- Parker has been hard at work on the PinHeck Test Fixture. It uses over 160 pogo pins to test every single function of the PinHeck Pinball System. There will be an article about designing pogo pin style test fixtures for production in the next couple weeks.
- Stephen tested the Diode Compression opamp he built a couple weeks ago on the FX Dev Platform. Sounds great!
- Noisebridge, A hackerspace out of San Francisco, wrote an article about the large variety of fuse quality. Cheap and gray market fuses tend to not break the circuit and melt. Potentially catching fire. Something even as simple as a fuse should be tested in your product to ensure everything works correctly and safely.
- Arduino.cc released a new ARM build of their popular IDE. Users can now compile code and develop on their raspberry pi style devices. Parker thinks an Android version would be great to tweak and push code up over Bluetooth or WiFi.
- Semiconductor materials market fell 1.5% in 2015. Mainly from changing the die bond wires from gold to copper. Materials to make semiconductors are inexpensive. Labor and process driving the cost of semiconductors.
- Sleev is a Kickstarter that sells adhesive lined heat shrink tubing for $3 for 2 inches. You can buy the same stuff at McMaster for $5 for 4 feet. Goes to show that consumers will buy anything with a fancy graphic and video.
- Stephen thinks we should Kickstart the “SKRü”.
- If you have any silly ideas for Kickstarters let us know via twitter: Parker Dillmann, Stephen Kraig, MacroFab, INC.
Special thanks to whixr over at Tymkrs for the intro and outro theme!
About The Hosts
Parker Dillmann is MacroFab's Co-Founder, and Lead ECE with backgrounds in Embedded System Design, and Digital Signal Processing. He got his start in 2005 by hacking Nintendo consoles into portable gaming units. He also runs the blog, longhornengineer.com, where he posts his personal projects, technical guides, and appnotes about board layout design and components. Parker graduated with a BS in Electrical and Computer Engineering from the University of Texas.
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.
Host 1 00:10
Hello and welcome to the macro fat engineering podcast. We are your hosts, Parker, Dolman
Host 3 00:15
And Steven Craig.
Host 1 00:16
So this week, we didn't really work on any of our company projects. We've been really busy with some customer stuff. But I did get to work on the pin hack fixture. Jig tester thing.
Host 3 00:32
For those who don't know, you want to explain what the pin hack is.
Host 1 00:34
So the pin hack is a Pinball Controller is designed for spooky pinball LLC who are up in Benton, Wisconsin. And it's a Pinball Controller that controls all their controls the servos steppers solenoids, switches, lights, etc, etc. all in one package on one board. Yeah. Most typical solutions for pinball controllers are actually multi boards, you buy one thing, and then you buy another thing to the drivers and all that stuff. So this is all one board solution we've made. I think we've made like 400 boards or something like that, so far for spooky. So the problem was the lot, the old way of testing required us basically plugging in lots of like, little tiny boards via connectors. And so it took like, 20 minutes to test that a board, which is when you have 100 boards to do you know, it takes, you know, half a week to test all these boards.
Host 3 01:33
Well, and your test required a lot of input from a human where they had to look at something press a button knowledge, a lot of things. It took
Host 1 01:40
Eight different programs. Yeah, right. Yeah, to actually test fully test and program these controllers. And so over the past couple weeks, I've been designing on the side, a fixture jig, and this week, I got most of it programmed. And it can test all the I O functions and onboard systems of the PINet controller in one minute and think was like 26 seconds. Wow, how'd you pull that off. So basically, it does its own onboard microcontroller that actually talks to the pin heck, over a serial connection. And it talks via pogo pins. And so it's got pogo pins on all the IO then come off the board. If you flip the board over, it's got actually little gold contacts that that the pogo pins can push into. And so it's basically a bed of nails kind of thing, where the board sits on it, and then you put clamps down and then hit go on the computer. And you only need to use two programs now. A serial terminal and the picket three programmer software. I think it's called MP lab IP. IP.
