Scott Swaaley and MAKESafe Tools

Welcome to the macro fab engineering podcast. I am your guests Scott sweetly.

And we are your hosts Parker Domine

and Steven Craig.

This is episode 131.

Scott is an electrical engineer from California Polytechnic State University. He started his career rewiring 18th century tall ships. Afterwards, Scott designed electrical systems for hospitals and data centers before becoming a specialist in renewable energy economics. Scott then began a career in teaching. And after working with students and adults in maker spaces, he discovered a new passion for tool safety, and founded make safe tools.

So Scott, what other things

do you do? Well, I tend to have pretty fruitful but kind of disconnected hobbies, they they vary quite a bit from year to year. I think nowadays, I'm most often kind of tinkering around with programming. For example, I just wrote recently a a gear geometry calculator. So if you go to pulley generator.us, it's just like a fun thing for a project.

What's that? What's that written in?

That one was in JavaScript?

So you're a, like, multi discipline programmer, then?

Yeah, I'm, I'm just good enough to get things done in a lot of things.

The Jack of jack of all trades, right? Yeah,

I'm persistent. More than smart, I think. But I do that. I also do. I'm a little bit of an amateur machinist, I somehow ended up with a 1935, South Bend lathe in my garage, which is awesome to play with, and actually had the original shipping certificate, which is kind of fun. And then I'm kind of into a lot of random thing. I mean, some days I'll write a poem next day. I'll program next day all who knows what I've done stained glass the other day, just for fun. Nice.

So what made you get into let's say, amateur machining?

Yeah. So I, I used to be a kind of engineering slash STEM teacher. And I, I don't know if you guys follow click spring on YouTube. Yeah. Yeah. So my wife started, like, recognizing His voice instead of mine, because it was on so much in our house. And so I basically learned how to machine from watching that channel, and brought it into my classroom and machine with kids for a long time. And I'm no master at it by any means. But I like making simple things. I can do a little bit of CNC work. I like lathes manually, just because there's so much fun to play with. And it's amazing how much in the house like, I'm just like, oh, I can go remake that part. Remake like a bushing for my bandsaw instead of going out and spending like four hours online trying to find the right part.

spending four hours actually cutting materials.

Yeah, exactly.

Maybe 13 hours, but you know, regardless. Oh, I need to sharpen my own tools. How do I do that? Let me learn how to do that.

i The same thing happened to you when I started learning how to weld is you start looking like oh, I can fix NOx. Now I can glue metal the metal? Yeah. Yeah.

And, and then I realize, Wait, my house doesn't have a 220 outlet? Oh, I just have to figure out how to plug this in to where my dryer goes. Then you have those running through the house big little cables.

So I'm curious actually a little bit just back on the the clicks spring thing for a second. You like watching that one of the hallmarks of him is that guy must be probably the most patient man to ever exist ever does that? Does that apply to you

know, I am persistent. But watching just when he files a gear tooth. I can't imagine doing that once and he will do it perfectly for like 17 years. And that just blows my mind. I have no idea how that works.

Oh, yeah. Yeah. And and his voice is so soothing when he does it.

And his videos are so high quality I don't even understand like his shop is so small. Like where does the camera go in those shots? And then, like, I've had that same laid like, how can you see what's happening?

Mm hmm. If y'all like click spring, there's another guy called this old Tony. Yeah. And he does machining and so I've really liked his style of doing videos.

He's got funny ones. Yeah. But he's he does machining but he also will do welding and have you ever seen the one where I think it was this old Tony where he's, I think milling a keyway on the inside of a pulley and he's using his shaper and he fuzzes out part of it. Yeah. I don't know if I can describe that in too much detail. That's my guess. But it was hilarious.

It's a device that puts something inside of a hole and back out of it. And then he censors it with the pixelation.

Like my, my wife ran into the room because she thought something was wrong. I was like laughing so hard.

So yeah, this whole Tony, go check them out on YouTube and it gets click spring as well.

Alright, so let's get back to our guest.

If you look at enough, click spring, I'm actually mentioned by name in one of those. Are you really? Yeah, I had a question about tailstock. And he in one of the q&a episodes, he he mentioned me and I just about shattered brick.

Brass brick, right?

Beautifully machined.

Let's go back to like going your first jobs Scott, which was rewiring a tall ship from the 18th century. Yeah. So this is the thing though. An 18th century didn't have electricity on these boats. That's right. I'm sorry, ships.

So I don't know if you guys remember. Did you see the very first Pirates of the Caribbean? The film? Yes. Yes. You

mean the only good one.

The only one I've seen I think that there's the interceptor is the fastest boat and it blows up at the end. Remember that boat? The British boat? Yeah. So that ship in real life is the lady Washington, a recreation of a tall ship that goes up and down the West Coast. And I don't even know how I got hooked up with that. And volunteered and ended up spending two different summers almost, like about six months total living and working and volunteering on that. And being electrical engineer and kind of curious and willing to get myself in the trouble. I thought it was kind of odd that you know, we're taking lots of passengers out and sometimes the Start buttons wouldn't work. And sometimes you got to wiggle things. And it was a a lot of volunteers have contributed in their own way to its electrical systems

will say that it would just randomly catch on fire. And

I think bobbing is the word you guys have been using. So like I started, like looking at it, and it got to the point where like, we'd have guests coming in in an hour. And I'd have the whole like control panel off. I mean, this thing's got this humongous Volvo engine. I don't, I don't know what it was probably 150 horsepower, some crazy. But this huge, huge diesel. And you crawl around and there'd be like this mess of wires probably 80% of which go nowhere. And it just it just wrapped in duct tape. Oh, and and so coated in oil, you couldn't even see what was what I think my favorite was I once found a there was like, some kind of oil pressure sensor or something that had just a positive lead coming out of it that needed to go to 12 volts. And it was connected to 12 volts at the microwave in the kitchen. Or like, you accidentally touch your knee to the engine block. And you can feel a little tickle of grounding it.

