Transcript
johngrimsmo: Good morning and welcome to the Business of Machining episode 405. name is John Grimsmo.
John S: And my name is John Saunders.
johngrimsmo: I'm kind of sick, but I'm here. Yeah, yeah this is a podcast where two manufacturing buddies talk about the state of our business and what's going on and what's new, what's what's fun, what's not fun, what's on our minds, you know?
John S: What's on our minds? Yes.
johngrimsmo: Yep, always lots on our minds.
John S: Yes.
johngrimsmo: Yeah.
John S: ah Agreed.
johngrimsmo: First one, as I talked about last week, so my, what's it called, Gorilla Stand?
John S: Gorillacan. Okay.
johngrimsmo: The Gorilla Stand came in The indicator base, the ultimate one, made by Bastion in Germany. um And I sat on the package for a while because I'm like, I want to Like film this. I want to do a little unboxing. i want to show it. it's It's cool. So I did. i came back one night after the kids go to sleep. And I'm like, I'm just going to nothing else. Nothing else matters. I'm just going to do this.
johngrimsmo: And it was absolutely wonderful. It was just a good experience. I got a new necklace for my GoPro. So I can hold the GoPro.
John S: What?
johngrimsmo: You know, there's like the chest rig that people use, but that's annoying.
John S: Yeah. Nobody did use like...
johngrimsmo: Yeah, exactly. But this is literally clip-on necklace that holds the GoPro at a great angle. And it is the coolest POV angle.
John S: Whoa.
johngrimsmo: It's just really cool. So I filmed...
John S: And it doesn't seem all bouncy.
johngrimsmo: No, ah the new GoPro cameras. Mine's an 8. It's not that new. It's got such good stabilization in it that it works great.
John S: No, I hear you there. We have an 11 and I realize you don't need the gimbals anymore because they're so good, but it feels like it would be weird.
johngrimsmo: No, it's amazing.
John S: Is it a GoPro product or just an Amazon accessory?
johngrimsmo: Amazon, yeah, $30.
John S: Interesting.
johngrimsmo: Super cool. Super cool.
John S: Okay.
johngrimsmo: um like I wouldn't ride a mountain bike with it because it will bounce around. But for shop stuff, it's fantastic. Just walking around.
John S: Interesting.
johngrimsmo: So super cool. um Anyway, the indicator stand is impressive. the The packaging comes in a big case with custom cut foam.
John S: Hmm.
johngrimsmo: um It's got a steel base. It's really, really heavy. It's got lapped steel round washers for the feet. So three feet, each one swivels on its own little swivel joint.
John S: Okay.
johngrimsmo: um
John S: Interesting. So this.
johngrimsmo: it's It's super cool how we did it because The adjustment is moving the third foot of the triangle, which tilts the whole base, which requires all three feet to be able to tilt.
John S: Yes.
John S: I see. Of course.
johngrimsmo: As opposed to a lot of indicator stands are totally flat on the bottom and have the little lever on top.
John S: The flexure though.
johngrimsmo: Yeah, the foot or the flexure.
John S: Yeah.
johngrimsmo: um So it's pretty cool how he did it. And everything is just dialed, machined to the nines, brass, steel. um The rods are even ah like polished.
johngrimsmo: um
John S: Yeah.
johngrimsmo: even if you just put it a lathe and wrap and rub some sandpaper on it, it's, they're, they're nice. It's like, it's really, really cool.
John S: Mm-hmm.
johngrimsmo: Really well thought out. and And I, I locked down my 20 millionth, uh, Mar super mess indicator and zeroed it on a gauge block.
John S: Mm-hmm.
johngrimsmo: And then but went home and came back the next day and it was, it didn't drift.
John S: Yeah.
johngrimsmo: So that was, that was pretty cool.
John S: Yeah.
johngrimsmo: Um, yeah, so just, I filmed a little good video. It's beautiful product. And, uh,
John S: You haven't put it up yet.
johngrimsmo: Super excited to use the video. No, I haven't yet.
John S: i have not. This has been okay.
johngrimsmo: It'll take some time.
John S: No, that's totally fine. I have not been idle this week. I'll say that way. And I just see that you put up a few of the which I've not watched.
johngrimsmo: Yeah.
John S: And I was like, well, maybe I missed Stan video. but Okay. Awesome. look Well, I'm looking forward to it.
johngrimsmo: Yeah.
John S: We,
johngrimsmo: Yeah. I had you in mind, had other people in mind as I'm filming this. And, um, It was fun to be able to geek out on a very technical, very specific project.
johngrimsmo: Like not just unboxing, but showing how I use the indicator and some things I don't know some things I haven't done before. Like it's got one of those front bumpers, so you can do a, you put a ah cylinder square on it.
