MEP EP #161: Kalashinikaze: The Right to Fly Arms

Podcast Transcript

Host 3     00:10

Hello and welcome to the macro fab engineering podcast. We are your host, Stephen Craig and Parker Dolman. This is episode 161.

Host 2     00:20

A quick announcement before we jump into the podcast, Matt crab will be at South by Southwest in Austin, Texas this year, we are teaming up with particle.io to put together a hardware happy hour. It will take place on Friday, March 8, which is in like, two weeks or is it this week? It's next weekend? Yes. That's good, because I have to be there. From 4pm to 8pm. At the super cool jester King brewery. Check the show notes for full details and to RSVP. Join us for food and refreshments and network with fellow hardware nerds and kick off your South by Southwest weekend in style

Host 3     00:56

Is jester King. Is it a jester King brewery in the town or is it the official one that's a little bit outside of town?

Host 2     01:04

It's the one that's a little bit outside town.

Host 3     01:05

Okay, that one is awesome. Whatever. I can't remember the name of it, but there's a little pizza place that they have on the on the site. That pizza is excellent.

Host 1     01:14

They're providing the fruit. Ooh,

Host 3     01:17

Nice. Nice. Yeah. Jester king does a lot of like, stouts and they have ours our room and stuff. Yeah, they have excellent beers.

Host 2     01:26

And it's um, hopefully the weather is really good. Because it's that place is a really nice just the hanging outside under like the trees and stuff. It's very Hill Country. II.

Host 3     01:38

So like rocky ground and dead trees?

Host 1     01:41

Yes. A partial desert part trees.

Host 3     01:45

Yeah. Yeah. But it's, it's no, you're nowhere near the summer. So it's gonna be gorgeous, you know? Oh,

Host 2     01:50

Yeah, it's gonna be nice unless it rains. Okay, so I've been working on the parallax propeller dev stick. The Type C edition is what I've been calling it now. So it has like its own designation, I guess to differentiate from the previous versions Red Sea, like the original one that I showed you. weeks ago, like the original green one I had. That was the original prop dipstick I designed. And then I made the Redux version, which is like, by basically started macro fab, I made the Redux version. And now so this is type C addition. So the routing is almost complete. Of all the new stuff. I moved from, oh, 603204 twos to kind of compact design, I kind of want to reduce the width, it's like 1.1 inches wide. So 101,100 mils wide, okay, I want to like get that like 900. Round, it should be able to now. And I moved the qf P parallax propeller to the QFN, which I can't remember what the actual size differences, but it's like, it almost looks like it's 75% smaller, because I think it's like point five pitch 2.4 pitch, I think. Okay, so just everything's getting smaller. Yeah, everything is getting small on it. And then I changed the voltage regulator, from the NCP 1117 to eight through the AP 7361. C, both of them are the 3.3 volt variants. So, Steven, why would I change the voltage regulator?

Host 3     03:20

I was just about to say, like, I'm looking at our show notes. And beneath that one is just the question why? I'm wondering, was that supposed to be me that asked the question, why did it change? No. So why would you change that regulator? Well, I mean, frankly, I don't I don't know the difference. I don't. I don't know either of those regulators. One quick question. I know you had a favorite 3.3 volt regulator that you use for a while and it had pretty good specs and low noise and it was it the line regulation was good. Was that is or is that one of those two?

Host 1     03:58

No, it is not. Okay. Okay. So

Host 3     04:00

This is these are two separate ones. Okay. No, no.

Host 2     04:02

So the reason why I switched is because of the next feature, which was adding lithium battery charging support off one cell, I guess

Host 3     04:13

I should have just read. No.

Host 2     04:17

And so yeah, so I'm adding a single cell with the battery charger, on on the device on the plastic. And so you can have a you know, 3.7 volt battery on it. Now, to make that work, you need a ldeo or a switching regulator that has a very low dropouts, then you need like a s ldeo. A super low drop.

Host 3     04:42

I was about to say if you're going from three, seven to three, three, you have to have ultra high low. Yeah, regulate.

Host 2     04:53

We coined that term, that term here first now, ultra high low dropout. Yeah, analog

Host 3     04:59

Device. He says if you want to use that, you know, 10 bucks our way. 10 bucks, each of us,

Host 2     05:05

Each of us, yes. And so the cool thing about this new, it's not new, it's made by diodes Inc, the AP 7361 C is at 3.3 volts at about 300 milliamps, which is about what? Most things I use this thing for, it only has a drop out of point one four volts. Wow, that's tiny. Yeah, that's that Max to typicals like point nine or point oh nine, yeah,

Host 3     05:33

Typical is a lot higher.

