- Brewery Update
- Still bending tubes
- Going cheap and buying a roll of stainless steel tubing was a mistake
- Started looking into the Jeep Fan controller project again
- Evolving into a Body control module
- Investigating the best way to add wireless remote control?
- Pinotaur LED REV2 design is completed
- Using those APA-102C-NEW-260
- Working on a motor controller add-on for the Pinotaur board
- Socketed motor controllers allow stepper or DC motor control
- Rotary switches
- Three datasheets and all are wrong…
- VAAPA: Variable Architecture And Parameters Amplifier
- CNC Update
- Working on greater accuracy
- Most cuts are within 0.005 but certain situations can make it up to 0.01
- Got a O flute bit in.
- Feeds and speeds
- Analog tricks
- Clipping opamp in feedback
- Virtual earth mixer is no longer a virtual earth
Rapid Fire Opinion
- Arrow Electronics launches European FPGA Developer Contest 2020
- Entry via the Arrow European FPGA Developer Contest 2020 web page and choose from the following Intel FPGA development boards: MAX1000, CYC1000 or AnalogMAX /DAQ1/2/3
- Projects can involve anything from software, special IPs/interfaces and soft cores, to board modifications, adapter creation, adding a PMOD/Arduino, creating an IoT device, and programming algorithms for AI/CNN
Special thanks to whixr over at Tymkrs for the intro and outro!
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 2 00:10
Hello and welcome to the macro fab engineering podcast. We're your hosts Parker, Dolman.
Host 1 00:15
And Steven Gregg.
Host 2 00:16
And this is episode 219.
Host 1 00:20
So, Parker brewery 2020 is coming along, right?
Host 2 00:25
Yes, it is coming along. We missed the April 4, brew date first brew date though. I am still bending tubes.
Host 1 00:36
How much of a bitch is it to bend steel to
Host 2 00:39
So, so I have a proper bending dye and everything for happens to be bending, it's actually not a problem. The problem was, because I cheaped out and I bought stainless on a roll.
Host 1 00:52
And then you straightened it in quotes, Oh, it wasn't really straight. So it's not,
Host 2 00:58
It's fine enough to like bend it because you're you can like, tweak it, it's the tubes aren't completely straight, because they still got kind of a bend to it. But it's like, you know, over a whole eight foot section, you got like maybe two inches of bend. So it's not too big of a deal. The problem is, when you unbend tubing, it tends to not make it round anymore.
Host 1 01:22
Oh, no, you got like a football cross section a little bit. And that probably doesn't bend very well. Well,
Host 2 01:31
Bending again, it's fine. It's when you put it into the compression fitting. Oh, problem. Because now it's what's one cross section is basically wider or has a larger diameter than half an inch, so it doesn't fit in the compression fitting. Oh, that sucks. Yeah. And so I think I want to keep going and bending these tubes that already cut and stuff. And I'm going to order a half inch, just straight tube. So I don't have to unbend it and then cut that up, basically build templates out of my not so round tubing, and then cut the good stuff and then cut the good stuff and then use those as templates. You know, that's
Host 1 02:20
That's actually a really, I mean, overall, you'll probably be way happier with it will probably look nicer. And it'll probably be like exactly what you want.
Host 2 02:28
Yeah, it's just the one of those at least it's not that much money out because I think it was only I think I spent $80 on the rolls stainless. So it's not a lot of money. It's still one of those like, Man, I'm wasting material now. Because like you can't make it round again. Yeah, I guess the only way to do that is if you had a dye and you could redraw it. Yeah. But I don't I can't do that in my house.
Host 1 02:54
I'm wondering, I swear I've seen well, okay, so like, nevermind, I guess I was thinking I've seen some copper tube tools that can expand the ID of a hole.
Host 2 03:08
Yeah, they make, I guess tubing expanders? Yeah, that's far more like making fits.
Host 1 03:14
Right. But I was wondering if if there was something similar to that that would read circularized?
Host 2 03:21
Yeah, maybe that exists. But I guess I'll look after this podcast, that that's a thing. Because I wonder if one of those tools is basically cheaper than buying another 140 bucks a tubing?
Host 1 03:34
Host 2 03:36
Host 1 03:37
Because I mean, how much how much of the actual tube for in most cases? Is it going into the compression fitting? Like the last two or three inches? The last inch? Actually. Yeah. So could you just like put that on a mandrel and hit it with a hammer to make it circular again?
Host 2 03:51
I've tried like, tapping it round. Yeah, that it just takes.
Host 1 03:58
Yeah, it takes more than $80 worth of time. Yeah, I've
Host 2 04:02
Tried actually using a compression fitting that I've messed up because I I soldered it wrong. Because it has the right ID right. I actually tried to hammering that onto a tubing and then pulling it off. It still won't you need something inside. You need a mandrel basically you need something inside to push the wall out correctly. And I don't have that kind of tool.
