Parker's pinball controller has gone gold! Revision 3 is being fabricated! Stephen then explores the softness factor of diodes and the SSPS returns?
Parker and Stephen just talk about Projects... Just Projects for 30 minutes.
Parker and Stephen talk about the Analog board of the SSPS and clickbait engineering article titles.
Figure 1: Servos for the SAIM
Figure 2: MacroFabs’ My200 Pick and Place. Is this a robot or just a automation machine?
Parker is an Electrical Engineer with backgrounds in Embedded System Design and Digital Signal Processing. He got his start in 2005 by hacking Nintendo consoles into portable gaming units. The following year he designed and produced an Atari 2600 video mod to allow the Atari to display a crisp, RF fuzz free picture on newer TVs. Over a thousand Atari video mods where produced by Parker from 2006 to 2011 and the mod is still made by other enthusiasts in the Atari community.
In 2006, Parker enrolled at The University of Texas at Austin as a Petroleum Engineer. After realizing electronics was his passion he switched majors in 2007 to Electrical and Computer Engineering. Following his previous background in making the Atari 2600 video mod, Parker decided to take more board layout classes and circuit design classes. Other areas of study include robotics, microcontroller theory and design, FPGA development with VHDL and Verilog, and image and signal processing with DSPs. In 2010, Parker won a Ti sponsored Launchpad programming and design contest that was held by the IEEE CS chapter at the University. Parker graduated with a BS in Electrical and Computer Engineering in the Spring of 2012.
In the Summer of 2012, Parker was hired on as an Electrical Engineer at Dynamic Perception to design and prototype new electronic products. Here, Parker learned about full product development cycles and honed his board layout skills. Seeing the difficulties in managing operations and FCC/CE compliance testing, Parker thought there had to be a better way for small electronic companies to get their product out in customer's hands.
Parker also runs the blog, longhornengineer.com, where he posts his personal projects, technical guides, and appnotes about board layout design and components.
Stephen Kraig began his electronics career by building musical oriented circuits in 2003. Stephen is an avid guitar player and, in his down time, manufactures audio electronics including guitar amplifiers, pedals, and pro audio gear. Stephen graduated with a BS in Electrical Engineering from Texas A&M University.
Special thanks to whixr over at Tymkrs for the intro and outro!
Hello, and welcome to the macro FEHB engineering podcast. I'm your host, Steven Craig.
And I'm Parker Dolman. That was pretty good. Steven. I got it. Yeah, changing it up. Yep. Changing up. So this is episode nine. Wait,
no, no, what we meet where I'm supposed to be? Episode nine. Oh, well, this is episode nine. So what have you been doing this week?
So we've been working more on the super simple power supply. Okay. Bout ready to actually lay traces down? Right. Really close. Really close? There's a full on simulation, the analog front end?
That's correct. Well, the analog back end?
That's right. Yeah, it's the back end, right? Well, I guess it's the whole end. The entire the whole analog section is is pretty much up in multi multi sim, we were able to stuff the entire thing into
a single simulation. If you are having
issues earlier this week, we're running out of parts with the free version multi sim,
yet multi sim blew, the one that you can download from Mouser has a part maximum of 65. And I needed like two diodes, just two more diodes. So I had to go in and start hacking off some capacitors that they work for. For the real world applications for the simulation. They're probably not necessary. Yeah, just so I can add some more parts.
And and then you're into an issue with it was true. Was it? Computational complexity was too high?
Yeah. So it's, it's something to do with how spice takes its time frames, individual chunks of time for simulating. Its time base was too large. It's kind of cool, though, because multi sim has its own analysis where it will just keep changing variables and attempting to get a result works until it works, or it doesn't.
What was it called? I can't remember what it popped up on the screen. The first thing it does when you do the error checking.
Oh, yeah, no, the very first thing it does is it just does it again. Yeah, does it again. I think I think the term was something like confirm error analysis or something like that. So you just try it again.