John S: Yeah, the rolling. Yeah.
johngrimsmo: And I've never done that before, but I've seen Adam and other people do it. So I talked about that a bit.
John S: Tom Lipton has a great video on um showing the use of that.
johngrimsmo: and Oh yeah.
John S: Well, that's the, ah so we, we, we proudly have a gorilla stand cat 40 type holder.
johngrimsmo: I've probably seen it.
John S: And, um, I want to hear more about this, but I will say, ah like if you go Google gorilla fab or gorilla indicator stand, there's nothing. So it's like, um, I want to see, i want to see some content.
johngrimsmo: Oh, not that you can't find their website, that there's no juicy details out there. That's what you're saying.
John S: Yeah. Is that what you i mean? For the first time I hear about a product like this, like when CJ is talking about some vibratory Tumblr or Dennis is talking about the third quasar he bought this week, it's like, I'm going to go Google it and see what it is or YouTube it and see what it is.
johngrimsmo: Yeah, you look for stuff. Yeah.
johngrimsmo: Exactly. So hopefully our video will be will be that because it's a good deep dive to the point where I was texting with a Gorilla Fab guy and and I was like, oh yeah, I took it apart and I showed these things and he's like, you know maybe don't show everything inside because I'm kind of proud of some of those things.
John S: There you go. Good.
John S: yes Yeah. Yeah.
johngrimsmo: And I'm like, okay, fair, fair. That's fine.
John S: Yeah. The, to your point though, there is something to be said in some of this we learned from, well, trials or other mentors or specifically when Mitsutui came, came out here years ago and we'd filmed some stuff together.
johngrimsmo: Yeah, remember that. Yeah.
John S: And it's like, okay, you know, you want to think about how you present the part to the indicator. Are you moving the part? Are you moving the indicator base? Where are you leaning on the, because a lot of times we lean on tables or push on stuff.
John S: um With the this cosine error, these are all things that may not matter, but they may matter. And you at least should know. um Cosine angle, I think for sure being the most regularly impactful, which is just how steep is the indicator needle relative to the part plane.
John S: Anything over like 15, 20 degrees gets pretty significant. um
johngrimsmo: Yeah.
John S: But then um air you know air gaps, dust, like feeling how good the part, like it's.
johngrimsmo: Yep. And I, I went over some of these things in the video, um just kind of broad strokes, but like, you know, wiping the the granite down with your hand to feel any dirt.
John S: Okay.
johngrimsmo: um Even remember that Dave Arneson's 50 tidbits of the shop thing.
John S: Yes. Yeah. a Precision instruments guy.
johngrimsmo: So one of those, yeah, is, is actually stone your stone. Like,
John S: Yeah.
johngrimsmo: stone the granite plate. um So talked about that a little bit and I did it and it's like, yeah, skate's good. You know it's clean, you know it's good um because you want everything to slide around really nice.
John S: We've got the, we have a brown and sharp 50 million dial test indicator. And then we have, we have a bunch of tenths indicators, but the 50 million one, and then we have two millimuses.
John S: One is micron, one is actually, I think inch.
johngrimsmo: Interesting.
John S: Yeah. Those are probably the three most highest resolution hand tools, if you will.
johngrimsmo: Yep.
John S: But I'll tell you it is pretty easy to get varying results on those. um Even with the CeraBlock, just depending on, noise in the system. um Oh, we have a, we have a old school, actually tip of the hat to Tom Lipton who gave this to us, a digital electronic indicator, which is freaking sweet.
johngrimsmo: Yep.
John S: Oh my gosh.
johngrimsmo: yeah
John S: That is really nice, but um really makes you appreciate to making sure you get a repeatable measurement first on your test artifact before you start trying to measure your part itself.
johngrimsmo: Yep, even the how you slide the part under the indicator tip, you start to see variation or speed or bounce or whatever.
John S: Yes.
johngrimsmo: And I'm not the best example because our granite plate is an old CMM stand with holes all over it, which is not not a bad thing, but it's on a really crappy base.
John S: Yeah,
johngrimsmo: That's on four feet, but it's really only on three feet because there's posted notes under the forfe fourth feet.
John S: Got it.
johngrimsmo: It's not level. It's not balanced. If I smack the the granite, you see it in the indicator tip. The whole thing's vibrating for 30 seconds.
John S: Mm-hmm.
johngrimsmo: So I know it's not right, but I don't really want to spend $1,500 on a new sand ridge and base and all that stuff. um
John S: Yeah, the grade was only 800 bucks, but the base is...
johngrimsmo: Right. But you really want the bass to make it nice. And I want it for sure, but not right now.
