One faithful evening, Parker received a marketing email about gamifying the MacroFab platform and thought it might be a good podcast topic.
Chris Church returns to the podcast to discuss the recent outbreak of COVID-19 or Coronavirus in China and how it is impacting Electronics MFGing.
Lets bring in the new 2020 year with a MacroFab platform update for PCB Specifications and Hellhound preamps! Thanks to all our listeners and guests!
Figure 1: Screen shot of the new MacroFab dashboard.
Figure 2: Artist representation of the UN banning killer robots.
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 are your guest, Stephen Craig and Parker Domon. So this is episode number 48.
We're getting close to the one year anniversary. Yeah, it's
right around the corner. Yep.
So the main thing for this week, there's gonna be a new release on the macro fab website, we're going to have a new improved, you know, as seen on TV dashboard. And, yeah, so it's gonna consolidate basically all our little services like like fulfillment and, and inventory and all that stuff and kind of give you an overview of what you have going on at macro fab.
Yeah. And we're supposed to be releasing this. This next Friday. 29th, the 29th? Or the 30th 30th? Friday 13th. I think today's the 29th. The 30th? Yeah, Friday, Friday, the 30th, we will have a new improved dashboard. So the dashboard, give kind of an overview of everything that you have going on at macro fab. Yep. Just so I said. Right. So yeah, your recent orders, the status of all the orders, your inventory, your cost of inventory, fulfillment, statuses, that kind of stuff. Yeah, it's not kind of like a one one page, you can look at it, and just know everything without having to, you know, keep your own record. Yeah, it's
not going to be the default view yet. We're gonna see how people respond to it. And, you know, give us feedback on Oh, hey, you should change this thing. Or, you know, it looks awesome. That'd be that'd be the best thing ever. People say it's awesome. And we don't change it.
Yeah, that's what you hope for.
But then eventually will probably become the default. Do you at macro fab? So yeah. Cool things.
Yeah. Exciting. So last week, we had just some some minor issues with the minor, the audio. And so a bit of a bit of the podcast kind of cut cut a little short. So we're gonna pick up from where I was talking last week, because I was going on about our, our new Precision resistor project.
Yeah, the resistor of death, which is the current name, we still haven't come in the last over the week, we haven't come up with a better name yet.
So let me recap real quick that basically, we're doing a series parallel configuration of a bunch of resistors in order to get a single resistor that has much better tolerance than any one single resistor that you have. Exactly. So just a real quick recap, we're basically making an array of resistors such that you have the equal number of rows and columns in this array. And by doing that, you can have a cumulative resistor that that basically has a really tight tolerance. So let me let me go over just a quick little calculation thing. So there's a really simple equation to find out what the total tolerance is of an array of resistors and caps, or I'm sorry, of resistors.
It took like half a whiteboard for you get this though? Yes. Simple. Yeah, no, it's a lot of ours on the board.
That's, that's the wonderful elegance of math, you You start off with something simple, it blows up into something huge, and then it comes out. Again, that's something simple. Yeah. So the equation goes like this, to find the tolerance, the total tolerance of an array of equal rows and columns of a resistor. It's, it's basically the tolerance of an individual of each individual resistor divided by the number of resistors in either the columns or the rows. So So we're talking about doing a array of resistors that are 200 rows, what long or high by 200 columns.
So what why did you pick that? What Why did you pick that? 200 number?
So what what I did was I went to Mouser. And I looked for their tightest tolerance resistor that you can buy, which is point 00 5%.
Over on 10k.
That's great on a 10k. Yeah. And how
we calculated this last week, I forgot is actually how much deviation from 10,000 Was that that's a
50 ohms. Deal. Yeah, point 005 And 10k is 50. ohms.
Yep. 50 ohms. That's a little crazy. Yeah, it's real. plus minus, or just
that plus minus, okay, yeah, that's plus minus so 99950 to 10,050. Yeah. So so if you have a an array of resistors, each resistor being 1%. If you have that, erase that you have 200 rows and 200 columns, then you'll get an equivalent resistor of point 00 5%. Now 200 rows times 200 columns means 40. 1000 resistors. So we're actually going to build a board that has 40,000 resistors in series parallel.
You know, this is something that we forgot about talked about last time they got cut off was our my 200 Picking plates. So how many components per hour? Can it do?
You know, I think
he's like, 10,000, right?