Yeah, exactly. Yeah. I think I think that's what happens when, when you when you go on like your your computer, and you have network issues, and he goes trying to solve network issues in Windows, it doesn't actually do anything. So just wait.
Well, it's kind of the digital equivalent of tech support, saying have you tried turning it off and turning it back on?
You know, a little side note on that is last week, I was actually having internet issues. And I was on tech support with Comcast. And I was like, Oh, God, this is gonna be like three hours and I'm you know, talking to your computer has the time. Yeah, it was actually not that bad. Really, actually, the person I was talking to actually asked what I've already tried. Really, so I got the skip like all the BS. Have you tried turning it on and off again? Or like unplugging the coax cable and plugging it back in? Like all that stuff? I got the skip it. Wow. He went straight to basically you know that thing where they go, Oh, we're gonna try remote turning it on and off base. I gotta jump right there.
So you skip 3040 minutes of Yes. I've tried this. Yeah,
exactly. Yeah. So that was, I guess tech sports getting better.
Yeah. So yeah, actually, uh, and last week on the last podcast, we we did the live design on the SSPs. Yes. So we kind of translated that over. And as some of it so far, some of it Yeah. For at least for the way we were handling the digital communicating to the analog side. Yes. And last week, I believe that the conclusion we came up with was a zero to five volt control signal, correct. Were two and a half full was technically equal to zero volts. Yes. Well, it dawned on me this week, as I was working on the simulation, we can actually do a much better job if we ended up using two separate DAX and each one is zero to five volts. And each one controls a positive side and a negative side. Yes. So one DAC will control the voltage output from zero up to the maximum voltage and the other one controls it from zero down to the negative maximum.
Yeah, we just have to make sure not to run them both at the same time.
Technically, if you run them both at the same time, just get the difference. You get the difference, right.
So it's not a bad thing. It's just unwanted consequences. Well, Have a I guess
it just requires a little bit more dancing on the digital side? Or the software side?
Well, no, it's just got set up your tables right? When what you're sending out to your DAC. So it's not too big of a deal,
right now, it should be easy. But the cool thing is, if we use 16 bit, Dax will technically get 16 bits from zero up to our maximum and 16 bits from zero down to a minimum. So it's kind of like a 32 bit from all the way max to minimum
kind. That's kind of like, like video games from like, the late 80s. You know, it's got it's got we're basically you're making like a blast process, processor DAC, super power supply,
right? The cool thing is, though, if you look at it, you end up with half a millivolt of output resolution capability. In terms of pen on paper, I know it won't be that in reality,
we can just call it 16 by two bits. That's what it really is. 18 bits.
No, no. 16 bytes. Yes. I like that. I like that. That's
cool. Super bit processing. We'll come up with some silly acronym blast bit processing last bit processing
just like Sega Genesis, right? Sega does what Nintendo?
Yeah, they don't yeah. Um, and there's a little note here that says paneling, cancels wires. So what's that all about? But I think guards with the multi sim. Wait, say that one more time. Panning cancels wiring.
Oh, okay. So yeah, I think we've mentioned multi sim a few times on the podcast, we've pretty much had positive things to say about it. But I've found one thing that is just absolutely aggravating. And just, I have no idea what the designers were thinking when they when they implemented this feature, I'm not sure if it counts is that. So if you're laying down a wire on your simulation, and you middle mouse button, click to pan around, it cancels out, placing your wire,
you know, it's probably harkens back to the old days of basically, it's command base. And so when you're doing your wire, you probably use it runs the command wire. And if you're in the middle of your command, and then execute another command, ie move, yeah, it probably cancels out any previous command that you're doing. Probably AutoCAD 2000, which is where I learned all my drafting on was like that. Yeah, yeah, it probably harkens back to how old Multisim actually is.
Probably the weird thing is zooming doesn't affect it. Because zooming is probably just a redraw.
Yes, usually zooming is just a redraw of the, the GUI.