We check the oil pressure on the engine. Let's microwave popcorn,

right? Yeah, check the pressure. So we I did a whole bunch of work on there. And that actually turned into over time, me running a business called DC marine electric where I did electrical systems and control systems for yachts. Oh, that's cool. Yeah. And I was like, hey, I need to test your new control system, I'm gonna take your $3 million yacht out on a ride is great, though. I learned I learned a lot doing I bet

you the wiring was a little bit better in those right. It's questionable.

Amazing. So So one of the things I learned was that the the Number One Insurance Claim on boats is electrical fires by a significant margin, because you have heat and saltwater, and a lot of them will have sometimes three or four different voltage levels. And so people end up mixing and matching. And when you have water, you can do go. If you don't galvanically isolate, you can basically put a nice brass fitting into the side of your boat. And if the guy next to you is wired wrong and you're wired wrong, you can actually in like a matter of months, completely eat away that breast fitting and just have a hole in your boat. So I mean, it got it was really interesting, like simple electronics, but also like lots of little details.

And what people don't, I guess don't realize too is actually insulation on wires will actually have different, like temperature ratings, given if it's wet or dry.

And it gets really fun when people start using plastic. You know, what are they called the wire clamps. So then if he does, like if it's across the top of an engine block, and it gets too hot, then those melt and so the wires fall onto the engine block all sorts of goodies. Mm hmm. I'm actually I think, I don't know if it's still applicable, but at one point I was a certified marine electrician. What's

the number on that?

I don't know. I like 20

So it's got a What What does make safe tools?

Yeah, so it's okay, if I kind of tell the story about how it came to be. I think it makes more sense that way. Yeah, sure, please. Awesome. So, after doing a bunch of different kinds of engineering, I very kind of serendipitously ended up in teaching. And I taught all sorts of engineering stuff to kids, which I talked about a little bit. And I was running a shop with a team of 15 engineers at my last place. And anytime you're in a shop with shared people, humans level and children, it's, you know, having processes to keep everyone safe, safe is a significant challenge. And I started getting really frustrated with some of the things I'd see, for example, this is equally true in adult shops, you guys are pretty handy seems like with a bunch of tools. So like on a bandsaw. When you walk up to a bandsaw, to cut a material, what is one of the first things you adjust

the little guard that exactly? blade guard blade guard, you take it out of the way you put it all the way at the maximum and be like, oh, man, it doesn't go wide enough to cut material.

need this whole tree I'm putting in not fit? Well, you were right on the blade guard one because I know that most people, like first thing they do, even if they're doing everything right is they're trained to like go and adjust the blade guard to be a quarter inch above your material. And most bandsaw is when you turn them off, keep spinning for like two minutes. And so you're in a loud shop, you walk back up and you have a fine tooth blade. You can't tell it's spinning, you can't hear it. And the first thing you're trying to do is reach for the blade, it makes no sense. And so I wanted to put a brake on it, right? Oh, weld on a disc brake from a bicycle or something. And then that was like, No, I'm not going to do that. I have more chance of breaking it than I have do anything. So I started looking at a electronic brakes and seeing what was out there. And I saw like huge industrial stuff that was way outside our region budget. And so started tinkering. And over a couple months developed DC injection brake that worked for AC induction motors. And I just developed it for our shop and started using that on our band saws and our and our disc grinders. And people kept coming in and be like, Where'd you buy that? Where'd you get that? That's interesting. And I would do access control stuff. And people would always ask about them, and kind of accidentally realized over time that that that was a product. And so now i Our first product is a plug and play power tool brake. So you basically take any stationary power tool that uses an induction motor. So that's almost any tablesaw, bandsaw, bench, grinder, disc sander, anything like that. And all you do is plug it into this box, and you plug the box into the wall. And then when you turn it off, the tool just stops in a matter of seconds instead of coasting forever. And one of my kind of founding principles was I don't want this to be something an engineer has to install. I want it to be like the Apple thing, right? Like you just grab it and you plug it in, and it works. And so you don't need to you don't need to hire an electrician to install it. Exactly. And you don't need to be like an engineer to figure out which one to buy. And so it'll work with any induction motor, you can plug into a 15 amp outlet. And we're we just finished beta and we're selling them now. So it's pretty fun. Well, I

mean, this may be jumping down the line a bit. But but I'm I'm curious about how that has impacts with component or device testing and like UL marks and things of that sort, like, because you're no longer you're you're in between it and what you will thinks it's being connected to. So like, what kind of impact does that have?

So the one from the US I one thing is since we don't physically modify the tool and its simplest form, it's really just an on off switch that sometimes rectifies the AC

That's awesome. Yeah, that's that's like that sounds like one of the few times that you get to get around you ELLs rules. Like,

Well, it's interesting, you talks about a lot of us base requirement for at least ul 508, which is motor control, is kind of it doesn't say what it's gonna do is one of their biggest requirements. And so for motor braking, it's been around since like, Tesla age when he invented the induction motor. And so all that there's really no influence on the system, the things that you could get a little hairy with is if you had delicate electronics in your tool also. So if you got some fancy RPM meter and things like that, like in my manual, it says don't use it. And otherwise, the only thing it does is impact the duty cycle. So if you got an AC induction motor that's meant to, you know, it can successfully turn on, you know, seven times a minute without overheating. Now that's going to be five times a minute because each time you're also pumping some juice into to slow it down. So it just paid some heat in the motor. But for hobbyists, that's never an issue. And even for most industrial applications, it's not even close to the the temperature rating of the, the enamel on the coils.

So I the term DC injection, what does that mean? Because to me, to me that's like, for Well, I don't know what that is, it sounds like you know how in like, in sports, like baseball, you have like the little league teams that like, like the triple A, that go into the big leagues, like they feed into the big leagues, that's like a little band that's feeding members into the AC DC band.