Host 2     05:39

Whereas, like, the NCP 1117 is like 1.1 volts, which is kind of normal for lto. And so with, I had to switch over to that, to get to, so that a 3.7 volt battery can actually power the thing without, you know, the voltage regulator dropping out. And so I'm looking at for the lithium battery charging, I'm looking at the TI p q 24075. Mainly because I've already designed that part for project like, four years ago,

Host 3     06:09

I think we've talked about this multiple times, where you just look at your backlog and be like, well, I got that,

Host 2     06:15

I got that priority design. So I don't know, if I'm gonna change that yet. I'm gonna see basically, this T I have anything new in that family. And if it's better, it probably will be like better efficiency, that kind of stuff. But I mean, it's a, that that's a battery charger that's designed to be powered off of USB. For single cell applications. So it's, it's a pretty good part. And it works really well, if I recall with FTDI parts, and like their their power, the battery monitoring part of that so like, it can detect basically the FTDI like 230 X can detect it's been hooked up to a charger, like a dedicated charger. So you can pull more more juice, stuff like that

Host 3     07:00

Got that's cool.

Host 2     07:03

And then, so I haven't finished that running yet spa post pictures of how far I am into that project. And then I'm in testing my USB type C example boards. They worked. So I had the 80 Mega one. Somewhere around here. I think it's in a box. But it was working. I it was working great on the computer, all that stuff. And then I'm like a, let's see if it worked on the phone. Because the my Pixel has, you know, type C plugged it in, and it powered up. Hmm. And I actually use like the John, not John. J. Jason. Jason. Yes, yes. Hello. Yeah. I used his USB explorer and plug that into the phone. And the phone actually does point five. And that point five, five volts at point nine amps is what it can supply over USB power, wow. Delivery, which is like, that's like apparently, according to him, he actually commented in our Slack channel, I said that that's like the minimal that superspeed should provide. Okay, so Okay, so basically, it's providing the minimal, it can Yeah. And so so it actually will power up the board. And the thing is, it doesn't power it up immediately. It waits for a bit, I guess, as the phone's trying to figure out what just got plugged into it. And then reads the resistors and goes, Okay, it's a device, a dumb device, I'm going to provide power to it. So that works. But the problem is, I can't talk to it. Like, I can't open up a comm port, or I can't emulate a keyboard, which works fine on the computer, right? So I'm like, Okay, why is that? For I can't find like my AVR ISP programmer, so I can't override the bootloader. So I've been using like Atmel flip to basically like compile the code in Arduino. Find the dot hex file and like my temporary file a folder on my computer, and then use Atmel flip to upload that. And so that works, but it's kind of annoying. Once I find my AVR ISP programmer, I'll actually like put the arduino bootloader on it. But for some reason that's not working on the phone. And so I'm like, Okay, what's going on here? Let me try to figure something else out. So I took an Arduino Leonardo. Put the same code on it, but this time I could like compile it and upload it directly through Arduino since it has the arduino bootloader on it. And then I use a USB type A to type c converter board. So it has like type C plug on it and then a socket for Type A. So using the USB micro to USB, a normal cable plugged into my phone, and that works. And so for some reason, using the dongle worked, and so I'm like okay, well so special about dongle and so I plugged it into my USB type C, I have a USB type C like breakout board, so it breaks all the pins out. And so I started measuring all the pins. All it looks normal, except the CC one cc pins is pulled down with 5.1k, which is what you'd expect, but the other one is left floating. And so I'm like okay, why that's not even in the spec, the spec says the pull boat down. But it's only that connector is that the dongle is only connecting one side of the USB 2.0 data lines, because it has two pairs so it can handle the flipping right motion. But on the dongle only one was connected. So I'm like, okay, that's weird. And it's, it's on the same side of so like a five position a five is the CC pin on the a bank and there's a B bank. So a five and then a six and a seven are the data minus and data plus for it. And then on the other side is B five, b six B seven. So but only one of those is hooked up. So like a five, a six, a seven. And if you flip it over a course it's B five, b six, B seven. So why does that make any difference? Right? Yeah, then what I had, so I like cut the trace and left one of the CC pins floating and they cut the traces that connected the TX RX lines. And that takes our exercise that's

Host 2     11:32

Yeah, protocol, d minus d plus cut those lines. Plug that in, still didn't work. I'm like what's going on here. And so then I took the breakout board that USB type C breakout board and connected that manually to the Leonardo board Arduino Leonardo board. And that works on like that hardware implementation is what I put on this board and the board is not working. So I think for some reason, the default Leonardo a default 18 Mega 32. You for for bootloader is not compatible with the pixel in terms of whatever driver it needs. Whereas the Arduino one is, that's weird. That's the only thing I can think of. And I'm going to once I finally find my AVR ISP programmer, and make mine Arduino compatible with the bootloader I betcha it will just work. Hmm. Basically, I spent like, two full afternoons on the weekend, like pulling my hair out over this. And I think it's just because there's a bootloader firmware conflict some are going on. And like the startup sequence, yeah,

Host 3     12:42

Their startup communication is just Yeah. Which is

Host 2     12:46

Kind of frustrating, because I basically ruined my 18 Mega 3830 to you for board by like cutting a whole bunch of stuff. So I'm like, oh, man, I'm gonna use this for an article. It's gonna look ugly. And so I'm gonna have to like I basically want to re spin that board.

Host 3     13:01

Just just always pictures of the other side of the board.

Host 2     13:05

That's true. But yeah. Frustrating. Yeah. But hey, I think I found the solution. We'll know by the next podcast if it was or not. That's great. And the last project I've been kind of working on is that the wagons power mirror controller when that works correctly. I don't know if you've seen it, Steven.

