MEP EP#110: Dangling Transformers

Podcast Transcript

Host 3

Hello, and welcome to the macragge Engineering podcast where your host grab foam and Blitz. I will be ended

Host 1

That. This is episode 110. Yes, it is it is yes, it is no 111, isn't it? No, this is 110.

Host 2

Yeah, that's that's that's long recording next week.

Host 1

That's right. Yes. So Steven, that's me. Yeah. What project you've been working on. You've had me 3d print some brand new parts.

Host 3

I think we actually talked about it a few podcasts ago. But I designed a ribbon microphone. And had Parker print up some of the parts for it. And I built the microphone. And we are listening to it along with our regular mic phones right now. Yeah. So you have a it looks like a it's a 3d printed magnet thing, but a transformer kind of glued to it. And then yeah, so this is wired into the thing. This phone is not actually like in a case yet. But so so a ribbon microphone, is it's basically two magnets, and a strip of aluminum that's placed in close proximity to the magnets. And it's tension in between both the magnets, such that when air hits it, it vibrates and it generates a voltage on that strip of aluminum. And the strip of aluminum that I'm using right now is 1.8 microns thick. So it's like gold. You know, it's not, it's as thin as your, you know, stuff yet store. Yeah, right. Right. It's thinner. What? Yeah, it one point eight micron is very, very much thinner, thin aluminum foil from the store. I don't even think you could do aluminum foil from the shore because it the impedance of that is it's funky. Okay, so, a strip this, it's like, or doesn't vibrate enough for something, it has too much mass, okay. And it fights the magnetic field between these two magnets. So the strip that I made is one and three quarter inch long by about point two inches wide. Okay. And I got a little crimper. You, you texturize it. Next I was looking at it, it's got a pattern to it. Yeah, I got I got this little crimper thing from a place up in North Texas. And it's like an arts and crafts, like paper crimper. And you just pass the, the strip of aluminum through that and it gives it kind of a zigzag pattern. And you place that into the little frame that I had Parker 3d print for me attach it to a transformer and stick it on there. So

Host 1

Where are you getting the signals all from? Is it from the other ends of the so it's from the top and bottom of the aluminum foil?

Host 3

Yes, on the frame that I had you print, I put copper tape them. And I just made a basic bolting system, where you put the strip of aluminum on top of the copper tape. And then I have like a clamp system that screws in, I grabbed one side tension the other side, grab that. And then you just solder to both ends of the copper tape. So the signal is being generated across one and three quarters inch of aluminum. And that's just the right impedance to fit on the front end of a transformer, the audio transformer, which has a impedance impedance of or like a ratio of 36 to one or something like that, because the actual voltage that's being generated on the limiters, it's nothing. It's absolutely nothing. So you have to boost it up. And so I'm guessing what it's doing is it's usually because you have these two really strong magnets. Yeah, but aluminum is not. Ferro, what was it? ferromagnetic? That's right or not ferrous, I kept years, whatever. Yeah, yeah. So it's probably it's got a loon has this weird property when you like move it near magnets where you can induce an eddy current into it? And so that's probably what's going on? Well, no, yeah, I mean, you could you could do it with any conductor. I mean, that's, that's the principle of generating a voltage passing a conductor through a magnetic field. I mean, this is very similar to the way that dynamic microphones work, which is basically a speaker used in reverse, you're just talking into a speaker and it moves a magnet through a coil, then instead of a coil, you have two parallel magnets with this ribbon in between it. And it's, it's super sensitive, because I mean, the ribbon is actually only 1.8 microns. And you know, if you if you blow on it, you can actually break it and rip it. So it's just it's really sensitive, but it has a unique sound. Because of that, you know, it takes a lot more energy to move a dynamic sensing element. This thing like takes nothing whatsoever. The problem is it produces very low voltages, so you have to juice the game to get anything out of it even with the transformer stepping it up. 36

Host 1

So you'll get some random noise there. You do get some noise. Yeah. But you know, they will go.