Right? Right. So it's not an override command. No. And you know, actually, funny enough. Another thing, when I'm working in any software program, that's for schematic drawing, or PCB or anything like that, when when you do a copy, paste, if I paste, I want where, once I do the paste, I want it to be on my cursor, that the new item that I'm pasting, yes, it should be on my cursor, so I can place it wherever did praise doesn't do that. If you copy paste it, paste whatever item you're pasting on a fixed offset away from the item that you copied.
Yeah, I don't know. Is that offset dynamic? Or is it always the same offset no matter what, it's
always the same offset?
That's really weird.
Yeah. And it gets even more confusing if you are having schematic with multiple sheets, and you're copying from one sheet to another sheet, because it applies that offset to the new sheet. So if you have a new sheet that has stuff all over it, and you copy from an old one, you have to go hunting for where it
is and hopefully select the stuff that you actually need to select instead of the old stuff. Is that a little mic trouble? Ah, and then I've been working on the the IoT side a bit, right with the ESP 8266 Wi Fi module. So to do that, I'm basically building a small little dev board based off my propeller or parallax propeller development stick, little platform I designed. And so the new one is called the bit flicker because it's it flicks bits into the air. Wi Fi. Yeah. Basically, it's a prop dipstick with a Wi Fi module glued onto it. Yeah. I should get that done by tomorrow, when this episode comes out, and then I'll get that ordered. And so I can actually start writing the code for that. And actually, I have a friend who's porting this software called Blink, which is be a Why NK? Okay, which is a
like an Arduino kind of ish.
Not really. It's it uses Arduino. Okay. But basically it it. It basically combines your embedded world with your your IoT world with a very easy to use framework, it, basically you attach the libraries and it just works.
Okay. And it just makes your life a lot easier.
Yeah. And then the main thing too is it handles the framework for the app side on your on your mobile device, too. Oh, so it handles all that. And basically, it's like drag drop, where you want buttons and stuff. Oh, that's awesome. And I think it'd be really good for us to do that. Instead of trying to have to write us a app from scratch. Yeah, please.
Yeah. I've never tried that before. Yeah, that sounds sounds rough.
So my friend, his name is Roy. He's working on porting that library over to the parallax propeller. I think he's almost done. I think he actually has a talking right now. He just needs to flush out the library. So it actually talks with the the interface. Cool. And like all open source projects. It's the documentation is very weak. Basically, he has to go into the Arduino Libraries that they have and reverse engineer everything.
Oh, yeah. He's got a core it all out. Yeah.
But he got it working. There's a lot of stuff in there. That's really what he says. It's really weird. Like there's a, for each message that you send out, you chose to increment the counter for the message, but it doesn't, which is that's pretty standard. Sure. But it doesn't tell you what happens if you don't do that. Or if you like, go backwards, or the base doesn't tell you what happens if you do it incorrectly. You know,
correct me if I'm wrong, but but in the Arduino world, there's a lot of that where it just, this just works. Just trust it. Don't don't start questioning it.
You have to do X, Y and Z in the right order for it to work correctly. Yeah. But if you do X, Y, Z. Oh, they're gonna hit the fan. Yeah, they'll do it. Cool. Yeah.
So looks like the servos arrived for the XY platform. Yeah, the
really, really long wait on those servers have finally, I guess finally paid off. They're really smoothly run ball in the demo mode. At least they run really smooth, and they're very quiet. So I'm happy so far. They seem to have a lot of torque. We'll see. When I put the rig together tomorrow, if it actually moves. Yeah, it'll move. Yeah, don't move. It's got to get plenty of torque. I funny thing about that, too, is because I came on a on the slow boat from China. And they were supposed to arrive like a week and a half ago. And I emailed the guy and of course, they came in that day. I emailed him. Oh, of course. And he said, I'll mail them out immediately.
That just means they're sitting on a shelf for two weeks. Yes.
