Transcript
John Saunders: Good morning, welcome to the Business of Machining episode 458. My name is John Saunders.
johngrimsmo: And my name is John Grimsmo.
John Saunders: John and I have been going on this for about, I don't know, 10 plus years. um Highs and lows, fun, excitement, entrepreneurs, machinists, ah tinkerers, and we just had a quick talk right before we hit the record button and reminded ourselves we are not as young as we were when we started this, but now's the time to be bold.
johngrimsmo: Yeah.
John Saunders: it's time to have that conviction. John and I both talk a lot about, oh sorry, I think we both spend a lot of our personal time thinking about the balance of of intrigue and research and what gets our minds going versus the sort of day-to-day of running a kind of focused product-based company.
John Saunders: That makes sense, I paraphrasing?
johngrimsmo: Yeah, no, at this point, we don't run small companies like I have 11 employees. And it's like, there's a lot going on at all times and a lot to do and, you know, making sure the right things are getting done and that we are getting what we want as owners out of the business, both in time and money and whatever.
johngrimsmo: to make sure it's like the juice is worth the squeeze, you know, and to sit back and smell the roses every now and then.
John Saunders: Yeah, for sure.
johngrimsmo: But then, you know, with that you know nose full of of rose smell, get after it, you know, get back to work.
John Saunders: Yeah. Yeah. Well, it's interesting because You and I were talking about how if 15 year old burst version of ourselves could see our lives today, we would be very grateful and and proud and happy and we are, but like that does nothing to like scratch the itch of just what I think about on a daily basis. But i don't know what the answer is.
John Saunders: um Weird segue, like we had a local chamber commerce and then like a state affiliate from some business group that wanted to tour our shop and I was traveling i asked Alex to graciously host the tour and Alex was, it was just great to do.
John Saunders: Like i actually love having somebody else do that often.
johngrimsmo: Yeah.
John Saunders: and I asked him how went. He's like, oh, it was great. He's like, I tried to like show them how we're a manufacturing company, but we really, you know, really try to embrace technology. And they're like, oh no, you're a tech company, which is like, we're not a tech company in the sense of like that, but like, no, like from, from implementing software that we write to like how we're running the robots, how we're handling the automation, like, um,
johngrimsmo: Yeah.
John Saunders: And I think I'm not like trying to pat myself on the back. I think that's what everybody's going to have to do. Gone to the days of just like, oh, I have ah a networked VF2 and a hand code stuff and a bridge port. Like that's that's gone, you know?
johngrimsmo: Yeah. Yeah. I mean, your business is so much more than just making widgets. You're like solving problems and yeah.
John Saunders: Yeah.
John Saunders: Yeah.
johngrimsmo: Inventing things, literally inventing solutions to make better stuff, cooler stuff.
John Saunders: Yeah.
johngrimsmo: Yeah.
John Saunders: How are you?
johngrimsmo: i'm I'm good, man. I'm good.
John Saunders: Yeah.
johngrimsmo: um Yeah. The fog of Blade Show has finally worn off, and I'm kind of getting back into it. I slept a lot the past week, which was good.
John Saunders: Good. Yeah. Well deserved.
johngrimsmo: Yeah, it was good. um So yeah, getting hungry. And I've talked to you before about, i go through periods of time, sometimes a month or two at a time where either I'm really, what I say?
johngrimsmo: Either really creative and just want to create new stuff, or I just really want to hunker down and like solve current problems kind of thing.
John Saunders: Mm-hmm.
John Saunders: look
johngrimsmo: And I'm entering a phase of solve a lot of current problems and challenges and like projects, things like that. Things that are on the plate that I know will help immediately to either specific people or the company as a whole. And I'm getting like really jazzed to do a bunch of that.
johngrimsmo: So that I can exit that and get in my creative zone and like, you know, start running hard on something new.
John Saunders: Yep.
johngrimsmo: And I like how that feels for me. So yeah.
John Saunders: Good.
John Saunders: But you say that as if you are almost subconsciously wanting to fight yourself on like, is it okay to be the phase I'm in?
johngrimsmo: I, Yeah, it's the first time I've said it out loud you know in recent times.
John Saunders: Okay.
johngrimsmo: so I'm just like, you know.
John Saunders: Yeah,
johngrimsmo: But yeah, exactly.
John Saunders: Yeah. Just go with the flow, right?
johngrimsmo: You've got to figure out how you work best. And I mean, at the end of the day, everything I do is has a creative bug to it.
John Saunders: Yes.
johngrimsmo: Whether it's solving a problem in a creative way or whether it's like making something new, it's just what I do.
John Saunders: Yes.
johngrimsmo: And that's okay. That's cool.
John Saunders: I think that's what we... have the potential to rob ourselves of of is the ability to come up with elegant, great solutions by thinking and using our brain because we're as just self-imposed business owners.
John Saunders: Like where're we we are, it's not a good thing that we take on as much as we do. And so you ultimately end up just wanting to like get through stuff when in reality, sometimes it's like, no, let's think through,
johngrimsmo: yeah
John Saunders: like yeah i'm looking at this air compressor system on my desk like do i just want to get that off my desk or checklist or do i want to look at like no it's a really interesting way to handle this filter or no i don't like the way that fitting is going to disrupt the air so let's find a different fitting
johngrimsmo: Mm-hmm. And that is subconsciously important to you, you know, to find an elegant way to do it, to, you know, flex those muscles a little bit and be like, I know I can do it better.
John Saunders: yeah
johngrimsmo: and And why do it twice? I'm going to do it once and it's going be right. And then I can not worry about it again because I have this comfort in the back of my head to know I did it right and it's going to last for a long time. Yep.
johngrimsmo: Versus, you know, we've all done slapdash solutions and then you always have that in your back your mind, like, is it going to blow up yet? Is it going to like, I could have done it better, but...
John Saunders: Yep. Yep.
johngrimsmo: Yeah.
John Saunders: What's felt good. Um, I, I'm really, i am proud of myself. Um, I, Finally, I'm coming back to a lot of stuff that I wanted to stop putting off.
John Saunders: um And that includes days off in the shop. i've been I've done it total failure of the first half of the year or whatever, but until the year to date. But past two days, I've done it and it's felt great.