Now it's a DC injection, you actually take really bad superhero movies, and you put them in a syringe, and then you inject. So what you what you do is, it's basically just so I had to learn a little bit about AC induction motors, which are cool, because they don't have brushes or anything, you know, go Tesla. And so DC injection, you basically are just putting it don't even have to be filtered just can even be just full wave rectified DC into the same motor. And instead of creating a rotating magnetic field, it creates an fixed magnetic field in the stator, which induces the opposing magnetic field in the rotor, and makes it want to stop.

Oh, okay, so yeah, so you're turning the base of the outside coils of the induction motor into a permanent? Well, electromagnet. Yeah, exactly. That freezes the inside. rotor. Yeah, exactly. All right. Okay, cool. Cool. That makes a lot sense now, and that fun. And the cool thing about induction motors is there are no magnets in them. Oh, the other

all electromagnet on both stator and rotor, right.

And, and even the rotor is not electrically connected to anything. Ah, nifty magic. It's all it's all induced, hence, the induction motor.

So, so your little your little box senses when you've turned the product off, and then immediately begins rectifying that.

So right now we have just an alternate, off switch. So instead of turning your tool off, normally, you either hit a pedal or hit a little button that's just mounted right next to your off switch, and it knows it disconnects AC power, puts it in safe mode, breaks it till it stops, and then you're done.

Okay, so it's, it's a separate physical switch. So you leave whatever your device you leave that on, and then use yours as the master switch.

Exactly. And we're and we're pretty close to having a microcontroller version, I've got all the kind of peripherals ready. And I'm in that like, oh, which microcontroller? Do I buy phase right now?

It's a tough, tough phase.

And how do I isolate it means without, you know, killing my bomb cost.

So how does it detect

when the motor stopped running? So after you, you're injecting the DC into it? And how does it detect when it stopped? So there's two versions. So there's the our current version and our beta version, our current version, you just have a torque and a time adjustment. It's a one time adjustment. And so it says how hard it breaks and for how long. And then our beta actually does active current sensing. And there's a change in inductance as the motor slows, so you can actually detect the phase shift between voltage and current. But that's not quite out yet.

This one the current, I would guess, when the current spikes is when your motors stopped or correct.

It's actually it's the phase you look at so. Okay,

so So I'm curious about, sorry, we're sort of jumping all over the place. But I love this. It's great. It's really interesting. So I'm curious about how, how your testing has gone on this. Have you just kind of like taken it to people's garages and like hooked it up to a bunch of random stuff and said, let's give it a shot.

Well, early testing or recent testing? Were these different answers.

Just curious about all of them? Yeah,

he probably tested those 1935 South Bend lathe first.

I have actually done that. That's one of the things that leads to my one of the big sections in my manual, which is about reverse threaded spindles and why not to use DC injection on reverse threaded spindle back in 1935, they I mean, that would never exist on a lathe now, but if you have something massive, like a big Chuck, it wants to keep going. And so it will unscrew itself right and chase you across the garage.

And that doesn't that's that has a little bit of gravity and it weighs a bit right. Yeah,

just just a bit like eight pounds of cast iron sitting on that pick. Yeah, yeah. But uh, I ended up so I had a couple. I mean, I have this poor little delta bench grinder and this harbor freight disc sander that have been my kind of testing devices for God almost two years and they have experienced every possible failure case that exists and are still going strong cool, which includes like go into a go into a trade show and running it like cycling it like six times a minute for four days straight.

Now that you know that that was actually a question I was, I was gonna ask you have you done like just the most brutal test of like, get take it to full RPM, break it take it to full rpm and just like how many times can it withstand that? You're your product and the other?

I've done 10,000 And it was okay. Wow, that's impressive. The thing that starts getting worrisome at that point is motor heat. Because at some point, you start, I think, like the the lowest standard enamel rating is 155. C. So if you get above that you risk internally shorting the motor. And that's, that becomes an issue way, way before my door. I says trouble.

Well, okay, so if if that actually does happen, where you short the coils in the motor, does your device handle that?

No, not? Yes, it is not.

Sorry, I'm not trying to put you on on this, buddy. I'm just curious.

I think I think you might have to, I think might have to worry more about like the cord that's plugging the device into the wall, or your breaker. At that

point. I think just either the breaker on my device or the upstream breaker would trip because it would lower the impedance so much.

Yeah, you might get some nice smoke coming out of the motor too.

Sure. Yeah. Yeah. This is intended for 110 15 amps. Right?

Right. Our next version is 220. Also, but this one's all 110. Oh, awesome. And it's crazy, though, like I never put this together before I was doing all this work that you think about like your little, like, say that Harbor Freight disc sander. That's, you know, one and three quarter horse, you don't really think about that much. When it starts up, it pulls 40 amps. I mean, 40 amps at 110. I mean, that's a huge amount of juice. And I mean, my neighbors probably hate me, because, you know, apparently, the the wireless router in our house has the lowest, or excuse me, the highest low voltage threshold. And so without tripping a breaker, I can power cycle my router, if I'm doing like, really, really high current stuff in the garage. So I'll just get like angry stomp on the floor from my wife, like stop testing,

trying to watch Netflix. So I'm Scott, rewiring 18th century tall ships to designing your own product. How'd you get there?

