On this episode, Stephen talks about his new bias test system hardware and software. The bias test system is a purpose built test hardware system!
Is grinding out math problems just busy work? Is the current state of Math class curriculum hampering the real life deployment of engineering skills?
Bias Test system update
Checked against 4 multimeter and getting values “in range”
Next steps Finish a first version of the arduino code Merge the skipi communication with my code Solder into an actual amp to get real world values Schedule next python thing
Fixing circuit boards
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 Mac fab engineering podcast. We're your hosts, Parker, Dolman.
And Steven Craig.
This is episode 291. So before we jump into everything, just before the podcast, I don't remember how we got on the subject. But I've looked up free PCB, which is the PCB software design EDA tool that I use starting out, like, what, 14 years ago or something like that at this point. And lo and behold, it free. pcb.com is now a HostGator parking website for the domain. So I was like, oh, man, it's gone? Well, if you looked up free PCB in Google, there's a link to free PCB dot Dev, which looks to be a fork of that project. And I'm like, Oh, cool. Someone's actually working on it. And then went to the GitHub, and it was updated four hours ago. So it's an active, someone is actively working on free PCB. Just makes me feel warm and fuzzy inside
the good old days. Yeah,
I'm gonna have to try out. Free PCB,
Parker was super excited because they now have the schematic editor Yeah, has the
schematic editor so you don't have to build netlist by hand anymore. So
I've got a little bit of an update on the bias test system that I've been developing, took a little bit of a few weeks to work on it a little bit more. But well, first of all, I made a little daughter board so I can access all the pins now. But the goal of this board was also to test IDC cables, ribbon cables to see how much voltage I can put on them before bad things happen. And so I've got a 16 pin extension cable that goes off to my little daughter board with just a whole bunch of screw terminals.
But wasn't the idea that you could that your amp or your device, your testing has that connector on it already. So you don't need the breakout board?
That's right, yeah. So just for development, I'm going I have it such that I can basically wire this hardwire this directly into one of my amps. But eventually, the goal is to have a test connector on the amps. So I just plugged my bias test system with an ADC cable into the amp. And then I have access to all the signals that I need. And, and actually, I want to talk a little bit about that. Well, let's just do it right now. So the the, the original goal was to make my own kind of connector for it something a little bit unique. And then I was like, you know, it's probably easier to do just a header pin header. I didn't want to originally do that. Because I'd have to populate a header on every amp. And that header would be used, you know, once ideally. And so like if I didn't have to pay for that I didn't have to spend time soldering it. That would make sense, right? But so I was like, okay, cool. I'll just go back and solder a header in place. And then I realized, I have potentially up to a few 100 volts on a point one pitch header. So I started looking at clearances between pins, and how could I design one that would work out? Well, well, it just doesn't really work out well given. For almost all my signals, a point one pitch header would work fine. But there's a few signals that can get up to like 700 800 volts. And I just the clearance is I would hate for one of my ads to fail for our connector that I put on there that I would use one time for production, it just doesn't make any sense to do it that way. So I gonna have to go back and develop my own connector for this, get get a little fancy with it. So that's sort of a hurdle that's a little bit down the road because there's nothing really pressing on this is more like, for fun in the future. I don't I don't need this right now. Because everything that this bias test system can do, I can do by hand, I just really want this to be computerized. As such, I can have a database of everything. And if I ever trained somebody to do it, I don't have to have them calculate power ratings of parts or anything like that. They just go on the system and it says turn this knob until everything is green on the screen. You know, that kind
of thing? on making your own connector, how are you going to do that like but you have an idea of how you want to do that yet.
So I have already developed something that worked in the past for this, but I'm just going to make it a little bit better. So basically, I'm making a homemade version of a Tech Connect cable. So pogo pins that access exposed pads on my main PCB. So the goal was the on the main PCB that this connects to if I just have exposed pads, I effectively don't pay for a connector on that but The biggest thing is just making a connector that clips into the board and hangs onto the board and doesn't lose connection and doesn't slip or anything. Because there's many hundreds of volts, I don't want anything to short up, I would hate to have to replace a bunch of parts plus my little test system here because the connector slipped and shorted, you know, 500 to ground or something like that. So what I'm working on right now is a small board that has a handful of Pogo connectors on it, that will somehow mount inside of a 3d printed like finger gripping system, very similar to a attack connector cable. So I can just put pads on my boards, and then some holes on the board, and it kind of grips into the holes and hold the connector in place. That's sort of the goal. But I'm working on the hardware before I nail all that. Because right now I can do all this just by wiring things in place. And that's honestly, working on the hardware is a lot more fun than developing a connector.
Development connect sounds kind of fun. It's,
I'm enjoying this right now.
So the cool thing is, so far, I've tested high voltage on an IDC cable, and I haven't had any problems. These things are typically rated to about 300 volts, but I've put 450 on it. And I haven't seen any issues yet.
