Transcript
johngrimsmo: Good morning and welcome to the Business of Machining episode 473. My name is John Grimsmo.
John Saunders: My name is John Saunders.
johngrimsmo: And this is where two good friends have been doing this for a decade. talking about our businesses and our trajectories and the goods and the bads and the ups and the downs and you know kind of the the brutal reality of business sometimes. So we tend to keep some of it.
johngrimsmo: personal and private but um It is so nice to have somebody to talk with. who's you know doing a similar thing.
John Saunders: It's funny because John and I... probably talk somewhere between one to three minutes privately before every podcast is that fair
johngrimsmo: Yep.
John Saunders: And today we talked for half an hour.
johngrimsmo: Yeah.
John Saunders: So sorry, you guys weren't privy to that, but it's also. um
johngrimsmo: Like that's what we need personally.
John Saunders: Yeah.
johngrimsmo: um and then
John Saunders: Yeah. yeah
johngrimsmo: Yeah, now we've got everybody going, what did they talk about? well
John Saunders: Yeah.
johngrimsmo: Yep. Um, What's on your list for today?
John Saunders: I need to explain the tolerance conversation we had with Saunders. I have a great leadership quote, a really good win with Muse AI. um Insanely incredible, fun update on Barfed Mill.
John Saunders: And we have a probe issue, which I was proud of a creative temporary solution.
johngrimsmo: Nice. It's exciting. All right. On my list. tolerance and surface triangulation,
John Saunders: okay
johngrimsmo: So excited for my interferometer. run out on our lathe, your favorite, and grinding ceramics.
John Saunders: oh
John Saunders: Let's start with the demise of the late because I just that just what's the appetite?
johngrimsmo: So on our saga pen. We have the tube in the middle and on the ends of each tube there's a tiny little groove that's like a little detail feature.
John Saunders: Okay.
johngrimsmo: um Because instead of having when you. Screw the tip and the tube together. There's the seam like the where the two parts go together, you can see ah ah a line, right? And we're basically adding a line on either side so you have three little features three little detail lines which It's subtle, but it's kind of cool.
John Saunders: Oh. Okay.
johngrimsmo: It's pretty important. and if there's any run out in the collet or in the bar that groove when you turn it which is only two thou deep.
John Saunders: Sure.
johngrimsmo: becomes you know two and a half thou and one and a half thou or like Three thousand zero.
John Saunders: Yep.
johngrimsmo: And that's been a problem. So the guys are trying to figure out you know clever ways to do it. Is it a tooling strategy issue? Can you get a tiny little grooving tool that goes in there and doesn't blow out some of the other features?
johngrimsmo: Or do you just have to get rid of the run out? and um I think the solution is we just... have to get rid of the run out like
John Saunders: Where's the runout?
johngrimsmo: And it's either in the call it or in the chuck itself or maybe in the material circularity, something like that. I'm not exactly sure yet, but it's a couple of different ways. And we do the same feature on the main spindle and then.
johngrimsmo: ah identical feature on the sub spindle
John Saunders: Yeah.
johngrimsmo: on the other side of the tube and they all have to match and there's a lot of things that have to work to make it um
John Saunders: Mm hmm.
johngrimsmo: to make it work so um It's one of those little aesthetic artistic features that I put in That's like so hard to maintain, but I'm like.
johngrimsmo: I want it though. like
John Saunders: no No, but I don't disagree. It's almost like the canary in the coal mine because it's like. Some customers may not notice. You certainly will. But it's also like, okay, it's that tell sign of like, what's going on here?
johngrimsmo: Yep.
John Saunders: what um This is the NOC with the Royal QG65s.
johngrimsmo: Yep.
John Saunders: Okay.
johngrimsmo: Or the um, it's 5c call. It's whatever it is
John Saunders: Okay, five, six, got it. Ciao.
johngrimsmo: but it is the royal dead length.
John Saunders: Yep.
johngrimsmo: um call it system.
John Saunders: Okay.
johngrimsmo: And then even like how good of collets are we using? in the machine are we using just the cheap ones that are no name or using harding's ones we're using Special ones. I learned at, might've mentioned this already, but when I was at IMTS, I stopped by the Southwick Meister booth.
johngrimsmo: They make collets. They have forever. They're like in Connecticut.
John Saunders: Mm hmm.
johngrimsmo: and they had to call it with instead of straight slits they had zigzag slits that are wire edm cut And so I was like, okay, guys, tell me everything about it.
John Saunders: Hmm.
johngrimsmo: I asked them. at Southwick. And I was like, they said the zigzag is to help eliminate linear scratches in your bars as you're pulling it through.
John Saunders: Love it.
johngrimsmo: And for our saga tubes, we're buying 3750. ground stock.
John Saunders: who
johngrimsmo: with like a really nice grind i specify a beautiful grind good tolerance no gouges no nothing and we're not turning the od on the pens where we're relying on the grind company to do it So any scratches they put in, any scratches we put in.
John Saunders: mm -hmm
johngrimsmo: is a problem for us that we have to then sand out. um So like zigzag. slits in the 5C call, it might be a really good thing for us. but they quoted me 230 per call it as opposed to a 5c call it's usually like 40 bucks and still deciding if that's worth the squeeze but
John Saunders: I love hearing that from you because I've had the same debate in my head. And then you need to remind yourself that is like one saga break even.
johngrimsmo: Yeah.
John Saunders: Now, it doesn't mean...
johngrimsmo: and And something that... pays dividends like forever.
John Saunders: Yeah, exactly. Now, it doesn't mean you can say yes to everything that comes up like that, but I...
johngrimsmo: the call it lasts for years. Totally.
John Saunders: um I do think some of our frugality gets in our way sometimes.
johngrimsmo: It does.
John Saunders: But also are you having linear streaking that probably has nothing to do with the unless the college worn out do do
johngrimsmo: With run out. Yeah Nothing to do with run out, but it does.
John Saunders: Five C's wear out. i guess
johngrimsmo: Um. I don't know.
John Saunders: I mean, everything wears out, I guess. but
johngrimsmo: Yeah.
John Saunders: Seems odd.
johngrimsmo: Yeah.
John Saunders: Hmm.
johngrimsmo: it depends on what you're doing what you're cutting if you ever bump it if you ever like You know?
John Saunders: you Are there any other parts in the NAC that are would show if there's a, like, what if there's really backlash or a problem in the machine?
johngrimsmo: Think we've replaced them
johngrimsmo: There's definitely backlash that. It shows in different areas. it It might show here in a weird way but not in the way that's causing the problem. um Yeah.
John Saunders: Okay. What's the next step on it?
johngrimsmo: Um... I... I've said my suggestions this morning. so we just got to get rid of the run out i guess the next step is jeff like get rid of the run out
John Saunders: Okay, somebody else is working on this though.
johngrimsmo: Yeah, totally.
John Saunders: Yeah, sweet.
johngrimsmo: But I'm making sure I step in when I need to and like let them run.
John Saunders: Cool.
johngrimsmo: otherwise which is good.
John Saunders: Yeah.
johngrimsmo: I have more than enough other challenges to tackle myself, but that's good.
John Saunders: yeah
johngrimsmo: I want to make sure everybody's strumming.
John Saunders: um I appreciate everyone's patience. I have a new microphone. Hopefully that comes through.
johngrimsmo: It sounds really nice.
John Saunders: And so... in so many ways, and hopefully this microphone itself.
johngrimsmo: Mm -hmm.
John Saunders: It's the, I think, very prolific podcast, Sure SM. or not sure. mv7 plus so I'll leave it at that. If anyone wants to throw in feedback, good or bad, I'll welcome it. The bigger thing I want to share is we had a really good, probably six week internal conversation about tolerances.
johngrimsmo: you
John Saunders: And. Alex. runs our engineering, he sets. He has... full ownership of this process. especially as it relates to what our internal.
