is how I like to describe it.
EW (00:22:51):
<laugh>
CB (00:22:52):
It is an electromagnet without a core.
EW (00:22:56):
So the flap actuator goes between two magnets if you want it to go all the way back and forth. Is that right?
CB (00:23:07):
Um, yes.
EW (00:23:08):
Or does it only need one?
CB (00:23:10):
The current flap actuator that I have only has one, because I had a whole project trying to- Because at one point- So I first was making these actuators for fun. Then at one point I tried to create- I was getting into the robustness of these flexible PCB actuators. The reason for this was that I started observing some failures, when I was powering these coils for like one week or two weeks.
(00:24:00):
So the goal was to create a flap that could be powered for like a year or two. The advantage, the good thing about this, is that now I have a template for a flap actuator that can actually be robust and be used in projects like mechanical displays and where I am now confident that it would not fail for a long duration.
EW (00:24:41):
Mechanical displays. Say more about that.
CB (00:24:45):
<laugh> Yeah. So one of my future projects will be a display made out of these flaps. It will be like a mix between flip dot display and kinetic sculpture. Like I was saying, if I have not just focused on the durability of these flaps, going for this new project would be very bad, I guess, because it would not have lasted very long.
EW (00:25:30):
I guess knowing that they can go for millions of cycles, really helps the idea that this will last.
CB (00:25:37):
Yeah, exactly.
CW (00:25:38):
Pop-up books, electronic pop-up books, where the people can move and stuff. I am sorry, just thinking of other things you could do with this.
CB (00:25:47):
<laugh>
EW (00:25:49):
But I cannot just drive this flex circuit from my normal processor, right. I cannot just hook it up to an Arduino, and hope that I do not source enough current to blow up my chip.
CB (00:26:01):
Yeah. Arduino will not supply enough current. What you need to drive it is either a MOSFET, a single MOSFET transistor, but that would actually control the current and one polarity. You can also use an H-bridge, where you could reverse the current, and actually flip its polarity.
EW (00:26:30):
And that is how you get it to go back and forth.
CB (00:26:31):
Exactly.
EW (00:26:31):
When you want it to be close to the magnet, you have it go one way. If you want it to be far, you have it flip its polarity. You have a kit with those, the Flexar Drivers? Those are the H-bridges?
CB (00:26:45):
Yes. I have a website called flexar.io. There I sell a kit where people can experiment with this directly. The driver, what it does is basically has five functions. You can switch between each function to see. Because the actuator, for example, can also be used for buzzing, making a sound. It can either be used in a square wave mode, or in a PWM mode where it is driven with like a triangular waveform, and stuff like that.
CW (00:27:34):
Have you looked at reversing it, to make sensors? Like you could make a little-
EW (00:27:41):
Pressure sensors.
CW (00:27:41):
Paper keyboard or something?
CB (00:27:43):
Yeah, I am not sure if it can actually used as a pressure sensor. But it can be used as an inductive sensor. There are many applications and also development kits that are using PCB coils for inductive sensing-
CW (00:28:06):
Okay.
CB (00:28:09):
Where it can either perhaps detect a magnet or detect a metallic object. They have also showed that in some of my videos as well.
CW (00:28:18):
Like a paper microphone.
EW (00:28:20):
Yeah. We actually saw that.
CW (00:28:21):
Oh, did we?
EW (00:28:22):
I do not know that I saw it from Carl, but I saw something about it. Have you done a paper microphone?
CB (00:28:29):
I have not done it yet <laugh>, I mean, with these PCB actuators. Because I guess it cannot generate magnetic field. It would be too weak.
CW (00:28:44):
You would have a big low-noise amplifier.
CB (00:28:45):
And yeah, you would need a really large gain, I guess. It would be a fun project though, <laugh> planning to do that.
EW (00:28:54):
You show these on YouTube, and you sell these little kits, but they are all open source. So if I was not lazy, I could make my own. Do you make most of your money from YouTube?
