MEP EP#192: The Fourth Rule of Robotics

Podcast Transcript

Host 2

Welcome to the macro fab engineering podcast. We're your hosts perkiomen and Steven Gregg. This is Episode 192, which is eight away from Episode 200. Whoo. And so like Episode 100, which was a question and answer, QA episode, we're gonna be doing the same thing I think, right?

Host 3

Yeah, that's the plan so far. At least. That's what we've come up with right now.

Host 2

Yeah,

Host 3

It was fun last year. We enjoyed it. So I think we're kind of talking about doing some twists on it this time around.

Host 2

So what are those twists?

Host 3

So we would actually like you guys, the listeners to record your audio clips of questions and send those into us. And then we're going to splice them all together and make a question. So your audio will actually be a part of ours.

Host 2

Now, are we going to like pre screen these? We're just going to be like, soundboard it.

Host 3

We haven't talked too much about the semantics on it. It's just we had this idea. And I was like, Okay, that sounds like fun. Let's go ahead and do that. So I think we probably should pre screen them. Or maybe, maybe we will ask Iris to pre screen them. And just, you know, make sure there's no like naughty stuff. Yeah. So if you would like to ask us a question and have us answer it on the 200th episode of the mecca of engineering podcast, go ahead and record your question, and send it into podcast at macro fab.com. And then we will collect all of those and we will, you know, hopefully, there will be plenty enough to fill a hole that was

Host 2

That and we can answer them.

Host 3

Yeah, right. Right. Well, I mean, it No, no, no, no, no, but I mean, we will we will answer anything, right? Like that's, that is our promise right. Now, we will answer anything, you might not get the answer you want. So yeah, once again, record your questions, and send them into podcast at macro. fab.com. Cool. So what's been going on Parker? So

Host 2

I finished up that Tiger Shark repair the tiger shark pool cleaner pool cleaner. Unfortunately, not a tiger shark with lasers on its head. But it was yeah, the pool cleaner. So I put it all back together. I just like we talked about like silicone in the gasket because the old gasket just wouldn't fit anymore.

Host 3

Right the non serviceable part that you serviced.

Host 2

Yeah, service that was Yeah, I was able to actually like repair the circuit because last time it was I had just ordered the parts. So I was able to de solder the motor controller, which was a pain in the like, I've never had this problem like the soldering apart before. It had a big thermal pad underneath. It's like a T sock like 48 or something like that.

Host 3

Yeah, it was super long and had a ton of legs on it. Yeah.

Host 2

And like I could get all the legs liquefied, but I could not get the center pad to like go even with a board heater. And I like I you know, crank the hot iron to like 400 Celsius trying to get the thing to go. Yeah, just would not come off. And so basically I ended up just cutting all the legs off. And then since the part was broken, it kind of like crumbled into three pieces. And then I kind of chipped away at it until I got to the thermal pad directly. Like through the the material what would you call that? It's not epoxy, but the what they make the body of the ice out of oh, it's plastic. Yeah, that whatever the plastic is, I basically took that away, got to the thermal pad and then was able to get it off then

Host 3

God that's a pain in the ass. That's that's all it takes a lot of work to do. Yeah.

Host 2

And then when I put the new one back on, it went on perfectly. Like it reflowed great. Like, I have no idea why this old part just

Host 3

Did you did you reflow the thermal pad the solder that was on the thermal pin. Yeah, okay, like flux and re and flux. Yeah,

Host 2

I used some old school rosin flux, and I was using the dreaded lead solder. Oh, that's good. Yeah, so that's the I think that's why because I basically took the lead free off and put led it on. Yeah, it'll

Host 3

Melt at a much lower temperature. Yeah, that's always that's always the worst. I hate it. In fact, I was working with a I've got a Quad DAC on a board that I was playing with the other day and I had to replace one for some reason. And it's, it was a QFN that has eight pins on the side or four on two sides and then a big monster thermal pad like all of the the actual like connection pins are tiny in comparison to the the thermal pad and it was so hard to get it to align because each one of the legs As on this QFN we're not big enough to yank it into true. So like you're sitting there with a hot air gun you're like, oh, this

Host 2

Was controlling the, the surface tension. Exactly,

Host 3

Exactly. And if it has too much solder on it, then are you sure the legs are connecting and it's just like, Ah, it's such a pain in the ass. How to use solder components with a thermal pad like if you're doing it by hand.

Host 2

I will usually just put a I'll put just the tiniest bit of solder on and this is what led solder because hand solder I pretty much only just use LED at home, and that's why they don't let me solder at McAfee anymore. That's fine. I'll do an amazing job at it. Compared. I

Host 3

Mean, you gotta have some pretty good people there. Yeah.