Yeah, well, yeah, well, in the middle was a lot of really big scale engineering, like designing logical systems for data centers and things like that. So we'll skip that for now. And I got there honestly, by, like, I'm not as patient as the guy from click spring. But I am very persistent. And so, you know, I had an IE background, I've been a tinker for a long time. So I had the basics. And I my first device was literally a bunch of off the shelf relays, kind of hot glued to the inside of a metal box. And then over time, I realized that the amount of time it takes me to hand wire that is ridiculous. And so I started getting back into board layout. And I had done a little bit of board layout in college, I was actually a TA, and had done that for a while and then not touched it for 10 years, and came back. And that was lucky that I was at Hackaday super Con last year, which is an awesome conference, I highly recommend it. And I am someone that normally hates conferences, and that one was pretty good. And I was in Matt Bergman's little workshop was one of the guys that I think bought eagle from the AutoCAD side. And he did a little workshop on it, which was hilarious because the workshop actually didn't work because they had a server crash. But afterwards, I was asking a bunch of questions. And he basically spent like three hours with me showing how they had now mapped all the libraries to 3d models and how you can do your own 3d content. And then that automatically syncs to fusion. And so basically, I can do all my mechanical and electrical design in a free or nearly free software base, which is pretty sweet. Because my mechanical designs very simple. And so I basically would go through and make something work and say this is too expensive. And then what can I switch and realize that I should be using some solid state stuff. So let me do that. And then you figure it out, like with Triax. And I did a lot with looking at the benefits between electromagnet electromechanical devices like relays, and solid state AC devices like Triax or SDRs. And then hybrids and then I was going back and forth with that a long time and if you've ever dumped 110 volt DC into a large inductor and then tried to break that circuit, you will make an arc welder and so There's a few times.

We just talked about that last episode, I think. Yeah, cut too quickly.

What was the I think it was Parker that posted that picture of that little transistor that gave up the ghost on your, on my wagon? Yeah, that's my bench looked like that for a long time. You'd be amazed how gloriously you can destroy a relay. windings can actually shoot off sometimes. But I spent a lot of time going through that. And that is one of those kind of lost arts, like everything you look at online is his DC, low voltage stuff. So you start having questions about high power stuff, and you got a couple, you know, forums that talk about it. But then half the stuff they say on there is like not true. And so I ended up figuring a lot out and doing a lot of tear downs and isn't adventure.

That sounds like fun. So let's I guess let's break down your device. And if you want to, right,

yeah, I'm gonna give away all my IP right now.

No, how do you do the switching then. So what why, why is a really a bad idea, I guess for this high powered, you said like 40 amps at one 110 volts.

And so industrially, for these kinds of applications, they will use a contactor, which is basically a relay made to break larger currents. Because if you get a off the shelf 15 amp relay that will maybe be able to break break, like three amps of an inductive load. And so all of a sudden, if you if you start going from your general purpose relays like would be in your little, you know, smart switch that your lights connected to with Alexa, if you start going into inductive loads, all of a sudden, none of those work. And so you start getting into this world where you want really low heat dissipation. So you want just a piece of copper conducting electricity, which relays are really good at, but they suck at switching, especially with inductive load. So then you want something like a triack, which is a in the thyristor family, which is kind of like an AC transistor. It's a good like corollary. And those are very good at switching, like very different types of loads. But they dissipate one watt per amp,

is that just a normal is that because of the diode dropping them,

it's I don't know, to be honest. But the because they don't drop much. But I guess if they were if it's about a volt, that makes sense. And it's just, I mean, on my DC side where I use Triax. And I do firing angle Triax, we basically turn it on and off to only get kind of chopped sections of your your AC waveform. And if I do anything for more than about a second, the heatsink requirements just get ridiculous. And you have to start worrying about even junction temperature, because no matter what kind of heat sinks you have, you got to get the heat out so fast. That if you're a little to 220, like that's really hard to do. And so I started doing a bunch of research into both kind of switching, and then also arc prevention. And there was this, I found out that what most I won't say most what some people do is have a hybrid relay. So you basically have a relay contact and a triack in parallel. And the Triax does all the switching and the relay does all the carrying. So your triack never needs to be on for more than like two cycles. And so you basically you have no heat dissipation. And then your relay can sit there like a wire and do its job. And then before it breaks the track turns off. And I had, I had been playing with that for a little while trying to get the timing right, and found this product online, which I don't know if I should name or not, I'm going to give away all their IP to. And it's this magical, you just put it on the across any contact and it eliminates all arcing. And I was like That makes no sense because all their patent literature is about like RC snubbers.

It's called a diode.

It's just this little box that potted with two terminals coming out of it. It's a snub, it's a snubber circuit in a box. It's basically a snubber but RC snubbers on this, this high energy stuff basically do nothing. And so, but all their patent literature is about like using a metal oxide varistor or something. And it's like that makes no sense. Like how does that work so well. And so I unposted one just for fun. And which just sucks like unpacking things is so terrible, especially when you're trying to like preserve part numbers. And it turns out, it's exactly what I'm already doing. It's just a hybrid relay application. But the magic is and I don't know exactly how they do this is there. They have some logic in there. That gets powered by the arc itself. Ooh, because they have no neutral. Make can go across any contact. So they basically get they see an arc forming. And then they do a hybrid relay. And then it runs out of juice and turns off. It's, that's fancy. Yeah. But it's one of those, like, I'm already doing exactly what it was. And I'm like, What are these people doing that so much better than me? And then I look at it like, that's the thing. Yeah. So it's cool, though.

So. So for a little bit more explanation, I guess. So hybrid relay, what you're talking about here is so the relay and the track are in series. Correct? No parallel. Oh, okay. Completely different than I was thinking of. Yeah.

So they imagine that they're basically redundant switches. They're kind of like an order switch. And when you first turn on the circuit, you turn it on with the triack. So then current flowing through the track to

the load. Oh, okay. Yes, yeah. And then when the relay closes, it doesn't have the current surge through its contacts. Okay. Okay, that makes sense. Exactly. And then in reverse when you disconnect.

So that prevents the terminals from welding. Right?

Exactly. And you go, I have I can, we can upload some pictures to the blog of kind of the 10,000 cycles of a relay with and without, and with hybrid. They have it no pitting and nothing. And if you don't do that, and just do very simple AC loads, nothing fancy with Dc, you just destroy the contacts in 10,000 cycles. Hmm, that's pretty fun. Yeah, that's,

that's a great idea. But I mean, the usefulness of that, correct me if I'm wrong, but that sounds like it starts to take place when you really start getting into higher power applications, right? Like, that wouldn't necessarily make sense at low power applications.