Host 3     13:26

I have not. The looks like the back of your room is covered in plastic wrap. So

Host 2     13:33

Yeah, they've been doing some drywall work over here. But so I've got the board here. And I think we talked about this, like Mike Williams at macro fab made me some bats.

Host 3     13:44

Oh yeah. Mike Williams, he has a clickworkers lathe, right?

Host 2     13:48

Yes. And so he turned some little tiny aluminum toggle bats for these deep, the four way directional switches. Yeah, and I hooked one up kind of like manually to the like I tore my door panel apart and like hacked it into the the wiring harness and it works. It moves the mirrors up and down and left and right. And so next step on this project is to take the door panel back apart, and then measure the door panel. Like there's an opening there and the sheet metal measure that and measure the offset like how far back and you put the board and like Model A A brackets and then 3d print the bracket that this PCB will sit in, and then drill the holes out in the in the door card and wire it in. It should just work

Host 3     14:39

Nice. And those are just four way switches. You don't have any other stuff on that board right?

Host 2     14:45

Oh no. So the four way switches can only handle like, like 50 milliamps, okay. And those the motors can pull. Oh was like 150 milliamps okay. So I actually have motor drivers on the back Okay, I'll post I think I posted this schematic before, but it's got motor drivers and actually have an LTO. Because I couldn't find because I wanted to use really small motor controllers. So these are like SOC aid packages with like a thermal pad. But it's kind of weird looking. But I had to use an LTO on it to drop 12 volts to five volts, because that's what the IO on those motor controllers run at. Got it. Okay. But the actual motors are running at 12 volts. But for some reason, like you can't interface the, you know, the input pins and enable pins for some reason can't drive those with 12 volts. Just this particular motor controller.

Host 3     15:44

Got it? Okay. Very cool. But, yeah, so gotta get that installed.

Host 2     15:50

Yeah, that's probably going to be next weekend. I'll get that bracket all designed and printed up.

Host 3     15:57

You can get it 3d printed.

Host 2     15:58

Yeah, I mean, I got 3d printer and it can make any object I would dream to make. Is that going to be robust enough? It's to me out polycarbonate, so it should be okay. Okay, you're not gonna it's gonna be same thing that we made. You're the peristaltic peristaltic pump out of

Host 3     16:14

Excellent. Okay, so funnily enough, our, our listeners can't see this. But within reach, I have well now I'm dropping stuff all over my bench. I have a peristaltic pump that I purchased actually, from. I love it, the electronic store that's in Houston. We've talked about it a handful of times, EPO, they posted on Facebook, just an image they had like 15 peristaltic pumps. And I just, I have the phone number of the guy who owns the store. I just sent him a text and I was like peristaltic pump. Send me now. And so he actually yeah, we struck a deal. And he sent me one of these pumps, which is super cool. So yeah, sure, I have a 3d printed one, which is awesome. And I totally still want to use it. But I also have a laboratory grade peristaltic pump now, which frankly kind of sucks. Because it's it is a DC motor on the back. So I'm gonna have to rip that DC motor off and put a stepper on it

Host 2     17:15

To get what are you going to use that for because the tubing is not that large.

Host 3     17:18

So what's cool about this one is you could select whatever size tubing up to a certain level. This is a very thin tubing. This was used in a Gosh, this was a in a chemical engineering firm in Houston. And they went out of business and EPO got all the all the stumps. Yeah. So I looked it up before I bought this and you can get larger tubing. And this can deliver three gallons a minute at the fastest. And it has an accuracy of point one milliliter dosage. Oh, so yeah, if you want to if you want to get stupid, accurate, you can do that. Okay, so I've got a funny and

Host 2     18:01

Slightly so how do they do that accuracy with a DC motor?

Host 3     18:06

I don't know. i The thing is, okay, so this, this was sent to me, where it was obviously hacked out of something, there's wires hanging off of it. And there's a power board that has some of those. Some of those, like, you know, those transistor packages that they're meant to bolt up against a heatsink, and they have like 10 or 15, like angled pins that come out. Like, it looks like audio amplifiers, but they must be motor drivers. That's what I'm guessing. So, this this board is pretty simple. It has some high watt resistors. It has some big caps. It has one of those amplifier things but it has four little trim pots that say minimum speed, I our compensation, maximum speed and TQ limit. Who knows what any of that actually means. So there must be some kind of like timing circuit that

Host 2     18:59

I wonder if it's a optical encoder. Is there an encoder on it?

Host 3     19:05

No, there's not. It's literally just a DC motor with a transmission that goes into the heads. And there's no feedback that I can see.

Host 2     19:16

I mean, it could be doing EMF feedback. Geez, that's fancy, but that'd be kind of

Host 3     19:21

Yeah, kind of weird. Well, and this is the so this is still this is still manufactured. Let me see if I can get the part number. This is a where's it I got the customer number written on here? I don't see the part number written on here. I don't know I'll find it I had I had it somewhere. So funny, funny story about this. I bought this this pump and they shipped it up here. And I was playing with it on on the floor of my living room. And at the time I was watching like a nova documentary or something I don't I don't remember what it was basically I had that TV on and it was background noise. And they were talking about something about being healthy or something and human biology. And then they went into this, you know, sorry, this is a little gross. They went into fecal transplants, if you've ever heard of that, where like, you can transfer stuff from one person's body to another person's body and the live bacteria cultures can be beneficial. And there's studies going into it. And I heard that I looked up, and there was a video of a fecal transplant, at least just the equipment and there was this peristaltic pump. It was pumping poop.