Host 1 04:29
Hmm. Yeah, it's almost like it's almost like something that has a smaller ID you slide it in the tube and then you screw something into it. So it would expand
Host 2 04:37
Yeah, but the thing is, you can't expand it too much. So you need an outer die as well. Right but the expansion too much. So you might as well just bide Newt. Yeah, because just slamming a compression fitting over it with a hammer to try to like squish it back to shape does not work. That's that probably would work with copper. Because copper is so often enough, but the stainless is so hard. It just will not. It doesn't work
Host 1 05:05
And it completely defeats the whole purpose of a compression fitting. Yes. Yeah, you wouldn't want to do all that work and then find out that you have leaks because you it wasn't
Host 2 05:15
Perfect. Yeah. You know, we talked about this like, three weeks ago about like unforeseen consequences and projects that always tripped you up. Yeah, I thought I was so far ahead of the game. And I start I, I sent a picture to Steven on my first one. And that one, I was like putting the compression fitting on like, that feels a little, it should just fit right into the fitting. It shouldn't be like I'm ramming it into the fitting to make it fit. And then on my second one, I actually looked at it because they actually did not fit. It did not go into the compression fitting and I'm like, oh, that's actually like a square with kind of rounded corners on it. Yeah, is what it is, though.
Host 1 06:03
Yeah. It just gives you more time to do the project.
Host 2 06:07
Now I want to be done with this thing.
Host 1 06:11
And that's why I'm saying Delmon brewery 2020.
Host 2 06:14
Now it's going to be this year. I'm going to tie I'm going to order the either if I think I'm just gonna order the new tubing. Yeah, that sounds like the easiest way. I really wish I could buy it locally. But before all this pandemic Coronavirus stuff. I did go to like my local steel shop and they didn't carry this kind of stainless I need mainly carry industrial stuff.
Host 1 06:41
But think about it. You have 25 feet of practice beds before you before like game time. You know,
Host 2 06:47
That is true. And I actually was like, ought to send you pictures of my second one. It looks really good.
Host 1 06:52
So so are you. Are you eyeballing the bends? Or are you actually measuring and doing like bend radiuses and stuff.
Host 2 06:59
So what I am doing is I'm taking a taking like a coat hanger material to basically I kind of I guess draw it out in 3d space with the coat hanger. Then do all the bends and stuff. And then and then for the radius I am using the My My guess would be the ID died the bender because it's a it's a five d so it would be a two and a half inch radius. Yeah. And I'm doing that and then that then basically replicating that in the stainless and it gets kind of close. And then you kind of have to hand bend a little bit to get all fit just right. Because the coat hangers you know what? Less than eighth an inch in diameter. So it's, it flexes a little bit more. But
Host 1 08:02
So your bends. You're allowing like an extra two or three degrees for some slop. Yeah, totally. 88 degrees to get into that hole. Not 90.
Host 2 08:13
Yeah, but it seems to be fine. I've done too. And I only stopped because I just like you could not put the tubing into the compression fittings. Yeah, that sucks. Yeah. What basically I probably right, straightened it. I probably miss a line my the straightener fixture. But whatever. It just more time and money spring and a big hole and that thing in that project? Oh, yeah.
Host 1 08:38
Yeah, yeah. It's gonna look nice where it's done. For all of those who who don't know, homebrewing is not cheaper than just going to the store and buying there. No, it's not at best. It's the same cost. Yes. I mean, unless you're one of those guys, there's like, I don't know, hitting 4050 gallons at a time, you might get some economies of scale. They're like beer to drink, though. That is a ton of beer. Like yeah, that's That's enough to kill you.
Host 2 09:08
I mean, depends on the person.
Host 1 09:11
So last week, I was talking about those rotary switches and how I was going to order them. And well, I mean, I did order them and they actually arrived yesterday. And the very first thing I did was I was I bust open this package, and I get these rotary switches. And I started looking at the backside of them and all the pins and I was like, Okay, I'm gonna go find every datasheet I can find for these switches, which because I know there's multiple data sheets out there for these things. Because the Internet has history of older data sheets and what's currently available from the manufacturer what you can find on Mouser right now SparkFun sells these. I bought these from tubes and more. They have a datasheet well, they have an image, not a data sheet. And the funny thing is out of every data sheet I was looking for one just one that told me the story told me like What I wanted to know without having to purchase these switches beforehand, and I only found one that had any kind of information. And it only has information for one variant of the switch, which I've actually bought two switches about one that was a one pole 12 position and another one, that's a two pole six position. And I found a datasheet that shows the pins for the two pole six position, but it doesn't show the pins for the one pole 12. So you wouldn't be able to figure that out because they are different. And the funny thing is the only datasheet I found was like buried way deep in a Google search. For like, I searched under the family name of this, you had to go past page two. Yeah, like nobody goes that far. No, Google image searches. And I was like, Oh, my God, one of these actually exists. And lo and behold, it get this, it was the only data sheet that actually had pins numbered on them.