It's up to x safety factor working in your favor. Oh,
let's hope it's actually 2x. But I haven't tested an IDC cable filled with high voltage lines. I've tested like one or two kind of thing. Okay. Okay. So that's, that's the next step it passed first muster, which is just can I put 450 volts on any of the lines? And I've done multiple? And it seems to be fine. So yeah, so the been working on getting the code up to up to snuff on the on the board here. So the I actually have a question for you on this, because I'm curious about how you do it, I you know, I write firmware, not very regularly. So whenever I approach things, I usually have to kind of reinvent the way I do it. But so I got some Arduino Libraries for the ADCs that I'm using on this board. And they weren't fantastic. But they were written to communicate with one chip, I have to. So I have to add, chip select to be able to select for which one I'm working with. And so the way I accomplish that is basically on all of my functions. I just added an extra argument in there. So I can select which chip, it talks to. I'm curious about the way that you approach that. If you have multiple chips that use the same function?
Yeah, it usually would be I'd have a identifier of some court of some kind that you would pass. Yeah. Okay. Yeah, you could do it. In your case, I guess you're just passing a zero or one. Yeah. And then in that function, it sets that it just, it just takes that and shoves it onto your chips like climb.
Well, you know, here's the here's what's really annoying is whoever wrote this library, like they did a great job. But they went overboard, where it's hierarchical, hasn't many layers and things like that. So there's nested functions and things even down to like, chip select itself is a function. And they built it, I think they built it with the intent that somebody might do exactly what I'm doing. Because like, chip select on is a function and chip select off as a function. But if you go into their functions, it just toggles one pin. So I think they built it such that it could inherit something like I basically, basically, I have my top level functions. And my last argument in there is just throw a 01, effectively, for which chip I want. And then it had, I had to boil it down through all of their functions all the way down to which chips like that goes to now.
You might, depending on how that library is made, you might be able to make it like both of those chips are different objects in the Arduino code. And then that might solve your problem instead of having to do it that way. I don't know why the hudud library is constructed. But that might be the way to go. Because then you won't have to modify and add in a chip select top level the pass in.
Well, I've actually, I've already completed it. Oh, it's I run I drilled it all the way down. But But while I was working, I was like, There's got to be a better way of doing this. And it's a class I could have. I could have had a separate item on it, but I still would have had to have passed it on argument that was which chip it was and that would require heavily modifying the library anyway. It just as I was going through it, it seemed like it was like 50% built for this intent. But it's like man, why didn't you go all the way through so I don't have to go and like tear apart So I don't have to do any work. Well, I mean, like, I don't know. Yeah, I guess the thing is about it is like this, this, you know, it's a 24 bit add with eight channels, how many people with an Arduino are going to need more than those eight channels? I understand who's gonna have to have these chips. Right, right. It doesn't. Yeah, so and if you do need more than two, you'll go in and modify the library like I did. So it's just, somehow it feels like what I was doing was not right. But it functions and there's no problem with it. So it's like, okay, well, fine, whatever
your only problem is, if you ever had to update that library,
if I ever have to update that library, and there's and I have, like, if statements that feel unnecessary now, because it has to say like, Oh, if it's this chip, do this with this, you know, that kind of thing. And that just seems like inefficient and slow.
So I don't know how that libraries set up. But if I was designing from scratch, I would have a top level, like, Chip a, Chip B, Chip C, and then you would define those as like a, as a, what the, what the what the pin is basically, for the chip slots, your pet, you pass the chip select line for that chip in the top level function. So if you go read channel eight, right, and then you pass the chip suck line 13. And that's the one that uses
Yeah, yeah, that's how I would do it. Yeah, if I, if I, I could probably redo the library to be a little bit cleaner. But it since it's functioning now, I'm just going to leave it as is because I don't actually need to read like, unbelievably fast. So, and no one else is gonna look at this code other than me. So I checked this board against four different multimeters. And I'm getting values that are in range with all the other multimeter. So I'm happy with my measurements now. Like, I think I can trust this well enough to actually, like, do something that could be damaging if if I said it wrong,
it's calibrated to your NIST. I'm doing air quotes and this calibrated
Fluke well you know, I brought that that NIST calibrated Fluke home and I tested that against two other meters that I have here at home. And they were all reading basically the same thing. So
and then your other three meters you're checking are the free Harbor Freight ones. I was actually looking at getting a new meter. Oh, just to replace the one. Well, not really to replace it but basically to because right now I have I basically only have two meters. I have my old one that I've I have like the domin family heirloom meter. It's that Beckman industry
meet Oh, yeah. The pride of the family.