Host 3 02:51
IP. Yeah, IP. Okay. So, so pogo pins are actually really great solution for these kind of things. Yes, because adding gold contacts to a tee a PCB really doesn't add
Host 1 03:03
Cost. Yeah, since um, since macro fab, we all use Enoch, you basically have a gold finish already done. And it's really smooth. And so your pogo pins get really good contact on that kind of stuff. Yeah. And then through make sure that the the pin hex sits at the right height, I found these really cool standoffs that have swag, and swag and on it. So it creates a shelf that fits into the mounting holes that are already on the pin. Heck,
Host 3 03:31
Okay, so they help align the board onto the jig. Yeah, I'm using the
Host 1 03:35
Original mounting holes. So I didn't add any new mounting holes to this thing. Cool. So it just kind of sits on it. And then you put big clamps, toggle clamps on it, and it kind of just sits there. Awesome. So it's, it's fairly repetitive. It's got some issues because the clamps don't distribute the pressure evenly over the whole board. Yeah, it basically puts pressure right where the Cago clamps at. And so the next revision on basically take a piece of acrylic, since that's very hard and rigid, and mill out a cover jig, like a clamshell. Yeah, so you would put the board down, put the acrylic piece on, and then just use two toggle clamps to push the whole thing down
Host 3 04:22
And provides even pressure across the whole entire board. Okay,
Host 1 04:25
That's the next step. But it works pretty well. As is.
Host 3 04:30
So So if anyone hasn't actually seen toggle clamps before, go go and check them out. Go go do a Google search for them. They make fantastic Why
Host 1 04:37
Does Google Search we're just put it in the link in the in the blog. Oh, yeah,
Host 3 04:41
Great. Yeah. Even some pictures. They're really great for putting specific pressure on specific points in whatever your fixture is. You can get them for vertical applications and linear applications, all kinds of stuff in there. They're not expensive.
Host 1 04:56
No, I think they're $8 on Amazon. Yeah. But that's what Prime shipping so they come in two days.
Host 3 05:03
I've actually seen a lot of guys using Lin, C and C's for holding their, their workpiece pieces down
Host 1 05:07
Into jigs and that kind of stuff. Yeah, yeah, yeah, I've seen. My first experience with them was actually a woodworker, one of my neighbors when I was in high school, and he did a bowls. And so when he would glue up his, all the layers for the bowl, is he would use toggle clamps. Okay. And I think there were a lot more expensive back then, because we didn't have access to cheap parts in China. Back then.
Host 3 05:34
Yeah, yeah. I mean, $8 is pretty cheap. Yeah.
Host 1 05:37
Well, it's a lot of mechanical parts do for me, yeah, it talks a basically, there's this parallax propeller on the jig, or the fixture. And that talks. It's basically programmed up to run the entire test suite. Over basically a n one serial port. Okay, it actually talks to the pic 32 That's on the Pinhead board. And that's the only thing we have to program separately. There's a test program that runs on the pic 32 That basically sits there and waits for commands. And then the the fixture tells the pic 32 What to do, like turn this pin to this date, turn this pin to this date. Okay, and you just test. So when and then basically the pic 30 Does that sends an acknowledge back to the the fixture, the fixture goes okay, and then it waits a little time just to make sure you know things have settled down. Yeah. And then it's got a whole string of seven, four, HC one six fives, and then it basically pulls all the IO looks at the states and goes it did weren't supposed to do or didn't do. Right. Basically, it has a huge memory lookup table of this is what the states are when you tell it to do this. And then just it just it just comparing it just goes just a straight comparison. Okay, there's so so with
Host 3 06:55
With the fact that this thing controls lights and servos and matrix banks and things. Does the test just encompass pulling pins up and down? That's all it does. Cool. And so that's really easy. Yeah,
Host 1 07:08
All it does is you tell the pic to do this to pull a pin certain direction, and it doesn't. And then we read the state. And so that way, you can see if the state actually changed. And if another pin changed, then you can actually detect shorts.
Host 3 07:23
Okay, so when you pull a pin short, you check the pins
Host 1 07:27
Around it. All the pins. Oh, you check, you check the entire state of the board.