Host 1 11:02
Host 1 11:05
And in fact, one of the other data sheets, I think it's the one from SparkFun. And in no way my dogging SparkFun on this, but it has a footprint on there, it doesn't say if the footprint is from the top side of the board, or the bottom side of the board. And it's wrong. Like if you look at the bottom of the of this, and you design to that footprint, which I did on that project I did earlier in the years ago, it's incorrect. And on top of that the footprint for the datasheet on SparkFun shows a, like a PCB mounting hole, or in a mounting hole with like an anti rotation hole, which that doesn't even exist on the device. So it's just like, Oh, come on. So I mean, and here's the thing, like I purchased the rotary switches that have soldered lugs, not the actual like solder pins for through hole. In this particular situation, on my previous project, I bought the PCB pin ones, they're they're identical, like this switch, the only thing that's different about the switch is, you know, the actual pin pin head looks different. So, you know, this should work for the footprint, the sorry, the through hole, actual footprint, but I just think it's funny that these exist out there. And you can't, you just can't know unless you purchase it like I did. Yeah, and this is, the thing about these switches that I that I like about them is they're really beefy, they're really chunky, they can handle high voltage, they can handle pretty decent current, they're readily available, they have a lot of different options. So if you want multiple poles multiple positions, and they feel really nice, like they have a real nice like chunk when you turn detent to him. Yeah, like a really excellent detent.
Host 2 12:53
And just you don't know what you're getting when you buy him.
Host 1 12:55
Yeah, I that's that's the part that I just cannot understand. Especially like, okay, so Mauser sells these as a as a part, if you look at that, the datasheet for Mouser. It's just basically a link directly to the manufacturer, which was Alpha Taiwan. If you go to their website, I was like, well, maybe the link is wrong. So I went right to their website to find the part number. No, it's seriously just that data. So I don't know, maybe I should hit up alpha and be like, is there like a secret data sheet somewhere that is like all the information that I need to know. Regardless, I have them now. So these are going to be like, I'm going to store these away. And these are my like, every time I need a rotary, this rotary switch, which I've used these on multiple projects in the past, I've always done wire solder to them, other than the macro amp. But now I can just reference them, like straight up.
Host 2 13:51
Did you order a lot of them? Or is that just your reference?
Host 1 13:54
This is my reference. So now I can design around these.
Host 2 13:57
Unless you order another one. It's different.
Host 1 14:01
Yeah, you know that at this point? I wouldn't I would not surprise me at all, then you would
Host 2 14:05
Just use your reference part. Then the one that you ordered that was incorrect as your new reference.
Host 1 14:12
So I'm doing something a little bit ridiculous with this. I've actually I've actually made this before I might have even shown it to you, Parker.
Host 2 14:21
Oh, I'm gonna guess. Yeah. Is this the foam core board amplifier? Yeah, that was one of my favorite things. That's the part the most dangerous thing in your old shop.
Host 1 14:31
Oh, it was amazing. Well, I'm so I'm resurrecting that project. So basically what I did was I made I made a preamp, a tube amplifier, that every component was variable. So every capacitor was on a rotary switch with 12 different values. Every put every resistor was a resistor, a small value in series with a pot. And I built this entire thing on a foam core board looked like a science fair project. And it was awesome. It actually worked out really
Host 2 15:02
Totally look like someone in like 12th grade made it well, so not dogging your ability by No, no,
Host 1 15:08
Because like, the whole point of that project was to do it in an afternoon and I built that entire thing in an afternoon. And it was awesome. It was
Host 2 15:14
Just like, Sharpie marks all over it showing how the, the app was like, it was like the amplifiers schematic, the full schematic
Host 1 15:20
On it. Yeah, just drew lines to the pots and the switches and stuff. Yeah, it was legit, dude. So I actually I, I was thinking about that the other day, because I'm doing some more development right now, on some on some amp stuff. And I was thinking about it. With that particular idea, I built like a fixed preamp. I mean, even though everything was variable, it was still like, I couldn't change the order of how things were. It was for cascaded stages, amplifying stages. And I was like, Wait, what if I were to build each one of those stages, as an individual PCB, and a patch panel, and how well, sort of I'm not even getting that fancy, just like wire them together with like, maybe terminal blocks, or even alligator clips. Because I have this high voltage power supply, I can really, I mean, I most of that work is done, if I just have like a PCB that is a Triaud stage where I can vary everything. And then maybe another PCB stage, that's like a common cathode stage. And then another one that is a phase inverter stage like, I can just connect them however I want, and get amp designs going in, you know, 30 minutes, as opposed to having to design a whole amp just to find out if I like it. You just described a dev board. No, no, no, a if you did it that way, you basically could build a
Host 2 16:48
Oh. What's the the synthesizer modules? What are they called? Eurorack. Modules Eurorack. You basically making a Eurorack amplifier? Yeah, yeah, that's just module and you just like, crank,
Host 1 17:05
That's, that's exactly what I'm doing. And the thing is, like, a lot like, so things like fender amplifiers, they usually put their tone controls right after the first stage. So your signal comes in, it gets amplified by one stage, and then there's tone controls. And then there's amplification stages before it goes to the power amplifier. Marshall amplifiers do it the exact opposite way your signal comes in, it gets amplified three times at least. And then there's tone controls. So I was thinking, Well, what if I just had my tone controls as like a plugin block? And then I could say, What does it sound like if it's after the first stage or the second stage? Or what if it was the third stage, and there was another stage after that, like, I instead of having to just like spend hours wiring these things up to figure it out? I was like, what if I just modularize it so we're going to try it and see what it sounds like? Like I said, I've done this before, just not modular. I made one preamp with everything variable. And it was it sounded great. Yeah, you had one. You had one exact lecture. Right? If if you want to have those stages, I could break out into its own individual block you want to
Host 2 18:15
Variable architecture variable parameter. Right. Amplifier?