CB (00:29:08):
Yeah, <laugh> I make my income from two ways. From these projects and from the kits I sell on my website. Even though all my projects are open source, the good thing about YouTube is that first you are documenting everything in your videos, which would be very helpful to a lot of people. While doing this, you will be also building a community, which obviously by watching your videos will help you make more videos.
(00:29:51):
Then I also have a Patreon membership, where I have a lot of members that are supporting directly my projects. I also have a couple of sponsors that fund the PCBs and actually the projects themselves.
EW (00:30:16):
How did you get started? What came first? The kits or the videos?
CB (00:30:23):
The videos came first. My first video on YouTube was about the PCB motor. Then I spiraled my way down to PCB actuators, flexible actuators, and stuff like that.
EW (00:30:39):
How does a PCB motor work?
CB (00:30:42):
A PCB motor works like every other 3-phase brushless motor, but instead of having wired windings, the coils are printed onto the circuit themselves. Then you have a rotor on top, with a couple of magnetic poles. And by providing commutation signal, the rotor rotates.
EW (00:31:17):
So you are doing the whole three phase commutation. Where you are right now, it depends on what signal you get, because you are trying to throw the motor around, based on changing electrical signals.
CB (00:31:36):
Yes. It is usually in a closed loop. In a closed loop or orientation, it is based on the position of the rotor.
EW (00:31:46):
That is hard <laugh>. I did that once, and I am in managed to get some beautiful oscilloscope traces, and yet-
CW (00:31:54):
<laugh> The motor just kind of flails around.
EW (00:31:56):
No, no. The motor would go around. I just burnt out FETs like crazy <laugh>.
CW (00:32:00):
Ow.
CB (00:32:02):
Yeah <laugh>. It is pretty- When I was a student, I was creating a brushless motor ESC and it was pretty hard and challenging to do it. Because you have to have very, very efficient code, to get the motor rotating at very high speeds. Because if you have just a little delay, it will lose commutation. <laugh>
EW (00:32:33):
This is one of those cases where you actually truly need real time response. You cannot have any latency. You cannot do this in Linux.
CB (00:32:40):
Okay, hmm, I do not think so.
CW (00:32:44):
Depends on your Linux. I am sure if you applied enough CPU power.
CB (00:32:48):
<laugh>
CW (00:32:48):
<laugh>
EW (00:32:50):
I do not know.
CW (00:32:51):
I do not know either. I do not know what I am talking about.
EW (00:32:53):
Um...
CB (00:32:53):
I mean, for low speeds, I am pretty sure you cannot do it.
CW (00:32:56):
Yeah.
CB (00:32:56):
But, for very low speeds motors, I am not sure. <laugh>
EW (00:33:04):
It becomes untenable pretty quickly. I used a TI Piccolo, because those seem to be very designed for this problem. But I think in your video, you are using a PIC something?
CB (00:33:15):
Yes. My ESC <laugh> went through phases, because at first I started with a PIC16F. Then I switched to a dsPIC microcontroller. Then, for the rover, I switched again to the PIC16F, because the dsPIC was out of stock everywhere. <laugh> At the moment it is pretty <laugh> all over the place. But I am planning to make a whole video just about speed controllers, and how it can actually- I will go in more detail about PCB motors, and how you can select the most efficient ESC for your PCB motor application.
EW (00:34:13):
Cool. I look forward to that one. And please add some oscilloscope traces, because they are so pretty.
CB (00:34:20):
<laugh>
CW (00:34:20):
I want to make sure I understand. So the PICs are being used solely as speed controllers. So if you wanted to drive several motors, you are going to have a higher level microcontroller that is talking to all the PICs and telling them how to adjust their speeds.
CB (00:34:34):
Yes.
CW (00:34:34):
Okay.
CB (00:34:36):
That is how my rover was working.
CW (00:34:39):
Oh, okay.
CB (00:34:40):
So it had four PIC16F microcontrollers, and the ESP32 was talking to them, basically just sending the speed.
EW (00:34:54):
And the ESP32 is also talking to your Xbox controller?
CB (00:34:59):
Yeah, I was using a PlayStation controller.
CW (00:35:02):
Over Bluetooth?