Host 3

Cool. It looks cool. Yeah. What's the plans going forward with that? Well, we talked about, you know, did we talk about on the park? I can't remember we talked about Yeah, we did. We talked about that chip. That's like the all in one. Yeah. USB package. Yep. I was wanting to take this element, and or this design here and plug it into one of those and make a USB ribbon mic that's all controlled, all in one package. So I've proven now the analog side works. Yep. So now I've got to get that chip.

Host 1

So you would basically plug the end and said, plugging that into our preamp here, you'd plug that into the USB. Microphone, Chip?

Host 3

That's right. Yeah. Cool. So it's just kind of a fun thought to maybe make a USB mic for whatever, you know, yeah, can go. And you know, for, like, I've never done this before. I've made microphones but never ribbon mic. I just kind of like looked at pictures online of ribbon mics, and I went to Google SketchUp and made something kinda like it. And it straight up works. Yeah. So the the amplitude that we have on that we're recording is pretty low, because I'm not actually talking directly into my mic, or in the ribbon mic, because I don't have any protection over it. So that you know, my breath could break it. And if you talk directly into it, there's a lot of pops. Yeah. So y'all, we don't have a foam filter or anything over. Right, right. So I wanna I want to get a foam filter and gonna figure that aspect out. But I've got it. So it's sensing. So that's cool. Yeah, it looks like we actually listened to earlier. Sounds pretty, pretty good. Actually. Yeah, it's not bad. It really isn't. And because it's super magnetic, I didn't have to, you know, build a mount for it. I just stuck it right to the mount that yeah, like it's

Host 1

Kind of stuck with a magnet. Yeah. So we've got a dangling transformer. I was got a picture of and we'll pass it up there.

Host 3

I'll do is I'll take the, if you can give me the sketch of files, I'll make an exploded diagram of it. Oh, cool. So people can hell, I think we should just put those files if you're fine with putting those files up. People can build them. Sure if someone wants to build this. There is one thing. So with a ribbon mic, you want the ribbon to be under a slight amount of tension. Yeah. And that's why those those ripples that you put into it help because you can pull it and put it under tension, put it under tension, right. The thing about it is, it's very, very, very difficult to do I you know, I probably went through seven or eight different ribbons just trying to make this one. I had an idea to put like a little thumb screw down on the bottom. You can put put it in not under tension, and then turn that thumb screw and it would it would Oh, expand it. Yeah. Tension it without you having to do it by hand. Yeah, it's such a gentle rip it. So maybe, maybe I'll design something like that up. Who knows? Yeah. So we will we can show this design. And then I think if you try that do the Pro model. Yeah, we can do like a hackaday.io. Project. No, that'd be cool.

Host 1

Yeah. Yeah. If you want to make your own ribbon, like, yeah, we'll head that up. That'd be really cool. Actually, we can do the whole thing, the USB part and

Host 3

Yet well, and that USB thing. It does have a headphone output. It can drive headphones directly. So we have to build your own head phones to you know, trade German. He's been on the podcast. Yeah. Two or three times? Yes. Twice. He's made his own headphones, through here at macro fab. And they were pretty cool. Because he just used like, fr for as like earpieces. Yeah. And it actually worked out pretty good. Yeah, they were pretty nice. Yeah. No, like, I wonder how you would make your own. If you can make small drivers that would that would work for that. Oh, like make your own cones and stuff? Oh,

Host 1

They don't be cones. You can make them. What's that other kind of speaker? They usually use them for tweeters. Later, like the like the silk dome things? No, they're not cones. They're not like a normal cone. But it's a coil that moves up. Like a piano buzzer or something? Yeah, kind of what it's the bigger. It's kind of I can't remember what they're called. There's a specific name for them.