Host 1 18:23
Well, and yeah, like so. Like take the tone controls for example, a lot of times in amplifiers in traditional inflators there's they call them TMB, treble, middle bass tone stacks, and it's called a tone stack because if you look at the schematic, there's three potentiometers on top of each other but that's not the only version of tone controls out there. There's lots of variants out there so you know, why not just make each one its own little development block and try it out? See what you like what you don't like valve Parker's writing something in app
Host 2 19:03
Variable architecture and parameter amplifier. Ooh,
Host 1 19:07
Zappa vampa I like that. It's like the DARPA but for
Host 1 19:16
Variable architecture and perimeters, perimeters parameter parameter amplifier. I like how I like how you you included the A from an acronym
Host 2 19:31
Host 1 19:32
Yeah, that yeah, you know, okay, because of that whenever I whenever I get these things onto a piece of board, I will put Bappa on there for sure.
Host 2 19:43
When I come to Colorado next year, you just owe me one beer.
Host 1 19:46
Hey, as soon as the road has gone, we need to get you up here.
Host 2 19:49
Yeah, are you down? You actually need to pick up some stuff here.
Host 1 19:52
I do. I have a I have a ton of crap back in Texas. I have like three people with was with my shit line around there. I need Going out good stuff. So.
Host 2 20:02
Okay, so I've been starting to look at that. So long, long. Oh, man, how long ago? Is this like two years now? Maybe longer? There was that Jeep fan controller project I made? Oh, yeah. And I never really ended up using it mainly because I just went with a different direction with how I wanted the electronics, the auxiliary electronics and the Jeep to work.
Host 1 20:27
Was that was that the board that had your like, 15 Stage pie filter on it?
Host 2 20:32
Yes. As we were, I was just experimenting, like how can like could use stick the dirtiest power input basically like automotive 12 volts, like, just the arbayter? Directly? Yeah, yeah. And could you filter that to something that you couldn't even tell? Yes, you can.
Host 1 20:51
It was like, it was like an LC LC LC LC LC filter, right? Yeah, something like that.
Host 2 20:57
Yeah, looking back, we probably could have done something a lot better, like an active filter. Oh, for sure. But this was just a fun experiment, just be like, Okay, what if we just threw a pipe filter and a choke, and then a lot, a lot of inductor and capacitance and kind of just playing around with it. And that did work. The only problem with the board was I just went with a different direction with the overall project. So I stopped working on it. Well, I started looking at it again a couple weeks ago, mainly because I was looking for, like aftermarket body control modules. I think I've talked about this before a couple, like a couple months ago. But a Body Control Module in a car is like, what controls like your window motors, like for your like going like when you press a button, the current for that motor isn't going to that switch, that the switch sends a signal to the Body Control Module and the Body Control Module goes, Okay, that means lower that window. Older cars, the switch is the switch that's activating the motor, which works until you get a little corrosion and that switch, and then now your windows don't work. So I've been wanting to kind of develop a body control module for the wagon, because it's the old school style switches. And actually for the the checker, the 65 checker and this one probably be using the checker first because it doesn't have any wiring in it. So it has to be rewired anyways, whereas the wagon, it all kind of mostly works, the wiring. So the trucker would be the first one. So I've been looking at this project again, trying to figure out okay, what do I need it to do, actually, in the end, because the Jeep fan controller project had a bunch of like sensor inputs and stuff. And I'm like, Okay, I don't really need that I have gauges, gauges telling me what the temperature is, or like, what the RPM is, I don't need that, I just need to be able to be like, okay, if I'm outside the car, I want to have remote locks. Okay, so I press a button, it unlocks. And I also would like it, if I had a button on the outside to where I could press a button, it lowers all the windows. Because I'm not a big fan of remote starts, which is, you know, you press a button, your car starts up, especially since this is a manual, and it doesn't have a neutral switch on the transmission. So it won't know if the transmission is in gear or not. Which would be probably bad if you tried to start it remotely. Yeah. And it wasn't gear. So I kind of I kind of want a, a remote control aspect to this. So that's what I've been doing some research on. And I started searching for solutions that exist out there. And there's not a lot of solutions. And they're kind of chunky, I guess is a good way to put it. They're not really designed that well, it doesn't feel like and they're also really, really, really expensive. Like the remote controls or cheesy looking plastic blobs and, and they're not really, I guess integrated at all either, like a lot of them just control. Other relays, none of them have realized, like on the board, or using MOSFETs for some reason, like you could just use solid state and it'd be completely fine. So I think that's where I'm gonna go with this project is I'm going to investigate a way to do a wireless control for it. And the big key is, the remote control also has to be really low power. So I don't want to be swamped. I don't want to have to come home and put this thing on charger. Right? I want to be able to like okay, a coin cell last, you know, maybe a year or two. Sure. So I haven't really found the microcontroller yet I want to use because basically it needs to have it only really needs one way communication. I'm not too worried about preventing Someone from happiness. That's only one of these is going to exist in the world. So don't really have to worry about that. And also don't want to do like the cheap way, which is like us 433 megahertz and just kind of like, blast out a signal, kind of like your garage door.