John Saunders: um We'll see if I get continue you that. I think I will because it makes sense as to why it's able to come back. All those machines are gone and sold. In fact, the horizontal, it's super weird. They're here as we speak, like taking the horizontal or apart.
johngrimsmo: Yeah.
John Saunders: um um and act physically the rigors will show up uh in two days to actually move it out but like it's it's no longer under power it's just if like that's that was the last thing um and for that in terms of outbound now we'll have to get the two new brothers that are ordered to wait schedule them to be received and and bring them online which will definitely take some work but that work doesn't need to be um There's no pressure because those parts are made native currently on other machines that are online.
johngrimsmo: Yeah.
John Saunders: So we we can do that at our own pace, which feels great. Yeah.
johngrimsmo: Yeah, yeah.
johngrimsmo: And bringing it in the next two machines, is it relatively straightforward? Like, like you you know what you need to do from the first machine. There's not a lot of problems to solve. It's just kind of like have your team install them.
johngrimsmo: Like, yeah.
John Saunders: Short answer, yes. One is again, an effectively identical replica of the first sort of lights out 700. So we know the same vices, same everything. There'll be, I'm sure quirks hiccups and like, you know, who's going handle networking the machine and tooling, like whatever, but it's just doing, it's not risk.
johngrimsmo: Yeah.
johngrimsmo: Yep. All straightforward stuff.
John Saunders: Yeah.
johngrimsmo: Yeah, it's just work.
John Saunders: Yeah. Yeah.
johngrimsmo: Yeah. Yeah, it's always the little stuff like,
John Saunders: That's actually good Sorry.
johngrimsmo: I saw Steve bending down, looking between the Speedio and the Auroa. And I looked too, and there was little pu puddle of coolant at the door transfer between the two machines. And I'm like, what's going on there? He goes, I don't know. Just started dripping.
johngrimsmo: I'm like, have fun. You'll figure it out.
John Saunders: Yeah.
johngrimsmo: like
John Saunders: Yeah, yeah.
johngrimsmo: But then it's kind of weird. It's the first time it's done that.
John Saunders: Yeah. How is that whole, any update on that system?
johngrimsmo: Yeah, I mean, ah Speedio is pallet changing every day, multiple times a day.
John Saunders: supporting Good.
johngrimsmo: We have the Fjell blade grinding so consistent now across five pallets that we don't even run it during the day.
John Saunders: Good.
johngrimsmo: We hit Cycle Start when we leave at night.
John Saunders: That's awesome.
johngrimsmo: I don't even, I'm not involved at all. So Steve just runs that. And then we're still grinding the bla the Rask and doing Norseman hard milling during the day because those are more, they're automated-ish, but yeah.
johngrimsmo: Rask hard milling bevels is my next one of my biggest challenges now is like optimizing that because it works on the field, but it doesn't fully work on the Rask yet. He still has to like fiddle with it and the codes are different.
John Saunders: Interesting. What
johngrimsmo: That's why. um
John Saunders: what do you mean the codes are different?
johngrimsmo: the The way the method, the probing, the comping, all that stuff, I fixed it on the field.
John Saunders: Got it.
johngrimsmo: I just haven't applied it yet to the Rask and it's
John Saunders: Oh, okay. So you just have to, you know how to do it you just have to do
johngrimsmo: I think I know how to do it. Well, I have a working system.
John Saunders: Okay.
johngrimsmo: I just need to like, I forget all the details. So I need to like analyze the code and be like, oh, okay, I did that there.
John Saunders: Yeah.
johngrimsmo: Now I do that on the RASC. And then, which brings me into one of my points was, I've told you this before, when we machine the first bevel of a blade, like the the one side, now you have a missing section, you slap it on the other side of the fixture and now it's unsupported, right?
johngrimsmo: Yeah. So up until now, I've been machining Delrin inserts and I'll surface them on the Kern and like to a theoretical conforming size, right?
johngrimsmo: But the thing with a back support like that, especially in a hard milling, grinding bevel operation where the blade gets really thin, it's almost never perfect.
John Saunders: It's not yet, sure.
johngrimsmo: It's either clearance and not touching, or it's, you know, ah too much and bending the blade. So there's, there is no like perfect, um, which is annoying.
johngrimsmo: And now, especially on the Fjell, since I've gotten the consistency of op one machining or grinding like to 10ths, like every single blade is, and I have months of data like proving that is like exactly where it needs to be.
John Saunders: Mm-hmm.
johngrimsmo: Now I can upgrade the support side on side two and make it more gooder um so that they're properly conforming, not bending, not whatever. And got me thinking a lot, again, about those low temp alloys, like the serobend and all that stuff.
johngrimsmo: And I bought some from McMaster, but then got scared and never used it because it has cadmium and lead, which are just things you don't really want in your body um or around you at all.
John Saunders: Oh
johngrimsmo: So then there are other ones, like, you know, Dan Gelbart.
John Saunders: oh yeah.
johngrimsmo: Yeah.
John Saunders: Well, nobody knows him, but everybody.
johngrimsmo: So he had like, yeah, nobody knows him. I've actually, I've actually met two people that have met him in person.
John Saunders: Interesting.
johngrimsmo: Yeah. um He had a great video, unusual shop tips, part two came out recently.
John Saunders: I started watching it. yep
johngrimsmo: Yeah. And he talks, he's talks about these low melt alloys and he talks about the ones that are lead free and cadmium free ah called fields metal. And there's all kinds of other ones. And,
John Saunders: Yep. You got to say it like, a yeah I can't even.
johngrimsmo: fiels meto
John Saunders: Yes. Yeah.
johngrimsmo: in his Austrian or whatever accent. um
John Saunders: I think it's Israeli.
johngrimsmo: It is Israeli. Thank you. Yeah, you're right. And no, he was born in Israel and then he moved to Germany and grew up in Germany or something like that.
John Saunders: is that right?
johngrimsmo: I think so.