Host 3

You know, actually, with 3d printing, you could you could probably make a a negative of a speaker shape. And a lot you can make speakers in the same way that you make paper mache. Yeah, if you like cut up a bunch of paper, and then you you make a press, you can press the speaker dome. So I mean, it's all it used to be done. Yeah. And so it actually wouldn't even be that hard. In fact, if you look at old speakers, they just look like a whole bunch of glue and paper. Yeah, they're not super nice. So

Host 1

We should try that. Well, maybe not exactly that way. But yeah, even going back to arts and crafts making our own Yeah, we can start sort of Hackaday dot I'll look into that tomorrow. Okay, maker hackaday.io for this. Cool. Sounds awesome. Sounds

Host 3

Like a bunch of fun. Yeah. So, next week, I will actually be out of town. Yep. In Denver. My wife and I are taking Got a trip for spring break to have some fun out in Denver. And I thought this could be a really fun time. If anyone wants to meet up and hang out and go get a beer and just chill, I would love to meet, I think they do other things and just get beer and well, that is not going to be part of our trip. But But there will be for sure. So if any of our listeners out in Denver, want to meet up and just hang out, I would love to come and meet you and just, you know, have a good time. So I actually created an email address called Craig hang@gmail.com. That's KR aighang@gmail.com. Send me an email. And we can work out a time I'll be there. The pretty much the 10th through the 16th. So sometime during the week, sometime, maybe this coming weekend, three o'clock in the morning, three o'clock in the morning. In the morning, we get let's go get a beer. Let's Let's hang out. I want to get in. That's Craig hang@gmail.com. Hit me up. Yeah, I'd also be in the notes if you want to check it out. Cool. Awesome. So I had my new DAC article come out. Part two, where we did the we talked about this last week about the tests and stuff. So if you want to see all those fancy graphs that I made and stuff go, you know, that tickles your fancy, go take a look at them on the blog. We pretty much talked about that last week. Yeah. And you you found out that for two weeks version was plenty acceptable on the cheap version people preferred. And there you go. So even you pick the cheap version. So yeah, some some reason ceramic mic, ceramic caps and the audio stream just sounded better to at least two people. So but the thing is, there was a difference between film caps and ceramic caps, there was an audible difference, and there was a measurable difference. And there was actually a difference

Host 2

In the audio when it was just changing out like bypass caps. Like in the digital side. There was still a difference. You can measure. It was slight, but you could measure a difference. Gotta watch out for Yeah, just interesting things were like, you think it's just BS or like, oh, the best. Like you will always hear a difference. It's like