Host 1 25:22
Yeah, that's that's the Arduino way, man.
Host 2 25:24
Yeah. Mainly because also that that that would actually probably be the easiest solution. But I'd rather not do it that way. I kind of want some kind of some kind of wireless stack that's running and running some kind of like authentication
Host 1 25:44
With not just hammering out bits over the Yeah.
Host 2 25:47
Like just blasting out the code to anyone who wants to actually like read read it, or like you're driving down the road and like, your car just starts opening up people's garage doors.
Host 1 25:59
But you know, I don't know exactly how, how true this is. My mother was saying when she was younger, garage door openers didn't have too many different codes. And in her neighborhood, she would drive down the street open other people's garage? Oh, no. That was like three different codes or something like that.
Host 2 26:17
Yeah, most of them had, you would open up either have three or four DIP switches that you would set the code to. Right. Right. So and most people would just like flip them all one direction.
Host 1 26:28
So yeah, cuz that that's easy. That's easy.
Host 2 26:35
Yeah, we actually had our garage door was opened that way before.
Host 1 26:39
Really? Yeah. Nice.
Host 2 26:43
So that's, I want to hear if anyone in our Slack channel or on Twitter has got suggestions on how to make that work. Preferably like maybe a quick example, way of doing it. Like, I don't want to sit there and like, code up like three or four days worth of stuff to make this work. Like, maybe something I can get working in afternoon with some dev boards. Just like the communication like, I guess, like 30 feet range is probably fine.
Host 1 27:16
Sounds like you want for 33 man.
Host 2 27:18
Yeah, I know. Actually, that would probably be the easiest way, it would totally be the easiest. Yeah. Because you just have to blast out of cereal stream of like, this is this is me, this is me. And then like, the cars just you know, always because it's got that big battery. So you could always be it's just listening all the time. Yeah, yeah. But then someone drives by with their their garage clicker and like, opens up my door. So Oh, yeah, that's also nothing is the reason why I need the remote control. It's not for automatic locks, because I don't really care about automatic locks. But it's because on the checker, it doesn't have its 65. So it doesn't have side mirrors. It only has one rear view mirror. Nice. And so the only thing on the side of the body is the door handles. And so I want to do is take those door handles off and shave them what's called shaving the door. And then you'd have to get in, you just press a button on the remote and the door pops open. That's great. So I need I need this to be able to do that. And they make remote kits and stuff for that. But I wanted to control more than just the doors. And like everyone's like, Oh, you have to buy this and then this and then this and now you have like eight boxes shoved up underneath your dash and the wiring harnesses are all over the place. And
Host 1 28:41
Come on, that sounds like something you would love.
Host 2 28:44
I want one box to control them all.
Host 1 28:48
Have you ever looked into the technology? Or do you know what the technology is behind? Like key sensing and things like my wife's car knows what I'm nearby. And, you know, it'll it's really true. Our side mirror mirrors move out. It's like Hello, Steven. You know that kind of crap. Like I freaking hate DAX. I
Host 2 29:07
Tried adjusting the seat and because we can't let you do that seat. You know, I
Host 1 29:11
Don't like the fact that my wife's car has like user logins. I don't need to log into the car like literally like I if we could set it up where it's like, oh, it's just me driving so it knows the weird my seat position is versus my wife and stuff.
Host 2 29:26
My mother's old Mercedes had that but it wasn't like you logged in you just press the button on the dash it had a one and a two. And so you could set up your own profile you're driving you're driving profile with one or two. And that was actually kind of cool. Because you can just sit in and then press one and then it would stop the mirrors and then your seat and your steering wheel just right.
Host 1 29:53
I just you know every time I see those things anytime I see like superfluous luxury in that sense all I think he's like, that's gonna break. And that's really expensive, because that's on a motor, and it's gonna cost. Like, I bet us 1000 bucks to replace one of those sideview mirrors as soon as it stops rotating, you know,
Host 2 30:11
But it's a luxury item, you don't care how much it costs, right?
Host 1 30:15
I do. You do? Yeah, I give a shit. My truck doesn't have any that fancy features. Mine has a nice little steel rod underneath the seat and you lift it up, and the seat moves wherever you want it to go.
Host 2 30:30
It does have a screen in it, though.
Host 1 30:32
I installed that actually. Because I wanted Bluetooth.
Host 2 30:37
But um, yeah, that's, that's the big thing is I need to figure I need to basically settle on like, this is the architecture, microcontroller architecture that I want to use. I'm hoping Silicon Labs has something I can use. But I'm not. I'm not holding out too much for that. So cool. So Steven. Yeah. What is going on with the CNC? Last week, you were cutting calibration stuff?