John S: On that note, that's the other thing. and I don't know if this is best practices or just personal preference, but when you have an an indicator, so a test indicator that has a a record needle, if you will, not a straight up and down dial indicator.
johngrimsmo: Yep. Plunger style.
John S: Yeah. um And if that indicator is presented at like 10, 15 degree angle, I will, and I know we're on audio from everybody else, not video. I will not bring the part toward the whole indicator from away.
johngrimsmo: It makes sense.
John S: I actually start with a part kind of underneath the body of the test indicator and I push it away.
johngrimsmo: Yeah.
John S: um don't actually know. I think that would, it does make a difference in terms of what that initial force is,
johngrimsmo: yeah
John S: um to the ball and like indicator tips where I don't know that I've ever seen evidence myself, but it's why they sell different tip types. Anyway, I ended end up pushing the part away from my body and away from indicator to get it underneath the the needle, if you will.
johngrimsmo: And these are all the tips and tricks that you and I have picked up either firsthand, just figure it out or watching Tom Lipton videos or just over time you amass this information.
John S: Yeah.
johngrimsmo: And I think about teaching this to some of our other staff who's never seen it before.
John S: Right.
johngrimsmo: what What is it? What does it do? um There's a lot of finesse with, you know, ah measuring sub 10th.
John S: Yes.
johngrimsmo: And it's a super fascinating. I absolutely love it.
John S: Can i change the conversation?
johngrimsmo: Yes, I'm done.
John S: um i have to use the L word, so I apologize to of our fans out there, but I do need to talk about lathes. um
johngrimsmo: I'm like, what's the L word?
John S: Oh, come on. What's the L word? um Why would a lathe not turn apart perfectly round?
johngrimsmo: That's a really good question. I've been thinking about this. You probably saw Silo's cylindricity video or something.
John S: I started, I've seen, he put out the last like 48 hours.
johngrimsmo: Yeah, i I fell asleep to one of them, and turns out I watched the whole thing.
John S: I've not watched all those two.
johngrimsmo: but
John S: So again, to to what I think is the quick backstory is we made a really well designed hard gauge to test diameter, but you also end up checking what I guess you would consider cylindricity or total indicator reading of of how something could vary.
John S: And we took a turned part off the will in it. So this is a... brand new machine that's a very expensive machine.
johngrimsmo: brand new, yep.
John S: I might say it's perfect, but like, this is not a clapped out clausine from 72 here. And we're seeing somewhere, I believe, between two and four microns of diameter variation across a one inch part.
johngrimsmo: It's like one to two tenths kind of thing.
John S: um
John S: Yeah, i think i can think that would be what it is.
johngrimsmo: Yeah.
John S: um
johngrimsmo: and And we see that on our Swiss too. We we definitely see a trilob of... um one to two tenths, usually one, but if you rotate the part and check again, you're like, wait a minute, you rotate the part again, wait a minute, what the heck?
John S: Okay.
John S: Yeah, yeah, yeah.
johngrimsmo: So what?
John S: if the bearings were, if the bearings in the spindle head slot, that would make sense.
johngrimsmo: I
John S: Otherwise, I'm not sure I understand kinematically where this comes from.
johngrimsmo: i don't either, to be honest, but I know it's a thing and I know it's real and I don't know why exactly.
John S: Okay, okay.
John S: Okay.
johngrimsmo: um in ah In a Swiss lathe, because there is play in your guide bushing where your material is literally rubbing and spinning in a fixed guide bushing or pushed against the end or something, I get some of that.
John S: Sure.
John S: Bouncing around.
johngrimsmo: um ah I don't know. I just know it's a thing. So I was thinking last night because I saw silos video and I was like, well, we have a CMM and we could easily take one of our turn parts and just do a cylindricity scan of one of our turn parts just to see what's actually happening.
John S: Mm-hmm.
johngrimsmo: And it's, it's the Duramax CMM, but it's plenty accurate enough to show this kind of thing.
John S: Okay.
johngrimsmo: um I don't know if we'll ever get around to doing that, but it would be really interesting to do, you know?
John S: Yeah. So you have noticed us on the very, very capable, high-end, nice Swiss machining, Tornos, and have you ever checked it on but on the Noc?
johngrimsmo: Yep. Yep.
John S: Okay,
johngrimsmo: Probably haven't cared.
John S: yeah.
johngrimsmo: Although a lot of those parts we used to make on the NAC, but didn't care as much five, six years ago.