I don't I don't remember what the spec was. I know, we were averaging something like six to 7000. Yeah. And that's, that's averaging across. You have other stuff. So yeah, this might be two hours of runtime.
Yeah, that's, I think we should do a time lapse of it just running. The thing is, keep keep going. Keep on well, okay, real quick.
Actually, the my 200 jumpers. I've picked
it. That's because you're getting there. And then I'm like, Oh, wait.
Okay, so let's check this out. Because we might throw a monkey wrench in it. So in order to actually do this properly, you have to have a Gaussian response across the resistors. If the resistors are not evenly distributed, then you don't really get this equation. So in order to help that, we're actually going to buy four reels of oh, 402 resistors. Each reel is 10,000. Yep, up to 40,000. Now, the funny thing is, you think that's a lot of parts, it's gonna be expensive. Each resistor on a 10,000 reel is only point 001. Dollars. Yeah, so a 10th of a cent a 10th of a cent. So each reel is only 10 bucks. Yeah, it's 10 bucks or 10,000 resistors. So we've got $40 worth of resistors on this thing. Now, that being said, if we have four different part numbers, the My 200 is gonna have to pick four different reels.
Yeah. Well, I think if we set up the my five, my 200 correctly, it will actually gang pick. It'll drop the head four at a time. Yep. And pick four up at once. And then place those
Yeah, yeah. A time left would be fun on this. Yeah. And
want to do is because you, you were bringing up the Gaussian curve, right? Is at the same time. Order, basically two reels, right? Yeah. And then I'm going to design a board that we can test the resistance of like, 1000 of each resistor, just to see what kind of spectrum we're looking at, and see what the actual lay of the land in terms of tolerances on these on these parts.
That's gonna be fun.
Yeah. Look, the multimeter we're looking at the test this thing with it has a record function. So we'll be able to plug it into the computer and hit record. And so it'd be like, put the, you know, put the probe on record, probe record and just do that, you know, a couple 1000 4000 times. But
you're gonna need you're gonna need to, you know, put on some meatloaf and, and drink a bunch of beer. Yeah.
I think me cool experience. I think, um, Dave, you told me Dave Jones has done a test like this before, with through hole through hole parts. And this is gonna be surface mount. I'm gonna be putting them on board, solder paste reflow. All that good stuff. Yeah, and then test that way. Cool.
It's gonna take forever,
it's gonna take forever, but I think it'd be pretty. I think we'll get a full data analysis basically, of this project. Yeah, it'd be really cool.
So I'm gonna have to get a little bit blasphemous with this project. And I'm kind of forced to do it in Eagle. Yeah, you're cuz I have, I have the extended version of dip trace. And dip trace is kind of my native EDA tool. I've just been using it for years. And I like it. And those who've been listening to the podcast for a while, probably know that I don't like eagle that much. But the extended version of dip trace only allows 2000 pins. So I cannot do 40,000 resistors. Because that's 80,000. So
you need the the scythe version of the of the trace, because limited power.
Yeah, I don't want to spend another $400 to build this. For this one resistor that's only going to cost you know, 6070 bucks. Dell said and done. I wonder
how big the XML file for Eagles wouldn't be for this project? Oh my gosh, it's gonna have 440 1000 instances of this resistor.
It's, it's gonna take so long to upload. But yeah, so Eagle doesn't have the limitations. The one thing actually, so I haven't done this yet. But Parker was telling me well, okay, so in dip trace, you can you can connect resistors by literally going from pad to pad. Each trace pad to pad or dip trace will allow you to just draw a trace across a full length of pads now and the thing about dip traces, it won't consider them connected. But you could still export the Gerber's. And so it's kind of Have a cheater cheaters way of doing it. I know they're connected, but the DRC is gonna barf at me. But regardless, it means I only have to draw one trace. And then I can copy it twice and then copy it again. And I and I can basically do 40,000 resistors. In no time flat Eagles nine
letter? No, no ego will let you do that. When really I thought I thought it couldn't. No, you can just draw traces wherever you want draw, okay, but
it won't consider them connected, right? Correct. It will not connect. Here's the thing with this. I don't care if it considers them connected, because I can just look at this and know it's connected. Yep. It's not going to be complex. It's literally just vertical lines on there. So it's going to be pretty easy.