Host 2

So yeah, put just a little bit a tiny bit, because you don't want to make it proud. No, because it won't sit flush. Because you basically need to make just a little bit of layer there. Wet the pad basically just barely out and then put a ton of flux down on top that pad put your part on and then heat that up until you get a little bit of that little wiggle move. You can basically no it's it's starting to float a bit on that pad. Right and then you know you got a good contact there and then you solder the legs.

Host 3

Yeah, give it a good like I like using the the sharp nose tweezers that are angled.

Host 2

Oh yeah the little like the like

Host 3

The 45 degree one I come snake fangs. Yes, big thanks. Yeah, and then and then you just tap the top chef sinks down. But But honestly, I'd never do Q offends or anything outside of a microscope. I always work on their microscope when soldering those things because it's just like that's the only way that I can feel comfortable doing it you know, other than like fully testing everything and I don't really want to do Yeah, so I

Host 2

Did have some interesting I guess problems I guess because that's just one problem I had with reloading or fixing okay. Because it was inside this sealed box that had water in it everything had moisture in it I didn't think about at the time and I just a moment I hate with hot air it like one section of the board which was not this part just like immediately the laminated and exploded

Host 3

Just blistered like crazy

Host 2

Oh yeah blizzard like a whole like quarter inch and diet like thickness. Oh wow. Okay, yeah. And unfortunately it was there was no copper or anything over there so I was able to kind of like squeeze it with pliers and get it back to

Host 3

Kind of normal so we did you try to like dehydrated or something? No, I

Host 2

Just went for it.

Host 3

You could have just put it in like an oven at like 200 degrees F

Host 2

Night well okay yeah Hey, it works and so I got it all back together tested it silicone the snot out of the the gasket put it all back together let dry for a couple of days

Host 3

Made it really non serviceable.

Host 2

Yes because now you can definitely not get back apart

Host 3

It's just goofy as hell right?

Host 2

Oh yeah. Cuz well because I I filled the gasket seam which is like a quarter inch by quarter inch full of silicone Oh, and then the bottom ledge. I filled that up and then the squeeze it and just went it just pooped everywhere as I was putting the screws in and then it kind of just wiped it up and just so that's that's probably good. Nice and what down the Galison and plugged it in toss it into the pool and no one died.

Host 3

Just toss it in and then you jump in right away.

Host 2

I think that's probably why it uses like 24 volt DC instead of running AC current down that line.

Host 3

Oh sure. Yeah, that makes sense. It's

Host 2

Like the unit itself has a built in GF CI so pretty certain Nothing bad's gonna happen.

Host 3

Yeah, yeah, you would think that they would make it somewhat even

Host 2

Stock it or funk factory it will meet eventually so and then those things are not cheap. I looked up the price on them there that unit is like 1200 bucks

Host 3

Yeah that's that's price him

Host 2

And so the basically what's I betcha this is going to last about a year until like like that PCB fails again because of like how crappy I did like taking that part off right?

Host 3

How much you blistered the board? Yeah, how

Host 2

Much I've listed that board and burnt it and rosin stained it

Host 3

Why did you not clean it afterwards?

Host 2

Oh, I clean it with a little bit alcohol and I was like ants by fine now. I didn't I didn't have any conformal coating the spray back on it so I just kind of like went on the garage Republican acrylic spray paint and just spray it what I had

Host 3

On it, it will Okay, so one year might be asking a little much.

Host 2

But hey, at least I tried to coat the board.

Host 3

Yeah. For sure.

Host 2

And then basically, if we get another year out good, then I'm just gonna, you can just buy the non serviceable motor unit for like 300 bucks. So it's not like we have to buy a whole new unit. But we'll see how long it lasts. Nice. They might be episode 300. And then we'll be saying, oh, yeah, that shark. Tiger Shark broke again. So it's working it runs. I think my mother uses it every day now. So so it was just that one part, right? Yeah, that one. Well, oh, and I replaced that capacitor. Oh, that's right. Yeah, it looked a little melty. For some reason. It was weird. Yeah. Yeah. Did you use a long way I can think of is like the motor like either capacitor failed or the motor controller failed, which caused the capacitor to get hot. You know, is one,

Host 3

One or the other? Right? Yeah. One or the other? The? Did you scholastic that that capacitor dead into the board?

Host 2

Oh, no.

Host 3

Well, you gotta pull it apart.

Host 2

It's like, it's actually like a whole like, oh, about a 10th of an inch off the board.

Host 3

Dangling their float, you know, flapping in the wind, right?

Host 2

A little bit.

Host 3

Yeah, well, I mean, but but this thing. This thing does, like wall climbs and stuff. Right? So that capacitor is gonna get all kinds of different flexing

Host 2

Very slowly, though. It's not like this thing is a speed demon underwater.

Host 3

Yeah, it's just really slow. tranq tank tread, right? Yeah.

Host 2

It's amazing, though. Because like how much noise this unit makes like when it's on land. But the moment you put in water, it's silence. So I guess something with the bearings, how they're just because they're like nylon bearings. I guess even just what the tolerances they kind of need some liquid in there to kind of take up the slot to

Host 3

Convince them to work a little better.