John S: Yes, for sure, sure, sure. It's just funny because, again, we do other work on that machine where we do tend to hold quite tight tolerances and you would think a traditional lathe stick tool where there isn't material, there's material amounts of tool deflection or the inconsistent tool deflection because you're turning a fully contact it's not interrupted cut in stable material and a good work holding setup you're doing it all really quickly so it's not like there's thermal changes throughout the cut length like just not sure
johngrimsmo: oh
johngrimsmo: Yeah. And why is it is it consistent like across the length of your part?
John S: and parts are short that's a good question
johngrimsmo: you know Say it's a trilob or say it's a two-lobe. Does that feature go along the length? Even if it's short, it doesn't matter. Or does it like evolve and rotate? Yeah.
johngrimsmo: Because theoretically, i mean this is what a lathe does. It spins your material really fast and a fixed cutting tool just starts chipping it away in a straight line.
John S: Yes.
johngrimsmo: you know You move straight in X or Z, I guess.
John S: Right. Z. Z. Come on, John. Even I know that.
johngrimsmo: It's been a while since I've programmed a two-axis lathe. But...
John S: But like if i if someone told me, hey, here's a million dollars, I need you to build a phenomenally accurate, incredibly well-designed lathe, but I want you to have it intentionally induce three microns of variation, trilobularly in the part, I'm like, yeah, i got nothing.
johngrimsmo: Yeah, yeah, yeah.
John S: Like, i mean, I guess bearing slop in the spindle, right?
johngrimsmo: Yeah, it's fair. Yeah.
johngrimsmo: Yeah.
John S: um You hear about this with cylindrical grinding in twist drills, but twist drills suck. like what We're talking about much higher levels of material removal capability machines here.
johngrimsmo: o
John S: And you hear about cylindrical grinding where it becomes trilobular. I do believe that has to do with some, I i hate to use the word wrong words here, oscillations or vibrations or harmonics of the material as it's getting gripped.
johngrimsmo: yeah
John S: think there's two lower g grippers and rollers and in a top grinding wheel and it gets fed through. um
johngrimsmo: Like for centerless grinding.
John S: so
johngrimsmo: Yep, yep, yep.
John S: Bingo. Yes. Is that what i said? Okay.
johngrimsmo: Whatever.
John S: Yeah. But like, I don't, I don't tend to hear about this. Like we a little Herod grind all, um, that has obviously super tight, good bearings. I assume if I put a part in that dial and stuff in as needed go run that under rotate that part underneath our surface grinder wheel, I wouldn't expect to see similar, uh, try lobby a Larry right.
johngrimsmo: I don't know. And then to measure a trilo properly, don't you need a V and then an indicator on top? You can't just measure it from two points, like with a micrometer.
John S: So we are, and that may be wrong, we built a gauge that is effectively the same thing as thread wires, but the gauge is, I give Alex a really nice job.
johngrimsmo: Okay.
John S: it's um It's got, this this this has the tiny linear rails from McMaster car on it.
johngrimsmo: Okay, yeah.
John S: And you know we tried to factor in Abbey's error and there's alignment things and preload things and all that. And it's we've tested it against a master and it's giving you it's giving you diameter at a single angular orientation. So if you just rotate the part and let it, the nice thing with the springs is it doesn't, it means you're not, how you hold the part isn't impacting the measurement value. And it does repeat very consistently. So we're trying to think through these kind of like half layman, half engineering aspect of like, let's make sure this thing is repeatable.
John S: And we do care about, ultimately we do care about its accuracy, but we at first care about its ability to be repeatable.
johngrimsmo: Yeah.
johngrimsmo: And the real point here, like, are you noticing something and now you can't unsee it? Or is it a problem in your calibration test assembly?
John S: i think um I think what's going on is we soft turn the A2 or S7 pull studs, send out for heat treat, and they come back, and then we do a hard mill operation.
John S: And the new gauge is helping us better check how they're hard milled.
johngrimsmo: OK.
John S: um And there's a public spec, and there's kind of an internal spec, and we're trying to understand that tolerance. And so Grant made a pretty darn cool, I think I talked about this last week in the podcast, an ID clamping mechanism for the Willowin.
johngrimsmo: Yep.
John S: And I think it worked. I think you used the 3D printed one and that was enough conviction to say, okay, let's make one out of metal.
johngrimsmo: Mm-hmm.
John S: um And then that just is all ties into the same goal.
johngrimsmo: So the critical feature you're hard turning, is it an ID or an OD bore? What is it?
John S: It's an OD angle taper on a full stud.
johngrimsmo: You're turning the OD taper pretty much.
John S: That's correct.
johngrimsmo: Okay, just so I wrap my head around it. And of course, you're trying to measure a taper with hard gauges and
John S: Yeah.
John S: The taper is why we have the... Oh, actually, so sorry. This is a great example because i when i some of the things I've been working on this week is making sure when you have to problem solve, just break stuff down.