Yes, this is the family heirloom. And then I have my, like $10 RadioShack, one about long time ago, like it's like Velleman or something like that brand. And you can get them for like $12 so on Amazon. That meter has been to the beach dropped and saltwater dragged out over the driveway, it still works fine. But I'm thinking about like actually retiring it to be just the garage and actually get a nicer meter for like in the electronic shop. Yeah.
What do you what do you think? And again, you're gonna go with the good old yellow goes,
I think I'm gonna get a five digit sequence. That's Top Model.
Cool. Oh, a desktop. Yeah, those are nice. Yeah.
And then I want a portable, actual an actual portable meter that that can fit into like in your pocket easily for like DEF CON and stuff like that. And conferences, because those are finally starting to kick back up.
That sounds like harbor freight.
Yeah, you can get like a TS 100 soldering iron, a portable soldering iron. But I'd like to have a portable meter as well. I did find some meters on Amazon haven't purchased any yet or really done any in depth review or looking at them. But they're like the form factor of your phone. And they got like, a plug in the bottom that you plugged leads into. So you can't do high voltage or whatever. But, I mean, if it's no bigger than your phone, it's not a pain in the butt to carry around. But the big thing I'm struggling with is the problem with having a meter like that is like the leads are now like the bulky part. You know, like what do you do with like, cuz you know, your multimeter leads, like always get twisted up and stuff. And it's just so awkward to kind of like, untangle that. It needs like every tractable like leads because you're not passing out a high current or whatever. And 90% of time you're gonna be just beat testing,
like the like those lanyards that that yes, yeah.
When you test leads are lanyards. Well not like lanterns. Were like well
laid like the flat cable stuff like if you do it Really simple stuff. He could totally get away with that.
Yeah, they just don't make anything that's purposely built like that. I guess you could take a because they make those USB cables that have the retractable senators. Those have four wires in them.
Cut Cut them solder all four wires on them. There you go. Yeah,
you can solder two of them together. So you have you have to a pair Now basically, and then put some leads on them. And
I was I would say I have two extendable oh, what yeah,
one for solder everything inside together. Yeah, that might work.
You be king of the nerds. If you had that. Yeah.
It would be even better to if you can make one plugs in with the doctor USB. Instead of like a banana Jack. Someone to make that happen. That sounds like a really awesome Obi Wan. Yeah, I would buy one Meili. But um, yeah.
Wireless probes, man.
It's like that. Remember? The wireless ESD?
Strap? Oh, yeah. Oh, yeah.
Think about that, as it had a one. It's a lot of one mega ohm resistor in it. That mega home
resistor was just in the air. It just dissipated into the ether into the ether.
I did do a thought experiment once of how could you pull that off?
Oh, we we did a whole thing about that. I mean, not on the on the cast. I remember sitting in the in the lab talking about could you somehow charge? Charge something that you would later go and like connected to something to discharge it?
Yeah, I think the, the problem we ran into is the whole thing is your body is the whole reason why like you have a ground strap is so that your your referenced to the bench, right? Whatever you're plugged into for you ESD strap your reference to that. Whereas if you had a wireless strap, you're not reference to it. So you could still build up. Like, if you try, you couldn't you can't charge anything up and store it later. Basically what happens? You just get evenly distributed over that charge of its static builds up.
I mean, just polarize, like the ceiling with a few million volts, and have it just like pull through the air?
Yeah, no, no, that's actually the one way we've we actually found out about ESD ionisers. Yeah, that's how those work is. So there's fans basically out there that discharge things by D is D, I think D ionizing the air. And so the ions get stripped off of your body. And so you don't build up static.
I actually use one of those at work in my office just as a fan. Like this is
what gets in the winter. It gets really dry up there. Yeah. So I would say that those are probably really good for your your manufacturing process.
Yeah, you know, it's surprisingly with how dry it gets up here. We have not run into ESD problems. We don't we don't have any failures attributed to that. Let's just put it that way we may have. But in terms of like, we don't see that happen often. And I mean, there's been times when it's like 12% Humidity here. Yeah. Like when you get down that low, things are uncomfortable.
It's like your skin is starting to shrink.
Oh, yeah. Like everywhere you go. You're slapping door handles and things like that. I cannot. I have to get like a towel. If I go like pick up by Nintendo because it will blast me. It's yeah, it's actually it's not fun. You get you get shocked by everything.
I wonder if there's a market out there because all you would have to do because you have carpet. And that's the problem. And actually, what I found is most people up there have carpet I guess because it's warmer to walk on then like because everyone a well not everyone but most people down in Houston and hotter climates have hardwood floors. I think since hardwood floor or or tile or concrete. Why because it's cooler. And so you don't get a lot of static build up regardless even though you got like 80% humidity as well. But if you had carpet and you carbon you built you put carbon in the fibers and then you could ground your carpet just dissipate and so you'd be walking on an ESD mat the entire time so you would never build up static charge.