Host 2

Yeah, because they're pretty, like the whole unit is pretty noisy. Yeah.

Host 3

I bet you if you stick your head underwater, it's probably noisy too.

Host 2

Probably. I haven't I haven't actually tried that yet. Um, no, I don't like

Host 3

Water transmits sound excellently. It's quite, it's quite good. But I think the transition from water to air is pretty damn bad. Right?

Host 2

Probably, I would imagine. So yeah. Even like white doesn't do that that well, either. I guess that's a wave and a particle. Moving on,

Host 3

Actually, so something to say about the back on that capacitor. Something I'm super excited about is I think we're starting to reach a little bit of that critical mass with the with the podcast where when we say something like, Hey, we should get an application engineer on to talk about electrolytic capacitors. We actually got an email this week that maybe that might happen in the future.

Host 2

Yeah. From Kemet. Yeah,

Host 3

We already had some Kemet capacitor stuff. A while ago, Louis. Yeah. We're trying to get someone on to talk about electrolytic capacitors. So that will be awesome. I'm really looking forward to that.

Host 2

Yep. I cannot wait. Hopefully, I'm thinking probably within the next two months. We'll have them on.

Host 3

Yeah, I think it'll be really great.

Host 2

So I also worked on the pennant tar Rev. Two. This just has slightly out tweaks from Rev. One. I fixed the relay problem, which was basically set it up as normally closed. It's been great. Yeah, I reversed the logic on it. So I fixed that. And then some of the silkscreen labelings was incorrect. So I fixed that. And then when it had an order it and I ordered them as like, P surface mount only boards, because I can solder the through hole.

Host 3

Right the last time around you guys were we're doing quick turn. So you ordered OSHPark, right.

Host 2

Yeah. This is the funny thing is though, so we ordered because there are four layer boards. We ordered them for the OSHPark. And they took took seven business days to get them which is pretty quick, that but then it took like three or four days of just soldering them onto the boards. And I'm like, Man crap says we'll do in 13 days. I'm just gonna do that this time. It's like the same price. Right? I'm like, and I'll have to actually do anything. That's like the only time I've like pitch my own company on this podcast.

Host 3

You know, that is something that that's kind of funny. The the sort of illusion of speed, especially when it comes to like fast turn stuff, like, you get this idea or it's like, Oh, I could shave off a day or two off my project. A day or two is nothing like yeah, you can wait a day or two, you know

Host 2

If it's, if it's a personal project, like, for me at least I'll just do something else for those couple days while stuff comes in. Yeah, I've got like eight bazillion projects that need to get done.

Host 3

Yeah, for sure. So yeah, I don't know, like, there's there's been a couple of times when I've been at, in a design team where they're like, Oh, we've got to do a quick turn on this thing. And then like, use end up spending the extra $3,000 for a quick turn. And then it sits on someone's desk for four days. And

Host 2

Like, you know, that one heart shows up, or that person's out of town or, or whatever

Host 3

Reason, or it got to the point where like, as soon as the boards arrived, you're like, oh, yeah, well, the customer was like, Oh, we don't really need it that fast. And it's just like, come on. Yeah, I mean, I can mission critical, I get it, but 99% of the time, like, you didn't need to go that fast. Yeah.

Host 2

And then the next kind of project I'm working on today, actually, after the podcast is. So at work. The sales team has got a gong. So when they close the sale, they get the ring the gong, or hit the gong that's on you. You hit the gong. Yeah, they hit the gong. And I was fine with that. Because it's like, on the other side of the building, right? And then and it only happens, you know, couple times a day, whatever. Well, developments got a bell.

Host 3

I'm assuming it whenever something good happens, they do something. Yeah. So

Host 2

Whenever they have a software release, which is a couple times a day, they get the ring the bell. So I'm like, Oh my God. Like, man, this is just this too much of this, like hoity toity like self pat on the back kind of thing. Which is,

Host 1

I'm not a team player. No,

Host 2

I'm fine with it. But I'm like, do we really need to have like that, like, Hey, look at me and have everyone at the company. Look at me at this cool thing I did. I mean, least that's my opinion. So I went to the extreme for the engineering team, I'm doing air raid siren,

Host 1

They're gonna hate you even more.

Host 2

Whenever we get to do something awesome, we get the get the turn on the air raids, like 140 decibel air rates. So I found a first version I want to do is I just found something on Thingiverse that I can just download and print. And it's a hand crank one, I'm like perfect. Don't even have to do electronics, just print it and put it together. Because originally, I wanted to be like hooked up to like a Raspberry Pi. So you can just send it a Slack message of like air raid on and then like it spins up. But I'm like, Okay, let's start with the first one. Because maybe I don't need to get to that point where, like, you know, I might be fired in like two weeks because

Host 3

We wait, we wait, okay, I'm gonna throw a monkey wrench in here to make this to make this potentially better. So air raid siren is a really simplistic mechanism that shield like a spinning its spinning portion that just basically produces a pitch based off of like wind blowing through the siren, right? Yes, is there? or would there be a way to control the pitch of that, such that you could play like it would spin up and you could play like an auto music through it, and make it a meatloaf siren, where it would spin up and then play bad at it.