John S: So to your point, one of the things we should do is stop turning the taper.
johngrimsmo: Yep.
John S: We should just turn a round part of the same diameter and test it for circularity to see if there's potential the fact that you then have the rotation plus two axes versus just one um plus obviously a round part is easier to hand mic or measure versus the tapered part it's great call john thank you like break it down sorry
johngrimsmo: That's true.
johngrimsmo: Yep.
johngrimsmo: and And do think about the, if you measure it with a mic, rotate it, measure it again. Like that's where we see one or two tenths.
John S: who
johngrimsmo: of diameter variation, but there is something to that three points of measurement using a V block and a Mike Anvil. So you're, you're measuring a triangle.
John S: Okay. Sure. Sure. Sure.
johngrimsmo: Because if it's trilobular, that'll measure that if it's bilobular whatever.
John S: Okay. I'm with you.
johngrimsmo: So there's, that's kind of the where at with a lot of ultra um intelligent things. Like I know just enough to say a bunch of words and kind of figure it out, but I'm not the educated professor.
johngrimsmo: That's like, this is how you do that.
John S: Right. Right. Which i want to say tip of the hat and to the, you the Stephans, the Robins, the Tomlifters of the world, then the Silas for, I'm happy to stay in my lane as the guy who's built a successful business, but is ah student, not a teacher on a lot of these more advanced topics.
johngrimsmo: Yep. Totally.
johngrimsmo: Yep. Absolutely. Yep. And like I did with the indicator stand, it was fun to talk about from a, I guess I'm more than a layman's perspective, but I'm having fun with it all, you know, like this ultra precision thing.
John S: Yeah.
johngrimsmo: It's it's not my job, but it is kind of our business. um But we're having fun with it. We're learning. We're always progressing. We're not here to profess that we're, you know, know what we're doing, but we're getting there.
John S: Did you see breaking, what's it called? Breaking taps video on the hinges.
johngrimsmo: Breaking tabs? No, I saw the thumbnail, but i haven't seen it yet.
John S: Oh, it's really, he, he, it's great.
johngrimsmo: Is it good?
John S: He does a great job. And, uh, I thought it was an awesome example of the community too, because he was trying to improve the feeling. This is such a Grimso thing. He's trying to improve the feeling of these, um, springs, like torsion springs or whatever springs.
John S: And uh, got a tip from Adam, the machinist to put lapping compound on like a piece of paper and shove the paper in between the spring coils and then run the spring over the lapping thing to lap the ID mating surfaces of the core of the torsion spring to improve, just basically knock off the grittiness because these springs are made a billion at time.
johngrimsmo: Oh, no way.
johngrimsmo: No. and
johngrimsmo: Yeah. yeah
John S: And now you want one that's really feels really good. And, uh, forget, I forget breaking taps actual name, but he was like, oh my gosh, this definitely helped.
johngrimsmo: I love it.
John S: All right.
johngrimsmo: That's cool.
John S: Yeah, don't start hand lapping the Socket Springs though, please.
johngrimsmo: Yeah. No.
johngrimsmo: yeah
John S: Yeah.
johngrimsmo: Although...
johngrimsmo: um I had another fun interesting thing that I did the other day. I've had this um on the Speedio we're trying to hold tents on certain features and we're seeing things drift up and down throughout the day and being a C-frame machine that's not liquid cooled that's accurate. I've got the sub micron the whatever high accuracy package so I have five digit inch of movement but is you know what's what's actually happening but does the machine warm up does it move does it um because we have both a y feature that we're grinding like the bevels of the blade so we're grinding those and my guide says that the first one of the day always cuts weird and then it gets better and then second one's pretty much fine and i'm like well is it thermal growth in the y-axis like your spindles on the head the casting
John S: Yeah, right
John S: here. Okay.
John S: Okay. Yeah.
John S: Yes.
johngrimsmo: And as that gets hot, is the casting growing growing out in Y half a thou, a thou maybe?
John S: judent Normally I feel like what you see is the head nods down.
johngrimsmo: um
johngrimsmo: Sure.
John S: so your Z is going to cut deeper into the part, I think.
johngrimsmo: Sure. And I agree, i've and I've never really heard about the Y thing, but we're kind of seeing it.
John S: You wouldn't think that Y shouldn't grow.
johngrimsmo: So I wanted to kind of figure this out. So I made a probing routine. that probes, I looked at the table, I'm like, what's a fixed thing that never moves and is never in the way of something else?
johngrimsmo: I can't just probe this fixture because I could change it and have another fixture. So I used the laser body. And I made a probing routine that just comes down and touches the same point on the laser in X, and then touches a web in Y, so it measures the width and center point, and then touches the top of it in Z.