You know there's a there's a simpler solution that people do they buy whole home human modifiers and just put that in your ducts. And then human Yeah,
but you gotta feel like, oh, I guess if it's in your ducts, and you would never have to fill
it up. Yeah, right. Yeah. It's just part of the system. Such that your heater doesn't dry your house even more.
But I like your idea. Yeah. carbonized carpet,
Carbonite, it's called I actually liked that idea a lot. To make it. That sounds actually really good. Because because the problem with like, a humidifier is, I don't know, I've been in a couple of places with the humidifier, and they just don't seem to work that well. Like they don't, they don't get the humidity up enough, in my opinion to actually counteract static. So
I actually have a humidifier that just above my couch, I just leave it on a shelf there. So when I'm sitting there watching TV, like it just makes a little blanket of, of moisture around me just over your head. Yeah, just yeah, just there. And it works in the winter. Yeah. Okay, so yeah, back to the bias test system real quick. Got some, just few things on the next steps. I'm gonna finish my Arduino code on this, which basically, that involves making sure that I'm reading everything the way I want to, but also merging in the Skippy code that we have. So that this thing can actually talk to the computer. And then I'm going to actually solder it into one of my amps over here. That's a known working amp. So I can get real world data. And then we'll schedule the next Python event where we can merge the two together and start building a GUI a GUI for it and apologize to everyone. We it's been brought to our attention that the original Python chat we had got deleted, unfortunately, the video for that. So we
didn't delete it. Amazon deleted it. Right. So hate them. It's not our fault.
The code available, I thought, I think we made that available somewhere.
It's available in the slack, which could also be deleted as well. Well, I have the code. So you have to go. Because so what happened is tWitch will only keep a VOD. It's called a VOD video on demand, I guess, for two weeks. And that happened like four weeks ago. So it's, unfortunately not. So it is my fault. I didn't download it.
No worries. We'll make the before the next Python thing. We'll make the code available at what what it boils down to is this code works. You we can give it to you. It's pretty easy to read. So you don't need to watch us ramble for an hour as we write code over the internet. You could probably just read through this and catch up in 10 minutes or however long. Cool.
So today in the Slack channel, someone posted, I say someone let's see who it was.
I think there was bumbled monk. Yeah, bobble monk,
brought to our attention a company called Java Jeeva Jeeva Jeeva materials. So anyways, it's the self proclaimed the world's first fully recyclable PCB substrate. So it is a it looks like stained bamboo PCBs. Yeah, it looks like wood. And so the whole idea is, when you soak the PCB in hot water, they don't say a hot water in the video. It's, it starts with the laminate, and then you can process it by burning it or composting it or very variety of methods to get rid of it, right?
It's it's a little bit more eco friendly than epoxy and fiberglass.
The whole idea is on one part is it's easier to recycle. But the other thing is, its materials are more eco friendly from the get go then epoxy and fiberglass. So like I think it was like the stat they're using. I haven't read the there's like they have a source. I haven't read the source yet. But they're saying like basically the carbon footprint of just the materials is 60% less. So just the materials, so that doesn't include like recycling it later, right?
Or that's kind of manufacturing with it.
Yes, yeah. So it's called solute board. And so I just went through. There's not a lot of data about this yet because they don't have any technical specifications. It says email us or if you want to work with us, so actually, we're thinking about email and just to see hey, is build some boards out of this stuff kind of looks weird. They do have For, like their example board that they showed in the video is basically an Arduino version, all through whole version, but doesn't have any traces on it on that, on that in that video. So if you go into that video and see the board, there's no traces on that board, which I thought was weird. But then they show a video of a board working with like, with like, those ws 2812 LEDs smart LEDs on it. Yeah. But what I can't find about this stuff is traces go on it like the traditional manner. So like plating and etching, copper, and like, you know, you have to bond the copper first to the board. That's all done the traditional PCB way. So it's just they're just changing the substrate. So why question is, how does the soaking in the hot water when D laminates a substrate? How does that? How do you do the other stuff? Like the silkscreen, and solder mask and the copper? How does? How do you post process that off?
Yeah, and like plating internal vias and things like that? Yeah, a lot of times, those are baths of things.
Yeah. And they do say that it does work with, it's compatible with aqueous PCB fabrication processes, I would suspect that you can't do water wash, because water wash, water wash with water wash. Flux, you probably can't do that because you have to wash that with hot water. But I'm gonna bet you most aqueous PCB fabrication processes are talking about is like just like rinsing or something like that with cold water. So you probably have to use no clean only, or rosin. Rosin actually is a very, would be very interesting because it comes rosin comes from trees, and this is a tree
you know, okay, so one thing I'm curious about, it's, it's really cool. In terms of like, the aesthetic, it looks, it does look cool. It looks like excuse me farmhouse PCBs in a way. But, but one thing about Fr. Four, that's, that's a bit nice is the surface roughness is it's not particularly rough. And it's not necessarily fibrous, either. Whereas if you look at some of the pictures of this solute board, it really does just look like wood. does. And so the the surface roughness? How does that have a huge impact on flatness of traces and flatness of parts? And things?