Host 1 31:09
I think that was actually two weeks ago as well. Okay, so I've been cutting lots of calibration. You know, that's one of the things that really sucks about a CNC unless, okay, now, if you've ever built a CNC from scratch, and that includes like assembling kits of CNCS, then you'll probably understand exactly what's going on here. Like, every little thing you do impacts its performance, right? Like, the trim of the head, you know, the whatever rails system you're using, like has to be bolted on just right, like everything has to be right, for everything to be right. And like, even if you're cutting in one spot on your table, like is your linearity, right? Is it is it cutting the same size in the opposite corner of the table kind of thing. So I've been playing a bunch of games with that, like, technically, I would say that the CNC has been done for a while like, but I've just been getting super anal about getting as much accuracy as I can out of this thing. And from for the most part I'm getting, I can easily hit five thousandths of an inch, I've been consistently getting 3000s or less of an inch for most of my cuts. And then in a couple of situations I get worse, I get like 10,000 errors, which for the majority of the stuff that I'm doing is so far beyond what I need to do. But it's still like, I'm just chasing after that dragon of accuracy. And it's really fun, I've just been having way too much fun with it, it just does kind of suck because
Host 2 32:41
I'm just imagining is a dragon holding a micrometer.
Host 1 32:45
The the overall rigidity of my machine is okay at best, which means that I can't really take it super fast. And that's not necessarily a problem. However, if you with feeds and speeds based on the material that I'm cutting, the majority of the feeds and speeds that I'm going for are faster than what my machine would accept in terms of moving the actual gantry. Know, I can move it that fast. It's just the accuracy suffers because it it relaxes. And yeah. So the, the solution that I'm coming for at this is to go with bits that have fewer flutes. So if I have fewer cutting surfaces, I can actually run my bits slower in terms of the actual feeds, and my my RPMs I have virtually anything on tap. So that's not really an issue. And I actually picked up a you know, I'm confused because I've always called these one flute but I've been recently been seeing them called Oh flutes or zero flute. It's basically the think of taking a razor blade and twisting it into a helix and then sticking that on on a you know shaft. I use these things almost exclusively at work for cutting aluminum. And they're awesome because they have one big cutting surface and a really deep channel in them. So as you're cutting, it makes a big chip and then it throws it and it gets rid of it and gets it out of the pocket and the more flutes you go the kind of like the kind of the spacing in between each flute gets a lot smaller and it packs more chips into there. So if you have a CNC like mine where I have like no chip evacuation whatsoever, I don't even have compressed air on it. I basically have like a weak vacuum pulling chips out of it, like having a bit like this is going to be significantly better because I can take bigger chunks, make bigger chips and then throw them out and hopefully throw them in the direction of the vacuum and get rid of them. Well It'll be nice. So just arrived today. I'm going to start doing some test cuts later tonight on this. And I, you would be surprised. I trash went out last Thursday and my trash bags were just filled with six inch by six inch squares like just tons of them. All of them are like almost six inch by six inch.
Host 2 35:21
Yeah. The Trashman is out there measuring them.
Host 1 35:26
Yeah. So I actually, the machine was not cutting 90 degrees, it was slightly off, I think I measured it at like, I don't know, it was less than a degree off. If across 39 inches in the y axis, it was less than a 16th of an inch off. But that's enough to make me annoyed for what I'm going for. And it really should be pretty much perfect. So I actually I came, I came up with an idea where I, I cut a slice of MDF in just the x axis. So I have a straight line. And to that I clamped a 90 degree triangle those 90s that I actually machined at work. And then to that 90, I clamped a straight edge that went across the table, and then I could drive the bid across the straight edge. It see how far it was off. I cut the shaft of the of the gantry moved, while the motor is actually still held. One side of the gantry is fixed and the other can slide so I can basically adjust the angle and the gantry. made it perfect and then welded it in spot. And it's pretty nice now. So fun little you did that. Honestly, that's been like the best part of this is like coming up with solutions on how to actually fix the issues with the machine. It's like, every time I cut something, I'm like shit. There's another problem. Now I got to figure it out. But no, it's good. And I'm hoping with this with this bit. Tonight I'll kind of commission the machine and call it Dun dun dun as in like, I'm done messing with it. Now I can make things with it.
Host 2 37:12
You're gonna take that that champagne bottle and hit it and it's going to skew the bass.
Host 1 37:16
Yeah, somehow it's like not 90 anymore. Yeah. No so So last week, and I actually cut a cabinet a, like not a cabinet but basically a shell for for an amplifier. And it came out pretty nice. There was a few small issues with the Z height not being perfect. And probably me not measuring perfectly my raw stock before I put it on there. So you know a little bit of fixes there and I'm gonna read cut that same cabinet this week. And I have the old one and then this week and I'll do the new one I'm going to do a comparison to see like, Are all these things I've done every single night on this damn machine making it and actually better or is it just worse in a different direction?
Host 2 38:04
No, it's gonna look exactly the same. Oh God,
Host 1 38:06
I'd be so mad if that was true. No, I can already tell that it's it's different. You know, one of the things though is like the I had to come up with a new clamping system to actually mount the spindle to the Y Z carriage. And the thing is, I came up with something that's an aluminum plate that has an 80 millimeter clamp for holding the spindle, but my YZ plate is not perfect. So they the head isn't perfectly in trim. And so the bid is not exactly 92 to the table and I actually measured that today at work I brought it in and put it under a microscope because on on only like if you cut a square on one side of the square only one side you can see very small step down edges which means that in that axis the bid is angled just ever so slightly, and it's like a 10th of a degree I don't even know if I'm gonna care about about adjusting that but what I've been doing is loosening the the gantry and sliding feeler gauge shims into it to just ever so slightly change the angle of the bit isn't at the table and I think I right now I have a 22,000 spacer which is a pretty big spacer. But I think I overdid it a little bit so I might just slide a 20 in there and not even test it and be like I'm done. We're good.