CB (00:35:04):
Yes.
CW (00:35:04):
Oh, neat! Okay. I think I missed that from the video, is how that was connected. So that is cool.
EW (00:35:11):
This rover that we are talking about, everybody is thinking like a mouse droid from Star Wars. Something, you know, at least calf high, but this is tiny!
CB (00:35:24):
Yeah. <laugh>
EW (00:35:25):
How big is your rover?
CB (00:35:28):
I do not think I ever measured it <laugh>. But it is quite small. It is around- The battery was, I believe, 4.9 centimeters in length.
EW (00:35:40):
And the battery is pretty much the length.
CW (00:35:42):
Yeah.
CB (00:35:43):
Yeah, exactly.
EW (00:35:44):
I think the wheels go a little beyond, depending on which you have printed up.
CB (00:35:48):
Exactly. It is around five centimeters in length, something like that.
EW (00:35:53):
Yeah, it was tiny. And that of course causes its own problems, because it is not heavy-
CB (00:35:59):
Yeah.
EW (00:35:59):
Which makes it easier to go, but harder to get traction.
CW (00:36:03):
Okay. So the rover, the hook of the rover is that it uses these PCB motors, but actually the structure is also PCBs. So it is like...
EW (00:36:11):
It folds into itself.
CW (00:36:12):
Folds into itself, right?
CB (00:36:13):
Folds into a cuboid.
EW (00:36:15):
Did you ever try tracks instead of wheels?
CB (00:36:19):
I have not. But it is one of the main points that is highlighted for the next version.
EW (00:36:31):
<laugh>
CB (00:36:31):
But before thinking about that, I am just focusing on the motor, and trying to squeeze a little more torque out of it. My first step for this next rover is going to be just focus on the motors, and trying to get more torque out of them.
CW (00:36:54):
How are you trying to do that? I can see that the PCB presents some limitations about how physically things can be structured. So what things can you adjust to get more torque?
CB (00:37:08):
Yeah, there is actually quite a <laugh> long list. First of all, the most obvious thing that can be done is adding more poles. So currently my motor has six poles. Going for like a nine pole motor, would increase the torque. The only problem with that is the area. I am still not sure if nine poles can actually fit in that area and if not, how much bigger does the motor needs to get.
(00:37:51):
Then there are a lot of other stuff that can also be tested. Something like, for example, a lot of people suggest Halbach arrays. My rotors are quite small, so I am still not a hundred percent sure that the magnets would fit inside that, and have that orientation. But I am still experimenting with a lot of stuff, currently.
EW (00:38:23):
You put this experimentation style of development on your YouTube channel. You also put your mistakes on. How do you decide what you are going to put up? How do you decide what mistakes are too embarrassing to show on video?
CB (00:38:44):
I actually try to put everything in the videos, as long as it obviously follows the same arc of the video. But there are some things that would make the video last forever. So for example, for the rover, I was having a problem with the PlayStation joystick communication, because I had the joystick that was like a mockup and it was not working with it. I spent weeks trying to solve this problem and I eventually did not solve it. Then it comes to how long the video is going to last, and is it important for the build? So there you have to make some decisions, what to leave out, and what makes the video interesting.
EW (00:40:09):
It must be difficult to have the balance of making the video interesting and fun and entertaining, but also showing the engineering in a way that is understandable to the people who are interested in that sort of thing.
CB (00:40:25):
Yeah, it is very difficult. <laugh> I am always not a hundred percent sure how video is going to go. But I try to keep them as fun and as interesting as possible. Because my goal ideal is just to build an engineering community. With that you have to make the video interesting for engineers, and for students and people that are just starting out.
EW (00:41:09):
Have your videos gotten more technical over time or less technical?
CB (00:41:14):
Mm, I am not sure. I think they have become more technical, but I am always trying to explain the technical stuff in more fun ways, so that the students and people that do not have an engineering background can still understand them.
EW (00:41:41):
And that is why your self-soldering video made me hungry.
CB (00:41:44):
<laugh>
EW (00:41:47):
Let us see. I have a couple listener questions. MattyC wanted to know where you get your inspiration for your designs.