Yeah, like, we were talking about, just last week, about BGA A's. I want to say it, forget about it. I want to see VGA on this board. Oh, to be honest, though, if this solves a big problem, because the thing about Fr. Four is, it's you can't recycle it. I mean, once that epoxy is cured, it's, it's that going to be that shape folds life until you burn it. So if they can get rid of part of the waste, that's, that's a good thing. And if you can, and the other thing is, it doesn't have to replace, I think a lot of people think about because I saw some of the comments, I started looking this up online. And a lot of people are like, Well, you can't like we just sit there, you can't do a BGA because it's because it's probably too rough for like certain like, well, then you don't use this for BGA is you can use this for your dwee nose, right? You can use this for all the PCBs that are in your like household appliances that are all through whole lot of times or are one sided boards, like and especially since most of that stuff anyways, or throw away products that you you buy them for two and a half, three years and it breaks and then you throw it away. Well, this is the solution because these boards probably don't last. That's the thing about fiber. Fr for it's a forever material.
Oh yeah, if you have these boards in Houston, they might last two or three summers maybe maybe.
Well, it's it does say it resists humidity, but I mean
it seems beyond humidity. It's just hot water all the time swimming in the swamp. You know, okay, so another thing is they it almost looks like there's a veneer on the
that's what it looks like. Yeah, well, it looks like a bunch of veneer, you know, but what it is? It's okay. I hate because we're just why would we are just shooting from the hip now because we there's not a lot of information on this. Yeah, it does look like veneer with a polymer binder. Yeah, all pressed together.
We'll see Okay, so here's the thing, it very clearly has a grain direction going on, it's wood it has, it has a grain direction, I really hope that they like plywood alternate that direction, because this thing would be very weak in one in one stress mode, right? That's another thing about it like ever for. Its pretty stiff. It's decent in that way, and it doesn't have a clear grain direction, which is nice. Hopefully, this doesn't either. But then again, like you said, if it's Arduino, then it'll work for that. Right?
Yeah. So I want to see where the where this goes, can it? Can it take place have two layer boards, for appliances, and that kind of stuff and high consumer grade items, like printers and stuff like that,
you know? Okay, so So when it comes down to it, I think there's actually one major thing that I, I questioned from seeing their their layouts or their their PCBs with this. If they can have plated through holes, I think it could be successful. If it doesn't have plated through holes. I think that kind of shoots it right away and just makes it like, a cool thought concept.
Yep. No, I agree. It has to be able to play through holes. If it can, then yeah. You know, why not? Yeah,
for sure. Saw you and it may be a way, you know, like, if you got a hobbyist level thing that you're trying to get out and sell some, that could be a marketing thing. We use all your board. You know, when you're done with this, you can recycle it.
Yeah, I just want to know how my biggest concern is all the examples. Well, here's one with silkscreen on it. Okay. Is when they bought that copper to that fr for? It's pretty on there. How are they handling recycling that part? That's that's or do they just not? Because does burn it? Which is fine. It just, this would definitely burn a lot easier and a lot cleaner than fr for does. And by the way. Most electronic recycling in quotes, they just straight up burn it. Sorry, people out there.
Yeah, it doesn't look like they put on one of these a bunch of yellow. silkscreen. Yeah. Now, a lot of their examples here I'm seeing are one layer. So it doesn't look like plated holes. But you know, if they can do it, that'd be awesome. Well,
it's also I bet you. They don't have a huge PCB factory, right? It's probably it's probably being done in someone's garage.
Oh, probably. Yeah. Well, this is all Yeah, let's keep an eye on it. This is a lot of proof of concept stuff.
Yeah. And try not to shoot in the foot. I'm, we're just trying to go off what we are reading on their website right now. And it's still just not a lot. I want to see a nice technical document, like, you know, actual environmental testing and that kind of stuff. Because it's gonna be actually really great for IoT stuff. Because most IoT stuff is only a couple years, or phones. People only keep their phones for up to three years before they get a new one. Or they break the screen and have to get a new one. Maybe we should reach out to them and see.
Yeah, see if they want to come on. Yeah, that'd be fun.
But I think I'm gonna reach out just to see if I can get boards made and like, be like, Hey, I'm willing to throw these into the reflow oven here. Cool. As long as I catch the reflow oven on fire element, it's one of those. It's I think it's exciting because it's something that's actually different. In the PCB material, space.
Yeah. It's not like just a new way of making fr four it's a completely separate material.
Yeah, it's a completely separate material. So maybe, oh, oh, man. So Steven, get this. You can make your your amps and stuff out of this stuff, which will give it a warmer wooden tone.