Host 2

So I don't I

Host 3

Mean, like, if you could make a controllable read inside of it are so yeah, you'd

Host 2

Have to be able to control the basically well. So how they work is by they have fins inside of this drum and the drum has a different protocol. Basically the how the designs work is they have a different number of of a openings in the drum portion. First, the inner portion has a different number of drums, right? Different number openings, I mean, and that ratio is basic, and then how fast you spin it is the frequency.

Host 3

It's sort of it's sort of like, you know, like a like a like a whistle like a coach whistle has one of those veins and you blowing through it produces produces a standing wave within the thing and then the little ball that like jiggles in there, like disturbs it so much ringing Yeah, you modulates it and rings it well, this air raid siren is basically the same thing that's multiplied by however many vanes are in the outer drum, right?

Host 2

Yeah. And so you would have to be able to change the number of openings on the fly. So I know you could probably make that work on the outside because it's stationary, but the inside one might be really hard to do that.

Host 3

Well, let's see. What I was thinking is is instead of like doing it the traditional way, use the spinning internal thing to just act as like a turbine that just sucks air and blows it through. And then you control the vein openings, such that you could control the pitch then, in other words, you're making like a really unique, weird wind woodwind instrument, you know?

Host 2

Yeah. Well, that's all I'm saying is you could control the outer drums openings is the only real way I think to make that work.

Host 3

Yeah. But to but to do it at audio rate would be very difficult. Yeah. Yeah.

Host 2

Well, you'd have to, you'd almost have to probably use pneumatics to quickly open stuff.

Host 3

Yeah, that might not even be fast enough, though. Yeah. I could type in let the low turnout in, in Slack. And it would just

Host 2

Be you could just I mean, you can do it how horn like air horns do it and you just have different. Like, all you have, like 20, air raid sirens, all different frequencies. The trick is spinning them up fast enough. Yeah. So you'd have to predict they take a while to spin up or you just have a big motor that can instantly turn on and then a break to turn it off.

Host 3

But like, I'm thinking like, a, like a soft stop break. Yeah, it stops. It's just like hammers a rod into addition to the vein. And then explodes. Yeah,

Host 2

That only works like three or four times. Yeah.

Host 3

How fast is an air raid siren typically. What's like the speed that it spins at?

Host 2

I think like, depending on sometimes it's like 60 Hertz. That's why it's, you know, AC motors, most AC motor spin.

Host 3

Right, right. So 60 hertz, but the, but the number of veins is like a multiplier, right?

Host 2

Yeah, something like that. But it doesn't really matter on it doesn't really matter on the speed you spin it. That's mostly the volume of it. The veins really dictate the frequency.

Host 3

Yeah, but I think like you hear all those air raid sirens, they do modulate in pitch. And I think that's due to the speed right? Think so.

Host 2

I don't know everything about the physics of air raid sirens. Let me see here. Here's an article from 1942 about air raid sirens.

Host 3

Okay, yeah. 45 rpm. With 10. Yeah. Okay, so it's 45. And you just multiply it by the slots. So okay, a 10 slot on a 45 RPM gives you 450 hertz. So you could modulate the speed a bit? Yeah. But you'd have to have a motor, a motor that responds to audio speed rate changes, like that's some serious inertia moving right there

Host 2

When people do stepper driver music. Yeah, but that's

Host 3

Just the music of turning the shaft not turning all the extra mass that you'd have to you'd have to have a beefy motor to do that. Or a beefy stepper.

Host 2

Or a little tiny Airaid.

Host 3

You right, like this little tiny whistle thing? Yeah. Yeah. I think you would have to be screaming fast for that to work.

Host 2

Well, let's start with, let's start with the hand crank one and see if I still have a job in two weeks.

Host 3

Like I said, they're gonna hate you. Yeah.

Host 2

Okay, so what are you been working on?