John Saunders: Yeah.
johngrimsmo: um I forget. But then he came to Canada and loves it. Yeah.
johngrimsmo: Basically, ah found the sources last night, like Roto Metals is a good place to buy it. And the lower the melting temperature, the significantly more expensive it is.
johngrimsmo: So if you want the stuff that melts at 140 degrees Fahrenheit, so like a warm cup of tea will melt this stuff, it's 16 times more expensive than the stuff that melts at 250 degrees Fahrenheit.
John Saunders: Yeah.
John Saunders: Yeah.
John Saunders: Interesting.
johngrimsmo: um which takes effort. It's like boiling water is not going to cut it anymore.
John Saunders: But you it's infinitely reusable.
johngrimsmo: It is infinitely reusable.
John Saunders: Who cares?
johngrimsmo: Yeah, exactly. So question is, you know, do you need the super low melt stuff if you're going to reuse it a lot? um then that could be helpful.
johngrimsmo: But I want to just basically make a casting, put the blade, like mount the blade perfectly rigidly, pour this stuff in once and then let it harden. And then it stays you know forever until it wears out or whatever.
John Saunders: Oh, so you you aren't you are not envisioning having to re-melt it for every blade.
johngrimsmo: Yeah, it'd be too much work. Like it is an option for sure.
John Saunders: Okay.
johngrimsmo: And that would be like the perfect ideal because then every blade is perfectly conformed every single like that's kind of awesome.
John Saunders: Yes.
johngrimsmo: But I think it's unnecessary for what we're trying to do, um especially if I can keep up one accuracy bang on, then it's fine. So.
John Saunders: Yeah, this is the whole like betting betting a rifle barrel or potting something in. i get it
johngrimsmo: Exactly.
John Saunders: Can i can i ask you to pause for second?
johngrimsmo: So. Sure.
John Saunders: Okay, because I think this is like the problem that you and I both get into is like, we just, we want it we want it done, we have an idea and and you don't fishbone it right, or whatever.
johngrimsmo: a
John Saunders: And I think if you were to hand this to a professor that ran a engineering lab or ah or a but grad school engineering lab, and they had 10 teams of three people each that were super, super, super, super smart, hands-on folks, and were like, come up with applications to this.
johngrimsmo: Well...
John Saunders: What I'm thinking, i' I'm not... in that group. So forgive me for presupposing that i'm going to be able to offer solutions, but it's like, wait minute here. You've got rubbers, you've got hydraulic systems, you've got a spring systems that can be individual Springs.
johngrimsmo: Yep.
John Saunders: You've got hydraulics plus pins where they can self balance. There's like a lot of different, interesting ways. And ah the real question is also how much force do you need on that blind side to, to stop the harmonic?
John Saunders: Cause all you have to do is arrest the harmonic, vibration of that thin blade tip. You don't necessarily need it um a lot. Does that make sense to soak up or dampen?
johngrimsmo: Yeah.
John Saunders: I'm not using the correct answer your dampen, et cetera.
johngrimsmo: Yep.
John Saunders: um
johngrimsmo: And I don't know the answers to those force vectors.
John Saunders: Yeah, sure, sure.
johngrimsmo: um It is calculable if you're a lab and want to like spend the time to do it. I don't really care either.
John Saunders: Yeah.
johngrimsmo: I just want like a solution. um But you're right. The question is, you know I've had lots of ah lots of chats with my friend, GPT.
John Saunders: Yeah, yeah, yeah.
johngrimsmo: um about this in depth and i thought a lot about it. And and through those conversations, you know, it kind of expands my brain to exactly what you were just saying of like, oh, I never thought about like that or like castable epoxies or urethanes or things like that.
John Saunders: Mm-hmm.
johngrimsmo: How much, how hard does it really need to be? Well, we're using Delrin and it's this hard. um There's like low metal thermoplastics that are softer and I was playing with those last night. And like my fingernail can make an indentation in those and it can't in Delrin.
johngrimsmo: So, okay, Delrin is quite a bit harder, you know? And then it's like this Fields metal stuff is is quite a bit harder than any plastic, even though a lead-ish consistency. um And i I had discounted it a few months ago because I was like, it also pours like water.
johngrimsmo: So in this in it's awesome.
John Saunders: crazy.
johngrimsmo: But in this cavity, I'm like, oh it's going to leak everywhere. So I just kind of gave up on the idea. And then exactly, I didn't really realize that like three months ago when I bought it and didn't use it.
John Saunders: Plumber putty dam it up.
John Saunders: Yeah.
johngrimsmo: So then I was thinking about that. I'm like, well, it was just clay or like butyl tape is the one I'm going to, or butyl putty, the strips, whatever.
John Saunders: OK.
johngrimsmo: um That's what I'm going to go with. And I'm like, oh, that solves that. And I could even redesign the back pocket on the next fixture to like conform it exactly how I want.
John Saunders: Yeah, sure, sure.
johngrimsmo: I'm like, okay, this low melt alloy stuff is like, I think the ultimate way to go. um
John Saunders: Bothers me to think that like, you're still you're still either 100% perfect or you've got major errors.
johngrimsmo: Yeah.
John Saunders: One thing I'll say and then I'll shut up is do your casting idea where you create that perfect, um um what's the word? Like mate, like conforming mate, but then, well, yeah.
johngrimsmo: Conforming, yeah, mate, yeah. To one blade, to be clear.
John Saunders: yes but then instead of it being mounted in a perfectly rigid backstop if you had it on a pivot or ah a spring plunger that way you pre-load the blade up against it the but bow the blade shouldn't bow because you're you've well you've got the whole surface area of that so it should be relatively even so it's not like you're putting a spring in one or two spots you're pushing against the whole thing um just just
johngrimsmo: The blade does bow pretty easily.
johngrimsmo: Yeah.
johngrimsmo: Yeah, yeah, hear
John Saunders: again, you may not need much and then that that allows it to have, and you aren't over-constraining an unnecessary degree of freedom.
johngrimsmo: Some compliance and... Mm-hmm. Yeah. Yeah.
johngrimsmo: Yeah, a layer of rubber behind it or something, but then you're creating a compliance layer for the machining. um Because a 3 8's end mill is coming in there and side milling this blade and putting who knows how much force sideways into it.