CB (00:41:57):
Okay, so some ideas I come with them in the shower or something like that. The self-soldering circuit was one of them, for example. Because like I said, this has been five years in the making this idea. But a lot of other projects just start by creating other, finishing up other projects. So for example, the PCB motor project, I started working on the new version as soon as I finished the rover. A lot of my projects I think are created that way.
EW (00:43:00):
Do you have any that you have put on the shelf, that you wonder if you will ever get back to?
CB (00:43:08):
Yes, actually. I have some projects that happened before YouTube. So before I was posting video, I was still working on some other stuff, that maybe someday I will revisit.
EW (00:43:31):
That brings up another question from a listener. William asked, "What motivated you to start a YouTube channel?" And you have mentioned community. What is the maker community like in Malta?
CB (00:43:45):
Mm. <laugh> Pretty zero, I think. The first maker fair that was going to happen, got canceled because of COVID. But I think it is improving with time. Let us say that, because currently they are trying to motivate a lot of young students to get into the STEM world. So it is improving. But when I was young, it was <laugh> hard to find a friend that was into this kind of stuff.
EW (00:44:28):
What are some of the other motivations to start the YouTube channel?
CB (00:44:33):
Yeah, so basically, when I finished my engineering course, I start a startup with my friend. And during the time we were experimenting with a lot of interesting things, but the startup eventually failed because of financial issues. But I remember saying to myself, like, "We have done a lot of research, and documentation was not very good. If perhaps we have put all this stuff on YouTube or Hackaday or something like that, maybe it would have been beneficial to others. And maybe also to us, because you actually have to show what you are creating." And that is when I started the YouTube channel, I think. After that failed startup experience.
EW (00:45:43):
In school, could you have imagined doing YouTube as a job?
CB (00:45:49):
No. <laugh> I was very shy. If you watch my first videos, <laugh> they are horrible to watch. But when I was at university, my passion was always trying to invent new things. So at one point, we also had a group of four friends, where we weekly discussed new inventions that we could invent, and create as a startup or as a business.
(00:46:28):
I think YouTube is the perfect place to do that, because when it comes to hardware, it is very, very hard to start a startup hardware. First of all, any hardware project starts with R & D, and obviously, during that phase, the whole project could fail. Or from the research you have made, it would be unfeasible to do. So YouTube is the perfect place, I think, to do that. Because even if something does not work, you still showing people what you have learned, which I think is very entertaining as well.
EW (00:47:27):
The education part is admirable, but if you want to have a new startup with a new invention, if you are giving away all your ideas, how will that work?
CB (00:47:40):
You are referring to us putting them as open source? Or?
EW (00:47:44):
As open source. As explaining them on YouTube. If you invent something that you want to have a startup about, do you think you can be as open about it?
CB (00:47:55):
Yeah, I think you can, because as you go along, people will start also putting their ideas in your comments. And suggesting if the idea is feasible, or what sort of applications would they use it for. I think, by putting it as open source, you are first giving it to the community, so anyone out there can build it yourself.
(00:48:34):
Like I said, by having this community, if someone, for example, copies you or uses your idea, people still- For example, one time it happened on one of my projects, people will comment, "Have you seen Carl Bugeja's video about?" So there is still the feeling of a community behind it.
EW (00:49:10):
So on your videos they comment, "Have you seen your videos?"
CW (00:49:15):
<laugh>
CB (00:49:16):
No, no. On other...
EW (00:49:18):
Oh, okay.
CB (00:49:18):
If, for example, someone tries to create something similar, more stuff like that.
CW (00:49:27):
It does happen though. People do comment on it.
CB (00:49:29):
Yeah, yeah.
CW (00:49:29):
<laugh> When people showing stuff and say, "Hey, we should check this guy." "I am that guy."
EW (00:49:33):
I have had people ask me interview questions straight out of my book, and it is like, "Uh, maybe you should check the book."
CW (00:49:40):
<laugh>
EW (00:49:42):
Let us see. William had another couple things to say, but also, "Please let him know, I have been watching his channel from almost the first upload. I really respect his enthusiasm and creativity."