He forgot to ship them. Anyways, we have now the really nice build quality is pretty good. The bearings are smooth. The stickers are applied straight.
Is this the the Parker quality?
Or the stick? That's what bugs me a lot when you if you buy something and the sticker is not even put on straights. Like do they actually even care about putting in the device together? If they can't get the sticker on straight?
Yeah. You know, the, the funny one is when you get something and it's got that reflective gold QC sticker on it. Oh, yeah, you know, instantaneously. It's like, okay, this thing is probably not that great. Those stickers on it.
Um, or QC stickers on stuff that's like doesn't even need QC. Yeah, like plastic pieces and stuff. I completely forgot where I was gonna go with that QC. Oh, yeah. You're just talking about the hologram? QC stickers. Yeah, it's really funny when when you see like, it's only authentic when it has a hologram sticker but I'm like, you can go to LA Express and buy whatever hologram sticker you want for like cheap, right?
But it gives you that warm fuzzy feeling that somebody looked at it and put a sticker on it.
Well, I'm talking about counterfeit stuff.
Same thing Yeah.
Just because as a hologram doesn't mean it's legit. Yeah.
All right. Well, on to the rapid fire opinion section. Yeah,
the RFO fo Okay, so we got the transistor wars. vacuum tube Strikes Back. Ooh. So this is an article that you found earlier this week.
Actually, a buddy of mine sent sent me this. And he goes, Do you think you can use this in a in a guitar amp? So that's how we stumbled upon this. Yeah. So
it's a, a new type of transistor that uses it's similar to vacuum tubes and the fact that the only thing that's the same that there's a vacuum inside of the device, sure. But instead of having a normal silicon gate substrate in your, your MOSFET, and transistor, it basically just has a vacuum. That's not really a vacuum. It's filled with helium. There's nothing vacuum about it seems.
Yeah. Well, okay. Yes. If there's if they call it a vacuum, because helium, the the atoms are so sparsely populated, that the chance of an electron actually being impacted as traveling through helium is very minimal. So they get away with calling it a vacuum bag, get
away with it. Yeah. The main thing to get take away from it is how fast these things can go. Yeah. Because since you don't have the since electrons are going through essentially a vacuum, they can stop and start really quickly. Whereas you if you have, let's say, a substrate for the gates, that's they put a gallium on arsenic. And it's I can't remember which one it is in effects. But basically, it takes time to turn it off and time to turn it back on. Right? Because you have the saturates the substrate with, with electrons basically. Is Mike fell back down again.
I need to I need to learn how to tighten these things. You turn it to the right, turn it righty. tighty. Right. Yeah, right. I got it. It's all good.
So I guess what was essentially since your substrate doesn't exist anymore. Your time on and off is very short. Right? And they're saying 460 gigahertz? That's fast. Yeah, almost half a terahertz. I wonder what's the fastest scope you can get?
It's got to be in what? Maybe 10 gigahertz? Maybe maybe a little bit higher than that?
Because I bet well, they got to have some kind of a frequency counter that can go that high, at least.
Well, okay. So if you're talking about like, a digital scope, probably not an analog scope might be able to get up there. Yeah. It did. The thing that is just crazy, in my mind is, in what I've never really dealt with the above gigahertz range electronics, other than just reading about them, but I mean, a few Pico farads of capacitance just completely ruins your whole design. How do you even get to 460? Giga?
Are they're saying they can get these things up to a terahertz. Wow. Which is just insane. speed wise, it's it's almost unimaginable. Thinking is something that can go that fast. And how it's one thing to go that fast, but it's also how do you make sure you're going that fast?
Well, it seems like chicken and egg thing, you have to create this product. So you can use it to test
not you have to make a better version of what you make. Yes, you have to have, you know, Nyquist because you have to go at least double. So if they make a one terahertz, you need at least a two terahertz scope. Yeah. Or basically something that can sample two terahertz to see that signal. Yeah, yeah.