Now, one thing I'd like to see is this thing's actually designed to look like a giant s&t resistor. So the PCB has Yeah, it's the same like, you know, ratio as an Oh, 805. Yeah. And then like, you know, he NYG plate the edges around the corners. Yeah. And then like when told done, mask it off, and like spray paint it black. And then do some like vinyl lettering on it. Right? 10k? Yeah, yeah. Having the plus minus point. 005. Exactly. I think we Awesome. Well, okay, so
here's the funny thing. So we're going through all this rigmarole to get this resistor. So earlier, we talked about we Hey, we searched for this really tight tolerance resistor on Mouser. Yep. That point 00 5%. Resistor on on Mouser is $21. So this resistor is going to be a lot more expensive than that. But it's, it's cooler. That's cool. It's yeah, it's much cool. Some handle more power. And it doesn't need to be laser trimmed. No. So I think actually, one of the biggest takeaways from this is yeah, we're going absolutely ridiculous with 40,000 resistors. But the equation works, you know, for anything down to two resistors, basically. So if you're doing something like feedback gain and an op amp, you can get a tight tolerance by you know, paralleling in series, four or eight resistors. Yep. And actually get something better. And it's still cheap to do that. Yeah.
So that just might be the only thing you have at your bench. Right? There's like some 5% resistors. And you need 1%.
Shoot, you could always a stack them. And so yeah, skyscraper so
yeah. New York, or was it Hanson style?
Oh, yeah. Like the the old radio guys. Yeah. Old dead bugging? Yeah. Yeah. So I mean, yeah. If you need a tighter tolerance, just use that equation. And there you go. So yeah, we'll, we'll come back with an update on that. And next week, and a massive, or a PCB with a ridiculous amount of components. Yeah,
it's as far gonna only be like, I think we're gonna use Oh, four twos? Yes, those things only gonna be like, how big was this gonna be?
I don't remember the final size. But I think when I was starting an in depth trace, the height of it, in the y direction was 10 inches,
so it's gonna be 10 by 10. And then, well, if you want a ratio of the ratio, it's probably gonna be like 10 by Oh, no, not 20. Every 10 by like, 16 is gonna be huge. Yeah, that's pretty cool. Yes, we arc.
So we set up as
like an art installation with a meter attached on the end with like, 10k by 10.0000.
So so we need to mount this on the wall and sell it for $40,000. Exactly. Yeah. So actually, the one thing I'm actually going to adhere to, and I don't have to do this, but I want to, I'm going to use correct spacing in between all the components. For manufacturability. I could go and just cram them together as
I'm stoned. Yeah.
Fixing random tombstones in 40,000 resistors would be horrible. Yeah. So yeah, no, we'll use proper spacing on everything and do it right. And it'll be all up on GitHub. Of course, the one thing that we're gonna have to build this thing, we were the only people who are dumb enough to build this kind of stuff. But but the one thing that is is important, is we want to kind of keep an eye on the resistance of the traces in between resistors because those will have an effect,
especially over a board that big. Right? Right. So
I don't know how we're going to do that yet, but we'll figure it out.
Yeah. Math. I'm afraid we have to use math. Okay, I'm on the phone to the CFO. So last week, the arpha got cut off. So we're going to cover the topics we were covering last week. Hopefully, we come up with the same jokes we did last time. Or better ones.
Yeah, hopefully better. And then there's a couple
extra at the end. So this is gonna be pretty lengthy segment, I think. Yeah. Okay, number one, cherry switches. The company cherry. They were sold. Oh, they got bought out by the ZF brand, which is a German sensor switch company.
Isn't the cherry brand German?
I don't know. I have no idea. Anyways, they got bought out like in 2008. So there's a long time ago. Yeah. But they're now basically changing the name from Cherry switches to be under the ZF brand. So those nice mechanical keyboards will not have cherry switches will have
ZF switches, there's there's still the same switch, just
switch just set the part number might change and the manufacturer logo in the datasheet will be changed. I really
hope they don't change the color scheme. Yeah, that's one thing they should leave alone because it's established.
Yeah, for all the switches. One thing I didn't know there was a cherry actually makes like sensors and stuff. Really. I don't know what sensors I didn't go that far into their catalog. It just makes it says switches and sensors and i i look at all their switches stuff. They're very
clicky. So sensors,
very clicky sensors.
Which which carries which do you prefer?