John Saunders: dimensional drawing PDFs are and in talking to the machinists and so forth. um So here is. Okay. there There were some examples that I won't get into that because they're just boring about how to use spec.
John Saunders: parallelism. on a fixture plate and is it across the whole plate or is it across certain sections and you can look at industry terms that's boring. Here's the question I wanted to... raise. Let's say we have a tolerance. And just again, for the sake of this example, let's say it's half a thousandth.
John Saunders: total has to be within half a thousandth of an inch.
johngrimsmo: over a 30 inch fixture blade or something.
John Saunders: Okay. Doesn't matter. Sure.
johngrimsmo: Okay.
John Saunders: we we I always use VF2 plates. they're popular sellers and it's a good example. 40 inch by 16 inch plate. And let's say we ah we want the whole thing to be within half a thousandth of an inch. And let's say we nailed it. And it's like within two to three tenths, like it's very much.
John Saunders: a great plate. but on a very, very small section and usually this sort of thing happens to be on a corner. of the plate due to lead in or or extreme whatever mac let's say it's one point or let's say it's 0 .8 thou So. Our tolerance is half a thou, and it's 0 .0310s outside of our tolerance. But it's an area that, say...
John Saunders: a postage stamp or even smaller. Um, By definition, That plate fails QC. Now, the way this tends to happen tends to be thinner. And so, and I say that because usually um that's okay in terms of functionality. So if it were thicker and it would then cause the rest of the plate to then lose its flatness, that would be a real issue. And we wouldn't really argue about that. But my argument is, or the question I raised is.
John Saunders: if it's if it's effectively immaterial. And it's... um relatively small. These are all subjective. and here's the real ringer, and reworking the plate. probably actually reduces the quality of the end result product like the whole thing might be within tolerance but it might be now twice there the the deviation.
John Saunders: um And it was really fun to go through this process and think through it and and argue about it, debate about it, both from a... because we want a process to follow we want to make good decisions there is a business aspect to it know do we want to really rework plates do we want to sell them um
johngrimsmo: As blams are like, yeah yeah
John Saunders: Okay. Yeah. And and we we have, I actually need to get like the final. summary. We need to, we need to process size this in, but basically we.
John Saunders: Um, reached a point where the team and QC can make decisions about a scenario like that. Oh, here's the other point that I sort of enjoyed, and it's a little bit of a...
John Saunders: pedantic point but If you took a... one thousandth of a ah thick blade mic. in measured parallelism, you're going to get a really different result than if you took a micrometer that had a two or three inch.
John Saunders: anvil size right so there's some intrinsic averaging in how hand tools are used
johngrimsmo: Yeah.
johngrimsmo: Mm hmm.
John Saunders: um That being said we take a lot of pride in our quality and and customer service and is the practical aspect of not wanting to. reship. heavy plates frayed across country if if a customer finds something that's unacceptable and we're proud to say we really have had an excellent track record on quality customer service.
johngrimsmo: Yeah.
John Saunders: so the punch line is is the compromise that we we kind of expected. But what I didn't anticipate was a better framework for it to work in. so now We have the ability for QC.
John Saunders: the QC team themselves to acknowledge a deviation like that. And we'll probably end up selling them as like a class B plate.
johngrimsmo: There you go
John Saunders: Kind of like what like Starrett does, you know.
johngrimsmo: Yep. Gauge blocks, everything.
John Saunders: Bing up.
johngrimsmo: Yeah. Okay, that's a good way to put it.
John Saunders: Because it's not a blam, and we don't want to like make customers think it's a blemish, but it's also 100 % not the same.
johngrimsmo: Hmm.
John Saunders: value and quality and offering that we want to stand by on our normal product.
johngrimsmo: Yeah.
John Saunders: Um,
johngrimsmo: it's sort of like the like the rego fix up call it's the ultra precision call it's like Somebody told me they just make all the collets and they inspect some of them and sell the good inspected ones as EOPs.
John Saunders: And that also...
John Saunders: yeah It's funny. Yeah.
johngrimsmo: you know.
John Saunders: The other one I love that I heard is that Swarovski is actually like a optics manufacturer for the industrial world of like lasers and in scopes and that all the. extra glass they just turn into tchotchkes for the airport
johngrimsmo: Is that true?
John Saunders: I have no idea if that's true.
johngrimsmo: I don't know.
John Saunders: Yeah. um
johngrimsmo: That's funny though
John Saunders: so it's a pretty So that was the simple beauty answer is like that gives the team a framework to work around that doesn't involve. alex or another corporate manager going in taking time and doing an actual override it allows us to communicate to the customer that they're buying something that is different
johngrimsmo: Right.
johngrimsmo: Hmm.
John Saunders: um And then. That we still have a process where. If there's truly a flaw, and this happens, you know we've had plates where a tap will break. And so one out of a thousand holes is compromised.
John Saunders: That's probably something that we would truly mark as a blam.
johngrimsmo: Yep.
John Saunders: Many customers are happy to exchange.
johngrimsmo: Yeah.
John Saunders: a buy that at a lesser price because they're like a good grief you actually can still use the threaded hole it just doesn't look as good like you get the idea um So it was a really good conversation that I'm glad that we tackled because it wasn't just like.
johngrimsmo: Hmm. Yep.
John Saunders: the whole like well no one will really catch it so Once it's mounted, you're not going to pinch mic it.
johngrimsmo: Yeah.
John Saunders: So let's just ship it.
johngrimsmo: Yeah. And the problem is you and and us to a certain degree. You're selling to machinists, 100 % of your products go to machinists.
John Saunders: Yeah. All right.
johngrimsmo: a large percentage of our products go to Machinist as well, which is fantastic. um but we have a very diverse so discerning customer base.
John Saunders: so
johngrimsmo: especially your customers.
John Saunders: Tell me no one's miked your thing, though.
johngrimsmo: like
John Saunders: but
johngrimsmo: Oh, people have definitely miked our... Is...
John Saunders: That's funny.
johngrimsmo: I saw somebody, you know, there's the quantum mic and then there's like the super quantum mic, the big fatter one that measures down to millions.
John Saunders: Mm hmm.
johngrimsmo: Somebody miked a saga tube and they're like three, seven, five, zero, zero, zero. All right.
John Saunders: That's funny. wonder if that's ah I wonder if that's the go -to over at Danny's Dow factory.
johngrimsmo: Yeah, probably. He does have one.
John Saunders: Have you ever seen... thermal he does right because if the thermal it looks like
johngrimsmo: Yep.
John Saunders: It looks awkwardly chunky.
johngrimsmo: yeah Yep, yep.
John Saunders: That's funny. Can I change topic real quick?
johngrimsmo: Mm hmm.
John Saunders: Part of what you and i are talking about before you hit record. I think tied really well into this I was reading ah blurb. like a Saturday blurb from a guy named Scott Kirby, who used to be the CEO of. American Airlines, or excuse me, used to be high up in American.
John Saunders: one of the head role 10 years ago didn't get it and left and now runs united and is the ceo or maybe it's vice versa whatever it doesn't matter He did one of these like Saturday talks about.
John Saunders: his life as a leader and in everything from AI and his job and so forth. um And he had a quote in there that um made me think of you and our discussions in this podcast.
John Saunders: people confuse management and leadership. Management is about compliance. Leadership is about inspiration and vision. As a CEO, if you're making too many decisions, you have to make all the decisions and everything gets punted to you. My job is really to make five decisions a year. the big decisions with no clear right or wrong answer that only the CEO can make.
johngrimsmo: Whoa.
johngrimsmo: When you were saying five decisions, I was like, I thought he's gonna say a day Like a year?
John Saunders: but
johngrimsmo: Holy cow.
John Saunders: But that's what without...
johngrimsmo: ah Steve. Yep.