John Saunders: sure.
johngrimsmo: And then the grinding takes zero force the way I'm chop grinding it. It's like.
John Saunders: Maybe you separate them.
johngrimsmo: Yeah. How? Why?
John Saunders: Because you can rough it in rigid environment and then, I know you don't want to move it between fixtures, but I'm just talking loud and then allow it to be unconstrained or or more, yeah.
johngrimsmo: Yeah.
johngrimsmo: I just want it to stay straight. Yeah.
John Saunders: Yeah, let me think about it. Yeah, keep playing.
johngrimsmo: Yeah. yeah For sure. And i i told you a few months ago how I improved the way I clamp the blades so that they're not bowing.
John Saunders: Uh-huh.
johngrimsmo: The RASC is still doing it the old way, but the Fjell is now like basically perfect.
John Saunders: Uh-huh.
johngrimsmo: So I was on the speedio last night with like indicators and like pushing on the blade and loosening different screws and learning even more about this clamping induced deflection.
johngrimsmo: which is, it's intuitive once you see it, but until that moment, you're like, no way.
John Saunders: Mm-hmm.
johngrimsmo: this I'm clamping it. I'm clamping it hard. I'm clamping it flat. Well, loosen these two screws and watch the indicator. Like, holy moly.
John Saunders: Sure, sure, sure.
johngrimsmo: Like, holy cow. And then I can also, because the way it works, I can push on it with my finger until it like hits. So I can see how much movement. And and it's only about five pounds of force to go from, you know, a thou and a half of bow down to zero till it's like hitting the fixture.
johngrimsmo: But it is bowing by a thousand and a half, depending on which screws are tightened and whatever. So, yeah.
John Saunders: What, the blades are stainless? Okay. Like, its almost like, could you machine a ma machine a magnet and then turn on and off the magnet that's perfectly machined with stainless, isn't generally not very magnetic.
johngrimsmo: It is possible, but it's a lot of work. Yeah.
johngrimsmo: It's a high carbon stainless. It's got a lot of magnetism to it.
John Saunders: Oh, that I would.
johngrimsmo: Yeah, it could work, but, and I've already designed a system for that with like switchable magnets.
John Saunders: Yeah, sure.
johngrimsmo: It's it's just a lot of work. And I think there's, it's an unnecessary amount of complexity um to have these magnets in the fixture. And then you got, you know, do you deal with chips? Do you deal with whatever?
johngrimsmo: And how do you pot them flush? And how do you like, you still, it's a cool idea. It's a really cool idea, but yeah, I think we can avoid it
John Saunders: I hear you, but but fish, but like, take you're smart. like Take the time to like lay out all these, if anything, just to save your own um notes and and be able to look back and brainstorm on it.
John Saunders: If this works great and we're done, then great. We've wasted the time thinking about more further.
johngrimsmo: Yeah. yeah It's fun. um
johngrimsmo: Your turn.
John Saunders: I'll read my list and then you can ask me what you want to hear about. um Avoiding sole source and material, home assistance back on track coming together. Will and robot is actually was down. It just got repaired. I like four seconds of ago I got a message. um Measuring material with our robots and reevaluating process bins.
johngrimsmo: Let me hear about the Willy robot first.
John Saunders: Grant was just like, Hey, it's not working. And, um, we actually had this problem happen right when it was brand new and and then came in and like, I forget what they did. We don't even know what they did. It was, everything was so new to us then the machine and the robot and all that. Um, and but it was the same air where the robot arm was losing communication and connectivity with the controller.
johngrimsmo: Interesting. So is this a a UR problem or a Willimon problem?
John Saunders: Only a UR problem, almost nothing to do with the Willamon.
johngrimsmo: Okay. Okay. But I guess so who who fixes it, like
John Saunders: Um,
John Saunders: So funny enough, um we have gotten so much better or so good at UR stuff. And Willowman is just, um unfortunately, Marcus had come back. Marcus, I believe, has now since not stayed at Willowman.
John Saunders: He was their guy. So, yeah. in no way am I getting cocky here, but like us, we had reached out to a few people at Willamette, they're just like, you you guys know more than us. And especially now that weve we have four UR robots, we've been talking to Graham at Lights Out, we've gotten so much more competent on it.
John Saunders: Like we're we're good, like I love it. but
johngrimsmo: Yeah.
John Saunders: what ble What really frustrated me was that Grant took the, it's a hanging robot. And so he told it pulled it off the hanging pendant and did something. I think we were able to hook it up to an extra controller that we have for the next brother that's about to come online.
John Saunders: And for, I don't know, five minutes, we got it working again. And then we've lost that communication error again. I'm like, okay, that's... very edifying. This is a wire.
johngrimsmo: Yeah.
John Saunders: This is just a loose wire, which I believe the loose wire was the problem two years ago.
johngrimsmo: Okay.
John Saunders: And so Grant and I had scheduled a time at three o'clock today to sit down and like look at it together, brainstorm, sort of just pick through stuff. Um, cause like I still want to send it back to them and get a multi-thousand dollar repair bill.
John Saunders: Like I just was like not interested in that right now. And then Grant just messaged me. He's like, yep, found the loose wire. in The second joint is back up and running.
johngrimsmo: So internally to the UR robot.
John Saunders: Yeah.
johngrimsmo: Interesting.
John Saunders: but we had heard we've never had to do a joint replacement, but Dennis has mentioned, um, I believe he can do one in either 30 minutes or two hours and like not a particularly big deal.
johngrimsmo: Wow.
John Saunders: Um, so, um, it was, yeah, I'm very, very happy and proud to, that's great.
johngrimsmo: Awesome. Yeah, good for you guys
John Saunders: Yeah. Yep. What's next on your list?
johngrimsmo: guys. An interesting problem that we have had for almost ever, for six plus years, on when we assemble our Norseman knives,
johngrimsmo: they either work or they don't. As far as detent, lock up, lock fail, things like that. Eric does an operation where he calls spine whacking, where he opens the knife and he smashes it against a wooden block to see if the lock will fail.
johngrimsmo: And I don't recommend anybody do this, but we do it it internally to like prove that the lock is good. And for the longest time, for years, It either works or it doesn't. And he has some that fail. And then he'll switch around parts, handles, blades, inserts, whatever. And he's never been able to see a pattern of what's what's happening, why it's happening. Some work, some don't. We've done all kinds of machining. We bought a CMM to solve this problem.
johngrimsmo: um
John Saunders: This is the carbidized, carburized.
johngrimsmo: and We don't do that anymore, but that that's how we used to do it.