Host 3 07:31
Okay, how many? How many pins are on the board? Or how many test points? Are you? Oh, 160
Host 1 07:36
Host 3 07:38
So every time you change a state you have to compare all 100? And
Host 1 07:42
Yeah, but the parallax probably it's on the fixture is a 32 bit. Yeah. And so you only takes like, what was that? That'd be like, six ops? Yeah, yeah, it's pretty fast. You actually, I actually spend more time waiting to make sure that the pic 32 did all that stuff they should do then actually comparing. So it takes longer just waiting.
Host 3 08:06
So So what happens when a board passes?
Host 1 08:10
So when it passes, the serial terminal will spit out board was passed it nice and sweet to the point. And if something fails, you actually have to go back into log and see what failed. But it gives you an indicator. Yeah, something failed. Yeah. So it will do is it will actually, it'll spit out, like solenoid for failed. And so then you can just look at the board and see, you know, what's wrong with that, eventually, the test program won't give you a, a good indication of what caused the failure, okay, by need more data from all these testing to figure out what, like, if it reports this, it's this issue, I need to build up that database first. And then I'll program that into the test fixture. And then the fixture will actually tell the line operator, oh, this solenoid failed, and it's probably the FET forte. And now I can actually program look at FET you know, soul 15.
Host 3 09:07
And you'll know exactly where
Host 1 09:08
To go exactly where to go. Okay, so
Host 3 09:11
Shoot, I lost my train of thought here. Oh, so when when it gets a failure, you can have a failure from a couple different failure modes. But you already said that it can detect a short, it can detect a short and it can also detect a just a pin not changing state. Correct. When it when it receives a short. It obviously flags that as an error. But does it indicate it as a short?
Host 1 09:34
Not yet? Okay, well, though, because usually when you have a short it will, because basically two pins will go hi, yeah. And then if it detects that and goes, Okay, it's probably a short between those two pins. That hasn't been programmed in yet. Yeah. But yes, it can do that.
Host 3 09:48
Yeah. When it comes to programming, these kind of test jigs you spend X amount of time just getting it working. You can spend eight times as much programming and in all the cases is for errors. Yeah, exactly. Yeah. So
Host 1 10:04
I think that's everything about the test jig. Eventually we want to have that I'm going to write a blog post about this test jig. Cool See, help other people develop their test jigs for their production runs. Awesome.
Host 3 10:15
Yeah, I think that's that's a great idea. Especially because this isn't a board with like, three test points. This is a massive one. And showing success with that, I think is a great idea.
Host 1 10:25
Yeah. I think they started testing this morning, and they got through, like 30 boards. Wow, already. Yeah. So it's definitely an improvement over the last test method. Yeah. So, which basically was an engineer, one of us sitting down and testing these boards, right?
Host 3 10:45
What's the size of the board?
Host 1 10:47
For the tester? Both of them, the pin hack is a six layer board. That's 14 inches by six inches. And the the test fixture is 16 inches, which I think is the max we can do. By what I mean? Yeah, eight, nine.
Host 3 11:08
That's a big board.
Host 1 11:09
And it's to layer it. Yeah, cuz
Host 3 11:11
It didn't need much. No.
Host 1 11:13
It's just if you look at the board, there's a there's a prop on the top. Some connectors, you can connect up to it. And then like a big ring of the one six fives and five, nine fives. Yeah, boy around the board.
Host 3 11:24
It doesn't it didn't even have bottom load did it? I think
Host 1 11:27
Most of it is on top layer. Yeah, I think the bottom is just ground. Because because the
Host 3 11:31
Jig is just looking at states of a couple pins. 100 Something
Host 1 11:36
Pins. Yeah, nothing. Nothing fancy.
Host 3 11:38
Nothing too special. Oh, cool.