John Saunders: oversharing it's like you know when you and I talk about the team and our company and our machines and our product line. That's the CEO's job. That's your job, John.
johngrimsmo: Yeah. Yep.
John Saunders: I also love. that you and i both do our uh, interferometer and similar type projects but like What is that? it's interesting when you think about you and I are trying to really do three things.
John Saunders: That's probably stupid. We're trying to be leaders. We're trying to be managers and we're trying to be technicians. This is like. 100 % goes right back to email.
johngrimsmo: Yep. yo Yep, yep. and I think I'm a lot more picky about which of those I'm taking on. uh in the current business with 12 employees and like we're I can be picky about which of those I do, but I don't.
johngrimsmo: You know, I'm still not maximizing my usefulness in in the ones where I belong. And I'm still being the technician when I. I spent the entire day yesterday staring at toolpaths. but I'm like. Pretty good at that.
johngrimsmo: and this is the higher end tool path that i'm that i'm working on but um
John Saunders: Yeah.
johngrimsmo: Yeah, that's good.
John Saunders: It's just simple, the simplest thing. And I don't know if I'm just forgetting my own past, but like, I don't remember somebody explaining it so simply, the idea of like management versus leadership.
johngrimsmo: Say that again, management is compliance.
John Saunders: management Management is about compliance. Leadership is about inspiration and vision. CEOs shouldn't be making too many decisions.
johngrimsmo: Yep.
John Saunders: And I see that, talking very openly, when I'm not here even for... There's been some times where I've not been here for weeks. everything like i'm not bsing everything runs great you know alex make decisions If Yvonne's here, she can help out.
johngrimsmo: Yep.
John Saunders: like The team makes decisions. I get. relatively few messages. like it's It's a very proud moment.
johngrimsmo: Mm hmm.
John Saunders: When I'm back, there are many times where I'm like... Oh my gosh, every 15 seconds. i've got a phone call an email a dm a question And it's, I've got to be careful and look for something.
John Saunders: Some of our employees listen to this podcast. I want to be like.
johngrimsmo: Yeah.
John Saunders: very balancing like truthful and kind old but also not being rude but like sometimes like i want to have that open door policy and talk i've also tried to say hey like let's make sure we're batching up like if the questions don't need to answer let's batch them up we can sit down and so forth and avoid the staccato interruptions.
johngrimsmo: Yeah.
johngrimsmo: It's so easy for, oh, you know John's in. Let me go ask him. right Right now, because it's on my mind right now. Yeah, I'm a big fan of the matching too. then make a list, write it down. If you don't need it immediately. but but you want it tackled, like text me. Make a list, like bring it.
John Saunders: And I tried.
johngrimsmo: I don't.
John Saunders: I try to do it.
johngrimsmo: all right
John Saunders: Sorry. i
johngrimsmo: What? i'm very conscious of my focus these this past year or two, which has been very interesting.
John Saunders: Okay. Yeah. Husser.
johngrimsmo: Um...
johngrimsmo: Even to the point of like... ignoring people that are standing next to me waiting to talk to me while I finish the thing.
John Saunders: Oh, I do that all the time. in a
johngrimsmo: I know.
John Saunders: Like, just I hold my finger up at it because I'm like, nope.
johngrimsmo: Yeah. Exactly. um What were you going to say before?
John Saunders: I try to make sure. You know. I, as the boss, can have every right to have a double standard, but it's not a cool thing to do.
johngrimsmo: Yeah.
John Saunders: So, like, I'm trying to make sure I don't do the same. And, like, sometimes I do have to lean on Alex a bunch or Yvonne or anybody, actually everybody, seriously. um But I also try to remind myself, hey.
johngrimsmo: Right, right Okay.
John Saunders: you know. that can wait to our sometimes it's funny sometimes you're like that can wait to our weekly meeting or at the end of the day or combine like batch it up um
johngrimsmo: Mhm. Okay.
John Saunders: Sometimes I'm in like, hey, block and tackle mode. I need to get this off my plate. Let the team go do their deal. I don't know.
johngrimsmo: Yep.
John Saunders: um
johngrimsmo: Yep Yeah, you try to lead.
John Saunders: But that's all. Sorry.
johngrimsmo: Try it. lead by example and not interrupt people like you don't want to be interrupted as well. um But on the other hand, you know Stuff's got to get done.
John Saunders: is it the same It's funny, I just have no idea what a true day in life is like for you. You get in. Are you? Do hours go by when people don't talk to you?
johngrimsmo: Um... Rare, but sometimes. Yep.
John Saunders: Yeah. Okay.
johngrimsmo: especially if I'm in the zone. um Like yesterday I was at my computer on the shop floor. just fine tuning some tool paths back and forth. And then, you know, questions would fly in and out.
John Saunders: Yes. Yeah.
johngrimsmo: and But for the most part, it wasn't too bothered. Or I'm... gone deep with a couple employees for like an hour or two. um Figure something out plan something things like that, but it's kind of all over the place
John Saunders: Yeah.
johngrimsmo: um But rarely do I like... i don't I don't do production anymore. Um, Which is cool because the rest of the team handles that I Set up stuff.
John Saunders: Yeah. Yeah.
johngrimsmo: I fix stuff. I break stuff. I design new things And yeah, and that's where I'm best at is like problem solving.
John Saunders: Yeah.
John Saunders: Cool. All right, next on your list.
johngrimsmo: home
johngrimsmo: Yesterday, I would have used the interferometer 12 times.
John Saunders: ah Well, it looked like last week it looked like it was close to working.
johngrimsmo: and And not just me. like Yeah, yeah Not just me, but like three other people on the team. would have used it yesterday as well, which is I'm so excited. Yeah, the the first version functions, but only on flat things.
John Saunders: Oh, that's right.
johngrimsmo: Um,
John Saunders: Okay.
johngrimsmo: and obviously need round things dude i i saw on zygo's instagram they posted a halloween thing they Use the white light interferometer to scan pumpkin skin.
johngrimsmo: inside and outside.
John Saunders: What?
johngrimsmo: And I was like, I didn't know you could do something so matte. like I thought it needed some reflectivity.
John Saunders: Oh. machine.
johngrimsmo: Which, because you're... you're relying on the light to bounce back off the surface and interfere on the beam splitter. And like it has to bounce back.
John Saunders: Okay.
johngrimsmo: And if the thing is like Vantablack, like nothing.
John Saunders: Yeah.
johngrimsmo: you know it won't work and I was like Pumpkin skin really that's great because I was wondering if you could scan a piece of paper like that's kind of the ultimate craziness to me and i think it is theoretically possible but
John Saunders: Huh.
John Saunders: Yeah.
johngrimsmo: i don't know We'll see how my system comes together. Obviously, I'm really only scanning shiny metal things. but
John Saunders: right
johngrimsmo: um there's they're shiny and then there's dull metal. and if i can scan dull metal i'm flying.
John Saunders: Hmm.
johngrimsmo: Um, So yeah. So whether it's surface finish or... form of a micro feature. um like this little groove on the end of the saga tubes Um,
John Saunders: Yeah. Yeah.
johngrimsmo: That would be super helpful. And. Yeah.
John Saunders: Is it just... is software though to go from flat to curve.
johngrimsmo: um No, it's it's mostly all lens.
John Saunders: Oh.
johngrimsmo: um the numerical aperture of my lenses right now. because it's such a big viewing angle. is like very... Narrow, I guess Um, and the microscope objectives that I bought for this that came in.
johngrimsmo: um have a wider numerical aperture so they can like take in more light and see more angles.
John Saunders: okay Okay.
johngrimsmo: basically Because mine... That's why it only scans flat things is because the light bounces straight back up into the lens. The lens can't take the scattered light that goes off to the side.