CB (00:49:55):
Thank you very much.
EW (00:49:58):
Where do you want this to go?
CW (00:50:03):
Self-soldering, self-folding, self-driving robot.
CB (00:50:06):
<laugh>
EW (00:50:07):
Or YouTube star with ten billion followers?
CB (00:50:12):
No, I do not think so. <laugh> I mean, my goal in life is just to keep inventing stuff, I guess. Even if some of my projects I think cannot actually be turned to a viable product. So my goal is to just keep improving some of them, and try <laugh> to create a better future, I guess <laugh>.
EW (00:50:44):
One more listener question from Peter. "What kinds of projects would you recommend for someone looking to learn more about designing foldable circuits?"
CB (00:50:52):
Okay.
EW (00:50:52):
Foldable or flexible? Well let us go with both.
CW (00:50:56):
Foldable.
CB (00:50:57):
Foldable? Okay. My suggestion would be to first start with rigid PCBs, and use connectors or soldering bridges to connect the PCBs together, if you have not obviously done a foldable circuit before. Because that way it will give you a good idea of the clearances and stuff like that of the PCBs. And things that you have to be careful, that perhaps do not pop to your mind when you are designing things. I think that is a good way to start.
(00:51:42):
If you have seen my foldable rover, if you have seen the two-wheeled robot, so before the foldable robot, there was a two-wheel robot. That robot was actually connected with connectors. So it had three PCBs in total, that sort of formed a U-shape, and were all connected with connectors.
CW (00:52:10):
That raises a question that I had in my mind earlier and forgot to ask. Are you doing anything with the coils, or with any of the projects? Some of them you are doing on flex circuits, some of them are doing on regular PCBs. But are you doing anything where the manufacturing places for the PCBs come back to you and say, "What are you doing? This is weird or this is difficult." Or...
EW (00:52:35):
One of the sponsors is the PCB manufacturer.
CW (00:52:38):
Well, that does not-
EW (00:52:39):
Do you get special treatment for that? Do you get to have special circuits?
CB (00:52:42):
No, I do not think so. I mean, they sponsor <laugh> everything that I ask for, but they do come up with like- Because their marketing team is different than their technical team.
CW (00:52:59):
Yeah. <laugh>
CB (00:53:00):
So, when it comes for example to flexible stiffeners, stiffeners for flexible PCBs, I tend to either- For example, for the rover, had around six of them, so they do come back just to verify if they have understood the file correctly, like "How many stiffeners do we have?"
(00:53:33):
I am working on a new project for example. It is just a small actuator. This actuator has a flexure. I am not sure if you know what flexures are. They are used in plastic normally, the result of 3D prints currently with flexures and stuff like that. They are really popular and I am trying to integrate them into PCBs. And for this specific design, I had the same model, for like ten times. And for them, seeing all these stiffeners got them crazy.
EW (00:54:22):
So stiffeners has been one of the areas they have asked about. Anything else?
CB (00:54:29):
Yeah, some of my shapes, for example. I made a wearable heated vest, which was around 70 centimeters big. So imagine a PCB almost as big as one meter, very close.
CW (00:54:53):
<laugh> Wow.
CB (00:54:54):
So, that was strange to them.
EW (00:54:58):
<laugh> Yes.
CB (00:54:58):
I once made a Christmas tree that sort of unfolds, <laugh> that also raised a lot of questions on their side. The RGB ball also had a very weird shape.
CW (00:55:18):
Right, right. Well I imagine they are getting used to it by now.
CB (00:55:22):
<laugh> Yeah, for every project I submit, I expect a follow up email from their side.
EW (00:55:32):
Going all the way back to the flap actuators, and thinking more about my personal projects <laugh>.
CB (00:55:40):
Okay.
EW (00:55:42):
Usually when you have them, they are flapping back and forth to their full extent. They are making a large motion very quickly. Can I make that same motion but slowly, or does it have to be fast?