You know, there's probably someone out there who's in like, some defense contract or something like that. And they think that a terahertz is slow or something. They have some like, super transistor that, yeah, some crazy. Yeah. It's
I wonder if you will get to the point where do bleeding edge technology, you don't actually know if it actually is functioning correctly, unless the outcome is correct. Of what your device does? Hmm. Yeah, I mean, two plus two equals four. And if it does that correctly, you don't actually have to know if it actually looks good on the scope. You just
thumbs up. Let's go. Yeah. That day. I don't know. I mean, technically, in terms of the mathematics, there's not an upper limit to this kind of stuff. I mean, you can just keep going higher in frequency.
What's the Well, eventually you would just run into speed of light issues when you
Yeah, right. If you're, yeah, there's, I suppose there is some kind of there would be an upper limit. There's a boundary somewhere Yeah, somewhere, but I think I think all of our materials are going to Will they, they they will dictate that well, before we do speed of light. Yeah,
it's it's similar to the maximum heat problem. What is there a maximum temperature you can reach?
Well, it's kind of like the same thing in the opposite way. Is there technically a minimum? I mean, obviously, there's zero Kelvin. But the word reach can you actually yet
they actually reach there? I think this is getting to the point where it's beyond our degrees. Yeah. I didn't take a theoretical physics class.
No, no, no, I was turn ON light bulbs and
go back to blinking LEDs. Someday you'll be able to blink an LED at 460 gigahertz. Ah, and then the Atlantic had a really cool article this week about what is really a robots. And I thought that was a really cool question in the article goes into the history of what of robotics and some more of the more modern stuff, you know, like, autonomous cars and that kind of stuff. Yeah. So what these what the article states is, most people say anything that's automatic, is a robots, like the ATM or robots, like used as a broad word. Like on autonomous cars a robot Roombas. Robots. The arms in a in a car factory or robots? See, Threepio is a robot see three peers are robots. I think he's more robot ish than these other things, though.
Oh, he's an he's an Android.
He's an Android. Basically, they use robot to describe anything that's become automated
in the world. Okay. Seems it seems like a really ambiguous Yeah, that seems it is an ambiguous
I mean, by that standard, pretty much anything like your thermostat on the wall is a robots.
If you get a control McDonald's Happy Meal that blinks?
It's a robot. Yeah, by the standard. Yeah, but it does something automatically. Yeah. Which I guess you could say, crocs are automatic shoes, because you don't have to tie them?
Oh, God, no.
But it goes into other things where like, if you start to classify a robot as something that does something automatically, with out humans intervention, which makes more sense of what a robot could be?
It does, but But don't you have to interact with the robot with the fact that you design it. So you're setting forth all the interaction to correct beforehand? Yeah, so
this would this would be stuff. So that's of those automatic, like the Roomba you turned us turn it on, but someone actually programmed it to clean your floor. Mm hmm. thermostats, that kind of stuff. Stuff that's that the article says not robots like RC, a lot of people will say like, a drone is a robot, but it's actually flown by human. Sure. All it does is just, you know, take like it like a quadcopter The only thing the quadcopter really does is make sure you don't flip it over.
It keeps it level, but if it's doing that automatically,
but does that port on it, but you dictate where it goes. So it's the human has more control over it than it does because you can make it crash, it won't prevent itself from crashing. Hmm,
you know, a definition just popped into my head that I think might might shed some light on this. Think of it this way, a mechanism, a device or an apparatus that executes instructions during its normal operation without human intervention.
Yeah, I think that's a better description. Some people classify it as sense Think Act. Okay, it has to do those three things without human intervention.
Does it have to do all three? Well,
I guess it can sense and decide not to do something so it sends thought and didn't act which will be the same thing.
Well, what if it What if it thinks and acts without sense
what would be an example of that?
I'm not sure. Why would it Yeah, I can't think of anything that we you know what maybe maybe that's why it needs to be all three. Yeah, you need it
needs to be able to Have something that changed for to think about it.