John Saunders: It's telecentric lenses though.
johngrimsmo: it is yeah well the my old one is but the new ones are not
John Saunders: Oh, interesting.
johngrimsmo: Yeah.
John Saunders: Huh.
johngrimsmo: I know that's kind of where I got it wrong too, is that. most microscopes don't need telecentric lenses. Like for a vision system, you do.
John Saunders: Interesting. Yeah, sure.
johngrimsmo: but I'm not building a vision system. I am, but
John Saunders: Yeah, sure.
johngrimsmo: Not for the interferometer. um And. if the First of all, the lenses that I got, the Mitutoyo lenses are perfectly formed to be like, not have much sphere collaboration. around the outside. So like dish or whatever but i also brought bought a calibration Um...
johngrimsmo: test pattern from Thorlabs. that like basically puts a grid pattern on the entire field of view. of the image.
John Saunders: Like a checkerboard?
johngrimsmo: Yeah, basically like a checkerboard
John Saunders: Yeah, I bought one as well, I think.
johngrimsmo: Yeah, and then and then
John Saunders: Yeah.
johngrimsmo: you can just use software to like error map the whole field of view uh for that lens so like That's why.
John Saunders: Yeah. Yeah.
johngrimsmo: But yeah, so. Got an email from Thorlab saying that all my stuff should be ready to ship tomorrow, so I should have it shortly.
John Saunders: ah Okay figure it
johngrimsmo: And then big things left. There's definitely some software tweaks, but I can do that as I go. there's some machining i need to do to build the like the frame and how all this stuff's gonna bolt together um I need a... a heavy steel base.
johngrimsmo: that rides on it. inner tube for vibration isolation.
John Saunders: Mm hmm.
johngrimsmo: and then how i mount the motorized z -axis column and how i mount the optics to that i need to like finalize that and then The first one, I built everything on a breadboard electronics wise.
John Saunders: Yeah.
johngrimsmo: And I'm going to like use cursor and either fusion electronics or.
John Saunders: Right.
johngrimsmo: key cat or something like that to uh to design a circuit board. and need to put my logo on the circuit board and basically combine circuit boards that I already have. breadboard you know pinouts and resistors and all this stuff and i'm just going to send it out to jlc pcb and in four days i'll have
John Saunders: Yep.
johngrimsmo: five new ones. and put that all together. It's gonna be sick.
John Saunders: That's such a great use because... All of the MCP fusion stuff is kind of silly for you and me because we can do it. And like, i but I have no business doing like Eagle. board layout.
johngrimsmo: Yeah.
John Saunders: but if you have the schematic or what you've made and like let ai do its thing and then you take the finished output put into a different ai be like what's wrong with this board interrogate it like and then for $13, you have a sample done.
John Saunders: It's like, holy cow.
johngrimsmo: Yep. Yeah.
John Saunders: That's awesome.
johngrimsmo: That's going to be great. And so it's all coming together. um I do have a goal of having it running by the end of November.
John Saunders: Awesome.
johngrimsmo: So.
John Saunders: That's what I was forgetting was you it wasn't that you didn't have time to focus on yours waiting for parts right now
johngrimsmo: Yeah, I'm waiting for parts and I haven't worked on it in the past week or so because I'm There's stuff I could work on for sure. but doing other stuff. So yeah, super excited, super, super excited.
John Saunders: We're having a... um sort of an internal debate of. this is going to sound like a random. point of buying one of the have you seen the laser welders that are now more common and prolific for
johngrimsmo: Yeah.
John Saunders: you know two to five grand that are
johngrimsmo: Yep.
John Saunders: insane. It makes you look like you're like a TIG welding dime slayer.
johngrimsmo: Yeah.
John Saunders: um actually had the chance to use one Which. Even doing it in person, you're just like, oh my God, this is crazy. um And. You and I have talked over the years about like, hey, stay in your lane, do what you need to do, like stay focused. And like, we've gotten rid of so much stuff and I love that we're focused on what we do. And so part of this would be.
John Saunders: I'm more okay with this if we... had the chance to kind of grow our physical footprint a bit which we might um but The other really big thesis behind justifying, I mean, it's a relatively small purchase for us these days, which I'm. It's nice to be able to say that. is Senko Send.
John Saunders: who will continue to crush it on providing sheet metal. And I don't think we'll ever touch welding. Because it just seems like welding now gets into lots of like complications and labor and assemblies.
John Saunders: and in in in
johngrimsmo: Mm hmm.
John Saunders: distortion and things bending so it's like if you can order relatively inexpensive stuff from Send Cut Send. It's all different thicknesses and shapes and profiles. It shows up and then some of it you want to...
John Saunders: obviously it's nice if you can just fasten and screw stuff together but some of it's like no you need to weld that If we could then.
johngrimsmo: Yeah.
John Saunders: either weld it or even tack it and then just take it to a weld shop that can like lay the major reinforcement stuff down on. I'm just. Chewing on it.
johngrimsmo: Yeah. and and whether it be production based like parts you're going to sell or racking. Or like.
John Saunders: it's Yeah.
johngrimsmo: you robot load ro stations or something like that. And that's where both the Senkutsen side and also the welding like really come together. as to... are
John Saunders: Yeah.
johngrimsmo: Important stuff, yeah.
John Saunders: We're doing, I mean, for the robots, automation, the infees, the outfeeds, the regrips, even how we've... like up -armored our table with the curtain vices. Like even I was really happy when the lights out guys like that's like legit.
johngrimsmo: Heh heh heh.
John Saunders: Because it's like instead of having cleaning as a maintenance task. like let's just spend a thousand dollars to get sheet metal That is just. sops the chips from getting in the current vices once because that's just like this is what we're doing now like
johngrimsmo: Yeah.
John Saunders: Yep.
johngrimsmo: that's that's the side of automation people don't really talk about but that's the fun like exciting part where it's like you know like you say hardening the process of it just works like
John Saunders: Yep.
John Saunders: Mm hmm.
johngrimsmo: you know the point of automation is to just work but oh it's not that easy
John Saunders: Yeah.
johngrimsmo: Like um I was talking with. um Phil from Send Cut Send.
John Saunders: Yep.
johngrimsmo: um Super smart guy. And he said... They've basically blacked out the windows on some of their brother machines because they're like I don't need the operator to look inside like the Robot's just going to load it.
John Saunders: So funny.
johngrimsmo: and parts are going to come out.
John Saunders: Yeah.
johngrimsmo: I don't need the operator even curious about what's going on in there.
John Saunders: Crazy.
johngrimsmo: Like, what?
John Saunders: It's so funny, yeah. Yeah.
John Saunders: That was our silly. We're having a little bit of frustrations with with our probing and Yamazin. and the brothers um I think it may have just gotten fixed before we hit record. but um we don't have a probe on one of our machines right now and we need that machine to be running and so we We could run it without proballidations, but...
John Saunders: it's also like hey you've been down this road enough to know that it's that's when the park gets misloaded and bad things happen and
johngrimsmo: Yeah.
John Saunders: You and I know. It's like, well, you then you damaged a fixture. You broke a bunch of tools. It's just stupid. so um i we talked about like okay do our fishbone diagram here. What are we trying to stop? We're trying to stop the machine from running when the part is misloaded. We actually can do that in fusion.
John Saunders: pretty darn easily. We're fortunate that we have a dummy. We have an empty tool pocket on this machine right now. We threw we grabbed a broken... uh drill i got a worn out drill from the carbide scrap bin And now fusion comes over. and it Bip. It a Z plunge on four sides of the stock.
John Saunders: and to make sure this because if the stock's there it'll break the drill bit um so there's
johngrimsmo: Oh.
John Saunders: Yeah, basically you walk basically you take the drill around the part and then you take the drill over the part by like five thou.
johngrimsmo: Okay.
John Saunders: and and then you do a tool break detect.
johngrimsmo: No way.