Host 1     20:44

So I was like, okay, cool company, huh?

Host 3     20:46

Well, yeah, no. Okay. So I will buy all brand new, like swab it down with alcohol before I brew with this to get brand new to

Host 1     20:57

Sterilize everything without call. Oh, yeah.

Host 3     20:59

So I honestly Well, I mean, if it's in Houston, it was passing oil and gas. That's that's the only thing like gas. Yeah, yeah. peristaltic gas pump. Gosh, how did we get on that tangent? Oh, yeah, you're polycarbonate. Polycarbonate. I was just, it's just funny because it was sitting in the sitting right on top of one of my speakers here. You know, we

Host 2     21:21

Should do a video on that peristaltic pump that we did design.

Host 3     21:27

You know, okay. Because we built it works. Well, we built it. And it works. And even above and beyond that, after we tested it the first time around, we built that backcap I installed a stepper motor, and I wrote some some quick Arduino code. And I was like dosing, specific amounts. And I mean, it worked actually pretty well. The biggest flaw in the design was that it didn't pinch the tubing enough it, it works fine, at generally low RPM. But if you if you really juice it, then it starts to slip. And so I was going to actually just put a shim on the internal radius to just kind of crushed the tube a little bit more, but you know, and then I moved. Yeah, we should just build another one. Yeah, yeah. That was fun. That was 20 something hours of printing out for one piece. Yeah. Yeah. I think I still have that. Well, you probably still have the model for sure.

Host 2     22:27

Yeah. I saw the model somewhere. It's on my computer. Yeah. Okay. But so besides peristaltic pumps, that pump poo, but he'd been working on Steven.

Host 3     22:38

So I've had a handful of work projects, like the work I get paid to do projects that have been consuming my time. But in the in the kind of the side done that I've had recently, I have two quick little things. So Josh Roser, or Roz, he was our guest for the last Star Wars podcast. He's been building guitar amplifiers for a long time, he and I kind of started at the same time. And he, he's got a new design that he's working on. And I've been kind of coaching him through a handful of electronic stuff. And we wanted to look at a new form of tone controls, look at you know, bass, middle, treble kind of stuff. And for the longest time, there was a architecture that I wanted to build up and try and I've had all the parts sitting in a box somewhere, and I was like, You know what, I'm like, I was gonna do it all like knights make a board and stuff. It's like, there's no need. There's like 10 parts. So I did a point to point like, little solder job and pulled up my old high voltage power supply that we talked about a handful of podcasts ago, you know, the one that has like, blown up capacitors and stuff inside? Oh, yeah, it's

Host 1     23:47

So worth it. Yeah, oh, it

Host 3     23:48

Still works great. I fired it up. And there's just, I'm not going to touch it, it works fine. You know, I measured everything in the noise is great. So it's whatever I'm, I'll just take it till it fails, basically, and then fix it. But regardless, this, so this tone stack runs on more needs to be able to run on high voltage because I don't want to do voltage translation down to low and then have to amplify it back up. I want to be able to do my filtering at high voltage. And there's this really great part that's actually made by microchip. It's a transistor called the LM D 150. I believe we talked about it in the past, but here it is, again, lnd 150. And it's a it's a high voltage MOSFET. So it can go up to I believe, 500 volts, but it can basically take a small signal of like one to or up to about four volts in and it can amplify the living bejesus out of it with a 500 volt supply. So I threw together this tone stack that came out of a PV amplifier that was well they have it in two different amplifiers. One's called the triple X and one's called the J SX, which the JSX is Joe Satriani, Cetera Jani. That's his like signature model. He's a virtuoso guitar player. If you've you've probably seen him before a little bit goofy,

Host 1     25:09

But I'll give you I'll take your word for it.

Host 3     25:12

One of the one of the things I saw back when I was repairing amps, I played on a handful of these. And I love this tone stack because it was so powerful like and when you turn the mids knob, like it worked. That's that's a huge complaint about old guitar amps was because they tried to do three levels of filtering passively in, like, in a real tight, compact, low component count solution. And what ends up happening,

Host 2     25:41

Your Oh, you know what low component count really stands for crappy? Cheap,

Host 3     25:45

Oh, well, yeah, cheap, for sure. But the thing, the thing that always sucked about old tone stacks is that you get three, you get bass, middle, treble, pick two, two of those are going to work, one of them will basically do nothing. So some amps, the bass control and the treble control are really powerful. And if you turn the mid knobs, you sort of don't even hear anything change. But but it's there. It's just really, really weak. And it's it's there the checkbox off on the design, basically, basically, but with this design of this tone stack, it uses an active feedback. So the the transistor, well, it inverts the signal, but it also feeds back into the tone stack itself. So each knob even though they they interact a little bit, they interact with quite a bit less than a passive tone stack. And they all do what they need to do. And in the JSX and the triple X BIPV. Those are both the feedback path is via vacuum tube, why don't I don't want to do that. I rather just throw something in that makes it a lot easier. Well, this l&d 150 Does the whole job and has a gain of 100 while you know, running on anywhere between 80 and 400 ish volts. So So I built this up on my bench and tested it out and everything works really great. So that was that was really convenient. Because it was one of those things where it's just like, fired up. And it's like, Hey, I've been wanting to build this for literally years, and it just works. So