Host 1 11:39
And I haven't had any issue with noise, like noisy pins or something like that. I haven't had any issues. I thought I was gonna have to do deep bouncing or something. But you don't even debounce you just I just read it in. I just wait for the states to settle. That's probably why I wait a little I wait a little bit. Yeah, but works. That's convenient. Yeah, then I think it's because I had that huge ground plate underneath. That keeps everything pretty quiet. Yeah, shield. It does have decoupling and all that stuff around the board. Yeah. But
Host 3 12:06
That's the standard, the standard stuff. Cool. So I was working on the FX dev board a bit more this week. Oh, yeah. Yeah, yeah. So finally got all the breadboards assembled onto the actual dev board,
Host 1 12:24
All 160 pins a solder times two,
Host 3 12:27
I guess we have two dev boards on there. So that's been somewhat of a fun adventure, working with the manufacturer, the board and getting getting the footprint right for all of those pins to fit through the board. But we we got it done. And it actually looks pretty nice. And I ran my first my first test on it. I've been testing the peripherals, but But this week, I actually built my first dev circuit on there. So I actually selected a tube screamers style of circuit. Okay, just because I one of the things that
Host 1 13:01
We can do you actually meet I effects pedal
Host 3 13:04
Correct? Yeah, an effects pedal on the board. One of the things that that you can engineer as much as you want, but you're not going to know until you actually try it out is what is your noise floor look like? Did you route your signals in a way that makes sense for everything. And I was kind of sweating a little bit until I actually built something. And I'm happy to say that it's it's very quiet.
Host 1 13:28
Yeah. The video he had on Twitter was really nice. Yeah, yeah. He actually brought his guitar in app and played some music. Yeah,
Host 3 13:35
It was. It was it was kind of fun to be able to, to just jam all day in and build audio circuits. So I actually the diode compression op amp.
Host 1 13:46
That was the discrete op amp we made. Yeah, yeah. We would talk about that a handful of podcasts ago. Yeah. About three or four podcasts go. Yeah, so I had
Host 3 13:55
Originally built that in anticipation for using it in this circuit. And that op amp has a really unique configuration in the fact that it has. It has diode feedback on the internal class A amplifier inside the op amp. And we can post some stuff on the blog to kind of give demonstrations of what that is, but But effectively, when you drive that op amp into clipping, it transitions from the linear to the nonlinear range in a much smoother way than a standard Pampas. So instead of just completely hard lining, it kind of rounds into it. It rounds and what I found is it rounds a lot. It's the only word I can think of for this is really smooth. This the pedal I built is not harsh at all. It was just really really smooth. It almost doesn't even get characterized as distortion. It's It's really strange.
Host 1 14:49
It will put a audio clip up on Yeah, on the blog post.
Host 3 14:53
Yeah, and I have a handful of pictures so we can we can show that off. So yeah, I was I was really excited because I got it got a proof of concept there and it worked beautifully. It's exactly how I wanted it to be.
Host 1 15:03
Yeah. The you had some, I think was film caps on there too.
Host 3 15:07
Yeah, when I do this kind of stuff, I, I'd like to use the least amount of electrolytics as possible. So even even for like my 10 microfarad filter caps, I made a mess. I used film caps for
Host 1 15:19
This. Yeah, I was actually, it's been a long time since I've seen film caps. I'm always surprised how large they are.
Host 3 15:24
Which is funny, because these are actually, in comparison. Pretty small. But they're, they're huge. I typically call them hot dogs, because they're, they're, they're big circular red. Yep. So
Host 1 15:37
Don't try to eat them. No, no,
Host 3 15:38
No, no, not fun.
Host 1 15:40
I wonder if you if you overvoltage a film cap, if it would turn. Like if it striped up like a hot dog on the grill?
Host 3 15:52
It would be hard because most of these film caps are in the 600 volt range. So you really got to do something to get
Host 1 15:57
You're not having a super simple power supply can do that. 600 volt?
Host 3 16:01
No, no, no, I have a I have a 400 volt regulated power supply back of my shop. And even that won't do it.
Host 1 16:11
Yeah. The the crazy thing was that that that setup that you built has actually how quiet that that? Pedal was. Yeah, on those breadboards
Host 3 16:26
They do one thing so so when it came to the layout the board is is 10 inches wide. And and it has the input and output jacks on opposite sides of the long length of the board. So doc style. Yeah, how exactly how big. So? So yeah, the thing is, you have to run a trace all the way across the board. And that's where I was kind of scared is my is the output trace is that so long that it's gonna pick up a bunch of junk. But I did my did my homework, and I put I put grounding shields around it and it just fine. Absolutely worked out great.