John Saunders: OK. Yeah, I remember that.
johngrimsmo: And that had its own problems, but it was more consistent than our current method.
John Saunders: if
John Saunders: Okay.
johngrimsmo: um I think we solved it. I think Eric and I put our heads together. I had a theory and I talked to him about it. um Basically, we make this tiny little thing that's an oval half inch long with four holes in it.
johngrimsmo: And the way I toleranced it is all four holes have five tenths of clearance to the mating bosses. And then there's ah an arc in the back. That's your fifth locating feature. So I have five locating features that are equally toleranced. So far, way over constrained.
John Saunders: Yeah, right.
johngrimsmo: And we're making this soft and then we heat treat it ourselves. And as we've learned, what does metal do during heat treat?
John Saunders: Okay.
johngrimsmo: Whatever it wants.
John Saunders: whatever Whatever it wants to do.
johngrimsmo: early I'm glad you came up that too. um Literally whatever it wants. You can stress relieve it. You can play with it. You can do whatever, but it's going to do weird stuff.
John Saunders: It's going to move. Yeah.
johngrimsmo: And these holes, I haven't proven it yet with measurements, but I'm convinced that they're elongating, stretching, moving distance between them is is not the same. And so I think what's happening is it's locating in the rear hole and the front holes have forward clearance. And in the front is where the blade smashes against this end of the lock insert during the spinewack test. And if those forward holes have clearance before they contact, this whole thing is compressing.
John Saunders: Sure.
johngrimsmo: And I think that is the cause of our failure when the lock fails during thing, because they're the lock insert is literally squishing, failing, and then not working.
johngrimsmo: And the ones that work are contacting on those forward posts first.
John Saunders: Oh, yeah, interesting.
johngrimsmo: And I was never ah under the microscope.
John Saunders: and do you How do you know their contact on the 401ks first?
johngrimsmo: I can see like, like gap in the back contact in the front.
John Saunders: OK. That's cool.
johngrimsmo: And I have a small sample size, but it makes so much sense. Um, and And then Eric has has literally hundreds of failed inserts that like are just in piles. And we're always running out of good inserts because he's always like swapping whatever.
johngrimsmo: And I'm like, well, what if we just like literally with a Dremel open up those back two holes so that they're only locating on the front two holes?
John Saunders: Yeah. Mm-hmm.
johngrimsmo: Let's try it, whatever, they're scrap anyway. Eric goes, they worked.
John Saunders: John.
johngrimsmo: I'm like, no way. Okay. and And he's been pushing me to just hard mill them full stop from raw material, which is definitely an option, but it's a lot of work and a lot of cost and a lot of time to hard mill everything.
John Saunders: Yeah.
johngrimsmo: And and you still have tolerance issues as tools wear. um But you lose the you lose the dimensional ah variable of heat treat because when you're milling it, it doesn't move after hard milling.
John Saunders: Yeah, it's been fine.
John Saunders: Yeah.
johngrimsmo: It's good. um and And you're right. I've had very good success with hard milling more and more and more on our blades from roughing and finishing.
John Saunders: Yeah.
johngrimsmo: The tools are very expensive, but it does work.
John Saunders: I guess I'm defensive because I know how game changer has been for us and like all the problems you face and downstream stuff and and the tools have lasted way longer than I thought.
johngrimsmo: Mm-hmm.
johngrimsmo: Think so?
John Saunders: And we're still roughing with a Haas tooling three sixteenths, I believe is YG1. And then we use the Maldino or somebody else for the, they're they're more sensitive.
johngrimsmo: he Yeah.
John Saunders: So I don't want somebody who's listening who maybe thinks they should be hard milling, but they're scared off. Like, don't like god do it and you make your own decision, but like, it's great.
johngrimsmo: Yeah. Yeah, just you know take all this for the grain of salt we're valuing it as. But hard milling is great. um But it's like i I have a relatively stable process except for the one variable of everything does whatever it wants in heat treat.
johngrimsmo: And I'd have to like reprogram all new speeds and feeds, load all new holders, buy all new tools. There's like 10 tools that make a lock insert. And I'm just kind of avoiding it. But I think at least short term, I can literally just in the program, open up those back two holes so that they are clearance and then exactly they're over-constrained and then just continue and we might be good, good enough for now, solve other problems, you know?
John Saunders: Yeah, because they're over constrained.
John Saunders: Sure.
John Saunders: You don't want to make it the way you make it now. Heat treat it I'm sorry, make it the way you make it now, except leave a couple of thou on those features and then come back and just do a quick hard mill with just one tool, basically.
johngrimsmo: It's very small and hard to hold. And I mean, I'm making up excuses. Sure. Some people could do that or some people would do that.
John Saunders: What I'm hearing is just you, you, you love not having to change everything except it doesn't work.
johngrimsmo: Kind of. Yeah, you're right.
John Saunders: So it's like, so stop, like fix it.
johngrimsmo: Yeah.
John Saunders: We've talked about fix it, fix it in a way that me may not love in terms of efficiency, but guess what? You're really good at then taking the thing that does work and backing into like, okay, let's be more efficient about how we make this, but like make it work first.
johngrimsmo: okay Which I think just opening up those two holes will make it work.
John Saunders: Yeah.
johngrimsmo: But you need them.
John Saunders: delete the two Delete the two. Oh, you need them to mate in the.
johngrimsmo: Yeah, they all have to do something.
John Saunders: Yeah, structure sure, sure, sure.
johngrimsmo: But they can be clearance. They don't have to do anything anymore. um
John Saunders: Yeah.
johngrimsmo: But it still does leave the variable of it moving in heat treat.
John Saunders: I
johngrimsmo: And it just kind of masks it and gives it enough room to wiggle more or less most of the time.