Right. And it's really the inductive load piece. So some people will even do that hybridization just to drive a relay coil. Because if you think about driving a relay coil, you know, like, you got to do flyback, diodes and things like that, to make sure they don't blow up your power electronics. And so depending on how people are doing things, when they do like ladder logic with relays and things like that, sometimes, though, they'll use it for that. And then oh, can I go on a total tangent right now? Please? Okay, awesome. So, other thing I learned recently, from a application note, and it's one of those like, you've read 100 data sheets, and this one says something that no one else has. So you guys know, activating a coil, right, just like little coil for like low power for a relay or something. If you do that with a MOSFET, or a transistor, or whatever, the voltage spike, when you try to turn it off, can fry your electronics. And so people put a diode. So it can call a freewheel diode or a flyback diode, I think and so and so it can basically dissipate its energy through that little diode. Everyone does it, right. And what I learned from this data sheet, which I can look up later, we can put in the notes, is when you do that, you're actually letting the magnetic field maintain itself just a little bit longer. Because instead of when it when it voltage spikes, it voltage spikes really fast. But then the magnetic fields gone. When you put a flyback diode you add, you know, half a millisecond or something where there's actual current flowing,

well, yeah, cuz it's putting that voltage back into basically backing the current back into the core coil,

right. And so you actually have a instead of the magnetic field collapsing quickly, you have a slow release magnetic field, which means that the relay arm, because there's actually a mechanical thing on a spring in there, kind of is a little sluggish and leaving and, and like most relay contacts are welding themselves like microscopically, all the time, in the spring action pulls it off. But what can happen is because the spring action is being fought with that, like diminishing magnetic field, you basically can kill your relay much sooner, because you're not letting it pull the contacts apart with its full force, which was like, Oh, that makes total sense. And they have some really cool waveforms to check that out. And so what they do instead, is they put a zener in series with the flyback diode. And so if it's a 12 volt coil, you can put a 12 volt or a 15 volt zener there. So it basically will be a flyback until it hits the Zener voltage, and then it just shuts off.

And then lets the magnetic field collapse naturally, right?

So it basically absorbs the worst part of the spike, but doesn't let it do its like little slow dissipation thing. So you still get most of your spring poll, and you get the you kind of shunt the worst of your voltage spike, but it's like one of those little things. It's like, Why didn't someone tell me that like 30 designs ago?

Well, and here's the thing, like, here's gonna be the engineering question that comes after that. It's like, okay, great. Everyone can probably get on board with that. But what value zener do you pick for your application?

Well, they actually have a really good note about that they say they say, I think it says like, quote, choose a zener value appropriate for your application. Thanks. Thanks.

That was the thing is I was I was looking at that. Because Scott actually, Scott sent me that application and I was reading I'm like, Oh, this is great. Because I was like, this would be really good for pinball. Because there's a big problem with with flyback diodes and pinball burning up and stuff like that. And I'm like, this will, this will reduce load on those guys. And, and the thing is in pinball use MOSFET. So you're not really worried about contacts, welding, and relays. But just having the actuators your solenoids will actually retract faster, because there's less field now in those coils. And so like, Oh, so you would have a, your, you could sell your game design as a snappier gameplay, even though it might not even be perceivable by human beings.

But microseconds where the difference

by microseconds, but it's there, you can call it like Uber diode or something like that. Uber diode technology in your pinball machine?

I just picked one that was like, near my voltage level for that the coil was rated for and it's been great.

I figured that's what it would be. It's somewhere near your, your voltage level?

Well, it's, it's, and it's one of those things where like, nobody's going to be able to tell you what voltage that coil will snap to when you break it. You know, it could be, it could be millions of volts. That might be a little bit hyperbolic. But it could be a lot of volts, like, you have no clue. So what zener just like,

and you can't read it on your scope, because then you load it and it really dissipated. Yeah, I've got it. What? What's the input impedance like 10 Meg, or something like that. And I guess that's enough, because I can never detect it. But then it keeps blowing up.

I think I think that that actually comes down to like gut feel, like knowing what the what the switching voltage of the coil is. And then picking a percentage within there and just being like, gut feel and then test it. If if our listeners

know how to measure, like the flyback voltage off a solenoid correctly, let us know in the comments below or in Slack. That would be really fun. I'd like to try to measure that. But again, I think my scopes are only 10 meg inputs as well. So it'd be fruitless. At best. Yeah.

And another tangent so scopes. One thing I've learned and relearned and refuse to learn is I'll actually phrase this as a shout out to Riggle oscilloscopes because I have shorted line voltage through the ground of their probes probably 30 times and I still have a working oscilloscope and I don't know how that works. A like 20 amp breaker,

they have they have really really really thick chassis like frame inside.

But it's like the little like 28 gauge wire on that little probe lead. Something I'd love to hear from people, if people have confined cheap differential probes, I can not find those that are like under $400 It's crazy.

I found one gosh, it was well man, this is probably three or four years ago. It it's a DIY project. But this guy did actually a really fantastic job, I'm gonna have to dig it up again. Because you can just get the board's made. And and he made a he made a really nice active differential probe.

Oh, I'm gonna have to look at that, because I can't afford an isolated scope yet. But the probes I should be able to.

Yeah, especially if you build them yourself. Yeah. So you're working on a new version that has a microcontroller in it right of your device. So what's the microcontroller going to do?

So right now, I have

actually actually we're gonna take a slight backpedal. The fact that your current device doesn't have a microcontroller. And it's how does that work

analog for the win?