So actually, we'll hang on. What about stripping way down low? What if you think of just like a like a timer? A timer has to? It's not sensing anything, but it does have to change a state? Well,
I think it's so you're so sensing that timer rolling over,
I guess, maybe. I mean, it's thinking and it's acting. But I guess I guess the sensing would be some kind of time input to
it. Yeah. I, I'd like to go even farther down down the hole on this robot thing in basically the when you can start calling the device not in it anymore. And you start it starts to do start to personify it, even if it's a robot like a Roomba. So you give them names, your once you give it a name or you give it you can call it a he or she or whatever. Besides it's it. That's when it becomes a robot robots, like an Android or a something that it doesn't have to have feelings or anything but the fact that how it acts, it crosses that a Turing test kind of thing. And your brain. Well, okay,
so let's just say the the ASIMO robot, if they've called in it, instead of ASIMO, according to your definition, that would not be a robot. Well,
when you watch it move. Well, the thing about that robot is it doesn't really move like a organic creature. It's really stiff. We're talking about the Honda one, right? Yeah, yes. Yeah. It's kind of that is still like, yeah, I guess it is a robot.
Well, I mean, see, Threepio is stiff. And he was a robot.
Yeah, but his personality is there. That he has a person he actually has a personality.
In the movies, it's like, like the Boston Dynamics. They're They're robots they make Oh, the like the the mule? Yeah, the mule and stuff, which are like, they are scary looking like a deer. How it moves around. They actually have a bipedal like almost Terminator style one now. That doesn't mean the cables hooked up to it, they can walk around the sky,
that's only a couple years away. Well,
we had the they hook up Boston Dynamics with the Microsoft, Twitter AI, the world's gonna end because there'll be a Hitler loving sex robot that can move around
and carry 300 pounds.
Ah, oh, yeah. So speaking of the boss dynamics is a really cool video, I'll put I'll put a link in the description of someone dubbed over the Boston Dynamics videos with with obscene words. It's really funny. I need to see that. I'll put it I'll put it in the blog. Awesome.
So have you heard of the new gallium oxide substrate? transistors?
Ah, no, I assume it has gallium and it's attached to oxygen.
That's actually probably a pretty good guess. So So with as we go further in material science, we're finding that silicon just doesn't really cut it. No, it doesn't. Well, silicon by itself, single crystal silicon. It's great. But it by today's standards, it's pretty slow. It has high resistance. It has not so great thermal characteristics. There's just a lot of things that that prevent us from from using it in high speed technology or high power technology.
Yeah, that's why they they dope it like crazy now.
Yeah. Right. Right. Well, I read an article just earlier today about new gallium oxide substrates, which is, which is really cool. It's a It's we're starting to get to the point where we're getting fringe materials, where you now pick your material that is really hyper specific for your application. And it does that application very, very well
instead of silicone which is very broad in terms of semiconductors.
Yeah, exactly. You can you can kind of throw silicone and anything it'll do the job. Yeah. But you know, if you're going for high speed you want this if you're going for LED applications, you want this. Well, gallium oxide is actually fantastic for high powered
applications. Okay. wasn't good at high powered. So,
it has a an electron volt bandgap of five EV, which is significantly larger than then say gallium nitride or silicon carbide, which both are used in power applications and really the electron volt bandgap. Basically, all it is is a measure of how much energy it takes to excite an electron within the crystal. And you would think that it would be better to have a lower bandgap voltage. But in power applications, if you have a higher bandgap voltage, you can actually have much higher isolation. And you actually are able to use much thinner substrates, which has an incredibly low resistance. So, if you're talking about, say a MOSFET, your resistance, you can get that next to nothing now,
yeah, since you're you're going through a lot thinner material. Yeah. Yeah. Which is that intro because we were talking about gallium nitride or gallium nitrate resistors for
the switching transistor. Yeah.