Host 2     27:16

Your brain was just thinking about the entire time. Yeah, yeah, it was just

Host 3     27:19

Mulling on it that entire time. Yeah, that little l&d 150 is super awesome. And he's super specific. And it's also like, why would microchip be the one who makes this? Because it's just so weird. But it seems like it seems like a good solution. If you are doing really high voltage switching or something like that.

Host 2     27:39

I don't know. They could have bought a company that made it. Probably.

Host 3     27:43

Yeah, I think there's I think there's someone else who makes it also. So I don't know. But, but if you ever need high voltage amplifier, and high voltage can mean you know, 100 volts, not it doesn't have to be like 500 then this little guy is great. And it's available on Mauser and is not expensive. So there you go.

Host 2     28:03

That's a good question we've never talked about before, what do you consider Steven as high voltage?

Host 3     28:11

High Voltage would be anything that I have to use a transformer or a switch mode power supply to go up from so in other words, anything above 240 Because 240 is readily available and I don't have to I don't have to do anything special to get above that.

Host 1     28:28

Gotcha. Okay,

Host 3     28:29

That's high voltage.

Host 1     28:30

I consider High Voltage anything over five No, anything over one?

Host 3     28:37

Oh, yeah, you're right. You're right. Yeah, the

Host 2     28:38

Digital joke there. No 50 volts is when I would consider anything over that high voltage got it? Mainly from that's a pain aspect. 50 volts is a pain to

Host 3     28:49

Get or to generate because you'd have to just you'd have to

Host 2     28:54

Pinball machine solenoids typically run at 48 to 50 volts

Host 3     28:58

Sure but But what I'm getting at is like it's not common. So that's where it sure would become a pain it's not one of the stock value.

Host 2     29:06

Oh no, no, I'm not saying like pain in terms like it's hard to do. Pain is and like, if you get hit by 50 volts that's the threshold of pain of like, Oh, that was not fun to get hit by 50 volt AC Yeah,

Host 3     29:18

Yeah. I don't know off the top of my head. I but I know like, I know, I've touched 24 volts without feeling anything. What what what is in general, what's the threshold of voltage before you feel something? I mean, if

Host 2     29:37

That's the thing is around 45 At least for me, it's around 4550 volts. Okay,

Host 3     29:42

Okay, you know, I

Host 2     29:43

Consider I think that's what I consider basically is not comfortable to handle that voltage anymore. So you have to start taking precaution.

Host 3     29:53

Yeah. Okay. Okay. I you know, it would have to depend on like 1000 things like how salty your hands are in the eye. How's nasty sweaty? You're Yeah, for sure. Because I mean, one vote could hurt pretty bad if you if you get hit by that. But yeah, but ya know, I'd like I know 24 volts, you know, unintentionally I've touched it, but I've also been like, Oh, I'm touching 24 volts and I didn't even know, you know?

Host 2     30:18

And then what would you consider high amperage,

Host 3     30:21

High amperage, over over a half of an amp because it takes, in general, the threshold to stop your heart is a half an amp. So, but from left hand or right hand, you know, half an amp. So that's what I would say.

Host 2     30:38

They mind is one amp, because a 10 mil trace can carry over Most boards, an amp. And so if you go higher than that, then you have to be like, Okay, I know I'd actually pay attention to what this trace is doing. So 10

Host 3     30:52

Mil on one ounce copper, right, unless you like standard Joe Schmo. Yeah.

Host 1     30:57

Yeah, Joe Schmo stuff,

Host 3     30:58

You know. Okay. So quick, quick tangent. How often do you do trace calculators?

Host 1     31:03

Only when it's over an amp?

Host 3     31:07

On the only one you think you'd need to right?

Host 2     31:09

Yeah, no, really, it's like, okay, when I'm doing a board, I'll usually pick 10 mils, unless I need to do a really dense design, or there's some impedance stuff. I'm like, man, 10 mil is going to be enough? Because I know it's a microcontroller. And it's like, it can at most put out 25 milliamps on that pin. Yeah, well, 10 mils is gonna be plenty. So when I started doing power, like for the, like, the, the pinball controllers i are designing. Yeah, the power stuff. I'm like, tackling that out. But usually, it's like, I'll take the size it recommends and then like double that sometimes just to keep heat down if I have space, because there's nothing wrong, increasing the width, if you have space for it.