Host 3

So actually, this week, I've been putting a bit of thought into that old, old project we picked up a few weeks ago of the so you want to pick a part? We just we've been talking about it a bit on the Slack channel. And I forget, I apologize. I forget his name. Who's in charge of getting the websites and everything. Chris Mullins, Chris Mullins. That's it? Yeah, Chris. So Derek, from the Slack channel actually came up with the idea of IE Park picker.com. Which is pretty awesome. And we've like, within like seconds went and got that with Chris Bowens, like it was like, Oh, that's

Host 2

A good idea. I'm buying it. Yeah, it was

Host 3

Just like, Derek was like, What about EPR picker and like not even like a minute later, we already had it. So we're thinking right now, II part. picker.com is the name of the website. And just as sort of a recap, we're kind of using this as like a Wikipedia page for engineers to be able to identify and pick parts and, like, have a good, deep understanding of the nuances of each different type of parts. So I've been thinking about like, what should come first, like, what's our next step? We sort of have a website, we got people on board with this, but it's like, okay, what do we do next? And Parker and I were actually discussing it. We were talking. We probably need to create like the very first one as a demonstration to everyone saying like, okay, So this is what our vision looks like, this is what we think a page should look like. And I would think that it would evolve, you know, with with time and stuff, but like, sort of, I think Parker, you mentioned it, you called it like the grandfather page. And I like yeah, like, let's make the grandfather page. So,

Host 2

I gold standard,

Host 3

I think we should do resistors as the first one. It's really simple. There's not a lot of like snakes in the grass or weird things hiding in there. And people can fact check us pretty quickly on resistors. So we're not going to do anything crazy. And we can sort of build it as a template. So if we have the resistor page, the resistor wiki page, we can build it up, figure out what it looks like, kind of get an idea, a hack and chop some things, maybe we'll even ask some people to do some edits and reviews on it, and just make sure that it looks good. And then we can start pushing that towards other things. Because like, it's funny, you know, like resistor is a pretty easy thing. I would think that capacitor would be a pretty damn difficult one, you know, because there's a lot going on there. In fact, it's kind of funny, I've, we're getting a new testing person at work. And the person who was doing testing is now actually going to go off to repairs. And I was asked to get that person up to speed on electronics, because that person has done a lot of testing in the in testing as in like turning trim pots and watching an oscilloscope and things like that, but this person doesn't have a huge background in electronics. So we've been learning from the learning the basics, and she's like, Hey, let's, you know, we were I've kind of had a good feel of resistors. How about going to capacitors? I was like, not yet. No, you're like, No, this. Capacitors are weird. Like, just, let's, let's get resistors down a little bit more,

Host 2

You know, wait till she asks about op amps.

Host 3

We've been we're actually doing op amps before capacitors. Because Interesting. Okay, there's capacitors, when I started off with like, basic electronics, like, here's current, here's voltage, blah, blah, blah, then we went into resistors. And now we're doing op amps. And then eventually, I'm going to be like, now everything gets really weird. As soon as you put capacitors in there. And again, as soon as as soon as someone goes like, what does the capacitor do? I was like, you really can't ask that question. Like, because like, there's like, you have to ask the question, like, what is the capacitor do in this situation? Like it always works out that way? But it's like an inductor to exactly I don't even want to say that word to her yet. She doesn't she doesn't even know those exist.

Host 2

Think about it as a long extension cord.

Host 3

Oh, gosh, yeah. No. Too many analogies with electronics? Yes. But yeah, so So with EPR picker, we're talking about throwing together a resistor page. And I don't want it to be onerous. I wanted to just be like, really straightforward. So we'll see how that comes out. You know, as as, as we go through just kind of like throw things throw ideas down. And really, I want to use it as, like I said, kind of a template for the rest of the grandfather page. So keep your eyes out for that. I'm sure we'll, we'll kind of like past that by all the slack people first, because we've got a lot of people who are way more intelligent than both Parker and I in the Slack channel. And they will be able to be like, Man, this, this article is full of bullshit. Here's how you make it better. Cool. And then, and then I've been also messing around more with the fermentation controller project. But mainly, that's just getting more parts in,

Host 2

We were talking about the, you might have found out what's up with the screen,

Host 3

Maybe fingers crossed, maybe the screen, I was driving the screen where the logic side of the screen with 3.3 volts. And technically the screens looking for five volt logic. So there's a level shifting transition in there, I did get some level shifting chips and put all those in. So I'm doing true five volt. And I'm actually getting exactly the same results. So it wasn't although yesterday, a second screen arrived, and this is the screen that's meant for Parker's unit. So I'm going to plug in the second brand new screen that has not been touched at all and see if I get the same results. Because maybe I've just been barking up a tree this entire time. And I've just had a bad screen. Yeah, because we

Host 2

Actually did find in the not the screens. datasheet the screen datasheet said point five Vcc was a high. Yeah, right. But then we went into the controller and the controls like yeah, on the data lines, but on all the control lines. It's like, I can't remember what was it? I gotta look it was

Host 3

Higher than normal three, three, voltage. It was a seven five or something like that

Host 2

Point seven five Vcc was a high

Host 3

Right. So this so three, three wouldn't work in that case? Correct. But I've done everything. I have rebuilt this circuit like 10 times. But yeah, so that's not working either. And like I said, I'm getting exactly the same result. So it still keeps pointing back to something in my code is, is wrong. And I know it's gonna be something really simple and really straightforward. And it kind of it's annoying. This is the this is one of the, this is one of the shitty things about using someone else's embedded code on a different processor. That's where it gets kind of shitty, because I don't know any of their initialization stuff. I don't know what clock speed they're running it, I don't know, the light. In fact, I don't even know what their pull up structure is, or anything like that. I'm running all my own stuff. And I'm running all of what is recommended, but you know, they have delays and things like that in their code. But if I don't know what their clock speed is,

Host 2

I have a right there. What they write their library for clockspeed.