John Saunders: Right.
johngrimsmo: Okay.
John Saunders: Easy.
johngrimsmo: I like it
John Saunders: And like I don't care if it breaks don't care if it breaks the broken drill again and I don't care if the machine stops tonight at 8 o 'clock.
johngrimsmo: Yeah, yeah
John Saunders: I just don't want to have it Run if it's misloaded.
johngrimsmo: That's clever. I like that.
John Saunders: Thank you.
johngrimsmo: Yeah, here I was thinking you're going to use like axis load feedback monitoring and like literally bump it or something like that.
John Saunders: Go.
johngrimsmo: I'm like, no, okay.
John Saunders: It will. When I was like... block and tackle it, solve this problem. Like go grab a chopsticks from the kitchen drawer. I'm like, actually wood is terrible. like yeah it soaks up like we've broken carbide or worn out there you go
johngrimsmo: Yeah, it'll ban that, yeah.
johngrimsmo: What sort of diameter must be smaller?
John Saunders: i Grant took it.
johngrimsmo: Like... Like. gotta be under quarter inch for it to actually like break
John Saunders: Carbide's pretty brittle. um the Yeah, the thinner the better for here, for sure, yeah.
johngrimsmo: Yeah. That's cool. And you guys have. just the Z tool set of probes. Like like the up and down that's all it does
John Saunders: Correct, you have the bluesy nanos.
johngrimsmo: Okay. Yay.
John Saunders: mm -hmm Mm hmm.
johngrimsmo: Cool. boy, I gotta say speaking of um Since I upgraded the air. filter quality on my Speedy o li Bloom laser.
John Saunders: Oh yeah.
johngrimsmo: Zero problems.
John Saunders: Good.
johngrimsmo: I haven't even thought about it for like months now.
John Saunders: That's great.
johngrimsmo: And I'm like, hmm. Interesting.
John Saunders: Yeah, right.
johngrimsmo: That used to plague us. i gotta go tell the guys like
John Saunders: Yeah.
johngrimsmo: It's been good, right? Like, oh yeah.
John Saunders: Mm hmm.
johngrimsmo: Surprise, surprise, clean filtered air. on a optics system. Like, reduces all problems. Wow.
John Saunders: Yeah.
johngrimsmo: That's right.
John Saunders: Well, good. You learned.
johngrimsmo: I actually bought ah three more of those filter two more of those filter setups and we're going to put them on the air blast guns by the two critical sinks where we're like. cleaning and drying off parts.
John Saunders: Uh, Great call.
johngrimsmo: And... you know, we're definitely spraying. Did not. perfectly clean air on these parts that are getting cleaned. dried shipped off to dlc coding where they're supposed to clean them themselves but like the cleaner parts we can give them the better.
John Saunders: Yeah. All right.
johngrimsmo: And then, you know, we're blowing off parts before anodizing, but if you're blowing. moisture, dust, whatever.
John Saunders: Oil.
johngrimsmo: Oil. um you know, maybe you don't know if it's a problem, but all of a sudden it gets better with cleaner air. So the guys are putting together those those air filter sets right now. and It'll be good.
John Saunders: Whatever came of the McMaster... What was it, John? Was it wipes or a pen that you bought that could tell?
johngrimsmo: Mm, the dine. pen or i'm not It comes in a pen, but don't buy the pen.
John Saunders: Yeah.
johngrimsmo: So I bought the little jars. I have the liquid. I haven't cracked it open yet.
John Saunders: Okay.
johngrimsmo: um The SDS is a little scary when you read it.
John Saunders: Oh. Interesting.
johngrimsmo: um and And Angela especially is making sure we're very like. aware of sds readouts things like that um
John Saunders: Sure.
johngrimsmo: Haven't tried them yet, but absolutely will soon. And it is the perfect way to test this air quality issue.
John Saunders: Mm hmm.
johngrimsmo: Basically take... A clean. glass slide or like a clean anything. um It tried with the dying fluid. and you'll see if it if it either wets out or if it breaks into beads
John Saunders: That's what it was. Okay.
johngrimsmo: And then, you know, get it dirty, try it again. clean it in your normal cleaning procedure. Try it again and you'll see how clean your cleaning procedure actually is.
John Saunders: Okay.
johngrimsmo: Um, So I have it. I have two different levels. of uh dying. Dead. tolerance, whatever you call it.
John Saunders: Yeah, what do you call this? Oh, yeah.
johngrimsmo: It's...
John Saunders: Purity check.
johngrimsmo: I don't know what the term would be it's a level dine level done
John Saunders: D, Y, and E.
johngrimsmo: Amount. D -Y -N -E, yeah.
John Saunders: Okay, yeah. Hmm.
johngrimsmo: and super fascinating stuff. So yeah, i'll keep I'll keep you informed once I eventually try them. But they're like on the receiving table. They're out. I walk by them five times a day. They're there.
John Saunders: Yeah.
johngrimsmo: Nobody's forgotten. button.
John Saunders: I'm not even trying to give you a hard time. I'm just laughing because I'm like, oh, that was such a fun subject.
johngrimsmo: No, for sure.
John Saunders: I'm like, wait, whatever.
johngrimsmo: I know, I know And, and, i tend to do that i'm like I do the full research. I buy the thing and then I never touch it because it's like a preparatory purchase. It's like, I absolutely will need that one day.
John Saunders: Yeah.
johngrimsmo: Today's not the day.
John Saunders: ah Yeah. Yeah. It's funny.
johngrimsmo: Um, Grinding ceramics. so On our knife the detent, which is what holds the blade closed.
John Saunders: Mm hmm.
johngrimsmo: And so you get that perfect flip. It's a ceramic bearing ball.
John Saunders: Okay.
johngrimsmo: 1 16th diameter grade 3 which is the tightest tolerance around this spherical licity that's like humanly possible And. what we do is we have the current push it Push it into the hole.
John Saunders: mm -hmm
johngrimsmo: drop it into the hole with our um baller installer made by uh made by wells brent wells And it's fantastic. So basically we press the ball in.
johngrimsmo: And now it's in. And we press it to the exact right depth. We probe check, probe check.
John Saunders: mm -hmm
johngrimsmo: I want the stick out to be an exact within a tenth amount.
John Saunders: Mm hmm.
johngrimsmo: And we hold that every time. and then i want to grind it flat i want to grind
John Saunders: mm -hmm
johngrimsmo: basically 10 ,000 off the... diameter. and have a big flat on there. So we've got a diamond grinding wheel that was custom made. And the problem is it tends to chip out the ball. sometimes.
John Saunders: Oh interesting.
johngrimsmo: And. Weirdly... It does it it does it more way more on the Norseman than on the Rask. It rarely happens on the Rask. It does it more often on the Norseman. And the codes are similar, but they're they're very complicated handwritten.
johngrimsmo: like step down, grinding codes, probe, check, probe.
John Saunders: Mm hmm. mm -hmm
johngrimsmo: variable, dress the wheel every so many minutes. I made it like five or six years ago and I went. kind of full nerd john on making this code and even i'm reading it now going what are you doing? Like that's.
johngrimsmo: That's silly. I would do it differently these days. But it was interesting because the two codes are like. Rask and Norseman, they're just different enough that I don't really really want to sit there line by line and be like, what's different between the two codes?
johngrimsmo: So I loaded them into cursor and I said, what's different between these two codes and it made me a nice little printout and it said uh you got the same feed same rpm slightly different step downs between the two that's interesting um
John Saunders: Okay. Yeah. yeah
johngrimsmo: And just sort of. got me going down the rabbit hole of like why is it chipping out what am i doing um A couple of things that really stood out are. My step downs, we'll use microns for for talking here.
johngrimsmo: um my My grit size is 80 micron. So every diamond in size is 80 micron.