Host 3

If you're trying really hard your mind, maybe and then it comes down to just being subjective. Yeah, it's all subjective. Yeah. So PINet grab eight update. Yeah, so we got the prototypes all built, tested. They did not blow smoke everywhere. They all work. Hey, thank you. I got the use those backside tested. I got those hooked up to the pin hack right now. And so it is outputting lbds over to the Raspberry Pi into our screen that works great. I square C works great. The communication to the pic 32 works. And the eye to s to the PCM 5122. Works great. Did you just have a wire jumper over? I built a little cable? Okay, can I build a little custom cable because the connector changed and revision base I changed the connector to have more ground isolation between each pin instead of having like a like a, you know, power ground and then the data. I did power ground than alternated grounds in between the data just to keep you know, EMF down because you know it's pinball machine. It's tons of EMF, you got cylinder coil coils, find newer ways. Yeah, large current changes everywhere. Like goes from eight amps to zero and like half a second. Yeah, right. Right. So the only thing really had to do between our earlier software testing, so we were doing testing like raspberry pi three. And the new stuff was we had to change a lot of device tree overlays. Because the lbds takes up like 28 pins to you know, you got eight bits per pixel basically plus horizontal vertical sync plus end of you know display plus this tons tons of other signals, you need to basically turn it into an LVDS signal to drive that display. And so you need to enable like i square C and I to s for the DAC. Well, the drivers use the original pins on lesbi pi three, but we had to move them to like farther down the GPIO line and use alternate functions to get them there. But that chip that it's like the was at the BCM two seven something is whatever that broadband chip is a support program. I'm not sure remember the manufacturer that chip anyways, you had to move the pins and it's like it's almost like a Coming to Silicon Labs crosslink. Where like, you can just say, Oh, this, this, you know, hardware CO or hardware function needs to be on this pen now. Yeah. So you can cross link it over. But there's not really any no one really does this online, like when you go to like the Raspberry Pi forums and stuff like everyone's just like, and no one really uses the compute module for, like, the people who use the compute module don't release their secrets about it is the best way to share, share, share. And so we had basically figure it out. And so here are the secrets. Oh, yeah, the secrets are so if you need to, like move these pins around for like a square, see, we need to move them to look at my cheat sheet, we need to use the move that 28 to 31. No, that's it. Yes, yeah. When I two s 220 31. It's usually like, pins like two and 234, or five, something like that. So it's like, on the upper end, but you know, the, the LCD takes all those guys up. So we had to move them down. And I basically looked into the, the source code for these device overlays. And they usually have override functions that are at the bottom, but it's each one has a different way to override it. Oh, gosh, yeah. So like the I squared C, you actually have to pass it like SDA one, underscore pin equals the new number. But there's only like certain numbers you can use. It's like, why did you do it that way, but whatever. So you have the past basically three variables to make that device over, like change its pin, so you have to change it to tell what SDA what SDA Sel is, and then what the alternate function is. So you have to calculate the ultimate function of that pin that you're changing into. So like that pin can be like eight different functions. Hmm. But it's not the number that it is. Because it's like, for for this to be the use like pins 44 and 45. For is square. See, you have it's like all function. Three, I think. Okay, but you can't you don't give it all three. You give it all six. Why? I have no idea why. I could never forget why you just did it until it work. No, no, no, no, I figured it out by when you look at look into it. It's like all it gives it like all three for the default. So I got Yeah, it was I went to buried in the code is very Nikodem basically looked and said, okay, the default is this. But when you go look up the in the chip, like data sheet, it says it's this number. And so you just like add three, no idea why. But that reminds me of like, the days of like, sitting there and just ripping through tons of config bits for microchip chips. Yes. Where it's just like, okay, so this byte needs to be 0111200 You know, you're not, or you're automatically set by a comparator and you can't do a GPIO function. Yeah. Or it'll do like that one function where it's like, you can write to this chip one time, and then it locks itself. And then you got to go buy a new one, you know, so? Oh, yeah, he like done that before. Yeah, absolutely. Yeah, way. It's certainly, it's, uh, ya know, that can be a giant pain. Yes. Yeah. So that was one. And the other one is like, they actually had a separate device overlay to make it work. And so that was cool. But yeah, and a good way to check this stuff. So after you boot it up, and like it's not working, you can install a GPIO function on the Raspberry Pi, I think comes standard in Raspbian, where you just type in GPIO. And then space read all, and I'll tell you, all the GPIO like, what their states are, and what their alternate functions are set to. So you can actually go in and say, Okay, I actually set that to I square C correctly, it'll tell you what it is currently set to. Yep. Okay. And you can actually you've done it, yeah. And you can actually go in and manually change them to, like at runtime, which is kind of cool. Now, it doesn't help you when you're loading drivers, like the BCI that BS BCM PCM 5122. It won't help you at runtime, like puts when it's loading that driver in it has to have all the pin set before it loads the driver. So yeah, let us know I was able to debug all this stuff and get it working. And so we're going to use the this pin hack rev eight boards as like our test mule for the for the guitar system, because it's almost the same. So we have on the because like the Raspberry Pi side it's pretty much going to be the same for pin ATAR. But the pic side is going to change mainly because we're upgrading to that 200 megahertz the pic 32 MC. So we'll have more speed there and We're gonna change a lot of the surrounding architecture, hardware architecture around it to simplify it a lot. And the good thing about that is our pick three to code, we have like three abstraction layers that we, we coded in on that layer. On that side, we have like a bit of a game layer, we have a board layer, and then we have a hardware layer. So abstracted all the way. So basically, all we need to do is change the hardware layer, which is basically the, the hard coded basically registers to, you know, it's like your header file that you get from like, microchip. We basically have our own config bits, basically, yeah, we have our own custom one for our board, right. And so all we have to do is change those and then we can use whatever chip, because you suck them in whenever you need. Yeah, yeah. So we can Yeah, so we can upgrade to this newer chip without having to rewrite our entire pic, 32 kernel, or have to do just changes underlying, you know, one file, and it should all work in three. Seems like that's the the most efficient way of doing it. Yeah. So yeah. Cool. That's what I've been working on. I've got a funny story today. Well, from earlier today, it just popped into my mind, because I mentioned microchip and then thought of this. I was over at Home Depot earlier today. And I was at the self checkout, you know, they have like the, you know, you can do so? Well, one of the one of the computers was down and it had windows up on the screen. But on that it had a window with the big microchip logo on it. And it was talking about like reflashing whatever. Oh, boy, it's that. Yeah, like something was way wrong. And you know, nobody knew anything about that. And I'm in a line with like six other people. And I, I had to fight the urge so much to pull up my phone. I wanted to just walk over and take a picture of it, but I knew that would have been so suspicious. What the hell is this guy doing? You know? But it's cool that these microchip chips and those things. Yeah, I don't know. That's cool.