Host 3

Exactly. I have no clue. That's why like embedded applications, or embedded example code is great for Arduino, because it's always the damn same. But for anything else, it doesn't, it doesn't work, right. So I think I'm shooting myself in the foot too many times with this, I should just write all this code from scratch. But I'm like, I don't want to do that. And I don't know. So it'll, I'll figure it out eventually. But I kind of showed this section for for a moment to go back and start looking back at the the enclosure. And I bought a second enclosure for Parker's version or Parker's unit at a second screen. And I ended up going with solid state relays. So I'm going to purchase those soon. But those are all like Amazon things, you know, the 40, amp, solid state relays that you can actually drive directly from 3.3, which is awesome. So I'm gonna I'm gonna have all of those go. And so eventually, we're coming up here soon, I'm gonna get the schematic laid out, minus the whole screen thing. Because like, I got to figure out do I actually need all these, you know, what level transitions am I going to need and all that all that crap. So

Host 2

I'm just more what other Mauser order you need to order?

Host 3

No, there's probably more of that. Yeah, for sure. So I have been playing more with this. DLA I got these fun this thing is this thing is really nice, actually, the La 2016.

Host 1

I mean, for for, for like a home gamer like me, it does everything I need it to. I can't foresee a project where I would need anything bigger than this exactly better than this. And I specifically went for the 16 channel one, because I'm using 13 of the 16 right now. And it just makes Oh, it has all the decoding built in. So I can say like a at this time, I'm reading this on a parallel add line. And it just makes it super easy. I like it.

Host 3

Cool, under the RFO onto the RF Oh, rapid fire opinions. So this first one, I found actually earlier today, and it was like, I wish I would have known this a while ago. But it's super cool. Found on Texas Instruments. website, there is an analog engineers calculator, which is sort of based off of an app note that they

Host 2

Made a time ago. This is This replaces you right? Yeah, entirely. Yeah.

Host 3

A couple of buttons. Why even paid me there was an app note that was released got from awhile ago, which was like the analog engineers handbook or something. I think,

Host 2

Parker, we talked about it before. And yeah, when

Host 3

We were looking through it back when I was at the Fed. So this is like just a standalone GUI app that that was created based off of a lot of that information. But it has a ton of really good stuff in it. So it does your basic calculations for amplifiers and things like that. But it also includes a bunch of stuff for filter calculations for calculating your parasitics around an amplifier also doing your Add Noise calculations and things like that. And it will back calculate ratios of resistors and things such that you can say like, Hey, I want to hit I want to have an amplifier with this target gain. You know, what, what is a good ratio for that?

Host 2

But it gives you a ratio, which I like, because sometimes you're like, hey, I want to use like a 10k. What's the other resistor I need to use to make that ratio work?

Host 3

Right, right. But but but if you're like, oh, okay, well, I want to bump that up to 100k. The ratio stays the same. You can Yeah, you can still play around with it. Yeah. Hence, heads ratio, right. In fact, it's funny. I actually had a customer who sent in some stuff this week for a design review. And he used 1k resistors like all over the place on all of his op amps. And I think it was it was a general mistake because I've done design reviews with this guy before, and he hasn't done that before. But I was like, Hey, man, you know, one case is going to is really pretty damn low for an op amp. It's like, the edge of an op amp even working properly. You know, and it's stuff like, you know, I'm sorry, I'm going on like eight tangents here. But this is the kind of crap that they don't teach you in college. Like, in college, you're like, Oh, well, 1k would work, right? If you just plug it in. It's like

Host 2

They treat and they treat in school. They treat everything like a sphere of spherical cow. Right? That's, that's the physics examples. The spherical cow. Right? Right. Where like how gravity affects it? Yep. Yep. So Magna magnet is I can't remember. What was that example? It doesn't matter anymore.

Host 3

Yeah. Assume a cow, is how every product

Host 2

Is round. Yeah, perfect. It's perfectly round and weightless and frictionless and friction. That's

Host 3

Right. But it's also a blackbody and emits radiation. And yeah. So yeah, regardless this, check out this analog engineers calculator. I think that's actually not think that's the name of an analog engineers calculator, we can put a link up to it. It's actually pretty big. It's like 270. Meg's download, but if you do any kind of analog design, it actually makes a lot of that stuff easier.