John Saunders: don't love that.
johngrimsmo: Whereas hard milling is is the ideal solution would just make perfect parts out all the time.
John Saunders: that
johngrimsmo: I'm just saying i have so many other things to do. If I don't have to go down that rabbit hole, I'd rather not.
John Saunders: What's this is, this part is you're describing it It's like a stick of Trident gum, if that makes sense. or like
johngrimsmo: It's like half of a mini stick of tried income.
John Saunders: They're like, okay. Yeah. Go order, um go spend a hundred, 200 bucks and order five or 10 of them EDM'd by somebody.
johngrimsmo: There's, it's not just a 2d feature. There's, there's heights and there's surfaces and there's arcs and counterbores and engravings.
John Saunders: Oh, it's not. Okay. Okay. Okay.
johngrimsmo: Yeah. Not to say it can't be like sinker EDM because it could be sinker EDM, but yeah, sure.
John Saunders: I was thinking like isolate proof. No, no, no. I'm thinking wired. Just as, if anything, just as like a test, um you know, this is, this has come up here and other, it's going kind of quick back to this fishbone diagram of like what with you were doing with like, oh we switched it to the ones that had the back holes drilled out and the knife now worked.
John Saunders: And then go back to one that's over-constrained or like over-force it on the back ones or force it sloppy and make sure the knife fails again. Like you're really isolating out.
johngrimsmo: okay Yeah, like prove of the theory, right?
John Saunders: Mm-hmm.
johngrimsmo: Yeah. Yeah, I think we're about there to add more data to the conclusion and then figure out figure out the best path forward.
John Saunders: Yeah, awesome.
johngrimsmo: So it's good. It's really good.
John Saunders: That feels really good.
johngrimsmo: Yeah, it does.
John Saunders: Good for you
johngrimsmo: And Eric and I were having this headbutt a couple days ago, which was actually productive and and happy to do. of of he's like, why don't you just hard mill it?
johngrimsmo: And I'm like because I need to understand the problem. I need to understand what's happening so that the next knife I design doesn't have the same problem.
John Saunders: you. Yeah, sure.
johngrimsmo: um And I'm not always just choosing like the hardest, most expensive solution for the problem without still really understanding what's going on. So i'm I'm glad I think we have a good theory. And yeah, hard milling might still be the answer. That's what I told Eric too. I was like, it still doesn't like answer the problem.
John Saunders: Yeah, I'm with you.
johngrimsmo: Like I need to know why the 5,000 inserts we just made or whatever, some work, some didn't. you know As a manufacturer, like that drives me crazy.
John Saunders: Yeah.
John Saunders: Yes.
johngrimsmo: And I didn't actually know it was the insert as the source of the problem until now, if it that is the case. But yeah, so that was good.
John Saunders: Yeah, that's fine. good Yeah, keep keep me posted.
johngrimsmo: And as I was telling you earlier, that's that's one of those like like current game-changing problems that I want to fix right now. um Rask bevels is another one.
johngrimsmo: Lock inserts is one of these.
John Saunders: Yep.
johngrimsmo: and These are like big things that if I can just wipe them off the table, like Eric's life will be infinitely more productive making Norsemen.
John Saunders: Oh, yeah.
John Saunders: Sounds like those are both doable, though, John.
johngrimsmo: Oh, totally. i'm I'm doing them now for sure.
John Saunders: Yeah, good.
johngrimsmo: They'll be done in the next few weeks. But like it will like Eric spends, I'm making up numbers, but somewhere between 20 minutes on a good Norseman and 200 minutes on a bad Norseman to like finish it up.
John Saunders: Possibly, yeah.
johngrimsmo: It's like dumb.
John Saunders: Yes.
johngrimsmo: So
John Saunders: Good. That's great. Did I share um Courtney or Courtney and Grant working on measuring the raw material with the robots?
johngrimsmo: You mentioned it, like bumping it and and getting a sense.
John Saunders: Yeah.
johngrimsmo: Yeah, is it working?
John Saunders: Yeah. No, I just wanted to make sure I shared it because it's kind of relevant to how we're looking at ties into the other issue we're having of sourcing material where we've we've had gotten, we've kind of fallen love but fallen in love with certain vendors and you end up either knowingly or unknowingly sole sourcing.
John Saunders: And I'm like less willing to do that now.
johngrimsmo: Mm-hmm.
John Saunders: We have had some hiccups and problems and kind of to like just put on your indifferent hat. It's like, no, if i'm if I'm really auditing Saunders from an outside standpoint, it's like, no, you need to have multiple vendors making sure you're not And the like same thing, it's kinda like we're going back to your whole, like, what's the problem here? Like we're having problems of material availability, quality, so other stuff. And it's kinda like, why are we having this problem? Like no vendors actually like seeing there thinking like, oh let's just like, let's just screw this up. So like, where's the problem happening? And a lot of vendors,
John Saunders: themselves are using subs to do other services. so like we're reevaluating all of that, but some of it ties back into like, okay, great. If we can have,
John Saunders: Some of also just ties back to the way the ModVice material is cut from plate. There is no real bar product, if you will. And when you cut from plate in but in volume, there's going to be variances and being able to account for those variances with a robot that can either adjust the vise or if it's over a certain value or under, you just have the robot put it in a fail bin and move on with it.
John Saunders: We can just eat that or we can go back to the vendor and say, hey, these are way outside of the spec like you owe us or whatever. Feels like a big win there.
johngrimsmo: yeah um i'm going to throw a wild card at the table that I'm sure you've thought of. So a modvise body is what you're talking about, right?
John Saunders: Yep. Correct.
johngrimsmo: It's like a 4 inch by 4 inch rectangle of material or something like that?
John Saunders: Two by six for the bases and then the top jaws are effectively the size of a deck of playing cards.
johngrimsmo: inca
John Saunders: Two and a half by four.
johngrimsmo: 2x6, eh? They're really longer and wider than I thought.
John Saunders: Yes.
johngrimsmo: I don't know. i'm just thinking about round bar and how it's typically like more consistent and extruded and like cheaper sometimes.
John Saunders: so she
johngrimsmo: And like not out of full rounds, but out of slivers, like end slivers of a thing. I'm sure you thought about it. but Yeah.