Host 1

He got one of those yellow vests on, like gone out got yellow vests and come back and just start unscrewing it and working on it. But no, no, like, they would never guess. Yeah. Scissor Lift, drive it all the way over, put cones around.

Host 3

And then start working on Yeah, just to take a picture of it. Because I was I was looking at I was like, Man, that would be great for the podcast. If I could get like a better picture of it. Yeah. But no, so be like,

Host 1

No Gear Solid. What? You'd have like being under cardboard boxes? No, no, like, you can be like, you can be in disguise. Yeah, yeah. Pick him up, put him in bathroom.

Host 3

But that would be one of the nerdiest reasons to to do it. Like stealing anything. You just like trying to look at it? Yeah. That's fun. No. Okay. So I don't know if we talked about this on a previous podcast. This is a shortage of passive parts. We may have been you may have, but I saw a lot of comments on Hackaday on like, recent articles about this. And someone actually mentioned on Twitter that he can't get point one micro farad caps right now and Oh, four to wait. WHO and WHERE is another? No, he's, he's one of our customers actually. Oh, oh, okay. Okay, so he's basically having like, redo his whole bill of materials for that one part because he has a very specific point one Makerfaire Oh, gap he wanted. Okay, I was about it's not like jelly bean point. One. Makerfaire. Ah, I was confused at that. Because I was like, I'm pretty sure you can get those. Yeah, you can get them except can't get his Yeah. So I started looking more into this, basically, right now is manufacturers are they're not producing parts right now as fat like they're not basically scaling up their their manufacturing. Is there a reason for that? Margins? Okay, so right now, like margins on passives. And stuff is super low. Because there's a lot of manufacturers for it. And it doesn't make sense for someone to spend a couple billion dollars to spin up some more, you know, line space to build more of these. And so people just like, No, no, we're just gonna run out and just be like, Okay, we're gonna run out. So the type of component that is in every electronic device is now like, the baseline component is not going to go up in price. Yo, yeah, that's it. That's the thing those like Hackaday does this, this article on it, where they were looking at like, the same part from like, a long time ago, and it's like the same price. So it's almost like this. There's so many different manufacturers building these and driving the cost down that way. That means the margin is so small and so that these companies don't have enough basic capital to build new lines. That makes sense. Yeah. Well, and if you're in the passives game, you have to sell like so many to make any kind of money you know, you have to be selling them by the by the ton. Like literally just by weight. I can imagine it's like Scrooge McDuck like diving. Gold. He's diving into like, chip capacitors and stuff.

Host 1

That would be great. I guess imagine like them getting under your fingers and stuff. Oh, that would not.

Host 3

If there's one thing about electronics, there's never like, a nice feeling piece of electronics. So there's not like a soft piece of electron like, everything is hard. Everything is sharp, everything, like smells bad, everything will cut you. I don't know. I'm not sure I'm actually trying to think of any soft materials in like icy design or whatever I

Host 1

Can be maybe a flat flex PCB might even then it's so rigid enough to like it's sharp on the edges still. Yeah, yeah, exactly. Yeah. No. Electronics are offensive. Not like being offended. But like, on their on offense all the time. They're ready to stab you and cut you. And