Host 2

It definitely looks like a engineer designed it. Oh, yeah. No, it's like, here's a list of lists or calculators, fields, the plugin diagram, right, and then a button that says calculate, and it's all windows 98. style. Yeah,

Host 3

I mean, it's awesome.

Host 2

You know, that's actually been getting me is recently windows 10. Get rid of the old school. That's been there forever. Sound Manager. Oh, I hate that. Yeah, now they got now you have to go through the, like the windows 10 app version. It used to be you can go you could still get to the old school one, which is how I always tested if my microphone was working or not. Yeah. I think got rid of that.

Host 3

Well, also, Windows has killed off Ms. Paint. It is no longer supported.

Host 2

Really? Yeah. So I still have it. It's paint. It's just called paint.

Host 3

Right? But in there is I can't remember what version it is. There's an update to Windows 10. Where like, henceforth no more MSPaint

Host 2

Ah, yeah. Pour one out for our homies. Yeah, over MSPaint

Host 3

Helped me get through college. There's there was many diagrams for for lab reports that I created.

Host 2

So the our next one is, or next RFO is inspect AR or inspector or inspect. inspect our that's a this is an

Host 3

Ar.com Yes, this

Host 2

Is the work bench of the future. This is super cool. And this is yeah, this is really cool. I can't remember how I came across it. I think someone posted it in the Slack channel. And it's basically augmented reality for PCB assembly for doing like one off prototypes and stuff. And so I was looking at it. I'm like, I want to use this for like, doing like quality control.

Host 1

Yeah, it's like basically, you put you put your board underneath a camera, and you can overlay like Gerber layers or like you can, it can identify components on your board based off of like copper, things like that. So you can you can say like, Hey, show me C 15. And it'll pop up like a box around C 15. Or you can say show me on my board where all the 10k resistors are and it'll draw boxes around all the places where 10k should be. That's super neat.

Host 2

Like I don't, I don't think they have like a example where you can like rotate the board and I tilt it and it like it's augmented reality so it keeps the perspective of the Gerber file over which I don't really find useful. It sure looks cool though. Oh, I look Yeah, looks awesome. And looks like like one part it's like you can select nets as well which is kind of cool. So I really want to I need to send this guy an email see if I can try it out. Maybe we can get him on the podcast

Host 3

That would be awesome. You know like selecting the net thing is super cool to say like hey, where's all my power or Where's all my ground and it would like automatically show all of it that's super cool. This seems like it could be really helpful sort of in a macro fab way because I remember we discussed things of this caliber many times at macro fab for the QA team for them to be able to just like look on look through glasses or whatnot anyone say like this part should be this you know or something. Yeah,

Host 2

I'm imagining like, like everyone is just like in a pod. kind of like ALA. What's that movie? Minority Wally? No Wally?

Host 3

Oh, that kind of pot. Okay. We're

Host 2

Like, but like, everyone's got VR goggles on. And they're just like manipulating PCBs in 3d space.

Host 3

Yeah, well,

Host 2

I do quality control any with

Host 3

That. But that's virtual reality. That's not augmented reality, right hands can be both well, but with augmented, you actually have to, like, use your sense for both the reality and the not reality, right? Isn't that but augmented?

Host 2

Yeah, but you can display that over in a VR headset. I have another camera that suffocate like, on the forehead. Yeah.

Host 3

So I just think this is pretty cool. I'm not entirely sure the extent of the use right now because I mean, it looks like they're in a beta program within inspect AR right now. So they have three different tiers, like the engineer that that would, I guess that's sort of like the you and I on our bench kind of level, there's the r&d Lab, and then there's a full enterprise. So they're, they're offering openings for, I guess, requesting access to this right now. So it's still in beta testing. But it is pretty cool. I might need to reach out for a beta test myself, because this could be kind of cool for doing QA at our stuff. Yep.

Host 2

Especially at like, you know, macro would do tons of prototypes. And like, we're a normal visual inspection machine just doesn't have the bandwidth, basically to handle prototypes at scale. We do.

Host 3

Yeah, like, that's good. This looks like it could be really good for like one offs, you know? Yeah, yeah. prototypes. Yeah. Yeah. Yeah. Cuz I guess you just suck in a Gerber layer. And then it identifies the Gerber layer. And somehow, overwrite looks like

Host 2

You also give it like placement data and stuff.

Host 3

I guess you need to pick and place or something like that. pick and place file. I mean, yeah. Cool.

Host 2

Then our last topic is the little robot that lives at the library. And I haven't been able to fully read this article yet. But I skimmed through it when I first found it. And I like a me like, I'm in love with this, this idea. And I just never really thought about it until now. But basically, it's giving the illusion of personality to robots, to make them work better in society. Hmm,

Host 3

You know, this reminds me of a I can't remember where I was reading this, but like it was, it was a general article about humans interacting with

Host 1

Robots over the phone. And just like, the very, very much depending on how, like the tone of the robot would change the human's behavior, purely based off of the tone of the robot, like you could you could very easily make a human being absolutely hate and have a horrible experience, or be like, very sympathetic and kind to the robot based off of the tone of the voice that was played through the speakers.