I like the brown. They're they're not super clicky Yeah, they got a smooth action on
them. Oh, okay. Okay. I think I have blues in my keyboard. Yeah, you
got a little snap action. Yeah, I,
you know, I'm still not used to it cuz I bought it. I don't know, six months ago. And it's still like, so clicky that it's like, wow, this is crazy. But but they feel so nice. Yeah.
mechanical keyboards are awesome. But there'll be ZF keyboards, not Churi keyboards in the future. Yeah. I kind of think it's, I don't know, I don't I don't really have an opinion. It doesn't really matter to me what they're called, because I already know what they're called. But I think I think cherry switches are better than ZF switches.
Well, maybe maybe there'll be the ZF switches, but it's the cherry series
cherry series that that would work well. Okay. Extracting sounds with acid and UV. This is so cool. This is a really cool article that was on Hackaday. And it's about basically recording or getting the original sound files out of old arcade machines. Yes, and, and a lot of these a bit are, they're on ROM, but they're bit protected. So you can't actually read them out. Yeah, they had lock fuses losses on them. And so what these guys did, and there's like, this is for the main project, which is like a emulator for basically every single arcade machine ever,
right. And there was four games that they did not have the bass sound for,
and they had samples for them. But if you know these guys, these guys are all about emulating it down to the hardware level. And so they wanted the original masks basically, right? For these ROMs. And so what this guy did is he acid etched off the top of these ROMs and then masked off with I guess like some kind of optically, not transparent, optically. opaque, opaque. There it is opaque mask and then shot UV at the one bits, the right protect bit, flipped it and then read out the data. That is incredible. Yeah. I'm just amazed that worked.
So when people say hardware hack, a lot of times they're like, Hey, I soldered a wire somewhere. No, this is hardware hardware. Hardware right here. And actually I think the UV protection stuff he used is the same stuff used when when making the die. Oh, so
he he put over the resist did he okay. Oh, okay.
Yeah, makes sense. Cuz I they have an article. So the Hackaday has a link to the article where he actually does this work and he shows it and it looks just like the UV resist stuff. Man it is so cool.
Yeah, it doesn't seem that he was able to do that that accurately cuz I think is like, what 45 nanometers wide or something like that.
Yeah, I mean, the dye is maybe one millimeter by one millimeter. It is tiny. Yeah, but that one bit that he has to flip. Yeah. And of course he's got magnified versions up and he's like, yeah, there's the fuse right
there times 1,000x
Ray I don't know if you saw this but further down the line. They actually had broken gold wire bonds to the dye and they fixed them Oh with conductive paste. They literally just put liquid they glued it back together they Yeah. Ridiculous these guys are insane.
Really cool article. I'll put the link in description. Yeah, check that out. Yeah. Alright, um, this is from last week Danger Close fancy bear tracking of Ukrainian fuel artillery units. This is crazy. This is this is really good. Basically fancy bear is a like, I guess it's a hacker group out of Russia. And you can find they basically used malware to inject into other devices to be able to get data from, you know, sources they're not supposed to get data from. I guess that's the my interpretation of what the underlying cause Yeah, underlying concept, there's probably way more security jargon you can use for it. But this is the map not not a software security podcast. So these guys basically, they have an infected Android APK that's got their malware in it. And someone in the Ukrainian army use this infected APK probably accidentally or didn't know, to make an app to control or not control but to basically give guidance to a mean of their howitzers, their artillery, artillery. So first of all, the fact that they're using an Android app to basically do their calculations to aim howitzers as low interesting as imagine, like going on, like Ukrainian like Google Play Store, and there's like, like, Howard's are aiming at basically. So basically, yeah, it's for their old old artillery doesn't have a built in you have to, like, you know, move dials and stuff. Yeah. Or in the wheels. And so it's for their D 30. Mainly, which is that old Russian artillery. Anyways, so apparently, this app is infected with this malware, which led to these houses being destroyed by the Russian army.
That is crazy. Yeah. So
the app was leaking positional data, probably too fancy bear and the Russian, you know, country. Crazy.
That's absolutely nuts.
So yeah, just the fact that a howitzer is controlled with an Android app that just blows my mind
doesn't 16
You know, I wonder, like, written just like a microcontroller that spat out the data on the like, an LCD screen.
That's what I would done. Yeah. But I mean, the ABS gathering the data for you?
I guess, because it could get GPS and you have to do so little work at that piece of the punch in the coordinates of where you want to the artillery
at well, or turn the cranks and dials and things?
Well, no, no, I'm talking about you just punch in the Android app. Like what you where you want to hit. Yeah, yeah. Hey, yeah, interesting stuff. Um, why the United Nations must move forward with the killer robots ban. And I picked this topic because on the Star Wars special we were talking about the the robots that the trade Army uses and how Josh didn't like him because they're really silly. Right? I thought they're awesome, because they're goofy. Because if I had a killer robot, I would make it crack jokes before it killed you.