Ah, there you go. Now, how about this? Some DEF CON badges that some of the secrets require you you put it in a bathtub and D Lam parts of it.
peel it apart?
Yeah. Like there's stuff inside. Great. You have to recycle your your DEF CON badge.
I actually liked that idea a lot. People would hate it.
Yeah, sure. That would be super fun. No, no. Well,
because people like to, like display their badges. So that you had to just you have to recycle your badge.
At the end of the conference, well, okay, you could have an FR four badge that is somehow bonded to one of these. And in order to access parts of the other bed, you would have to de Lam this off of your fr for.
I know, I kind of liked the idea of like, you must return it to the earth. It's not a really Duff county theme, but no, it's not common themes of any kind of kind of hippie. Recently,
def cons got a lot of like, robots and things like that
going on. Yeah. The next year is 30. It's going bonkers. Yeah, you should come out for DEF CON 30.
Well, let's see what's up. Yeah.
Okay. Next subject to Oh, so I was working on a project at work. Loan hole, hold another tester, test fixtures, but kind of move on. We're working on a lot at work. And I had a, I was doing some experimenting with high voltage and high currents with this tester. And I had accidentally hooked up backwards. And basically, I dumped like, close to a kilowatt into a little relay, little signal relay. And it it stuck. Basically, it was testing fine. And then that part of that event happened.
That relays no longer right,
that really became no longer relay. It became a fuse.
I continue continual trace, it became a trace. And, and the thing is, I hit the current limit of the power supply and safely failed all that good stuff. And I was like, Well, I and I kept testing and I kept failing there. I'm like, Okay, what's going on? So I beat tested the board and I'm like, okay, pulled the relay off. Unfortunately, basic design beat tested, really, it was continuous, even though with no power. So I'm like, okay, that's probably bad. So I'll replace it. So I replace it. Board was working fine after that. Unfortunately, through the process of thermal cycling the relay while y'all was a while was desoldering it the stiction came loose. Oh, you broke the weld bead. I broke the weld bead. But I did I didn't manage take a picture of it. So let me I'll show you all the picture
is show until today. Let's see if this work. Oh, I can see it. Okay.
So I'm gonna try to make it so the stream people can see it, you know,
when so what well, Parker saying that by the way, we do stream these episodes every every week on Twitch. So if you want to watch us live and comment and get on Twitch,
get on Twitch. So you can actually see the, the where the little arc happened and welded those two paths together.
Nice. How much how much current was flowing through that?
you said a kilowatt worth of energy.
It was like 600 volts it's probably ampere Halfmann
Okay. Well, that's so thermal cycle cycling it broke that well, that's interesting.
Yeah, it was able to pop it I actually tried to hit the relay to see if I can make it snap out. But it took Oh, it was thermal cycling. They finally broke it though. But I was able to I slice it open with an exacto blade to open up the enclosure. Have you
have you tried using this relay again actuating it you should try actually waiting to see if it's if it recovered?
Let's put it that way. It would probably work the problem is one I removed the CEO because the CEO to relay so now it's exposed but two, it would probably eventually arc again because the codeine is gone from that spot. Yeah. It probably would work fine for low voltage applications, but I would never trust in that in a production tester.
None and I'm just I'm just curious if anything else was damaged in the event.
Oh, like the um I did power it up so the coil still works. Oh, it has in the in lease may continue. So it functions fine. Got minus the arc weld. I'll post that picture up on on our podcast notes. Everyone else can see him. So one last topic for today. And this kind of man, this is all everything's been kind of connected today. It's been great. It's fixing circuit boards. And kind of like the method How you go about fixing them. And here's what I mean by this is this is not like diagnosing, like, oh, I have a board that's failed, and why did it fail? This is more like, Okay, you blew something up and it made something. How do you go about fixing that? Or like, or acid damage from batteries? Stuff like that? Yeah,
yeah. Or, or a straight up crack board? Yeah, yeah. Yeah. Certainly seen those before. This is based off of a Reddit thread that you saw, right?
Yeah, there's based off of red thread, I'll post a picture. I'll post the link in the podcast notes. But here's the picture for the people in the Livestream.
So yeah, this picture. Yeah, this board is roasted.
This is pretty rough.
And yeah, and the court, the caption for this image is possible to fix
the question mark. My favorite, though, is just looking at the like, this was been on the beach somewhere. It's got sand in Google. It's grit. I bet you're actually the inside of my multimeter looks like this. Probably.