John Saunders: Whoa.
johngrimsmo: which is fairly fine. Um, And... You think when a grit is is embedded in a resin, housing like a grinding wheel let's say
John Saunders: mm -hmm
johngrimsmo: uh of that 80 micron how much is being held by the resin. like 80 of it So. almost nothing sticking out, right?
John Saunders: Okay. Okay.
johngrimsmo: So maybe 10 or 20 microns sticking out. what is my depth of cut in microns oh my depth of cut is 40 micron
John Saunders: Okay.
johngrimsmo: And so like i'm my I'm asking the tool to cut way deeper than the grits could possibly stick out.
John Saunders: Yeah.
johngrimsmo: And. Therefore, I'm... creep feed grinding or i'm like side grinding as this cup wheel comes apart or comes across the part i'm not using the tip the bottom of the cup where the grits are sticking out i'm kind of using the side of the wheel And hmm.
johngrimsmo: my biggest blowout is. on the far side of the ceramic ball. where the side loading force pushes the last bit off because ceramic is super brittle.
John Saunders: Push it. Okay.
johngrimsmo: And then there's an SFM issue that I'm spinning way not fast enough. I'm spinning like a thousand SFM when I should be like.
John Saunders: Oh.
johngrimsmo: 6 ,000 SFM. um And I'm only spinning at 6 ,000 RPM. I could easily do 30. you know.
John Saunders: Sure.
johngrimsmo: So there's things I can do. that i've just been ignoring and it's like everything was built upon the way we ground on the Maury in 2015. when we just tried it and figured it out.
John Saunders: Yep.
johngrimsmo: And it's just been an evolution ever since then without the full deep dive of like. no grinding technical ceramics is like an industry like there are research papers on this And ChatGPT found them for me.
johngrimsmo: Yeah, so a bunch of little things are coming together. Higher SFM, way lighter step downs.
John Saunders: Okay.
johngrimsmo: um So instead of going down 40 microns, go down like five and just go zip, zip, dip zip, zip, zip, zip, zip, as opposed to like.
John Saunders: Yeah.
johngrimsmo: like really slow. So I haven't tried it yet, but I'm at the point today probably where I'm going to. try the new faster lighter. step downs and should hopefully get amazing results.
John Saunders: Still seems odd that your the Norsemen were... great and the flask was problematic or vice versa.
johngrimsmo: Rask is problematic. But yeah, I think there's... either enough difference in the way the codes are separated. Like the step downs were different, I think.
John Saunders: Yeah.
johngrimsmo: um But I'm not exactly sure this would also be an amazing thing to scan with the interferometer
John Saunders: Oh my God. Right.
johngrimsmo: Right?
John Saunders: Well, so if you have a. sorry this is an awkward question to ask on here but if you have a ball if you have a little bit chip out Does that mean the whole blade is ruined? Can you pull it?
johngrimsmo: It can.
John Saunders: How do you pull the ball out?
johngrimsmo: and and You don't. You just scrap the whole handle.
John Saunders: Oh, yeah, it sucks.
johngrimsmo: Yeah. So what is the tolerance?
John Saunders: Sure.
johngrimsmo: What is the, does it matter?
John Saunders: Yeah.
johngrimsmo: Does it not? So I had this discussion with Eric and Gabe and some of the other guys and, you know, who's catching it, who's scrapping it, who what's the limit. um scan this with an interferometer all of a sudden you have that full screen on the page you can measure anything you want but What have we always been doing?
John Saunders: I can't see anything!
johngrimsmo: like What?
John Saunders: you You can measure, but like let's not have something to measure there. right like The point is not to measure the flaw.
johngrimsmo: Exactly.
John Saunders: It's like to stop.
johngrimsmo: Yep, yep. Exactly.
John Saunders: But yeah, here.
johngrimsmo: Um, so So I was asking Eric, I was like, what's our scrap? Rate what's the scrap? reason what What do we do?
John Saunders: Yep.
johngrimsmo: Does a chip actually cause a performance issue?
John Saunders: Mm -hmm.
johngrimsmo: or not.
John Saunders: It's a good idea.
johngrimsmo: And. I strategize the tool path to... if it does blow out It's in a region that is not being physically used by the detent. Um,
John Saunders: Oh, interesting. That's really smart.
johngrimsmo: So. That's cool. like even if it does chip out, it still functions just fine.
John Saunders: Mm hmm.
johngrimsmo: But if you look, even with your naked eye, you can be like. That's weird. That's... you can see the chip ever so slight.
John Saunders: You see it, yeah, right.
johngrimsmo: So we don't. tend to ship the ones with the bigger chips. um just because they look bad. But now we are scrapping. tens of dollars per handle in material and time and
John Saunders: Oh.
johngrimsmo: you know effort and They're like fine
John Saunders: Yep.
johngrimsmo: They're e class B products, you know like you said.
John Saunders: Yeah, I know but this is also like I totally get the idea.
johngrimsmo: Okay. Yeah. it's cool
John Saunders: Let's just fix this
johngrimsmo: It doesn't work for us. so i need to fix it so i'm i had a really fun deep dive on friday spent probably half the half the day just researching and figuring out and comparing the codes and um going back and forth between cursor and chat gpt and my own research and and looking at a fusion and figuring it out and looking at the code and just thinking and sitting back and staring at the ceiling and just going what is happening here like you know sometimes you get down to molecular level and you go if i were the ceramic ball And then...
John Saunders: No, that's what I'm sitting here thinking.
johngrimsmo: the grinding wheels like coming at me like of course this makes sense like I'm like... I'm a delicate. Brittle little thing. like it's going to chip out, you know?
John Saunders: There's just like this visual in my head of, yeah.
johngrimsmo: What would I want?
John Saunders: Yeah, that's trippy. Well, so let's talk about that. This is a one millimeter ball or smaller.
johngrimsmo: 16th of an inch, so 60 thou.
John Saunders: 60 ,000. Yeah. So 1 .3 million. It's actually not tiny, tiny.
johngrimsmo: Yeah.
John Saunders: Why wouldn't you want to do a spiral in cutting pattern? That way you're never. shoving the tool. across the plane and leaving the last bit unsupported, which is always going to, I'm not a scientist, but like it's always going to fracture before it cuts.
johngrimsmo: Oh interesting.
johngrimsmo: Amus. with the current cup wheel it's one inch diameter three three quarter um It's too big to do that. There's clamps in the way there's stuff around
John Saunders: Oh. Oh okay, I see.
johngrimsmo: Yeah, it's a good thought though. Unless I go with like some 6C grinding tool.
John Saunders: Can you come at it from both directions. one direction, the other direction. That way it's.
johngrimsmo: Okay.
John Saunders: two separate tool paths that overlap in the middle, but then again, you're never grinding.
johngrimsmo: Mm hmm.
John Saunders: across that unsupported. uh cliff
johngrimsmo: And causing it to blow out the cliff. That's that's interesting. Um, I don't know
John Saunders: if i were as If I were a tiny little ball, and you were coming you were coming at me like a...
johngrimsmo: i If you were a tiny wallet, yeah.
John Saunders: I always think about...
johngrimsmo: You'd want to be circled in.
John Saunders: this is super off topic, but like, you ever think about what life's like if you're a bluegill? and you go back and you're like no look guys you're not gonna believe it i like went into this altered universe there are these people that were like 40 times bigger than us it hurt there's something in my mouth Sorry, this is like really weird, but like, it's crazy, right?
johngrimsmo: Yeah, yeah. yeah
John Saunders: anyway so I'll stop.
johngrimsmo: Yeah, the ceramic detail model is going, you're never going to believe what's happened.
John Saunders: Hmm.
johngrimsmo: Um... Yeah, I think going going.
John Saunders: Interesting.
johngrimsmo: Higher RPM. Much lighter cuts will just do the thing. Yep.