Host 2

Yeah, and, like, in through this thing, like giving the robots more human features like eyeballs that you can, like humans could read a quote, I'm putting doing air quotes here, emotions from the robot, right? Because it's a software right?

Host 3

No, it's really

Host 2

One that's going to die in the robot overtake overlord. Like next like

Host 3

Terminators with googly eye.

Host 2

Like, they have this whole like, like, emotion matrix of like, what the robot looks like, or is trying to convey to people which goes from like, angry and sad to board to happy and excited and content. And I was thinking about it as a hey, if, you know, if these devices have more human connections or actually have emotions, people might actually not like vandalize them or be you know, like, people like lizard people out there that are not good people. And will that will vandalize public things? Like you know, benches and subway stations and stuff like that? That I think that might be the solution is we just got to put googly eyes on everything the public owns. And then people won't be you know, shitheads

Host 3

Parker solves all of society's problems with googly eyes. Also, I like I like the fact that like you could see my ugly eyes and everyone knows what it is.

Host 2

Right. Yeah, I'd like to prototype with eyebrows. Oh, yeah.

Host 3

Yeah, that's awesome. You know the part that's kind of good. That's kind of annoying about it is like, just a pair of googly eyes and some angled eyebrows is enough for you to feel something with a robot. Right? Yeah, like human beings, we like to think we're complex. But in so many ways, we're really simplistic. And this is one of them. Like, there's three pictures on this website about a robot that's like inquisitive. It's like, what's going on? Or it looks angry. And it's purely just the angle that somebody glued eyebrows up. And do like you read it easily?

Host 2

Yeah, I do like this This subject is some of the customers were worried that the robots were taking librarians jobs. And the research shows that those people's attitudes towards robots may become more negative. IE, you know, picking up the robot and thrown in the toilet.

Host 3

Although, here's the thing, though, think about this. Like one, one function of a law librarians job is to understand in general, the source material that is surrounding them, and to provide suggestions. legitimately, a librarian is there for you to ask questions, and they can give you suggestions, a robot there to help you start your research. Well, but but even if you're like, looking like I want to book on fictional crime or something like that, oh, well, here's a great example of that. So that's less researcher, but it's more like a, but a robot could instantaneously access like data on that and be like, here's the top 10 list of these that you like faster than an actual librarian? Because I would, I would assume, in a lot of ways, a librarian who's just going to Google search after crap anyway.

Host 2

Well, the robot would go look up your personality Matrix Online from like Google and Facebook, and figure out and pick the top one for you.

Host 3

Personality matrix. That's a good one. I have not heard that one before. Yeah.

Host 2

Wow.

Host 3

Yeah. And then and then it would it's Googly as well just dance around and you get the book you won.

Host 2

Yeah, it just spits it out that.

Host 3

But at that point, why even have like, a librarian robot, when you could just, it would just give you the book on a tablet, you know?

Host 2

I don't know. Maybe, maybe. But like, some people like paper.

Host 3

Yeah. And some people want to actually interface with a googly eyed robot.

Host 2

Or people actual humans. Sure. And so we get to the point where robots and humans just look the same.

Host 3

Probably not that far away. Although

Host 2

The verse the Vert. I got it the first like, humanoid robot, or Android or whatever, that looks perfectly like humans, except that still got googly eyes.

Host 3

So like, Is it is it there to like, I can't remember what it's called. Like. Like the prime directive. I know. That's not the right word. But Prime Directive of robots is like number one can't can't harm humans.

Host 2

Are you talking about the three rules? Three rules?

Host 3

Is there like a fourth one must have googly eyes.

Host 2

Yeah, that'd be the fourth rule. Right? That's

Host 3

In the future. That's the only way we could tell humans and robots guys just evolving already, isn't it?

Host 2

Yeah. I think we've got the end. Yeah, we're

Host 3

Gonna did the fourth rule of robotics.

Host 2

Googly eyes. Okay, so you want to close this up? Yeah, so that was the Mac fab engineering podcast. We are your hosts for now until the robots take us over Parker, Dolman.

Host 3

And Steven Greg. Let everyone take it easy.

Host 2

Thank you. Yes, you our listener for downloading our show. If you have a cool idea or project or topic or googly eyes. Let Stephen I know. Tweet us at Mac fat at Longhorn engineer with no O's or at analog E and G or email us at podcasts at Mac fed.com. Also check out our Slack channel which we'll be talking about the E part picker.com in the future. If you're not subscribed to the podcast yet, click that subscribe button. That way you get the latest episode right when it releases and please review us on iTunes as it helps the show stay visible and helps new listeners find us