But they crack jokes and never kill anyone in the industry. Yeah,
well, the kill log Gungans yeah, there's not people.
They don't have so yeah.
So anyways, the the main thing is that I want to talk about is this quote, if we don't get a ban in place, there will be an arms race. Well, there's already an arms race for military stuff, but anyways, and the end point of this race will look much like a dystopian future painted by Hollywood movies like The Terminator, and quote, so the UN
is now using Hollywood, ladies. Yeah, the Terminator to determine robot bans. Oh, killer,
killer, killer robots. I guess that makes sense. Because that, um,
maybe they know more about Skynet than we do. Or they
are Skynet. The UN is kind of non un. Skynet. That's silly for this podcast.
Well, I mean, we already have the internet of howitzers now.
Yeah, the IO h. That is gonna be the title of this. Um, yeah, um, I guess it's a good thing, because war is bad. Sure, um, but yeah, I guess that's my opinion of this piece.
Well, I mean, okay, so they're, I guess the UN is looking into the future. They see a whole bunch of development happening on drone technology inside. So they said they're like, let's just nip it in the bud right now.
Yeah. Well, that's the thing. Is anyone that's going to develop a terminator is not going to care about what the UN says
Will, Will. But at the same time, there's probably a lot the UN knows of that we do not. And they're privy to information that we don't get to see what like, and they may already have information on killer robots. They're like, Let's ban these right now. Yeah,
because it's because drones right now, they still need a human to click, you know, the death button.
Do they think so? I mean,
I have no idea.
Apparently, the UN does
have it. Yeah, it does. So yeah. Good things wars bad. Don't do it. Um, flip pins. This is an interesting part that you can buy now. This is to add dip chip like pins to your PCBs for use in breadboards. So the like, so like, basically, if you took a dip chip, and then dremeled off the pins and solder those onto your PCB. This basically what this product is.
So you can make your own dip chips. Yeah. And
so I was asking my question, I asked myself, this is like, How is this any better than a normal 100 mil pin header? Because that fits in the breadboard? Right, right. And then I was like, well, maybe it will save dip sockets. When you because I 100 mil header is really wide compared to a chip, dip chip pin. Yeah. And so maybe I'll save wear and tear on that. Maybe, maybe, you know,
actually, so could total offshoot here. But this is just something because I know both Parker and I have made this mistake before, when you're making a hole for a header, go off the diagonal, do not go like make the hole size such that the diagonal of a square pin can go through a circle hole. I know it seems simple. But both Parker and I've made that mistake. But regardless, the whole point is you cannot stick a standard point one inch pitch header into a dip sock. Oh, it
just won't fit. No, it's too big. Oh, okay.
It's too big. So you can use these things to fit in a dip socket. Cool. So So actually, I actually have a board going through macro fab right now, in which I size the holes such that you could either use a DIP package or a header in the case. And the reason why is because long ago, I made remember that that diode compression op amp? Yeah, I want to be able to either put it as is it was a dip eight package, right? I mean, two of them, I actually made a single supply op amp, and I made a dual op amp. And I want to be able to flop this into my circuit or put in a regular dip chip. So in this case, that could be kind of cool.
Yeah, you could put these the the foot pins on your, your, your dual op amp, and you can be able to just plop it right into your audio circuit.
So we actually that. So I'm Mike remake my chip to use these use these gods. Yeah. Because then I could put it just right in a regular dip socket. How do I get them?
It's a it's through the OST chip. We've talked about these guys before. They basically have a it almost looks like a dev board. But basically, it's supposed to emulate or replace older style microcontrollers, I think, okay. And so you just buy it from them. Okay, well, I'm gonna
look into that, because that might work out. Well. I basically I have a guitar pedal that's going through right now. And I want to be able to swap in different op amps, including the op amp that I made discrete on PCB. Yep. So this could work for that. Yeah.