So actually, I because I'm my understanding of regulatories if you have to regulate firmware, it's just another step. I was trying to stay out of that as long as I could, and I'm an okay firmware designer. I'm not someone I'm like find a write it make it work, but I'll have someone else you know, look at it before I make it a product. And i i so I didn't want to do that yet. And I had to do all the timing and all the logic without it. So my original version was actually all relay logic, my original prototype. So all line voltage, three relays interacting with each other. That's it. And it would do like, basically a state machine, which was awesome. Except when it accidentally skipped a state right. So my version now is actually has a I have a few of my relays are 24 volt coils. And so I do some just 24 volt logic with MOSFETs. And so I think I have on my low voltage DC side, I have like 60 components that are just doing edge triggers and lots of different delays. Because if you think about how this thing works, I have my timing diagram has about nine steps for the for braking. And so I need to go through each of those, just with a bunch of RC timers on MOSFET.

I was about to ask is this all like accomplished with like, just 74 series logic and and timing RC timing circuits? Yeah,

it's super simple. I don't even have op amps or anything like it's all I have like two p channel MOSFET. The rest are ends and a bunch of little one microfarad Oh, eight oh fives and lots of magic and crossing your fingers.

Well, and you do get you do kind of get lucky. Because correct me if I'm wrong, but you get to rely on the fact that the mains frequency is 60 hertz. So you sort of get to like hang on to you hang your hat on that. You don't have to have a circuit that your circuit is designed around that,

well actually, I have complete isolation on the DC side. So all my the user interface basically takes a button and from that starts a sequence where it has two Opto coupled Triax and three relays that have to be switched on and danced in the right order. Otherwise, you short mains and that's a bad idea. And so it's basically this big long dance where I got to make sure that you know this timer triggers which Grounds this thing which starts the next timer, and on and on and on. So that's cool. But then the alterations get a little complicated when you got to do tweaks.

So you're gonna go you're gonna go forward and slap some smarts in it and have Have a think machine in it that will do this timing as opposed to RC circuits.

Exactly. And then I can be a current source instead of a voltage source, and then all becomes Wonderful.

So yeah, so the, I'm gonna guess what the new microcontroller version, you're gonna be able to sense the current when like, someone turns the device off and auto brake.

And it'll self calibrate by looking at the startup torque, or startup current. Ooh, fancy, yeah, lots of fun stuff can happen then. And, and so I was paying a lot of attention to your particle podcasts the other day, because I'm one of the things I'm getting from customers is that for industrial customers, they want to know what's going on with their equipment. So as an IoT device, I can then do access control, and tie into their little RFID badging system. I can predict blade changes, I can talk about who's using what tool for how long I can do reporting, if they're producing medical devices. And all with very, very simple logic,

you could also do what's really important for like, sanding and grinding is the load percentage of how much load you're putting on the motors. Because there's like an OP, I can't remember what percentages but there's a optimal load that you need to put on a like, let's say, a grinder, there's an optimal load that the motor is going to be good at running that and you'll get the most amount of work out of the adhesive or grinder or whatever your whatever tool you're using, like a like a flap wheel. Like sanding disc wheel, there's like a certain percentage load that you can get out of it, that you either load the motor up too hard, and it slows down too much. Or it's free spinning, free spinning too much. There's a percentage there.

See, I was thinking the motor would just like send an angry emoji to Twitter when it needed a new blade or Well, yeah, there's that too.

Well, what you could do is you could just pay your workers based off of how their machine is loaded. If they're in the optimal range, you give them like a bonus, you know,

and now now I'll be the union killing device.

Yeah, yeah. But I also I kind of I kind of see or foresee an idea of like, lockout tagout. You know, if you have a machine that you never want to start, you can, especially if it's IoT you could Just send it up never turned on device you know signal so

so that's that's in the in the near future but I'm overwhelmed with the amount of choice for microcontrollers right now

on on so when you're doing the current sensing how you doing the the isolation for your microcontroller, so that

was literally a call I had this morning. So the big thing I'm I have to decide right now is if I want to do isolation on the AC side or the DC side, because from UL or if I forget which standard it is, you can do either one so I can keep the microcontroller not isolated. And then just isolate the buttons and user interface. Or I can isolate on the AC side of a DAC or an ADC or I can isolate on the DC side of an ADC and I'm going through and like pricing all the different options to see what cost the list

is I've done isolation on the ADC side. So it was a DC isolation. I isolated the that the ADC itself

and just be a little like Opto coupler in the serial line or something. Yeah, basically,

I just Opto coupled Well, I did a it was magnetic coupled. Because I was doing it was I squared C magnet, a couple of some chips for that maximum makes, I think. But they're magnetically coupled into the optically coupled, I just chose them because it sounded cool. You could probably do optical coupled and be fine. But yeah, so I isolated on the DC side. But the ADC was isolated.

And I just learned today I didn't know this was a real thing. That little tiny isolation DC transformers are a thing that you can put on a data bus. Yeah,

I'm in little tiny power supplies that are like that, too. I think like I think the company's wreck calm that makes stuff like that, is it wreck calm.

There's there's actually a bunch of companies that make that in a previous life, I designed a temperature module for a large rackmount system, where power and data for all of the actual ADCs on the front end of this was entirely galvanically isolate isolated, it was actually like, we had a quarter of an inch of physical separation on our PCB for the front end of all, like the sensing elements and things away from the processor. And and we had to do both power and data across that galvanic isolation. And one of the things we implemented, or kind of just like the general vibe of everything was only isolate what you need to, because it actually costs a lot of money to isolate a lot. And when it comes down to, you know, maybe maybe the components are easy to isolate, but when it comes down to like, UL certifications and things like that, the galvanic proving process is is kind of stringent, and so only do what you have

to so then I should just do the two buttons is what you're saying. That

was good. Well, that was the that was the feeling that we had at a previous job. Some other people prefer to isolate the entirety of the whole thing. And that and that usually comes into performance, but if you but for your thing, like as a whole if the only thing that needs to be isolated is the buttons. I think that what sounds like a better idea and

then I mean I just get nervous What if what a voltage divider between mains and your ADC like that terrifies me?