John S: yeah know
johngrimsmo: And so I'm thinking to myself, the casting is cast iron. The bloom laser is aluminum, different thermal expansions, like maybe, I don't know, but this laser body shouldn't be so affected by spindle thermal growth because it's pretty far away from that temperature of the spindle, but the coolant and everything's no, not like the casting is around the spindle.
John S: Right, the aluminum's not getting hot warmed up just materially, so, right?
johngrimsmo: So anyway, so I made this probing routine and I've been and deprinting the results. Some, I'm not doing it automatically yet, but I'm probing periodically. I have some data from a stone cold machine, like not even warmed up.
johngrimsmo: And that's my zero. And then ah did it again after a warmup, not much change, you know, 12 minutes spindle warmup, not much change.
John S: so
johngrimsmo: But then after a running the spindle at 16,000 RPM for half an hour and doing a blade, um I'm seeing over a thou in and sometimes up to a thou or more in Z.
John S: But that could be, the Y movement could be the spindle casting itself growing outward, not the table casting.
johngrimsmo: Correct. And that's what I'm saying.
John S: Okay, okay. Oh, it makes more sense. Okay, sure.
johngrimsmo: Yep.
John S: That doesn't seem crazy to me.
johngrimsmo: It kind of makes sense, right?
John S: Yeah, agree.
johngrimsmo: Maybe.
John S: but
johngrimsmo: um Because it is, but we don't run it like balls to the walls for hours and hours and hours.
John S: It's probably stable once it's there.
johngrimsmo: We're not palette changing it yet.
John S: Sure.
johngrimsmo: So it like runs for half an hour and then it sits for 10 to 100 minutes. Then we're on another one. So it's swinging, you know?
John S: Mm-hmm. Oh,
johngrimsmo: And there's the line where you don't want to over-probe and over-rely on things because then you're jumping back and forth and back and forth. But maybe there's a situation where I probe that rigid feature and I reset G53, your external offsets, which adds a tenth or thou or something to everything.
John S: yeah, yeah.
johngrimsmo: um Dangerous. On the Speedios, it's actually, you have to get the factory to tell you where to put a physical wire jumper in the control panel to enable those external offsets.
John S: Interesting to do like a master override.
johngrimsmo: Like, yeah.
John S: Huh.
johngrimsmo: And, uh...
John S: They'll do it. These you have to know how to ask.
johngrimsmo: You just have to ask them. Yeah. So I asked my local distributor two days ago.
John S: Got it.
johngrimsmo: I haven't heard back yet. So I might just ask Yama's it.
John S: Interesting. Yeah.
johngrimsmo: um
John S: Right. Right.
johngrimsmo: Because I'm trying to edit these numbers and nothing's happening. I'm like, what the heck? then I find this practical machinist forum post where they're like, oh, yeah, you got to put a jumper in the back.
John S: Hmm.
johngrimsmo: So.
John S: Crazy, crazy.
johngrimsmo: so Because the y direction is affecting our bevels, which we're trying to hold to within less than a thou.
John S: you Right.
johngrimsmo: And we're seeing variation throughout the day. So like maybe this is a way to fine tune that. And then we have a z feature that is literally directly tied to the flipping action of the fjell.
johngrimsmo: So if that feature is 10 thou 5 tenths or 11 thou, you feel the difference.
John S: Right.
johngrimsmo: And this z growth of plus or minus the FAU throughout the day, depending on how hot it is, affects that. And I've been tracking that. We're manually measuring that feature. Every every blade, I'm seeing swings.
John S: Do you have, does the brother give you temp readouts or can you,
johngrimsmo: So it's interesting.
johngrimsmo: Good question. I feel like there is a temp sensor somewhere, but it's not like on the current where there's sensors everywhere.
John S: I think there are some adhesive style sensors that are cheap, you know, 10, 20 bucks from Amazon, but I would love to see not going crazy, crazy, but just a couple more data points around.
johngrimsmo: Yep.
johngrimsmo: Yeah, for sure. Temperature ends.
John S: Cause the answer is, I think you already know the answer. It's like, look, this is not a current and it's going to it's going to move.
johngrimsmo: Sure.
John S: our Look, our Willimon moves. We warm it up. But the trick is probably better have than when you end the program, just go straight into an 8,000 RPM spindle and it just stays on 8K.
johngrimsmo: yeah you
John S: So it's not, I'm guessing, may be wrong, that that's going to keep some heat, maybe not all the heat, but it's going to more than half cut your movement.
johngrimsmo: Yep. Keep some heat in there, you mean? Yeah, maybe. yeah
John S: um
johngrimsmo: Could be, yeah, that's a good idea actually. Simple.