And I was looking, I was also looking at this. I'm like, okay, cool. Maybe they solve the assembly problem, like, because soldering in 100 mil pin headers is a pain in the butt. Yeah. Because, you know, CIP single inline pin. The problem with them is they just flop around on the board. Yeah, so selective solder does not work too well, with those kind of connectors. It's basically the the fountain will come up and just the tension that the fountain will cause them to flop over. Yeah. And so you almost always have to hand solder these. It's always a pain in the butt cost a lot of money. I was like, Oh, I wonder if they solve that problem. And they, they didn't. They basically put them in. They put them in what almost looks like a female header socket, and then you solder that in, and then you pulled sleeve off. So it's a really good idea, but it doesn't fix the sembly problem of making it easy. easier.
Well, okay, so two things on that. First of all, these things are probably not meant for high volume. They're meant for you to hand solder them on stuff. The second thing though, I just thought of a way that this is actually works potentially really well. Let's say you have an old board that has a dip socket on it. But you can't get that dip chip anymore. You can get a surface mount version of that chip. Yeah, you can build a board and then plug it back in there. Or let's say there's like a 40 pin DIP, or you know, ROM chip on an old board, and you have an EEPROM that could work in that place. You could plop it down that guy and so it works for retrofitting.
Yeah, it works really well. That's just I don't know how you would solve this problem of that flopping over, but I'd like to see it solved because that's a lot of our customers have that problem.
Well, it sounds like it lends itself really well to the prototyping world it does.
But please fix some someone out there maybe maybe for pins or someone else that
will actually you know, but here's the thing, it really doesn't fix the prototyping world because most breadboards are spaced such that they can accept a header like I don't know I'm talking
about is is solder and a header down. Yeah, it this will flop around and the holes well, but if you're only soldering one, just do it by hand. Yeah, but I'm talking about like when the Kickstarter needs to build like 1000 Arduino shields. And now at the solder, you know, 4000 Sip pins, or SIP parts?
Well, good luck.
How? Well we do it every day.
That's why we know these things. flopper.
Yes, exactly. Um, maybe some kind of paste in pin where they just we can inject paste in it, but then they have something that support it. So it doesn't flop.
You know, I wonder Okay, so at the point at which the header meets the PCB, if they bent two little tabs out such that it rested on the PCB, then it wouldn't flop as much. And you'd have more surface area to solder down. But I don't think they do that. So.
Oh, maybe we should experiment with that kind of stuff. Make make SIP socket. Or sippin parts easier to machine
when I looked at these they look like they're just stamped. You know? Yeah, they're they look stamps. Yeah. So I mean, some big machine combined just stamps a billion of them at a time. And they check them out. Yeah. Oh, whatever. It's cool. Yeah, it's cool. For one offs.
Yes. And I think everything, you know, you said and is good applications for it. I just wish someone out there would I was hoping this would be the solution to that, you know, assembly problem.
Well, yeah. And great example, like an old arcade machine, you want to throw a different processor or a different ROM, or something like that. And you could do that with one of these. You can.
Alright, next one is, please want Alexa data, people begin to realize it's listening.
The Alexa is that network device that
it's the Android or not Android, Amazon cylinder that can like control everything in anyone.
Home Automation, sort of kind
of Yeah. And you can ask it like questions. It's like, it's like Siri on your iPhone or Google Now on your on your Android.
Yeah, my buddy has one in his apartment. And it's it's pretty cool. Yeah, I mean, all he does is he just asked to play music.
Yeah. But the cool thing is, is its API is open. Yeah, are open enough to where you can make it pay to pretty much do anything, right. But what happened was, there was a murder investigation that basically led them to ask Amazon for all the recordings that the Alexa device in that was in probably the murder role. Hang
on. Why is Alexa recording because it listens all the time? Listening is different than recording exactly, but
it listens all the time. But this Amazon didn't give out any recordings. So good. No one knows if it actually records or it's just listening all the time. That's what this whole articles about. Because always on voice is pretty sweet. Like you can just say, OK, Google, and there's my phone
that was that actually worked out.
really gonna work for phones a couple feet away from me. So yeah, it's pretty awesome. But us saying all this stuff besides is actually being recorded on the computer right over there. It should never be recorded.
No.
It should never be logged.
Well, okay. So unless it's working According to figure out your nuances, and figure out your voice so we can better listen to you. Ah, that might be where the recordings come in, maybe. But it probably wouldn't need continuous recording.
Yeah. Because a lot of a lot of these devices, what it does is when you give it your voice, like you say, okay, Google, and it went off again, because that's actually a sound processor in the phone, looking for your voice, right? And because you actually, when you set up your phone, you say, okay, Google, like four or five times, and then it records that and it has a profile. But then when you actually say what you want, like, let's go to the movies and get some hot wings, it will actually send that up to the Google server, process it and then send that stuff back. Okay, so it's only looking for the first phrase, yes, as long as the first phrase, and I betcha Alexa does the same thing. Because it doesn't have enough horsepower to, you know, do all that stuff.