It is kind of scary. Yeah. Especially when you look at things like switching power supplies that are just basically a capacitor and an inductor away from mains like yep, that's acceptable in some cases

so keeping going with the isolation is there any like physical board layouts to be compliant for that kind of stuff

and to do you guys follow Stephen

because Steven touched on that with like his galvanic golf go van. How do you pronounce that galvanic galvanic isolation, you had like a quarter inch of physical, I guess route out on your board,

we had an eight layer board. So we had four power planes and four signal planes. And I basically had this quarter inch like it looked like a river going through the middle of my PCB. And it snuck underneath these super wide chips that I had that were both data and power passing chips. And you could see all the way basically through the board because there there can't be anything in that channel you know? Yeah,

that's the the danger zone I'm in now. So do you guys watch the EV blog ever? One of the topic suits? Yeah, he like I his voice is a little hard on me sometimes, but but he has some good content and he did a lot of tear downs of multimeters. And so I actually learned a lot from from watching him and it kind of led me on the research path that had to the difference between creepy bid and clearance when doing board design. And so I gotta be careful cuz I got mains all over the place. And my new version is can do 110 or 220, which I thought was gonna be a trivial thing, you just double the space. Right? Right. And then but all of a sudden, you increase that space, and then all of a sudden, like, nothing can go in between two leads.

Yeah. How do you make this relay work when you when the space is supposed to be wider than the leads are?

Yeah, right. And so like, so all of a sudden, the routing got just way more complex. And I've been doing about cut outs. And so you do a little, you know, cut out in the board cut down on creepage. So now I got all these, these parts that go through where like, half the FR forest gone, because I'm having to cut out all these patterns. And then the all the traces kind of start parallel and then get away from each other as fast as they can. I actually had to go to two rounds on this last one, because it was so hard to route. And I couldn't have the width I had before on traces.

overcurrent care, caring, yeah. You thought about doing taking off the solder mask and then doing a, a wave on it.

So I, I feel like this might have been he vblog too. And I should verify this myself. But he did a semi empirical test of how much of a difference that actually makes with current carrying capacity, and found that it made very little, which was very surprising to me, which makes me want to test it myself. Because if you bought you know, Ollie Express is full of just hand soldered copper enhancement.

Well, the thing is, it would just be 10 instead of copper, and but it's thick. So it should lower it. That's

right. That's That's what I think, too,

you know, interesting that come that brings up a little bit of a can of worms. That little side tangent, but when you send a paste file off to a manufacturer, typically they're looking for apertures that fit over, you know, components, and things like that. But if it's totally valid to send a paste file to a manufacturer, where the paste file indicates that you want paste on a trace. And so that's sort of like a confusing gray zone, because a lot of a lot of manufacturers will look at that and be like, did they really want this? Was this a mistake? We're not really sure kind of thing. So I don't know, because because I I actually noticed we ran into that a handful of times at macro fab, where you'd see paste on traces. And it kind of holds everything because you like, what, what do we do? Do we contact the customer and be like, did you really want this? We're not entirely sure. Because we don't want to send you something that looks like a bunch of goopy solder.

Yeah, on that point. Because we are we saw that a lot. But also people will keep pace on their, their gold fingers. And for sure, they don't want that. And so yeah, it's one of those you always have to check it. And so at if your macro fab, if you put paste there, we're pretty much going to put paste there. It says, oh, yeah, it says so do it.

I guess I I, I lied a little bit, I do use it sparingly on like a to 220 package when I got to have high current coming out of it. I will I will move my solder mask back about a quarter inch until the trace widens. And so I don't have that. So I don't turn that little piece in refusing element.

Actually, in solder mask is a insulator. So it would actually will keep well from electricity, of course, but it will also keep heat in. So exposing the trace will help it radiate more heat. I don't know about how much but that's just something I read an app note a long, long time ago. Yeah, all

the calculators talk about, like internal versus external layer and I want them to have the option to say, like solder mask top layer.

I would think that the calculation would take that into account but who knows, maybe not.

If you did that you can do different colors, because different colors actually radiate red to radiate. Black Box radiation man.

Now we're starting to split hairs pretty hard.

Right, so last thing. So you mentioned that you were a teacher and you wanted to create a safety device to help you know, keep people safe. So what horror stories that you see, or in witness that was the creation of this device, if you're willing to share.

So yeah, I've actually been really lucky and Not really had any accidents on my watch with kids. I've seen, I think a better way to say is what are all the things since I never really learned from anyone. I learned from the like, Oh, let me see what that does for my entire life. And so most of it was me projecting mistakes I've made on two kids. But at the last school I was at, we worked really hard to make sure kids had access. So we'd have fifth graders that would direct supervision could operate a bandsaw. So I mean, we worked with kids of all ages. But uh, I would say for me, I've had just in my own life, so many close calls, that and especially as like a maker price similar to you guys, you work a lot of time by yourself late at night, in your garage, where, like, if you did something, you'd be on the floor for a day and a half before anyone even like asked for you. Right? Because they're like, Oh, they're just on a project. And, and I mean, I remember one of the worst ones for me was before I really knew what you could and couldn't buy at Harbor Freight. I was working on a boat trailer. And I had this huge Harbor Freight drill bit. I think the thing is like a three eighths drill bit, but I think the thing was, like 10 inches long. And I'm sitting there with my handrail trying to grind through like pry three sixteenths steel is a thing. And the thing bends over 90 degrees and spirals up my wrist. Oh, and I thought I'd slit my wrists open. And it turned out to be superficial, but I mean, barely. And so it was a pretty scary because I was out in the middle of nowhere. But some duct tape and paper towel fix it right. I was much younger than

Wow, that's incredible. Yeah, but

the the other thing is I've been going on the road a lot showing this to businesses. And there's a gentleman I was just installing a device at the San Diego fine woodworkers Association. They're one of the biggest woodworking associations in US

had three fingers

left. He more had, he had to like add fractions to figure out how many were left. I think he had tips. And he had talked, he was talking about how he'd been using a table software, 30 years. And just one day, right? He was a little less careful and got the tips of all his fingers cut off. And that's pretty common. Especially like, when I think about a table saw you get your piece and you got the little piece sitting there right next to the blade. And like everyone's looking at it, I probably shouldn't reach for that. And then they do. And so that I mean, it's those stories that make me think, Okay, I'm on the right track. I'm doing something that people need.