Host 1 17:03
And no, no 60 Hurt harm or anything like that. Right? Nope. You know where to be interesting is actually hook it all up, and actually put the scope on it and see if we can pick up anything in the 60 Hertz range.
Host 3 17:16
I'm sure if we tried we could make it happen. I put the lights near it.
Host 1 17:20
Yeah, fluorescent lights. And cuz you got to we got to experiment with all this kind of stuff. Yeah. And like for 80 megahertz. I bet you we picked that up. Oh,
Host 3 17:30
Yeah, absolutely. Well, and the thing that's, that's cool about it, too, is since it has all the peripherals on board. You don't you don't have to run jumper cables, which are just giant antennas to all your stuff. So the some, some of the most critical aspects of it are already taken care of, and shielded on the board. So it's kind of not surprising that it's quiet. But I it's a it's a really nice.
Host 1 17:55
Yeah. And those breadboards that we have for it have really good retention force.
Host 3 17:59
Yeah, yeah, they're, they're fantastic. So we don't get any
Host 1 18:04
Noise just from a connection being loose.
Host 3 18:06
No, no, the only the only thing that I think you would see that's a negative on those and it's a negative with every breadboard is you get a whole bunch of parasitics you get capacitance from strip to strip and inductance of the actual connections. But when it comes down to building guitar pedals, guys aren't going to care
Host 1 18:23
On doesn't really matter. Too much stuff unless you do an RF that matters.
Host 3 18:27
No, no. On to the RFO, Section
Host 1 18:32
RFO. So I found this article by northbridge, which is a hackerspace. In San Francisco. Yeah, it's one of the I think one of the biggest hackerspaces it's up there at least. And they were finding fuses for some kind of high amperage application. I can't remember exactly what they were using the fuses for. But they were they started to test some just see if fuses actually would break at what they're rated for. Hmm, because they had a whole bunch of 10 amp automotive fuses.
Host 3 19:07
That's probably one of the last things you would think to test tests
Host 1 19:11
Because you think oh, it's, it's a piece of pressed wire in a plastic enclosure, right? There's nothing special about it. Yeah. Or is there or is there so they were testing so they had some fuses from little fuse. And then they had $1 store fuse, and then a fuse from eBay. And they're all very varying in price. A little fuse was the most expensive I think they bought it probably from a US distributor somewhere. Okay, so it was an authorized little fuse. The little fuse broke at exactly 10 amps. Like almost instantly and it was rated for 10 and it's rated for 10 Okay. I think it was the eBay one took up to 15 amps. And it took about two minutes before it the blow. It didn't blow eventually, but it was old melty. Okay, That's not good. Yeah. And then the Dollar Store one kind of never broke. And I think they were pumping like 20 amps through it and it kind of got really melty and almost caught fire.
Host 3 20:10
So the Dollar Store one was just a just a wire. Yeah, but
Host 1 20:15
When you look at them, they look almost the same. Yeah. And so I'm gonna assume basically, it's the metal difference. The metals were different in each one, probably. And so basically, some factory in China copied the mechanical layout of little fuse, but not the outer, but not the alloy.
Host 3 20:37
Hey, do you know 20 amps? That's pretty close to 10. Right?
Host 1 20:41
Sure. Um, the crazy thing is we see all these like cheap 3d printers. Yeah. And they all use like the ramps board. And they use clones of the ramp board. Yeah. And the thing about the ramp board is it uses automotive fuses, blade fuses like these. For now, the legit ramp board that you buy here in the states, if you buy a real one, it's pretty expensive, but it has authorized, you know, fuses on it. If you buy a ramp board that was made in China, from, you know, XYZ Alibaba store, it's probably going to have XYZ Alibaba store fuses on it. Right, which will probably, you know, never break, what volt, what aperture they're rated for. And then you burn your house down.
Host 3 21:34
Fantastic. Well, a fuse is, if anything, I mean, it's only purpose is a safety application. Yes. In fact, all of its purpose is safety application. Therefore, it falls under a enormous amount of standards. And I'm sure that Littelfuse is constantly audited, they probably take a handful off of their line while they're manufacturing and they break them. I'm sure the dollar store and the eBay one are just hey, whatever. Wouldn't be wouldn't surprise me if it was stamped with a with some of the UL and CSA marking.