Host 3

Yeah, it's like if you cut like PCBs up and stuff, like the fiberglass gets all over the place. And it's all it's all super itchy. Yeah. So back to back to the shortage. Yeah. So these companies are just like basically saying, Well, if you won't let us raise the prices, were just gonna make the demand go way up. So supply and demand. So yes, so eventually, prices will probably go up. As it that's kind of like they should because, you know, even just like, it's the same price from like, six years ago for the same part. And, you know, at least it should go up with inflation, even if they got better at making them or whatever. Makes sense. Yeah. Oh, yeah. Oh, no. Yeah. It's looks good. It looks good. Well, here's here's the title for juicero. already found at Goodwill store. Yeah, so we talked about this. Oh, yeah. I was showing me a picture of it now. Oh, that is fantastic. Yeah. What I'm curious about is, did they say what the price was? 40 bucks. 40 bucks. Wasn't it like 600 $700 $700? So for those that don't know, you, Sarah was a Iot juice squeezer for $700. And it basically squeezed Capri Sun packages of juice. It was like a Capri. Adult Capri Sun don't. Yeah, they don't taste good of what I've heard. Yeah. Yeah. And now you can't, like they did like a buyback program for a little bit. And then yeah, but But um, what I heard is you were really lucky if he actually got anything. Yeah. And, and I stopped selling the packets. And so basically, this is like a 60 pound paperweight. And I guess someone dropped one off at Goodwill. Wow. So you can pick up for 40 bucks, you can pick up a $700 $700. Like, only like last like July. That's true. And they had a ton of money, too. They had a ton of funding. Yeah, they had a model where you could get these Capri Sun packs on like a regular basis. It was something like you get a handful of weak. Yeah, but it was like really expensive. Yeah, I can't remember the exact price for that. But and the argument I think they had was like every bag was barcoded a, they could trace it all the way back to the farmer. You kind of you got the feeling that you knew where your stuff was coming from. And it was safe and but you know, it's juice. It's Yeah, exactly. But the thing is, like, honestly, 40 bucks for that. Not that bad because the parts inside are awesome.

Host 1

So you can repurpose that into a I don't know. Oh, that's how you can get the last bit toothpaste out of the toothpaste thing to just crush it out. It's nothing puts out a lot torque. It does. So what else could you squeeze with something like that? What else what else needs to be squeezed like that? Oh my gosh, what

Host 1

Is a vacuum pack? Like lamb? What the hell is Iris is showing us a picture in the same thread layout. I don't think I'd want to squeeze that. Yeah, no,

Host 1

That's the mashed potatoes. If you now know. I was thinking when you brew beer, and you squeeze your green bag yet, but you're not supposed to squeeze a great bag, because

Host 3

You're supposed to just let it drip out awesome. Me being me being lazy has been a good thing then. Lucky let's put it that way. It's the same thing with tea bags. You're supposed to just dip a tea bag and then throw it at school. it then it releases the tenants which, you know, some people like the bitterness of that. And so you, I guess you could go that route, but what I've heard is for the majority of people, it's just not their style of Yeah. So what else could you then squeeze? Um, oh no. Because it cuz I keep thinking of like food. You know what? You could you could Yeah I don't eat a lot of JB weld and make like a mixer with it if it just like yeah, squeezes right around you squeeze it and then you know, like you put them inside that Oh, go in and out and squeeze and mix for adults freezes together until it's done. Yeah. Yeah, cuz like I don't eat a lot of food and tubes. Yeah, it's not the future yet.

Host 1

I don't know cuz it is a crusher of some sort. Maybe I don't think that's gonna throw to crush it can though.

Host 1

And it's really slow. It is very slow. I get a six pack of beer crushing six cans would probably be like you already.

Host 3

You drink a six pack of beer while it's crushing one? Yeah, that's right. Yeah. But yeah. Go to your local Goodwill and buy a $40. Juice arrow.