Host 2 39:38
Are we good? Yeah, well
Host 1 39:39
It's probably good that's that's a great way of putting it probably good. Close enough, right? Yeah. I'll put that on the commissioning side my name probably good. Like I said, like, the entire reason I built this this thing is to make things that don't require this form of accuracy at all. It's mainly more to have a machine that can make them so I can play video games. Well, it makes them and swear to God, that's what I've been doing, which is great. Beautiful part about CNCS Oh, yeah.
Host 2 40:16
I mean, that's the great thing about 3d printer is it you? Once you get it down, and you click print, and then you come back a couple hours later and your parts done? Or it's a big mess of spaghetti, either or?
Host 1 40:28
I mean, I could I could spend two hours behind a bunch of woodworking tools. Or I could spend two hours playing video games and make stuff.
Host 2 40:36
Yep. So the printer project, the LED rev to design is completed on using those APA one or two see new 260s. I tested. But mid mid last
Host 1 40:49
Year. So so give us the verdict. Are
Host 2 40:52
You happy with them? Yeah, they weren't great. We had 100% yield through our reflow process. Yeah, they weren't great. Nice. Pretty happy we tested them with we like took the strand and like wrapped it around, like the coils and stuff like tried to induce as much EMF as possible into the data connection. Because it's just, you know, five volt logic going out to these these chips.
Host 1 41:17
And it's right next to 50 volt high current stuff, right?
Host 2 41:21
Ginormous inductors that are being turned on and off really fast hammering Yeah, yeah. So you your emails all over the place. Yeah. It didn't care. Nice. So that's actually why I thought of I'm like, okay, the worst case scenario is you mess up a couple of bits. And the hue is slightly different. You're not going to tell? No. And so yeah, it worked great.
Host 1 41:46
Also, how often are you writing to those LEDs?
Host 2 41:49
A lot? Yeah. So
Host 1 41:50
If it was off for one cycle, or even a few cycles, who cares? Right?
Host 2 41:54
Yeah. And it's only be a couple of minutes off. So it wouldn't be I actually want to be able to like, because you can write bits out and then at the end, basically just write it twice, and see and then sniff the end, right? And see if you have any bit error at the end. That'd be kind of a cool project. But I haven't done that, because it seems to work fine. Maybe when we ever come across any problems with them, we'll do that. But
Host 1 42:24
You know, it's pretty, it's pretty cool to see it, you know, the old version having like, I don't know, 50% yield on a good day, and then going to this newer version, where it's like, okay, seems like they fixed it all. Yeah, I
Host 2 42:38
Think they just changed materials with lenses and the plastic enclosure. Yeah, they just change that material to handle the higher temperature. Yeah. And things work. Great. Cool. So I'm gonna get those ordered and made. And then I'm working on a motor controller for the pennant cardboard. So the Pindar board has got like four sockets at the top, and we design them so that you can pop in either a DC motor controller or a stepper controller, so you can control for steppers or you can control eight DC motors or any variety mixture of those. So this, this allows it to be more expandable.
Host 1 43:21
And most pinball games for like the toys and the
Host 2 43:24
Yeah, it's like rotating palaces and doors and stuff like that, right? Yeah. Yeah. We, the reason why we didn't put those on the board is because we wanted to be able to support motor, DC motors and steppers and not take up a loss real estate on the board. Like because they're pretty big. I mean, this takes up like 15% of the board, real estate alone, just these areas that you can put parts in. And most pinball machines will only use a couple of these. So we didn't want to have to populate, you know, all that that hardware, right? So we're making them like little modules you use kind of like a polu module for a stepper controller, a stepper motor. Very, very similar to that. And actually, we're, I think it's compatible with the stepper shows that the check this should be compatible. The only thing is not compatible with his DS, we couldn't find a DC motor controller that fit that same footprint. And so we're just going to make our own for that. But I think we it works with the standard blue stepper motors, that's cool. stepper motor controllers, which would be nice. If those are like so you're just
Host 1 44:37
Sending power and then step and direction signals and stuff. Right? Yeah,
Host 2 44:40
Yeah. And then and then it has a header that comes off the board. So you can plug in a you know, big chunky connector to it. Right. So yeah, we'll see how that works out.
Host 1 44:52
Like expansively next week.
Host 2 44:55
Yeah, hopefully next week. I have that. All working.