Host 1 22:11
Oh, yeah, they are. Yeah, yeah. They're stamped with them. But they're, they're not tested. No,
Host 3 22:15
No, no, clearly.
Host 1 22:17
And so this is, this is a good caution for using an unauthorized and gray market parts. even for something as simple as a piece of metal.
Host 3 22:29
The rule of thumb is just don't use the gray market. And
Host 1 22:32
Well, sometimes it's okay, in some parts you can only get in gray market. But something like this, if it's a safety safety thing. Make sure you get there legit stuff. Yeah. Or test your products by Gray by the gray market stuff and actually test it yeah, get
Host 3 22:49
Get samples we've had experience with the gray market at macro fab. Yeah. And some of its good some of its not as good so it's it's a it's a crapshoot. You gamble when you do it?
Host 1 23:01
Yeah. Every single time. Because you don't you just don't know. And you don't have anyone to run to when it turns bad. Yeah. Yeah. You don't have a distributor.
Host 3 23:09
There's no customer support for this kind of No. Yeah.
Host 1 23:14
And then a Arduino. I think it was arduino.cc. I can't remember what I think it's arduino.cc. Yeah. Which one is the one that we were supposed to? Like? I don't know. I just two of them. Yeah. Anyways, Arduino came out with a new IDE. It's same as the old ID except it runs on arm. And so on your Raspberry Pi, you can run the Arduino IDE now, which is pretty cool. Now that is cool. Yeah, you can use your $5 Raspberry Pi Zero
Host 3 23:44
Host 1 23:47
I know. Yeah. And they got me thinking is wouldn't it be cool if your IDE was on your phone? Okay. And so you could be you can change code in the field with just using your phone because everyone brings their phone but sometimes you won't have your laptop and then better would be if you could push the updates over Bluetooth or Wi Fi well the device
Host 3 24:15
Above and beyond that if you had the serial terminal available directly on your phone
Host 1 24:18
Yeah, cuz most phones nowadays have USB on the go. Right? You could just on the go over to
Host 3 24:25
It. Well, even even if it had Bluetooth, you could Bluetooth the serial terminal over that would be cool. Yeah, actually, I
Host 1 24:31
Think that already exists. It does. Yeah. Okay. Well, that's yeah, I have I have a serial terminal on my phone that can do serial over Bluetooth.
Host 3 24:40
Oh, go figure.
Host 1 24:42
But yeah, it'd be really cool. So you could look at your your Arduino code, so to speak and change stuff on the fly and just push the update over,
Host 3 24:49
You know, actually. So so. I'm part of an art collective here in Houston. And we do we do large scale school. For art here, and last year, I did a project where I had to wirelessly communicate to four Arduinos that were up in the ceiling of an art gallery. And I spent literally a week on scaffolding, programming them and making them all do what I was wanting. If I could have just sat on the floor in the gallery, and not had to climb up and down scaffolding 300 times, that would have been amazing. You
Host 1 25:27
Can do drive by programming.
Host 3 25:30
Sit in your car in the air conditioning outside. Exactly. Oh, I'll be nice. That would have been amazing.
Host 1 25:37
So yeah, this is a good thing. I be cool if more ideas come out like this. Like if parallax release, there's what the prop tool to be over more platforms, more more options is always a good thing.
Host 3 25:49
Yeah. Well, and the fact that with for, you know, $40 you can have a full desktop and I'm using quotes their computer. Yeah, Raspberry Pi. And do all your hacking on Arduino for $40. Have it all there
Host 1 26:05
Already go. That's pretty cool. And then there was a article that you linked this week that the semiconductor markets fell? Was it 1.5% last year? Yeah. Well, that's the material, not the market. So the chips chips have gone up? Yeah. Oh, yeah. Yeah, we build more more semiconductor chips, but the material cost has gone down.
Host 3 26:28
Well, and which is interesting, because you would think that, for the foreseeable future, everything's just gonna keep going up as, as humans consume more and more electronics, you would just expect that, but it looks like they're changing the bonding wires in internal to ICs. And that
Host 1 26:47
Caused where they change it from to what.