John S: not Not ruling all the power.
johngrimsmo: Yeah. Or lump parts together, or do a longer warm up, or... Yeah.
John S: think the warmth the issue, I think, i mean because this goes back to, you know, grinder, you don't turn the spindle off. Current makes you warm up a long time.
johngrimsmo: Yep.
John S: There's some story, I wish I could remember, but somebody bought a machine at like IBM back in the 70s and they never turned the spindle off in like the 10 years of owning the machine.
johngrimsmo: Yeah.
John S: Yeah, i I feel like it was like a Tyler Mitsui Siki story or something, but... ah That's your, that's, I would think, or you go the crazy other route, which is like, I've heard the people put um heating elements, space blankets around castings to drive, to control the temp that way.
johngrimsmo: Yeah.
John S: Yeah.
johngrimsmo: Well, it it really shows exemplifies the thermal cooling um theory and practice that current put into their machine, because it's like cooling fluid antifreeze is flowing through the spindle cartridge through the casting through the through the motor mounts.
John S: Yeah.
johngrimsmo: like you open the panel and you're like, oh, there's the FANUC motor or whatever, height nine motor and the mounting adapter, which I machined those from my original, you know, you know, Grizzly X2, which was quarter inch aluminum.
John S: Right, right.
johngrimsmo: And these are like inch and a half aluminum, but those are water cooled channel blocks.
John S: Yeah.
johngrimsmo: So water goes in there. So the, the heat from the motor is taken out. And I know some of the machines, the not Matsura, what's the other M brand? Yeah.
John S: Mitsui Siki.
johngrimsmo: Makino.
John S: Oh, Makino, sorry.
johngrimsmo: so Some of those, like the PS95, have liquid-cooled ball screws.
John S: Yep.
John S: Yeah. Cool through the middle of the ball screw.
johngrimsmo: which The Kern doesn't have that, but it's really cool.
John S: Yeah.
johngrimsmo: So, yeah.
John S: No ball screws grow for sure. So many of these.
johngrimsmo: I guess with an encoder machine, maybe they don't care, because it just sees that and captures
John S: Yeah. Somebody's explained this to me. I think we brought it up, but like and encoders aren't,
John S: Like there's a there's different a different approach of having a machine that's kinematically correct versus like the encoder saying, oh, you're not in the right spot. So what do you do? Like you effectively violate an error and then it figures out how course correct you back on.
John S: That makes sense.
johngrimsmo: it
John S: is it's not Otherwise, what do you do when your encoder tells you're out of position?
johngrimsmo: Well, yeah, you just for hope for it.
John S: go back you Go backward. Yeah.
johngrimsmo: Kern was telling me they try to mechanically get the machines as perfectly as possible, but it's never perfect. And then they do you know a full laser scan of the whole working area to fine tune the encoders or the map or whatever it's called to dial it in 100%.
John S: yeah
johngrimsmo: hundred percent
John S: It's interesting. I mean, linear motors have their own issues with heat, but otherwise seem solve a lot of interesting weight problems for sure on this.
johngrimsmo: Yeah.
johngrimsmo: And add hydrostatic on top of that. And it's like, oh.
John S: Right. Right. For sure.
johngrimsmo: But yeah, so I'm gathering data right now. And i think maybe maybe I'll experiment with probing that feature before every critical part and just see what happens, see if I'm getting more consistent output.
John S: We did that on our Haas lathe. And actually we did this and it legit works where the Haas lathe has a spindle temp value. I don't know what it correlates to in degrees. It's just a value in the hundreds or thousands.
John S: And we figured out when the lathe is warmed up and the spindles at a certain temperature, that value is above 1300, making that up.
johngrimsmo: Well,
John S: we actually have a fusion pastor where this program will not run if the spindle value is below that.
johngrimsmo: that's kind of cool.
John S: um Yeah, it's great because it's like, you don't like, I don't overthink this. I don't get cute, but you do need to make sure it's warmed up. And if it alarms out, you just know, oh I got to run warm with the machine for 10 minutes.
johngrimsmo: Yeah. Yeah, you can even program that in and have it warm up the machine until it is that temperature if you really want to.
John S: Yeah, exactly. Right.
johngrimsmo: Yeah. Yeah.
John S: Well, I'm curious to see what you learn on that.
johngrimsmo: Yeah, yeah. It's been fun because it's another one of those data points for the ultra precision of like, I know it's a speedio and it's not a occurring, but like, you know, I'm doing it. like
John S: Oh yeah
johngrimsmo: And I'm really close. You know, our issue is one thou of growth, both in X or in in Z and in Y. And I think I can probe that, comp for that I don't know.