John Saunders: It's, so yes, we have, it sort of could work work on a top jaw, except it doesn't because you end up, it just it's it's a great thought, it doesn't work. um It doesn't.
John Saunders: And we're also, we're really good about the way we make the mod devices now. We don't actually have that much material removal, which has been, a which is good. Like, less chips, less tool a life, less machining time, less everything.
johngrimsmo: Yeah.
johngrimsmo: Good. So with not single sourcing a vendor, ah won't you still get to the point where it's like you have one favorite vendor that's like the best price or the best quality or the best whatever, and you just prefer to use them?
johngrimsmo: Like, i feel like there's always going to be caveats if you have a top three, you know?
John Saunders: Yeah, so there are, um but knowing you have a backup, even if you don't use them a lot, is still good. And look, we went we ended up going back to a vendor that we haven't used in a while.
johngrimsmo: Yeah.
John Saunders: And I was a little bit self-conscious, if I'm being really honest, like, oh, are they going to be upset? And the reality is no. Like, they don't necessarily know or even care that we didn't buy it from them for a couple of years.
johngrimsmo: Yeah.
John Saunders: The reality is I'm being like, hey, ah RFQ for five figures of this material. like they Like, I'm being way too self-conscious about
johngrimsmo: Yeah, exactly.
johngrimsmo: Good.
johngrimsmo: Last thing I've got is quick update on the interferometer.
John Saunders: Oh, yeah, yeah.
johngrimsmo: Spent some spen some nice evenings on it. I got one amazing scan that's almost perfect, and then it got worse again.
John Saunders: Uh-huh.
johngrimsmo: So I've isolated the vibration from the shop. I hooked up my piezo vibration sensors to like everything. to see vibration and like, you know, tap on the table, walk through the house and my inner tube floating mount using the bicycle inner tube with the heavy weight on top works so well.
John Saunders: So cool.
John Saunders: Yes, yes, yes. Mm-hmm.
johngrimsmo: The fact that it's on my watchmaker's desk from the 1800 it's like wooden and adds vibration. If it were on granite countertop or something like that or on concrete, it would be so much more stable, but it's good enough.
John Saunders: Okay.
johngrimsmo: So it's not vibration. It's not my problem. It's I think the problem is computational. algorithm of how it's interpreting every pixel of the fringes as they scan across.
John Saunders: Oh.
johngrimsmo: So that's what I'm going back and forth with cursor like a lot with like, nope.
John Saunders: Okay.
johngrimsmo: And I literally screenshot the 3D model that I'm getting and show how like wavy it is, paste that into cursor be like, this was the latest result.
John Saunders: it was like the way it's it's, I'm assuming it's the software's doing some sort of a stitching.
johngrimsmo: Yeah, stitching is not the right word for it, but it's um analyzing what's it called brightness, something of each pixel so it can sense when the fringes are sweeping across the part light, dark, light, dark, light, dark.
John Saunders: Okay, okay.
johngrimsmo: And it's analyzing like a single pixel throughout every frame, thousands of frames of footage.
John Saunders: Yeah.
johngrimsmo: ah And then
John Saunders: and
johngrimsmo: Basically when there's no fringes, the pixel will stay the same brightness like for the whole video feed, more or less. But when the fringes go by, it goes light, dark, light, dark, light, dark, light, dark. And then it uses each pixel and mapped every pixel across the whole thing over every frame and stitches it all together.
John Saunders: Yeah.
johngrimsmo: And I'm just fine tuning that software side of things.
John Saunders: Dude.
johngrimsmo: So it's like, i it's so close.
John Saunders: This is awesome.
johngrimsmo: Right now I'm scanning a gauge block and I can see it's a 1.5 millimeter gauge block. So I can see the 1.5 lasered engraved into the surface, but there there are scratches on this gauge block that I can't yet resolve with my my program.
John Saunders: Okay. Okay.
John Saunders: Oh, interesting.
johngrimsmo: And so I can get a relatively flat plane, but it doesn't show the scratches. So I'm not accurate enough yet because it it is definitely supposed to.
John Saunders: You see the scratches, it just there's know what they are?
johngrimsmo: Exactly.
John Saunders: Okay, interesting.
johngrimsmo: So I'm close, but I'm not, I'm nowhere near where it needs to be, but it's fun.
John Saunders: Yeah, yeah.
johngrimsmo: It's like super fun. And I will know when I have it, you know, it's like, once I see those scratches and can tell that they're, you know, 300 nanometers deep or whatever, that's, that's a win right there.
johngrimsmo: So I'm close.
John Saunders: I was thinking of you because i William and I have gotten a little bit more involved with in our astronomy club.
johngrimsmo: oh
John Saunders: And we William and i built this telescope over the winter.
johngrimsmo: Good.
John Saunders: um Now, when I say build a telescope, we bought a mirror kit, we printed a lot of them.
johngrimsmo: Yeah, still.
John Saunders: It's a Newtonian reflector, but still, it teaches you to go through the process of how it works.
johngrimsmo: Yeah, it's huge.
John Saunders: And one of the guys in our astronomy club, i like i just I want to hang out with him more and more and more, because he has not only made multiple telescopes, but on almost every one of them, he has he has lapped in the parabolic mirrors himself.
John Saunders: And so he was talking about like the wavelengths and the acceptable errors.
johngrimsmo: Oh.
John Saunders: And I'm like, oh my God, this is so like, yes. Like it gives me, it's it's just fun to hear.
johngrimsmo: yeaha Yeah, that's what Adam's doing, Lany Machine Tech.
John Saunders: Yeah.
johngrimsmo: Grinding his own.
John Saunders: I should talk to him about that.
johngrimsmo: what
John Saunders: Yeah.
johngrimsmo: He loves talking about telescopes.
John Saunders: Yeah. It's really fun.
johngrimsmo: e Good for you guys.
John Saunders: Yeah. And sorry for my voice this week, folks. I've come over off a head cold and feel like I'm not sounding great.
johngrimsmo: You sound sultry.
John Saunders: Yeah. it'ss It's just, yeah, nothing personal.