Host 3 26:49
So since the beginning of ICs, we've always used gold, okay, because of its its malleability, it's easy to get in really fine capability. And its conductivity is awesome, and it doesn't corrode. So there's a lot of great aspects about it. But apparently, now we're using copper, which is cheaper. It's a bit easier to get than gold. But because of that, we saw a decline in the materials. Okay, interesting. Yeah, that's all just because of the little tiny wires that connect the pads to your die, the entire market saw a drop of 1.5%.
Host 1 27:28
It's actually kind of interesting that in one year, enough manufacturers change to that to make a big drop like that. It
Host 3 27:35
Must have been that attractive. And it was probably that easy for them to make a switch over. Yeah, somebody found it. I kind of doubt that it's just pure copper. But it had to have worked with everyone's process. Yeah, whatever. They already have just dumped copper in there instead.
Host 1 27:56
It's interesting. I wonder if you'll see a drop in pricing on chips by doubt it. I bet you the the bond wires such as very small price of that. Chip? Yeah. I mentioned most of it is the labor and attaching the bond wires.
Host 3 28:08
Yeah. Well, what's what's interesting is the materials, the the the dollar amount that for that the semiconductor industry spent on materials. The there was there was $24.1 billion spent on silicone, basically just your dye material. Yeah. And then and then packages in the the metal for the leads and, and all that that jazz came out to an additional 19 point something billion dollars. So we're spending, what's that $40,000,000,000.44 $40 billion in just material to create them. But it ends up being a however many trillion dollar industry. So it's a it's a drop in the bucket?
Host 1 28:51
Yeah. Kind of interesting that materials, I guess most of it's probably in labor, and QC and all in design, rather than the materials, which seems to be that's what most industries are starting to run into the limit of is just a labor costs and that kind of stuff instead of material costs. Oh, it's
Host 3 29:10
Also automated nowadays that Yeah, that's probably their, their big thing. And I'm sure this was a change that was initiated 15 years ago. You know, there's, there's some grad student who who'd been researching this for years and years and years, and they've finally made the change of climate
Host 1 29:24
Change. And then, for the last RFO there's a really goofy Kickstarter out there called sleeve and that's SL, Ev, no ESPN,
Host 3 29:39
Host 1 29:43
And it's basically a device that protects your cables. So there was a
Host 3 29:49
We really want to define it as a device, I guess. I guess they say that's what they say.
Host 1 29:53
Um, okay. So I, I think it was like maybe like this Back in our third podcasts we talked about just a little blurb about Apple was having all that those lawsuits for having bad cables. Yeah, basically the the strain relief was really badly designed. And so people were just having cable failures. And we actually had an guy at macro fab his power charger for his Mac stop working and it's got basically looks like a rabbit chewed on the end of it. Yeah. So anyways, this is something to preemptively stop that from happening. Okay. And it is a piece of rubberized tubing that shrinks when heated, and has a glue lining that keeps it in place.
Host 3 30:42
I've never heard of that before
Host 1 30:44
Heat shrink tubing.
Host 3 30:47
This guy still has a Kickstarter for catering to
Host 1 30:51
Sell heat shrink tubing with adhesive lining. So I went immediately to McMaster and you can buy four feet of this stuff for $5. Okay. He is selling two inches for $3.
Host 3 31:05
Oh, wow. Yeah, this guy must be making a killing. Yeah. Was it funded,
Host 1 31:09
Fully funded fully way over his is asking. Oh, geez. Yeah, it's one of those. We're in the wrong market. His profit margin is like 99.6%. I, he's getting away with it. Oh, yeah. And so I was thinking about this. And so he's not He never says it's heatshrink. Well, I think he says he trick but he's not selling. He never says he's selling heatshrink to mean, he's selling a application, which is what this guy's doing.
Host 3 31:43
So you use heat shrink tubing to accomplish something. So you
Host 1 31:47
Use heat shrink tubing to basically cover wires. That's what you use heat shrink tubing for additional strain relief. Yeah. So he's, but he's selling the application of protecting your fancy Apple cables. Okay, that's what he's selling. He's not selling heat shrink tubing. He's selling that application to you.