Host 3     31:52

Right, right. Well, and you know that one of the reasons why I kind of mentioned that is because those calculators can get really confusing. A lot of times, they barrage you with a lot of information, and you have to know it, they're not just like a one stop shop, you know, you put in your information. But you also have to know some other aspects about your board, like, is it going to be coded? And what does that even mean? If it's coded? Is this an internal trace, blah, blah, blah, you know, you have to know that kind of stuff. Because if you, if you just go to one of the online calculators and type in your numbers, a lot of times you'll be shocked at how big it says you actually need them to be, when in reality, they don't actually need to be that big, you know? So it's, it's certainly worth looking up. And the couple that I've done more recently have links to the standards that they use for that. Go read those that's actually worthwhile. I don't I don't remember the numbers. I think they're IPC dash something for the trace width standards. But there's like good reasoning as to why trace width needs to be XYZ. Correct? Yeah. So in general, with the audio circuits that I do, that are like between 12 and 24 volts, I do 10 or 12 mil traces for signal stuff. And then 20 or 30, for power stuff, but like power in just like the power supply area. So I've found that that that's a good balance between being able to handle enough current. And well, I mean, 30 mils can handle a ton of current, but like be really low impedance. And then 10 mils gives good routing without crazy clearances and crap. So Correct. Yeah.

Host 2     33:35

And in your world, though, is also like handling your trace, trace spacings important because you don't want that 400 volts, kind of like arc through the solder mask.

Host 3     33:48

Yeah, I had that happen once and made. I was sad, because I designed an entire board, and then realize that I, the last thing I did was put a copper pour on the whole thing. And I didn't have enough clearance. And 350 jumped across that. And left a nice leverage. It

Host 2     34:07

Looked pretty for about 30 seconds. Yeah.

Host 3     34:10

Kind of sucked. You learn now. Okay, so one other quick thing I think I talked about last week, the macro amp, I resurrected that project. So do the things I'm looking at the component layout, the basically where everything goes, I had an idea on how things were going to be and I'm sort of doing a little bit of an audible on that. One of the things was, this has two output transformers, one power transformer and a choke all in a small package. So I'm really focusing on getting all of those 90 degrees to each other. So with that, you have to have three axes of getting them 90 Just because I don't want any cross coupling or anything like that. So basically, I have the steel plate that is the Top of the amp, and I'm going to be throwing that on the mill at work here soon, and just slap some holes in it and get things rolling on it. And then I did a review of the board. And this was one of the first boards that I hand soldered at macro fab. And I use lead free solder and I just do not like the way it looks. So I'm gonna touch up a good handful of these joints, they all add a little lid, they look kind of like but so I'm gonna, I'm gonna fix them up a little bit. So yeah, that's, that's what I've been up to

Host 1     35:31

Cool. Yep, onto the RFO.

Host 3     35:35

So this first article I found earlier today, and I fell in love with it. This is called Know Your fits and tolerance. And it's a Hackaday article that it's one of the many articles that's basically just showcasing a video. So we'll post the link of this up on the show notes that you can go check. But this video is a great kind of like primer into mechanical fits and tolerances. So when you're designing something that needs to fit into something else, or fit together with something else, there's, there's almost an art to picking the right hole sizes, you know, one of the examples they give in the video is you got a 10 millimeter hole and you want to fit a 10 millimeter stud in that hole. You don't just say 10 millimeters, you know, and that, although that seems really simple, and that that concept is really straightforward. You asking the question? Okay, so what should the stuff be? And what should the hole be that ends up ballooning into a giant issue? You know, it gets really tough. So this video covers all different types of fits. And it, it really digs into the stuff you'll find in the antsy standards, that you can look up in the machinist handbook, which by the way, is 35 bucks on Amazon and is worth having a copy of that. You know, even if you're purely an electrical guy, like, you're still making mounting holes don't make mounting holes, you'll still you'll still interface with a mechanical guy. And at some point in time, like I you know, no matter how much board stuff you do, eventually it has to like become the real world and be a thing you know, so it will go inside of something, or it will connect to something. And so regardless, having the machinist handbook, it's been like Bible for all of this stuff. So there's like charts and pages and pages of types of fits in like R six, R three, that kind of stuff. They cover that in this video and show examples of like, this is what it'll look like. So they really boil it down for guys like me where it's like, okay, that's the kind of fit I want. And I see like it has this much wiggle. And I can kind of like work that out. So I really appreciate that mainly because I'm not a mechanical guy. By trade, I've done enough of it to like, have a good feel for it. But you know, it's not like I went to school for that. So go check this video out. totally worthwhile. Right. Want to watch it before I go to bed at night. Yeah. Also, they have a really great point in there for new engineers. Consider your tolerances. And one of the mistakes that gets made very often is that new engineers will just pick tolerances that are just extremely tight. You know they'll do because they want it to be perfect. They want it to be perfect. And they don't have that five millimeters and it will be five millimeters. Exactly. It's like you gotta learn the gut feel for how much fat you can add back into your design, like how much slop Are you allowed to give in your design, you know, and if your design actually does need precision, fine, but put that in the in the place that it needs precision. So, so you know, the exact same thing kind of comes up a lot. If you if you deal with like, precision analog, like where do I put my $5 op amp? And where do I put my 10 cent op amp? You know, getting getting a good feel for where those go and when to use them. That's valuable, really valuable.