Host 1

Good deal. Yeah. Send us a picture of your the guy who bought the $40 Oh, yeah. What? Oh, yes. I don't know if that guy bought it or not. So

Host 3

This is a question from our Slack channel. Is why is the board house putting a UL mark on my PCB? Oh, yeah. That's a good one. Actually. Funny enough. I have my synthesizer here. I brought in. We're not going to talk about on this podcast, but I brought it in for Parker discuss markings on it. And yeah, yeah, a handful of episodes ago, we talked about doing fr for face plates. I did fr for face plates for all of my synth synth modules. And I made one critical mistake, I did not tell the board house to remove the UL mark. So there's a UL mark on the front of every single one of my Yeah, that's what we should have said on that podcast. Well, now they know that you're going to use fr for from a PCB house, tell them to remove your UL marking, or put it on a side that it doesn't matter. Yeah, on the backside. And to be honest, most board houses are, you know, they're not just going to bulldoze your design and put something you don't want on it in a place you don't want it. If it makes sense, this place that I got these PCBs made, they clearly didn't care if you if you just look at the boards that I uploaded, you would know not to put the N on the top layer, they clearly didn't care and they just did it. So I always find that those that PCB manufacturers will put that marking in like, the most like things they want, they don't want to touch your like designators and stuff. So they put it in like a blank spot where like, you're going to put something there like a marking, but they put that there. Yeah, yeah, they always will find that spot. Yep. So the reason why they put that mark there, though, yep, is because that material that use needs to be UL recognized, to be used in the UL marked product, it doesn't make your board UL certified or whatever, it just means that PCB is recognized as a material that can be used in a UL certified thing. That's, that's right. And and there are some industries that are exempt from that. Mainly, there's some aspects of safety devices that actually don't require that. Which is funny, you would think that safety would require it. But in this case, in some cases, it's not. Regardless, for any UL certified device, you have to have paperwork that shows that it's you well, you have to have paperwork that shows that the fab is is you know, and if you get audited, it doesn't happen often. But an auditor can ask you to open a product to see if that's on there. And they will look for it. And so you gotta be you gotta be careful. And so most, pretty much every single board house just by default, puts it on there. Because they had to go through all the rigmarole and get certified and stuff. So they want to put that on there as a badge of honor. It's a badge of honor that they paid all that money to be certified that money and went through like a year worth of just absolute garbage auditing. Yeah. So that is why it's on there. Unfortunately, here at macro fab, we don't really have a way to like for our prototypes, because like the only way to really do it through fabs is like the whole panel gets marked or none. And so we just marked all the prototypes. If you do a production run, we we we can do that for you. But for the prototypes, it's kind of like, well, well, and the thing is 99% of the time, you know that the maker community is not going to want that on their butts when they do fancy art. Well, yeah, yeah, exactly. So So you know, just just be mindful that there's a good chance it will be there if you don't take it off if he's correct, but I haven't personally met a board house that wouldn't take it off that correct? Just say no to they will take it off if you wanted. And you can also have them because a lot of times they put another number there that's like the batch number or something like that. That's a That's right. I didn't find number for them. You can also have them remove that to second date code. They don't really like to, but they will. Right, right, right. I don't think ours puts that on the board. So so this, this might sound scary, it is not an it's just this is just part of the industry. Technically, they are allowed to modify your design, if you put a a design and they can actually change things around. without your approval. In some cases. It very, very rarely happens with today's you know, with technology today, almost never does a board house do anything without you knowing but technically they can. And most of the time, that's just like, they'll slightly move a designator such that it's, you know, if there's a designator on a pad, they might move that for whatever reason or whatnot, but technically, they can change things around and they do sometimes when they need to. Yeah, we actually here we have a approval process because we get the working files back from our, our fab center, and then we basically like overlay them and make sure that they're the same. Well and a while ago we set up with with our vendors here do they're not allowed to make changes without explicit. Yeah, then the Contact Us Yeah, right. Right. So that was but but we had to set that up. We had to set that up because we knew it could happen. Yes. Yeah. So it's, that's those are two good things worth knowing right there. Yeah, they anything can change. And there's probably going to be a UL mark if you didn't say no. Right by your logo. For some reason. Yeah. Yeah. That's ugly. Yes. Yeah. It makes it look ugly. Yeah. So cool. So I think they'll wrap up this episode of The Mac fab engineering podcast, where your hosts crap from and blitz have never again. Later everyone

Host 1

Take it easy. Hit me up at Craig hang@gmail.com If you want to hang out in Denver next week. I just did