CB (00:56:00):
Yes, it can also be slowed down. You just have to tune the frequency of your signals, and it can actually be made quite small. And if instead of a square wave, you create like a triangular waveform. The motion is very organic. It is like, for example, a wing for a bird and stuff like that, because it is not just on off, on off, it is slowed down.
EW (00:56:36):
Cool. I have an origami jellyfish pattern that I want to try. And so if I get a kit, and I get the Flexar-
CW (00:56:45):
If you get a kit.
EW (00:56:48):
If you ship the kit that I order-
CW (00:56:49):
<laugh>
CB (00:56:51):
Actually order it.
EW (00:56:53):
And I get the Flexar board and the flap. Do I need another controller, do I not? I need something else to tell it what to do.
CB (00:57:05):
The driver actually has an onboard controller, so you can just plug it in and it will work. But it also has an option to provide external signal. So you can also control it with your own microcontroller if you would like.
EW (00:57:23):
Like SPI, I2C, UART?
CB (00:57:26):
No, it is just the two PWM signals that drives the H-bridge module.
EW (00:57:33):
And can I reprogram the PIC that is on there?
CB (00:57:37):
Yes, you can also do that. In the data sheet, there is the description about the programming pins and what they are.
EW (00:57:48):
Do I still have a PIC programmer? I might.
CW (00:57:52):
I am sure we have a PIC programmer somewhere in a drawer.
EW (00:57:57):
<laugh> All the programmers. Have you tried to get like a heartbeat in a stuffed animal, instead of a flap?
CW (00:58:07):
<laugh>
CB (00:58:08):
A heartbeat? What do you mean?
EW (00:58:11):
Like the-
CB (00:58:12):
Uh, like-
CW (00:58:13):
Oh, you want to make a haptic engine out of this?
EW (00:58:15):
Yes. I want to make a haptic, a heartbeat for-
CB (00:58:19):
I think you can. You can do it with the 12 layer PCB coil. Depends on how much you want it to- How strong you would like the vibrations to be.
EW (00:58:34):
How come you did not make this about 20 years ago when I wanted to put it on a T-shirt?
CB (00:58:41):
<laugh> I was eight.
EW (00:58:43):
Can you go back in time and make this? <laugh> You were eight. <laugh>
CW (00:58:50):
To be fair, I mean the physics of this was there 20 years ago.
EW (00:58:55):
Yes.
CB (00:58:55):
Yeah.
EW (00:58:55):
But those little coin cell motors never worked the way I wanted them to. Do you think this can be a replacement for some of the haptic motors?
CB (00:59:08):
Hmm-mm. I am not completely sure because I have some experience using haptic motors, and some of them produce a large vibration. My goal was never to create a haptic motor. I just wanted to use this for robotic applications and other stuff. But I think if tweaked appropriately, the 12 layer coil can be used as a haptic device as well.
EW (00:59:49):
Well, if I get it working, I will let you know.
CB (00:59:52):
<laugh>
EW (00:59:54):
Carl, it has been really wonderful to talk to you. Do you have any thoughts you would like to leave us with?
CB (01:00:00):
No. It has been really fun talking to you guys. So thank you so much for having me on the show.
EW (01:00:07):
Our guest has been Carl Bugeja, Electrical Engineer and YouTube Creator. His channel is CarlBugeja, which is C A R L B U G E J A, and of course there will be a link in the show notes.
CW (01:00:25):
Thanks, Carl.
CB (01:00:25):
Thank you so much.
EW (01:00:28):
Thank you to Christopher for producing and co-hosting. Thank you to DJ at RIT for recommending Carl. Of course, I love all the YouTube origami videos you send me. For everyone. Oh yeah, thank you to our Patreon listener Slack group for questions. And thank you for listening. You can always contact us at show@embedded.fm or hit the contact link on embedded.fm. And now a quote to leave you with, from Akira Yoshizawa, a pretty famous origami guy.
(01:01:02):
"My origami creations, in accordance with the laws of nature, require the use of geometry, science, and physics. They also encompass religion, philosophy, and biochemistry. Overall, I want you to discover the joy of creation by your own hand. The possibility of creation from paper is infinite."