John Saunders: My last comment was um really happy again. I've been talking a lot in my own head and with the team here and even with you about Home Assistant and it hadn't really played come into the role I wanted it to.
John Saunders: And now it really is. We've got only two cameras, but the two cameras that we have on it, we're using all the time.
johngrimsmo: Really?
John Saunders: The guys are able to check it remotely. We can do automations on it, but we're kind of, we're playing it slow. Sorry, shouldn't say automations. We can do like visual detection with it, but I want to be very deliberate and not create a lot of noise.
John Saunders: on it We're doing, we finally have some sensors on it for like water leaks. We are putting, starting to move plugs off of the Alexa platform onto Home Assistant. But I have mixed feelings, not to go too far off topic, but like I have, I've had it at home now for a few months with my garage door open and closing based on when I come home or leave. And it works awesome until it doesn't.
John Saunders: like
johngrimsmo: Again.
John Saunders: uh once a month or more it'll get somehow get out of sync and it won't work or it's really slow or laggy and so when i think about for example um the most beneficial to us would be when every machine has stopped running to shut the ball valves on the air compressors just to stop any especially on the two or three day weekend stop any unnecessary cycling of the compressors but um I have a zero tolerance for it ever accidentally shutting the them off when we're actually running or or tapping or facing apart.
John Saunders: And we were already brainstorming on it. It's like, well, if it wants to shut them off, it could probably send us some sort of an alert or email and be like, hey, seems like everything's idle. If nobody responds in the next 15 minutes, I'm going to shut the compressors off.
johngrimsmo: That's not a bad idea. Yeah.
John Saunders: We're like just talking out loud because I want to make sure this works for us and we just don't get
johngrimsmo: Yeah.
johngrimsmo: Or even opposite. i need a confirmation from somebody, not a 15 minute, no, no quote, but a, somebody needs to click.
John Saunders: Well, no, cause this will be, well, like the brothers are now running till two in the morning and then at 3 a.m.
johngrimsmo: Yes. don't think so.
John Saunders: it hasn't run for an hour.
johngrimsmo: Oh yeah. Sure. Sure.
John Saunders: Now I want it to stop.
johngrimsmo: Yeah. That's a good point.
John Saunders: We'll figure it out.
johngrimsmo: Yeah. I know you will.
John Saunders: It's more, I'm proud that we've actually, we're actually executing on the vision we had of like, hey, let's use Home Assistant to work for us.
johngrimsmo: Yeah.
John Saunders: Everybody can log in, everybody can access it. We can set up routines, we can set up alerts and notifications. um Actually, that's a good point too, is you can, yeah, I guess.
johngrimsmo: Yeah, no, I mean, I sat in on Nick's class at the summit and I was blown away.
John Saunders: Yeah. it lu
johngrimsmo: Like I, I'd been sleeping on this home assistant thing. And I looked into it a little bit a month or two ago after the after the summit event and got all excited about it.
johngrimsmo: And then reality came back down and i was like, i have so many other things to do right now that this is actually not that valuable to me at this point.
John Saunders: Yeah.
John Saunders: Yeah.
johngrimsmo: So I put it back on the shelf, but I'm glad I know about it. I'm glad it's working for Nick. I'm glad you're finding excellent uses for it. And it's it's fun to watch you you know make wonderful use of it. And I'm sure a lot of people listening are like, oh, I've heard about it. I don't know what it is. I don't.
johngrimsmo: Yeah, for the right use, it's pretty amazing.
John Saunders: It takes time, like to be just, like it's kind of like a be a surgeon, like you actually, it's not plug and play.
johngrimsmo: Yeah.
John Saunders: It's not Apple ecosystem just worth. Like it's, you know, you buy a new plug and it's a Matter, so you have to sync Matter plug, you have to sync it that way, that like an app and the permissions. And then, you know sometimes you're modifying hacks code or YAMLs, which I thought was hilarious.
John Saunders: You ever heard of YAML files?
johngrimsmo: Maybe, but not.
John Saunders: y il it It stands for yet another markup language.
johngrimsmo: Yeah, yeah.
johngrimsmo: Oh, no way.
John Saunders: Like the the guy who invented it was a little sultry himself.
johngrimsmo: Yeah.
John Saunders: Anyway, it's been fun.
johngrimsmo: Good. get there So you said you got two cameras online.
John Saunders: Yep.
johngrimsmo: You're using them as web cameras, basically, webcams.
John Saunders: Oh, actually we have three online. um Yeah, web cameras that have really good, easy playback.
johngrimsmo: Okay.
John Saunders: They're super high quality.
johngrimsmo: Yeah, that's what's cool about the whole thing, though, is like from the app on your phone or on your computer, everybody in your team has access to it.
John Saunders: One...
johngrimsmo: You have those webcams. Question, has it been discussed the whole now you have security cameras in your shop kind of thing, like in the shop?
John Saunders: We've had them for, actually we've had them installed when I moved in here, which is just fine, I don't care.
johngrimsmo: OK.
John Saunders: I have checked our security cameras like three times in my life. Like um they are not tools to check or anything.
johngrimsmo: But I know some employees are like weird about that.
John Saunders: Yeah, we're not. I don't use them for any, purpose and these cameras are frankly aimed at and into machines.
johngrimsmo: Like add a machine. Yeah.
John Saunders: So for purposes of what some companies would look at from safety or theft, like that's not what these do. We have a camera that's aimed at certain spots just to be able to use it for security purposes.
John Saunders: um But I don't think it's a,
johngrimsmo: Yep.
John Saunders: i don't think it's a I'm not aware of why it's a real cons.
johngrimsmo: Mm-hmm.
John Saunders: I'm not, but I'm not worried about it because it's been my summer.
johngrimsmo: Good.
John Saunders: You don't use them. so actually probably should have asked this.
johngrimsmo: here
John Saunders: right Don't answer that.
johngrimsmo: Yeah.
John Saunders: Yeah.
johngrimsmo: Moving on.
John Saunders: Move on.
johngrimsmo: Yeah, that's all I got on my list.
John Saunders: Good. Good talk.
johngrimsmo: Yeah.
John Saunders: I'll see you next week.
johngrimsmo: See you next week. Okay, bye.