Yeah. Oh, yeah. Yeah. You bring up table saws, and I abuse the crap out of my table. So I think I cut aluminum on my table saw more than

you get the right blade. That's not too bad, though.

No, just a normal 24 teeth crosscut saw just normal one. Cuts Great.

One of the days it's gonna turn that into a missile for you.

I think I think table saws and drills are some of the scariest things that you can find in a shop anywhere because they're just like, they're just wide open death machines that are just like, well, if it's up to you, if you do it right or not.

And the Dremel is like the jump, right? Like you're in the middle of something thinking everything's right. And then all of a sudden, it's two inches from where it was a second ago. And you're like, yeah, yeah. 10,000 rpm.

Yeah, yeah. Yeah.

Yeah, I would say my right angle, like, five inch grinder is probably my most dangerous tool.

Because you get no guard on it,

do you? Oh, hell no. It gets in the way of every everything.

So first thing you take off?

That's the first thing Mike I actually I don't think any of my tools have their guards on there anymore. Not recommended do not do that. Kids. If you're listening to this podcast, keep the guards on.

Practicing this for legal reasons right now. You know, we actually had a SawStop at at a shop that I worked at for years. And that that was that was really cool. Although the thing, the only thing that sucks about safety devices, and this doesn't seem to apply to what, what you've created is that safety devices always seem to get in the way. Like, especially with the SawStop because I knew it was very expensive if I accidentally tripped it off. And and I wasn't, I wasn't worried about tripping off by you know, bodily harm I was more worried was like, Is my piece of wood wet for any reason? Or you know, you cannot cut aluminum on the soft stuff. But that's okay. Don't let Parker touch your table. So yeah, right. And it's just like, I don't want to make a mistake and be out 200 bucks to replace the cartridge, you know?

Yeah. And those things have you ever having to go off? They're terrifying.

I didn't but but my buddy, who actually he's the owner of that saw stuff. He was he was listening to music. And he was he was Cutting veneers from two by fours. So like eighth inch little strips and stuff like that. And he said he was cutting a piece he turned over to look at another piece look back and the blade was gone. And apparently had his music loud enough. They didn't hear it, but he was just like, wow, it's just gone.

It's like a it's like a gun. Yeah, yeah. I mean, I gotta tell another funny story about that then. So we had that the last place I worked. And I went through three blades in like, three blades and cartridges in two days. Oh, figure out what was happening because they were all false fires. And we just like we were going, we're just cutting up scrap, and we could not figure out what was wrong. Finally, we figured out someone had used the full sheet of plywood as a backer board when they were painting. And they were painting with whiteboard paint, which can contains metal flakes to make it magnetic. And there was just these little droplets around this piece of plywood. And if the blade happened to touch one of those droplets, it would go that took us so long to figure it out like $300 later.

Yeah, I would say just proper for those kinds of safety stuff is it's just knowing how to use your tool and being safe round tool. It's what what was it because I did a lot of oil and gas stuff, right when I came out college and the what the number one cause of injury is complacency and your workforce in so like every morning we would have a safety briefing and they would go over all the safety infractions that from the previous day and that's the first thing you did at seven o'clock in the morning before you had your morning coffee was go over all the safety stuff. And yeah, cuz that's that's a big deal. It's complacency is like you've used your table saw for 20 years. It's never hurt you and then at one time you do not pay attention it will get you

go talk to some roughnecks out on their oil rig and they'll they probably have some stories that tell you about some guy who wasn't watching a chain that was flying around or something like that. And he got his arm torn off or something. Sounds like he needs to make safe products. You see?

Good. Yeah. So we're Scott, where can people find your your stuff?

Yeah, so it depends which life you want to look at. So number one, if you want to look at safety products, I'd go to make safe tools comm you can see our first product get on our mailing list, and we have a few more coming out in the near future. We also have a pretty cool newsletter that goes out talking about some of the updates to the safety environment. If you're interested in kind of the makerspace side, you can check out grit lab.org. That's a educational consultancy I run or the documentary films most likely to succeed are beyond measure where you can see us making really cool stuff with kids. And then of course, there's a bunch of social media that I don't use as much as I should because I'm just barely not a millennial. But you can do Twitter at at s suavely or also on Facebook, but I'm really old school guys. So check out the websites you have unless you have a

MySpace then right.

I remember having one actually, I wonder if it's still there. So

you know, I'll inject this to you do have a YouTube and I'm saying that because I actually watched your YouTube and and so if you want to see your product in action, there are some videos up on YouTube. So I think there's a bench grinder and a rotary sander.

Yeah, absolutely. And if people are running maker spaces or educational spaces, I will make a sweet deal because that's a movement I'm very passionate about. So if you guys are interested in these kind of products and you want to keep some kids safe or teach people how to make things, email me stuffs on the website and works on that.

Also, thank you Scott for being a guest on the podcast.

That was the macro fab engineering podcast. I was your guest Scott suavely.

And we're your hosts Parker Dolman

and Steven Craig later everyone take it easy

thank you yes, you our listener for downloading our show. If you have a cool idea project or topic or cool thing that you design that you want Steven Knight to talk about. Tweet us at macro lab or email us at podcast at macro lab comm also check out our Slack channel. That's actually how we met Scott. If you have not subscribed to the podcast yet, click that subscribe button. That way you get the latest map episode right Warner releases and please review us wherever you listen as it helps the show stay visible and helps new listeners find us. And stay tuned for next week's podcast where we have a secret map thing going on