So okay, let's identify, Okay. And it's obvious to see that something bad happened to this board. But let's identify all the things that look like they are bad with this board. So obviously, there's multiple sections that are charred. Yeah, just so that could be oh, gosh, that could be a lot of things. But I would I would suspect that first that's a component that actually burned on the board and not the board itself. Yeah, that's probably on the surface, which is nice.
does look like the solder mask in some spots had vaporized and it's exposing because you can see actual copper color. So it got so hot where a component or it got so hot that the solder mask basically flaked off. Yeah, he's not getting enough copper color underneath, but did not get hot enough to de laminate the copper off the board. So it was in that sweet spot.
That magic zone zone. It also looks like Okay, so there's there's some spots that look like the lamb where copper layer is lifted. So that's no bueno bubbles. Yeah, yeah. And, and some through hole part is just not in there anymore.
Yeah, not in there anymore. That's part of the part that went goodbye.
Probably Yeah, cuz it looks like a whole bunch of high current traces that are feeding something and it's around those areas. Yeah, and then I think this is looking like this board was conformally coated. Somehow because it's so glossy. It's such a switch to the second picture. Oh, yeah. Okay. Oh dear God. Even worse. Yeah, this has like okay, woof. This is like everything there's charring there's solder mask missing. There's it almost looks like I think
that trace exploded. I think that's what happened. That's the original event. It was a trace got a trace exploded.
That's a lot more than a trace that's a copper pour
more copper port exploded. Yeah, taking out like all the capacitors and components nearing. Yeah, basically,
it's like vomited everywhere. Because the boards covered in something
copper slag probably. Yeah, have you ever noticed that I've had a resistor explode before and it it partially desoldered itself before it went? And so the liquid is solder had soldered re soldered itself to like the enclosure before it looked like it looked like old splatter, except that it was solder and there was actually stuck to it. So it wasn't like it just like I don't know it kind of like soldered itself to the enclosure. It was impressive.
Oh, okay. Have you looked at the third image in this list yet? That to me looks like electrolytic caps have I've given up the ghost there's like part of the wrap wrapping inside that has exploded also there's a ton of rust all over this enclosure which tells me that you know it was in not great shape before the event. Oh yeah, look at that. It's
all the capacitors are Yeah, they peed unraveled. Yeah. Awesome. Maybe it was a capacitors failed. And then causing is probably a power supply causing the rails to go out of whack and which caused the traces to
explode. Yeah. Yeah, it's it's certainly a cascade, but it's definitely
environmental problems first, because it's quite rusty inside there.
Yeah, like all. So I don't know what this device is. But there's there's a bunch of frames that looks like power transistors are attached to. Yeah, look at the Yeah. See, that's all the wrapping of the inside of electrodes, I think. Yeah, this thing is messed up and possible to fix the answer is sure anything's possible to fit what you want to.
I like how the first comment is, that thing is fucked. I think I would look at that and probably try to find a replacement. I wouldn't, I would not try to fix this thing. No, that thing's rough. With enough time, money and patience, it's possible to fix anything except this.
Yeah, it looks like both water and heat fire damage. According to someone else, I'd get all the above all the above.
So back to the question.
Fixing circuit boards. There's like, wire like green wire fixes. I would say. So like, I'm gonna list off like some tools I use, I have a fiberglass pin. And it's designed for taking off solder mask. So like, because a lot of times, if you have a cut trace, you have to be able to solder to the trace somehow. And it's covered with solder mask, so you have to scrape it away. I found like using like an exacto blade or like tweezers and scrapes all mess is just not a reliable way to really clean it. So get a fiberglass pen. And so that way you can actually like because you're basically sanding the solder mask off. But that way, when you actually break through to solder mask, you actually have a smooth copper surface that you could actually solder to it that's clean.
Well, and that's the big thing. Just scraping solder mask off is not enough. Like if you expose the copper, you'd be surprised solder doesn't stick to it particularly well, you have to clean that copper and get crud off the top of it. A lot of times you get through the solder mask and you think you're all the way through but there's still residue left on top of the copper. And trust me you're not soldering through that. Yeah.
Keep a variety of different wire gauges. I think the smallest I use is like 30. And then up to like 22 ish. And make sure it's solid core instead of stranded solid core doesn't the problem with stranded is when he tried to solder a little small stuff the strands play. There's probably a technical term for that. But
splay splay key. You know, honestly, if you have through hole resistors, cut the legs off. Those make really great little jumpers that you can you know, if you have a broken trace, you can jump across with those solder mask pen.
They're not yet the exact opposite of the fiberglass pin solder mask. And so what do you repair? Yep, mask it off. That's not tinned so that you prevent corrosion.
If you want to get really fancy, they make trace repair kits. And if you're a really good solder, you can lay down new traces, and put solder mask over top of it, and you can make it look pretty good.
I've seen some professional repairs, I couldn't even tell the difference. Yeah. But that they're actually at that point. They're basically scraping off the solder mask, laying down a new piece of copper that's got the bond material on it, and then pressing that down and then re re solder masking over.