John Saunders: Not in... Just an innocent question. Why not like a carbide ball?
johngrimsmo: um carbide i don't know i've never tried carbide steel definitely like a stainless ball can still rust
John Saunders: Oh, totally.
johngrimsmo: um And we've we've had that a lot of makers.
John Saunders: Sure.
johngrimsmo: Um, probably still use steel balls versus ceramic. And, you know. It's interesting what the customer knows because I've heard this from some of our customers that. reach out now why do you ceramic instead of steel i've got knives with steel balls and i've got knives with ceramic balls And.
johngrimsmo: There's a lot of variables in there to not make it an apples to apples comparison. but um
John Saunders: Yeah.
johngrimsmo: This carbide ball is pretty interesting. You'd still have the same grinding. well it wouldn't blow out nearly as bad as ceramic did.
John Saunders: Right, that's kind of what drove me to think about it.
johngrimsmo: a carbide ball is kind of interesting i don't know
John Saunders: It's meant. I assume they're available. I assume they're actually cheap.
johngrimsmo: That's a good question.
John Saunders: it loves It loves kissing a grinding wheel.
johngrimsmo: oh Okay. I'll do a little look into it.
John Saunders: shouldn't wear. You know, you should wear them. meaningfully in any of our lifetimes.
johngrimsmo: Yeah.
John Saunders: Mm -hmm.
johngrimsmo: Some of the beauties of the ceramic are the kind of the surface finish Um... The hardness. Carbide would do just as well Hmm.
johngrimsmo: I don't know if there'd be any benefit in switching from what we currently use. but
John Saunders: Sure, yeah.
johngrimsmo: Yep. But this would be a good thing for a new model to be like.
John Saunders: may not be worth it.
johngrimsmo: come up with a new knife model that's like we're going to try ceramic or a
John Saunders: oh
johngrimsmo: carbide detent ball on this new model just to like see.
John Saunders: I literally thought you meant AI model, which would be like, Claude, tell me the pros and cons of...
johngrimsmo: right
John Saunders: these two things like what am i not thinking of
johngrimsmo: Sure.
johngrimsmo: But this is where... I don't know. I want to say this is where someone's expertise and experience has a lot of value over an AI model. trying to describe these D10.
johngrimsmo: of a knife that is has these properties, you can still play with it, but I don't know. I feel like I have a lot of experience in intrinsic. touch and feel knowledge that like an AI can't just.
John Saunders: Oh, sure.
johngrimsmo: spurt out words that make sense. Um, Pretty good, I don't know.
John Saunders: For sure. i think my point there is like i don't so for me it's like let me
johngrimsmo: Yeah, yeah, totally.
John Saunders: enlightened me and then i find that even areas where i think i'm when I think I'm pretty good. I'm like, oh, that's a great point. So it's kind of like, okay, let's check your ego.
johngrimsmo: Right.
John Saunders: and like instead
johngrimsmo: Yep, yep.
John Saunders: Yeah. ah well we had a first I think it's my first actual true. intelligent. Uh... response from AI.
John Saunders: which I say partly faceious facetiously.
johngrimsmo: Interesting.
johngrimsmo: Yep.
John Saunders: so i've been using muse i talked about that grant started using muse some it has access to our email we were trying to reorder rego holders for the brother and like just just a stupid thing or for whatever reason we've had a trouble keeping track of what is the right dual contact PG 15 BT 30 holder with the right gauge that we want for the brother. We ordered some different ones, different machines. don't have like a right home for it tool libraries excel files the wiki the pot like the
John Saunders: ERP, all dumb examples of like, I don't know which one exactly, sir, I want. So I took a minute to look where I actually looked on the machine and the holders themselves. And I emailed Grant and I'm like, hey.
johngrimsmo: Yeah.
John Saunders: Sorry I don't have more conviction, but I believe. This is the one I'm currently using. I believe I want three more of these. Grant does all the reordering. I believe I want three more of these. Will you please check? against the last Rego order we placed.
johngrimsmo: Yep.
John Saunders: Before Grant. got to his email. Muse told Grant, hey, I saw John ask. and did want to let you know that on August 4th, you placed the order with Rego and it was the same holder that John was asking about.
johngrimsmo: No way.
John Saunders: So I was like, that's.
johngrimsmo: That's interesting.
John Saunders: actually a freaking win because I legitimately felt bad about bothering Grant but I'm like no I don't know he's on the one that placed the order right now those orders don't go into a shared like blah blah blah um
johngrimsmo: Yeah.
johngrimsmo: Yeah.
John Saunders: So then...
johngrimsmo: Okay.
John Saunders: I tried it again and it didn't work. we We're coming to visit you. And so. I... Yvonne booked a hotel in Toronto. And a couple of days later, I. sent her like a fake email like a test email it's like hey i know we talked about toronto did you book the hotel late for later in November.
John Saunders: And we waited a day to see if Muse would chime in. It didn't take the bait.
johngrimsmo: Hmm. Interesting.
John Saunders: Yeah. she obviously is using it so i was like darn i wanted it to be like hey john was asking about the hotel you did book the holiday in like it's you're good whatever.
johngrimsmo: Yeah. yeah
John Saunders: so
johngrimsmo: It's not some personal email versus work email that it doesn't have access to or something?
John Saunders: In this case. Great question. No, I was... we we're on the same. there.
johngrimsmo: yes great
John Saunders: Yeah.
johngrimsmo: Interesting. Yeah, I haven't played with any of those. Muse grok agents that do all that kind of stuff, but
John Saunders: Yeah.
johngrimsmo: Curious. but
johngrimsmo: It's good to hear your feedback.
John Saunders: Yeah. My other win and then I'll. Oh. get through my list is um I pipe coded a Fusion plugin called the... BFM, the barfed mill.
John Saunders: John. I freaking love it. It was again another. high in my like machine career. it is a plugin that lets me pick a internal circular feature on a part.
John Saunders: In this case, it's about a half inch hole. I just picked that. whole. It then asks me what tool to use out of the tool library. I want a quarter inch. It's a piece of quarter inch break line, basically brass tube or whatever.
John Saunders: um it finds the tangency point. Of that. half inch hole in the direction I want to move. It then calculates a lead -in move with 10 ,000 clearance, leads the brake line.
John Saunders: into that spot. and then at a controlled feed rate grabs the part
johngrimsmo: Oh. You're pulling
John Saunders: opens up The new yep.
johngrimsmo: Okay.
John Saunders: Opens up the pneumatic vise, indexes the part the barfed mill part forward by the the distance in the... gooey. And then. Clamps the vice, clamps the pneumatic chuck. retracts the tool off so you don't get the 5c you know strafing lines and retracts the tool up
johngrimsmo: Yeah.
John Saunders: And I programmed it at home. And. was swinging by the shop on a Saturday. And I thought. I'll give it a shot just so I can go. Like. rework on it at home. It was 100 % perfect.
John Saunders: for the first time.
johngrimsmo: a ah
John Saunders: Watching it is. like Because it's what the Willman does.
johngrimsmo: Yeah.
John Saunders: um And I'm just like, oh.
johngrimsmo: yeah
John Saunders: Heck. Yeah.
johngrimsmo: Does the Willimon? Do spindle. in a hole and drag the bar or does it use the vise to drag the bar?
John Saunders: Uh... Everything it can do. Well, it pushes against the you've you guys on yours. It pushes the bar against the
johngrimsmo: I don't have that because I don't have the bar loader hooked up.
John Saunders: Nub.
johngrimsmo: to mine so i'm only actually we're not even pulling bar we're just making one part at a time still
John Saunders: Oh. Okay. Yeah. We don't do what I'm saying, to be clear. You could, we don't.
johngrimsmo: Yeah. You could, yeah.