Switching transistors for the Google little box channel challenge, right? We go. So basically, that thing's obsolete already. Well, well, once they build something out a transistor out of this stuff,
right, right. So so this stuff, what's what's cool about it is, I mean, it's it's shining characteristic is its high bandgap voltage. But what really is weak is a thermal conductivity. It's something like four times less than silicon. But they found a really cool solution to it. Apparently, in LED technology, using substrates of Sapphire is really common. And sapphires actually easy to grow in a furnace. Yes. And it has excellent thermal conductivity. So they're actually making a substrate of Sapphire, and then growing gallium oxide on top of that, so they're actually able to alleviate the thermal issues by putting this on top of Safar. And so you get the best of both worlds, you could or three worlds, you get really low on resistance, you get really high isolation, and you get really low thermal conductivity. It's basically the best switcher you can
get. Yeah, um, I want to see that that the bond interface on their electron microscope between the Sapphire and gallium oxide,
you know, and actually, I was reading the article. And what's interesting is, they were mentioning something about the fact that the bond is actually not really strong between the gallium oxide. Yeah, that's
all I want to say. Because they're not gonna be the same size in the lattice.
And I don't know the mechanism behind it. I didn't I didn't fully understand it, but apparently, oops. Yeah. loctite. Five, a 13495495. That's it.
There's a lot of that in the shop.
But but but apparently, the fact that it doesn't bond well, actually age and its thermal conductivity, which seems counterintuitive to me. Yeah, it does. But But apparently, some smart person out there has found that out.
Yeah, you think the closer which would be a direct connection would allow the heat transfer to be better? It?
That's what logical? A logical path would take you. Yeah. You know, the heat in terms of electron energy needing to move from one place to another? You would think a really rigid contact would would help that. But apparently, that's not the case. Apparently not.
It's cool. Yeah. Pretty cool. Hopefully, we see some effects with this stuff in it. Probably two years. He just takes about two years for this kind of stuff to roll in.
Yeah, yeah. Yeah. And they'll probably be $60 A transistor when they first start out.
Well, if it's as good as it says, And the hurdle, the manufacturing stuff, you know, quickly, yeah. It could be in every fit.
Yeah, well, and apparently, according to the article, gallium oxide is not expensive to make. No, I the all the cost up front would be the fact that it's just not widely used. Correct. And like I said, the Sapphire substrate is already widely used in the semiconductor industry. That's already cheap.
Yeah. That's actually speaking about this manufacturing one jump back to the the vacuum transistor. Yeah. Is it could be built and current CMOS, fabs using the same technology. They build accelerometers with, um, because accelerometers actually have to be have a minimum manners. Yeah. They have a moving part in them. Right.
MEMS? Yeah, yeah. No micro electronic. Something something Yeah. So you know, it's physical move. Yeah,
device so they have a vacuum or near vacuum that enables that part to not destroy itself. And so they actually can use that technology to build these transistors. Wow, that's cool. Yeah. So I'm thinking we, I wonder if we will see first those are the gallium oxide part of the gallium oxide received first?
Probably because they probably have more application direct application.
Yeah, direct it's all it is. You can probably just swap out you know, your your gallium nitrate FETs for these guys, and just get a percentage increase.
Yeah, well, I'm sure it's not as easy as this. But yeah, I mean, I wouldn't be surprised if you could just slap together a dye and throw it in a to 220 package and now you got a gallium oxide transistor. And let's
just go down to down the shop and bust that one out.
Well, I think that'll do it for for this podcast. I'm your host, Stephen Craig.
And I'm Parker Domon. Have a
good day. Later, guys.
Parker and Stephen just talk about Projects... Just Projects for 30 minutes.
Parker's pinball controller has gone gold! Revision 3 is being fabricated! Stephen then explores the softness factor of diodes and the SSPS returns?
Parker and Stephen talk about the Analog board of the SSPS and clickbait engineering article titles.