Host 2     39:09

So yeah, I'm going to check out that video tonight. And looking forward to it sounds good. Yeah. So your friend made a video game. Oh, sorry. Yeah, weird topic. Sorry. I

Host 3     39:22

Just saw that. So buddy of mine at work. He is our firmware programmer, but he does. He does some other coding. Also. He's an excellent firmware programmer, and self taught on top of that, just brilliant guy. He made a game over the weekend called pillow fight. And I just thought I would showcase this because it's just kind of funny. So we'll put the link up three player game. It's a two player game. It was so he came over for a party I had not that long ago and we played the new Smash Brothers. And he he likes the game but he was also like, Man, I want to make my My own version of Smash Brothers that has the simplicity that I want. And it's a great little game. It's like a little it's like an old Flash game, in a way, but I don't know like,

Host 2     40:10

And that reminds me a lot of the old arcade game joust? Yeah, kind of like the mechanics of the idea is like those two characters that bounce on beds, and the floor is lava. And so they jump and hit each other and they knock each other down into the floor.

Host 3     40:26

Yeah, and some of the, you know, a lot of the impressive stuff is the fact that there's a lot of momentum and inertia that was taken into account. It's not like you press left and you're just going left or right or anything like that. There's, there's, your characters have weighed, even though they're little stickman. Yes, yeah, so pretty fun. I thought I'd just throw that out there. Go check out the description. Link in the description.

Host 2     40:49

Alright, so the last topic is, I can't even pronounce this company. Kalashnikov, the AK 47 assault rifle company, changed the world. Now there's an AK 47 Kamikaze drone. So how do you pronounce like

Host 3     41:03

Kalashnikov? And that's a guy not a company?

Host 2     41:08

Okay. Yeah, I thought it is a company though now.

Host 3     41:12

I mean, the original designer of the AK 47. His last name was Kalashnikov?

Host 2     41:17

Yeah, I think it is a company now though. Yes. Because I think they're also making so the making, they make lots of different kinds of weapons to sell to people. But they also make an electric car I think,

Host 3     41:28

Really? Well. I mean, it's a last name. So it could be you know, me anyone.

Host 2     41:34

That's true. Yeah. AK 47. Maker makes new electric car. Okay, is the car called the Kalashnikov? Yes, that's awesome. Every product they have is just the collection. It's like CV one or something like that. And actually is very interesting. I'll put the link in here too, since we're gonna talk about now check out that. Okay. From the website, D zine. Whatever the hell that means. So yeah, they're making a kamikaze drone, a low cost drone that carries bomb, like carries explosives, and you fly it and blow things up. So the whole the whole idea with this is they're, I guess they're building it as it's going to change. Warfare, kind of like how the AK 47 changed warfare. It's, it was a cheap gun, allowed rebels to rise up. And so this allows rebels to actually buy drone hardware now. And so I don't really want to talk about that I want to talk about kind of like, what this is like the first we're seeing kind of like cheap, RC kind of stuff being used for warfare now, like, from a not like buy like a drone off Amazon and put a bomb on it. But like, a company's actually weaponizing something and selling it at 10 at like intending it to be a weapon

Host 3     42:57

Now. Yeah, that will and broadcasting that. Yes.

Host 2     43:00

And broadcasting. So like the intention here is different. It's not like you're buying a drone that's designed to carry a camera and then putting a bomb on there, like selling this as a weapons platform now. And so what is this implement Kate? Lee? How is this going to like impact the drone community and RC planes and that kind of stuff. And it's also goes into kind of like 3d printers, because that was the whole big thing about four or five years ago, like 3d printers were printing guns. Now in EU the media went crazy. And, you know, we still haven't seen a ban on 3d printers yet. So I don't think this is going to really impact you know, the citizens, so to speak, in terms of, you know, it's not going to impact like RC planes, like, they're not going to make those illegal yet. Maybe we need to put drones and RC planes under the Second Amendment.

Host 3     44:05

The right to carry arms to bear arms and fly our

Host 2     44:11

Right to fly arms.

Host 3     44:12

Right. I like that. Yeah, right to fly arms. There we go. It takes an act of God to do a constitutional amendment though. So we would need a lot of support on this one. So I'm reading this article now. And it's interesting because in the first sentence, they call it a suicide drone. So this thing is intended to deliver a payload. It's meant to be the payload. Yeah, it is the payload. That's crazy. Okay. And it's supposed to be really inexpensive. It goes 80 miles an hour.

Host 2     44:48

But it's like, oh, yeah, this is a RC plane that someone just strapped a little brick a C for to, it's like, fun. Yeah, that's crazy. Now it's going to be one of those. You know, it's the unfortunate thing is like who's going to be? Who's going to use one of these first against, you know, people? Oh, that's the unfortunate thing about Yeah. Yeah.

Host 3     45:09

So it's interesting because right now on Hackaday, on their front page, they have an article about the the FAA mandating external registration markings for drones. I'm wondering is

Host 2     45:21

Going to be like, Wait, like, when you paint the registered numbers for your like, plane on the tail? Yeah,

Host 3     45:28

It's similar. I don't think it's as bold. Let's put it that way. So it looks like the FAA is requiring drones. And I don't know exactly the requirements on like size of drones, but they they use the wording small drones. So it says there's a weight limit? 250 grams, okay, well, but that's tiny. So,

Host 2     45:53

Like, how much does the paper airplane way?