John Saunders: Yeah. the for anyone who cares the woman has a harden. uh Still. pin on the side of its sheet metal spindle and the bar feeder pushes up against that so it's still using the x -axis as a master control of the feed rate and distance of the material.
John Saunders: That make sense?
johngrimsmo: Yeah, it's your it's your pull out hard stop like the material goes out to here kind of thing
John Saunders: Yeah. Big up.
johngrimsmo: Yeah. um Yeah, what you were talking about before putting a... ah thing in in the hole and dragging your... bar across the first time i ever saw that was kind of As I was getting into machining,
johngrimsmo: um This is like 2000. ape
John Saunders: mm -hmm
johngrimsmo: 10 something like that And. this company, I can't remember what it was, but they were making um You know the car, the AC Cobra? like the old school.
John Saunders: The super curvy.
johngrimsmo: Yeah, super curvy, beautiful, like 60s kind of.
John Saunders: mm -hmm mm -hmm
johngrimsmo: ultimate sports car kind of thing so they're making reproductions of those and they were making this a frame rail out of aluminum that's like eight feet long or something like that out of this one solid chunk of aluminum and they had some hoss whatever with the doors they go up on the side and
John Saunders: Oh my gosh.
johngrimsmo: They had this video series. It might have even been in the the magazine. um modern machine shop or whatever. And they would like ma machine a section. put a thing in there loosen the vise drag the thing over with the spindle clamp it again
John Saunders: Oh my gosh.
johngrimsmo: and we
John Saunders: Accurately then.
johngrimsmo: with the doors open and everything and i was like that's brilliant that's really cool one day i'll use that haven't yet but
John Saunders: No, no, yeah. Yeah, that's cool.
johngrimsmo: My head.
John Saunders: Yeah, so it was funny too because we the first time we did that, we probed the part. um to Two. Run it again. And I was like, let me look at the offsets.
johngrimsmo: right
John Saunders: And I was like, oh, my gosh, it's moving with the brake line. I mean, it's not that crazy, I guess, if it's a rigid brake line that could. could and will bend out of the way when something goes wrong.
johngrimsmo: Yeah. Yeah.
John Saunders: um And yeah, we don't even probe anymore. We were running parts with it now. It's like.
johngrimsmo: Yeah.
John Saunders: this is incredible and then this one
johngrimsmo: It just works.
John Saunders: Yeah, just works.
johngrimsmo: Huh. I wonder if you'd ever you want to use like a carbide bar. in that scenario where It could shatter if it really breaks.
johngrimsmo: um Or if you don't even need that, if it just works.
John Saunders: Brake line for now because if it bends we still do the tool brake detect on it So if it bends it'll fail the tool brake detect um we have actually done a lot of 3d printing and if you do the layer lines so that it will shear.
johngrimsmo: Hmm.
John Saunders: in the direction you want it's just a little cheaper than buying carbide or ceramic bars but you certainly you can get that stuff for pretty cheap
johngrimsmo: Okay. Yeah.
johngrimsmo: Yeah. Love it.
johngrimsmo: Hmm.
johngrimsmo: um Last thing I've been working on is... Spin. almost the whole day yesterday working on tolerances infusion, like in every operation has tolerance value and.
John Saunders: Oh.
John Saunders: Really?
johngrimsmo: You go into. well you know at the top of the passes tab whatever there's the tolerance thing
John Saunders: Oh, the tessellation though.
johngrimsmo: And then if you go into compare and edit, there's like 17 other things you can mess with.
John Saunders: Sorry. Mm hmm.
johngrimsmo: So I'm sitting there asking Kirscher, like, what do these actually mean? And then. Heidenhain also throws its own cycle. 32 .1 or whatever it is on top of that. And I finally really dug into what those all mean and what the Heidenhain is doing and how my post works in order to.
johngrimsmo: auto -generate the cycle 32.
John Saunders: Okay.
johngrimsmo: And because on the surface of our Fjell. It is a. three inch radius.
John Saunders: Okay.
johngrimsmo: on the surface and we're 45 degree, we're 3D milling that surface. so on all of our fjell and rasks that are smooth handles we um We're 3D milling that whole surface.
johngrimsmo: And if anything is causing the tool path to tessellate or...
John Saunders: Mm hmm.
johngrimsmo: are um Uh... there's another word I'm thinking of, that surface, you'll see it as it reflects light. And when our...
John Saunders: Yeah.
johngrimsmo: finishing guys go to like scotch braided or tumblr or whatever those lines don't go away um you have to like sand through them and i'm like this is some tolerance setting this is a machine thing that The current should be able to do.
johngrimsmo: what I want.
John Saunders: Sure.
johngrimsmo: um And so. I was able to. whether it's drag code into cursor and have it like compute stuff for me with little angles between passes. um learn all my tolerance values. I learned that my post was putting a floor on the the tolerance.
John Saunders: Yeah.
johngrimsmo: not going under 1 10th when i could easily go to 1 millionth so i'll remove that
John Saunders: caps it right Mm -hmm.
johngrimsmo: And all excited. I ran another one last night, came in this morning. better but still not perfect?
John Saunders: Okay.
johngrimsmo: and i can in In the light, like I'm literally sitting there up to the window, like getting just the right angle and looking for this faceting.
John Saunders: mm -hmm
johngrimsmo: And it's better but still not done. And this is the perfect thing that the interferometer would just show clear as day.
John Saunders: Oh, that's cool.
johngrimsmo: Right? Like this is kind of the stuff that I want.
John Saunders: Sure, sure, sure, sure. sure Sure.
johngrimsmo: um so even the tighter tolerances didn't solve solve the problem But I'm getting there. again there.
John Saunders: Yeah.
johngrimsmo: Yeah.
John Saunders: But that's like such a great example of. in this case, AI, but like. you own this unbelievable machine this current and it ends up that like there's probably a couple settings that you
johngrimsmo: Yeah.
John Saunders: could change and can get way better results out of it.
johngrimsmo: i've been doing wrong this whole time.
John Saunders: Oh man.
johngrimsmo: Yep yep. Yep.
John Saunders: Interesting.
johngrimsmo: I've got a parallel toolpath where it goes one way and then it goes 90 degrees the other way and you get this X in the middle. and where the x meets should theoretically in the machine be at the same point in space.
johngrimsmo: where it goes this way and then it goes this way.
John Saunders: Okay. Yeah, sure, sure, sure.
johngrimsmo: And then cursor was able to analyze the program and be like of the 3042. points where they X meet, 99 % of them are within two millions.
johngrimsmo: and then you know the 1 % of them are within 4 million. So like that's not your problem. The code is good.
John Saunders: Okay.
johngrimsmo: The code is good, but something is causing the mismatch.
John Saunders: Interesting.
johngrimsmo: And it turns out my cycle 32 tolerance was like. Eight. 8 tenths. which is plenty to like make this anyway.
John Saunders: I love that. I love that.
johngrimsmo: It's good.
John Saunders: Don't diminish that at all. That's super cool.
johngrimsmo: Yep, yep.
John Saunders: it's sort of sad when you start thinking about like, okay. in Fusion or SolidWorks, I create a so ah cube. And one feature on it. And then by the time it gets to the cutting tool, interacting with the material, it's been like tessellated and intoleranced.
johngrimsmo: Yep. And the tool is not the diameter I thought it was. And like.
John Saunders: Yeah. yeah and pushed to a g -code which then goes the control the control does its own motion control processes and then those get moved to a chip to the servos the servos like you're just like oh my god there's like 17 translators between these two things.
johngrimsmo: Yep. So.
John Saunders: Good talk.
johngrimsmo: Progress. Yeah.
John Saunders: Good talk.
johngrimsmo: Good talk. all right man
John Saunders: See you next week.
johngrimsmo: I will talk to you next week.
John Saunders: This is great.
johngrimsmo: Bye.
John Saunders: Okay.