Host 1 44:59
Nice. Thank you. Uh, so I got a cool little thing that I was working on, or just kind of like messing around with at work. And it's just like a quick little analog trick for circuit design, if you want to, you know, sometimes it's nice to start looking at things in a way that you don't normally expect. And for most applications, you wouldn't do this, but clipping op amps, and in feedback, so take like an inverting op amp, if you go beyond its rails, it'll eventually clip right. And if you take an inverting configuration, sometimes it's called a virtual earth mixer, you can, or virtual ground mixer, like if you just have just a regular inverting op amp, the inverting terminal is, you know, I'm using air quotes, here's zero volts or it's ground. And that's not actually ground, that's the op amp, you know, maintaining its feedback loop and keeping that at a zero volts effectively. So as soon as you get that op amp into clipping, it doesn't, it no longer has the capability to actually maintain that virtual earth. So that node that previously was seen as a different impedance to everything that's plugged into it, and now completely lost all of that, and is now now everything that's plugged into, it sees the impedance of all the other things that are connected into it. And so just in a creative way, I've been using that recently. And not necessarily with the clipping of the actual op amp, I've been using zener feedback, Zener diodes, or just regular diodes in the feedback, such that if you're in the conduction range of the diode than the op amp, it has control over itself. But as soon as it clips, you don't have control anymore, and you get a bunch of voltage dividers on the front end. So I've used that recently to make some really cool kind of distortion circuits that are have like Voltage Controlled clipping and things like that, such that once you're past a certain limit, then the op amp loses control. And you can do things with that. So just some cool little analog tricks. If you're designing weird things in analog, just remember that you don't always have to play by the rules. And a lot of these textbook things happen with particular conditions like an inverting op amp only is inverting if it's not clipping, or it's only like the feedback loop is actually only working properly, or all your equations for that op amp are working, if it's not clipping, so just keep that in mind.
Host 2 47:38
So you're talking about you're using the Zener diode to induce the clipping then right you will I use beating the rail, you're you're using the Zener diode.
Host 1 47:49
And the entire reason why I did that is because the rails are not defined in an op amp. And if you're if you're designing a product, you want it to be more consistent. So I you know, a two or 5% Zener is way more consistent than just relying on the rails of an op amp clipping. So you if the clipping level doesn't matter for your application, then you totally can clip the rails of the op amp. You know, there's other things to take into account, if you're clipping the rails, I'm not saying just go and abuse your electronics. But like, yeah, if you have a zener in the in the output, you can you can define what your clipping level is, and then exceed that and have fun in that territory.
Host 2 48:32
You should have a sound demo for us.
Host 1 48:35
Yeah, I can totally do that. Well, with this. There will be some sounds coming out here soon. There's something that incorporates versions of this technology.
Host 2 48:45
Ooh. I'm looking forward to that. And so this week, we have one RFO and this is the Arrow Electronics launches European FPGA developer contest. 2020 I guess with everyone staying at home, everyone's looking for stuff to do.
Host 1 49:02
Yeah. Well, apparently this went really well in 2019. So they decided to bring it back. So air electronics is doing I guess it's a design contest. Similar to the macro FEHB design contests. Yes. But under another one of those, you know, maybe that's a summer thing. Yeah,
Host 2 49:19
Maybe that's the summer I'll yeah, last year was in summer two.
Host 1 49:22
Yeah, we did. Yeah, we did late summer on that. Yeah. After the road. Oh, man. We gotta get past the road.
Host 2 49:29
Stay at home, everyone. Yeah, stay at home
Host 1 49:32
And start thinking about things to design. But I saw this earlier today this the air electronics contest. It's pretty cool. So it's specifically for FPGA stuff. And what kind of sets this one apart is that you, you apply to be in the contest. And if your project is selected, then they actually provide you with a development board and you get to pick which development board you want. So you come up with the idea beforehand, write it all out or write a general proposal, and then kick that out to them. And there's Gosh, how many is this 1234, I think four different development boards, all FPGA base, which is pretty cool. And these projects can involve anything from software serial IP interfaces, and soft chords to board modification, adapter creation, all kinds of other stuff on this list. So I think it's pretty wide open. So if you have an idea that is FPGA or could be implemented on FPGA, then there's a pretty cool little contest. Well, and if you live live in Europe,
Host 2 50:41
Now saying is like I don't live in Europe, but if I did apply to this and got a board it probably just go live in my dev board. Graveyard.
Host 1 50:50
You know, I have a drawer right here. And yet, I've got I've got your your ice stick over here.
Host 2 50:58
Oh, not poor thing. Here's mine. Oh, nice.
Host 1 51:01
Yeah, yeah. Right next to ya. Right. So I can always open it up and look at all the things I'm not doing.
Host 2 51:07
Yes. All the stuff. I have so much MSP for 30 dev boards. Oh, my gosh, back in the day, back in the day, ti would go to like maker fairs, and they would just hand them out for free. Right? Yeah. So I just you every day at the fair you'd go around and get your free MSP for 30. That's right.
Host 1 51:27
Host 2 51:28
Oh, yeah. Excellent. I was wondering, I wonder if MSP 430 or one of the MSP lines has a low power transmitter setup.
Host 1 51:38
Now you got more go research? Yeah.
Host 2 51:41
I wonder if you could do it'd be really nice if you would, if you if I could incorporate the switch press powered it on. And couldn't figure out that that was a button that it would you know, that button is like open right door.
Host 1 51:58
So instead of like being well, I guess it would be a soft power at that point.
Host 2 52:02
Yeah, well, you couldn't make it a hard power to Yeah, because it's just it boots up and transmits that signal on your button press. That's kind of like how a a single button you know, 433 megahertz style garage door opener works. Yeah, it's basically what it powers on because that tax which press it just beams its information out to the world. Right. Interesting things. So that was the Mac fat engineering podcast.
Host 1 52:32
We were your host Parker Dolan, Steven Greg. Later everyone take it easy.
Transcribed by https://otter.ai