It's very labor intensive. Very. So you only do that if if your board is very expensive, and you need to fix it that way. You
want to make sure your boss doesn't know you messed up.
Isn't that the rule of becoming an adult fixing problems before anyone else finds out? Yeah.
But then you say that you you then you say it was a problem. But you weren't mad. But don't worry about it. I already fixed it. It makes you look competent.
When another problem. The problem with that is if you say you fix it, and you actually show that you fix it, they're like, Oh, well, this guy can do it. And then you get a promotion to do more of that.
Like it's like setting the clock on your VCR.
You're a wizard.
Well now your parents always asked you to do that. Well, VCR programming clocks on VCRs isn't really a thing anymore. So I don't actually think like no, because we don't. No one really has blu ray players yet either. That clocks off. Yeah, everything gets streamed nowadays. lost to the age at times. It's like the Save icon being a floppy disk. I wonder what the cut off ages of people knowing that's actually a floppy disk or that's just the Save icon?
Probably born in the 90s sometime. Could be guess,
late 90s Maybe there'll be one person that will say it's zipped that drive.
Did you have a zip drive? I did. I did. Things were legit. So big floppies fat, floppy, hard floppies we didn't really cover that topic that really? Yeah, you know, okay, so in this situation, in this image here, which everyone should go check out this because this is just laughable that somebody's been asking if you can fix this. But if something like your copper poor, like in this situation has has exploded, you're, I mean, that's not necessarily repairable, you're starting to get past trace width, unless, unless you're, you're confident that you can put a wire across this copper for and, and just wiring it's not going to cause a issue. But in this situation, there's so much charred. Honestly, the FR four is burnt beneath the copper, that you're not going to be able to repair this really. And I'm looking at this too. There's, there's electrolytic pads that the pads are bad. So you would have to solder the electrolytic in a non ideal way. So you know if this was something where it's like I don't know if this is like an heirloom radio that your grandmother had
a shuttle power supply.
If this was something that you like, you could repair it gingerly put it on a shelf and never touch it again. Then yeah, sure. Maybe you could get away with some of those repairs. But this is this is beyond.
I mean, rusted it is.
It lived a pretty hard life. It looks like it yeah, it wouldn't you never know maybe this was like in a saltwater area and just kind of corroded.
It was like a motor controller for like a barge
Yeah, I don't know. What is that? It does kind of look like a power supply but it does also have big power transistors that are bolted to
you know, actually reminds me of car radios. Huh? Yeah, that's what it reminds me of.
It does look like that, doesn't it? Yeah, I bet you maybe this was a boat radio,
or car radio and it was just in the junkyard and with a broken windshield. So just got rained on forever.
Oh, so it clearly slipped into something you see down at the bottom in that third image. It has like a big power connector that looks kind of like something where it would slot into I don't know some other kind of mating connector on the other side. But it's it's larger. I'm
actually looking at that. I'm looking at that diecast heatsink this is totally a radio for a car. Is it? Yeah, that's I mean there's probably other things that use that style but it's got really cheapy uncoated sheet metal stamped enclosure, car radio. And that actually that connector reminds you like a car connector to for audio. I betcha that's what this is.
Yeah, it's and it has a big multi pin amplifier thing. It is a car ready.
Yes, I scroll down and the OP says what we have here is the amp one from a 2005 Saab nine three
that has the internet you'll get answers. Yeah.
Oh, so it wasn't a they were looking to get it fixed because it costs 150 bucks to get a new one.
Yeah, 150 bucks is I think considerably lower than the repair cost on this.
Yeah. I would say I would buy another new one for 180 bucks. But yeah, man, so rusted it is that surprisingly that that was in a car that was being driven around. Okay, so that was the Mac fab engineering podcast. Oh, do you have anything else Stephen?
We were your host, Stephen Craig and Parker Dolman. Take it easy.
we have two questions. From Twitch, and it is favorite mainstream beer. So I'm gonna guess that his favorite like macro brewed beer. Mine would be banquet. Coors banquet beer.
If montecchi colts neck can count. I'll pick that.
Yeah, I think that's so I think that's a craft beer is I don't think I don't think they're owned by a macro like InBev or anything like okay, well,
then I'll say Yingling. Okay, is that a macro beer? It is
who the who owns
them I don't know.
Second question is favorite malt liquor?
Oh, make grenades. Yankees grades come on like that's an easy one.
I didn't drink those at all until you. You were like we gotta drink Mickey's while like brewing beer. I think we're chasing craft beer and then like we're drinking Mickey's
guy 50 a six pack.
That's expensive. I think I got him for like 475 before. Anyways, let everyone take it easy.
On this episode, Stephen talks about his new bias test system hardware and software. The bias test system is a purpose built test hardware system!
Is grinding out math problems just busy work? Is the current state of Math class curriculum hampering the real life deployment of engineering skills?