Yeah, well, it's. So this is,
yes, logged in some way as you have to send your voice out to the the, quote cloud, unquote. And then it comes back with what you what it says you
did. The clouds recording you
possibly we don't know.
Well, I don't need one of these things.
Yeah, interesting stuff.
I mean, shoot my my cell phone was recording everything.
I did definitely record your location wherever you go, though.
Oh, yeah, for sure.
The Voice? Well, you're interesting.
So it all depends on how much intrusion you are willing to accept?
Well, and it's is Amazon or like Google? Is it actually recording when it sends your voice up and stores it on a hard drive somewhere at the NSA? Maybe? Maybe not? Well,
this is just like that Apple thing where they were asking for? I don't I don't remember the guy's name. This is a while ago, but the the access code to that one guy's phone. You remember what I'm going to hear?
Yeah. And basically, yeah, yeah. And basically just like, hardware, hacked the phone and just read the data off that way. Right. Right. Yeah. And like, Apple asked, I think was the was the FBI. I don't remember. An apple asked me. How did you do that? Why are we laughing at that? Yeah, that's a good thing.
So you know, if they're listening in, I give away thumbs down on this.
Well, if they're listening and recording, there's no way they're recording everything. It's probably waiting for the key word just like Google is
Josh's rolling his eyes over here. He Josh's thinking that maybe they are
I think there's too much data
now. That is cheap.
Yeah. A big server underneath a mountain somewhere.
That's right. That mountains called Skynet.
There, yo, hopefully they got get long. Go on camera, or mic? No, they probably heard you, Josh. Alright, last topic. It was not really an RFO. But we got this like new piece of software, I guess to like, because we have two conference rooms that Makerfaire a big conference room and a small conference room. They're really close in size, actually. Yeah, they're pretty close. It's just one gets used a lot more than the other one. Yeah. And so we got a piece of software. I can't remember what it was called. You room or something? Yeah, room, y'all really? Yeah. It's not something like that. You know? It's basically we're trying to figure out names for the conference rooms now because we have to name them in this, like, scheduling software, right.
So you can go in and see when it's booked. And you can make them yes, we've been having
this not about whatever the pieces off was called. But it's like, what are we going to call these conference rooms?
Yeah, so our chat blew up today. Yeah. All kinds of names for our two conference rooms. Yeah.
Like I said, you know, like, I like the big room because we use the most like the rancour pit because it's like, from Star Wars or Star Wars, because that's like, where everyone goes out each other's throats.
Yeah, we have a meeting every week. That's it's fun. Let's put it that way. And I say fun and quotes.
I like that meeting.
That's yeah, that's the one where we give status updates. And we talk about everything that's going on. So it's intense sometimes and it's fun. Oh, yeah. But yeah, Marine Corps pit. I think that's that's that's great. I know Chris the CEO. He's thinking something electrical based
layout electrical. I was thinking hurts. Yeah, because you get feelings hurt there.
I said, showed injure and Heisenberg. Yeah, that didn't catch on. Now. I think it's too long. And they're hard to spell.
Yes. I don't know how to spell
So any any suggestions?
Yeah, tweet them at us at macro fab or at macro ninja near that's macro Longhorn engineer.
Yeah, let's, let's see what what our listeners can come up with. Yeah,
I want I want to good names because these are gonna be printed out, you know in nice embossed like
plaques plaques
for the doors and stuff. So it'd be
great if the two conference rooms are like, in parallel to each other in some way or the names reference each other. Yeah, you know, Yin and Yang kind of thing. So, yeah, cool, but the battle of the two conference rooms Yep. So that was the macro fab engineering podcast episode number 48. We were your guests. Steven Craig.
guests.
We were your guests. Yeah, Josh is actually the host
we are your hosts Parker Dolan and Steven Greg. Take it easy.
One faithful evening, Parker received a marketing email about gamifying the MacroFab platform and thought it might be a good podcast topic.
Chris Church returns to the podcast to discuss the recent outbreak of COVID-19 or Coronavirus in China and how it is impacting Electronics MFGing.
Lets bring in the new 2020 year with a MacroFab platform update for PCB Specifications and Hellhound preamps! Thanks to all our listeners and guests!