Host 3     45:56

Well, not 250 grams, I can carry. Like, you know, you know, like the drones that like, you'd see like flying around taking pictures at a wedding like those kinds of drones that are like bigger than the ones you buy at a toy store. Because they actually have to carry weight, like that would be over 250 grams. So the requirement now is to actually have a marking on the device, and they say it must be external to the device. So it must be something that if the device crashes, they could recover it and see that with

Host 2     46:31

Our explodes in the case of these new ones.

Host 3     46:36

Yeah, right, right. Well, and they do give a caveat that you are allowed to mark the drones on the inside of the device, as long as the internal cavity that it is marked in is accessible without tools. So if there's like a motor access point, or a battery compartment that has just like a finger tap to open it up, the FAA marking can be in there. And it's kind of interesting that we're talking about this at the same time, because I'm wondering if there's some kind of correlation or connection to these kinds, probably, there's a little bit there. It wouldn't surprise me. I mean, I'm sure. I'm sure a lot of this is way bigger than what we know about. Probably especially like if you're broadcasting remote control,

Host 2     47:22

Bombs and missiles. Like the for sale. Like that's kind of strange. So somebody somebody has known about it for a lot longer than we have. That's true, but it's like this company, they make arms do sell so it makes sense that you know that they are advertising cuz that's their whole point. You know, they're trying to find people to buy it. Yeah, I kind of liked the design looks really cool, though. I wonder if you can get it in a non exploding version. You know, it looks like semi automatic drone. It looks

Host 3     47:53

Like a lot of those concept drawings for the new version of the shuttle. You know if you've ever seen that, you know, Oh,

Host 2     47:59

Yeah. From like the mid to late 90s. Yeah, yeah. Yeah. Yeah, very

Host 3     48:05

Just at a random curiosity. Have you ever shot a ak 47?

Host 2     48:09

I have not I've shot a m 16. Before but not a ak 47

Host 3     48:14

I have not an M 16. But I have shot an AK before.

Host 2     48:18

My favorite way to we know Yeah, we're getting we're but my favorite build thread ever. So this is like, this is like beats all the car ones electronics I've seen this is really good if it was a build thread of someone who built a ak 47 from scratch out of a shovel. So I had to push like,

Host 3     48:42

What do you say out of a shovel like take the metal and like smelt it down and read? No, no,

Host 2     48:47

He took the shovel flattened it. Yeah. And the sheet metal and made the receiver out of the sheet metal. That was what the shovel was made out of no lie and made all the internal parts from that sheet metal. The stock was the handle part of the shovel. And then he had a that he bought a barrel that was matched in the caliber. Wow, that's incredible. And it worked. And he said it was actually besides being really heavy. It worked really

Host 3     49:15

Well. Well, okay. So regardless of anyone's feelings on on weaponry, if you want to look at a marvel of engineering, the AK 47 is like a tested proven design that I mean, they I heard a report or an article not that long ago. That one was found in the desert that had been under like sand dunes for like 35 years. They pulled it out, like basically dumped the sand out loaded it and shot it like a and in relation to those tolerances that we were talking earlier. Like it's that's a great example of where you do not need precision. That gun is very unpressed sighs and it's built in, like a good portion of it is built like stamped steel you know, stamped and steel. It's not milled.

Host 2     50:07

It's it's designed to be very inexpensive to assemble, correct and manufacture. And now I would say that's probably its shine only shining point. Yeah. Yeah, you know, everything else about it is terrible.

Host 3     50:23

If you set it up, right, it is an accurate gun. But you know, if you just go nuts with it, it's not going to be accurate like, shovel. Yeah, have you been there? Okay, so another fun, quick little YouTube video, go check this out sometime. There's a comparison, a slow mo firing comparison of an AK next to an M 16. And it shows the AK, like, once the actual round fires, the gun like vibrates and slides and like jiggles all over the place. Whereas the M 16 is like very smooth and very, like direct. But there's a big cost difference and there's a lot of different you know, build time cost difference. So it's a really that's a fun one to go watch. So, yep,

Host 2     51:07

Well out to find that video and put it into the show notes.

Host 3     51:11

Cool. Well, hopefully

Host 2     51:12

We don't lose any listeners with our Gun Talk.

Host 3     51:16

Fingers crossed.

Host 2     51:18

Fingers crossed. So yeah, that was the Mac fab engineering podcast. We're your hosts Parker Dolman and Steven Craig. CO let everyone take it easy

Host 2     51:37

Thank you. Yes, you our listener for downloading our show. If you have a cool idea project topic. Let Steven and I know Tweet us at Mac fab at Longhorn engineer or at analog EMG or email us at podcasts at macro voices.com. Also check out our Slack channel. I'm sure all our listeners will complain about our topics this time. If you're not subscribed to the podcast yet, click that subscribe button. That way you get the latest episode right when it releases and please review us wherever you listen as it helps this show stay visible and helps new listeners find us

Transcribed by https://otter.ai