Host 3 32:08
Okay. Well, I guess he saw He saw an issue and came up with a solution.
Host 1 32:14
Yeah, I guess. So. I think it's a little dirty. And so we were brainstorming ideas for other Kickstarters that would be like this. So what's the idea you had?
Host 3 32:29
So the first idea I came up with was called screw. And that's SK r u. And that the EU has little EU lots over it. So it's screw?
Host 1 32:40
Yeah. And so we chose the OOM lots to because it looks like a smiley face. That's right. I miss the mix. The person who buys it happy. So
Host 3 32:50
You can use screw to, to fix two things together and hold them together. They're holding together. And I think I can make a killing on this. Yeah, I think I think I can get funded in I'm gonna say less than 10 minutes. Lesson time. Less than 10 minutes. $1,000 $10,000 minimum? Yeah. And you know, I think I think you should be able to buy screw for I don't know, you get a box of three of them for $8 I think that sounds reasonable.
Host 1 33:22
You know, I've always hated that when you buy fasteners and they gives you an odd number of fasteners. That it's like hot dogs and buns, right? You have to buy because most time you need either two fasteners Yeah, or four? Yeah, have a specific kind and you get three so it's like yours have one extra or one enough. Yeah,
Host 3 33:40
Host 1 33:43
So yeah, if any of our listeners have any ideas for silly Kickstarters that are like screw or sleeve give us a tweet. And it's like I think it's you can either tweet us at at macro fab or at Longhorn engineer with no o's and Longhorn Yeah. Or at macro ninja near
Host 3 34:03
That's macro underscore ninja near
Host 1 34:06
Yeah, and all those links are in the blog post. Yep.
Host 3 34:09
So yeah, if any of the listeners have a goofy Kickstarter idea, shoot it to us, and we will. We'll talk about it next week on the podcast. Yeah.
Host 1 34:19
I think it'd be a lot of fun Yeah, I think I think screws got potential squeeze gotta find that that one application that already has the solution of the screw but it's the screws that screw Yeah. Oh, I can't believe sleeve.
Host 3 34:37
I just it's ridiculous.
Host 1 34:40
That's just an insane profit margin man. And in fact that no one called not because Kickstarter is usually actually pretty good about this because you're not supposed to be able to resell a product completely
Host 3 34:52
Well, okay, so he's buying it in bulk and cutting it down.
Host 1 34:57
So he's moving the scissors once and then putting it in? into a polybag with a what looks like a business card for the instructions.
Host 3 35:05
That's that's, and his instructions
Host 1 35:07
Don't have been used the proper way. He's telling people to use a lighter to shrink it instead of a proper heat gun.
Host 3 35:14
Yeah. Well, so it even has a safety concern.
Host 1 35:21
And his a It's not even like he did even fancy stuff for the adhesive. It still has the problem that adhesive line teaching has is when you shrink it too much. That adhesive squirts out the end. So it actually doesn't look too nice. And it gets all goopy because so goofy. And so in fact, in one of his pictures, you can see the goop Yeah, it's like all over the Lightning connector.
Host 3 35:41
Fantastic. Yeah, fantastic.
Host 1 35:43
Well, maybe I maybe I'm a little jealous that he was able to score like $10,003.
Host 3 35:53
Man. Oh, so if here's an example, you buy, you buy a 10 foot two by four. You cut it down into one foot sections, and you sell each one foot section as Wu D would. And there you go.
Host 1 36:11
And use it as a as doorstops
Host 3 36:14
The perfect Kickstarter. Let's go.
Host 1 36:20
Terrible, terrible doorstop. Anyways, so yeah, have if y'all have any crazy ideas or silly ideas, preferably silly ideas. Yeah, tweet us. We'll talk about it. And we're trying to find a crazy application for your screw oil would. So I guess I'll do it for this week right soon. Yeah, I think so. And so we'll see y'all next time on the macro hub engineering podcast, where your hosts Parker Dohmen
Host 3 36:50
And Steven Gregg. Later guys, take it easy.
Transcribed by https://otter.ai