John S: yeah. 100% I love, love, love getting more out of a machine than you're supposed to. Like, you know, it wouldn't be fun if you just, someone else gave you money you could buy all Makinos and Kearns and you didn't have to it, you just hit the button, right?
johngrimsmo: Totally.
johngrimsmo: yeah It's too easy.
John S: Yeah, exactly. Exactly. Well, speaking of making ah making machines better, i got a tip of the hat to Ethan Patane from Centric Iterations. um He's been helping me on refining an Okuma post that gives us fusion-based probing to use our Okuma as a CMM.
johngrimsmo: Ooh.
John S: Recognize it's not not a CMM, but there's still a lot of times where would be really handy to be able to measure a small feature or a big feature and get a value of traditional CMM
johngrimsmo: Ooh.
John S: traditional prismatic part get a value of a boss of a bore the value between two points and the way he did it is great because you just pick all the features you want infusion hit post and then it just goes probes it gives you the value yeah so
johngrimsmo: No way. That's cool.
John S: Um, this idea came up at IMTS and I was like, dude, I would love that. Like, what do what do you want out of it? Like you drumming a business or you were paid? And he was like, dude, I just all about paying forward, sharing it. Like he is offering this up as a service.
John S: Um, going to do a little video showing off how we use it. I think he's going to share a lot of that Kuma, at least relevant stuff. Um, so, and then the other thing he included was a, and he shows,
John S: think we're gonna show how we did it, so hopefully folks can borrow as needed for their relevant machine tools. But um it now we'll throw an optional alarm if your gauge length in the control, in our case, we'd set the default to plus or minus 100,000, which is quite large. But if the gauge length infusion to a library is more than plus or minus 100,000, differing from
johngrimsmo: Yeah.
John S: well, I forget which I said, Fusion in the control. So if you, Fusion thinks it's a 4.25 inch gauge-like tool that's in an ER, you Mari tool, ER 16 by four, and your control thinks it's actually only 1.8 inch gauge-like tool, it'll alarm out.
johngrimsmo: That's kind of cool. Yep. Because I'm going through that now. I'm trying to stop using one of these end mills and switch to a different version, which will be shorter.
John S: Mm-hmm. Yep.
johngrimsmo: Might even be a different holder, but I've already got all these programs that already have, you know, the RegoFix holder, the PG-15, the Callit, this tapered tool, everything.
John S: Oh my God, I know.
johngrimsmo: And I'm like, if I just change the tools and don't change all the cam, it'll work fine. But now I'm breaking a system, you know?
John S: dude i i mean, I've given up on telling Autodesk that this needs to be fixed because the tool library is not a library. It's a one-way street and it's devastating for folks that want to build out libraries and run production.
johngrimsmo: Yep.
John S: What is interesting is um I don't know, I'm not sure this is like a one size fits all thing right now, but the tool path, the company is working on and has tools now called pre-tool that are in their tool library tools that you can use for free that can help.
John S: um Because one of the, basically, sorry, went rambling. One of the ideas here is like, don't worry about your tool library anymore. Let Infusion, let toolpath handle it. And then when you want to reprogram your part, it just reapplies your clean tool a library from toolpath and pushes it back down.
johngrimsmo: Oh.
John S: which actually could be great for us for fixture plates. Cause the same thing is like, oh, if I want to change a drill, we've got, we 785 fixture plate files for all the different models we make. Like, I don't want to go change the sand big drill from one diameter to another or through spindle coolant feature or whatever.
John S: And now you could just let it push it from there, which is really cool.
johngrimsmo: So the toolpath tool library becomes your master and your
John S: I'm not doing, I'm doing it a disservice, but if you want to learn more, anything about this feature, just go, just Google pre-tool from Toolpath and you can watch videos or see what it does. um
johngrimsmo: Cool.
John S: The backstory is Toolpath's main goal is CAM automation, soft clicking parts. It'll program your parts for you. You can't do that until you know what tools you have to use and they can give you the 6,000 tool library that they have created that's every tool in the world, but that doesn't help most people.
johngrimsmo: Accurately. Yeah.
johngrimsmo: Yeah.
John S: So you need a way to quickly say, no, Grimsmo wants to upload his Norseman library or his Mori library or whatever you, however you're handling it. um And that's a lot of the origin of this.
johngrimsmo: Sweet.
John S: Yeah, it's I've heard from some other ah kind of third party-ish folks that Pre-Duel has been actually super helpful for them. so
johngrimsmo: I'm gonna look that up. Let me weekend look up.
johngrimsmo: All right.
johngrimsmo: You working on anything interesting lately?
John S: um We took our first cuts on the new Tormach 1500MX. um We're actually, I think a joke waiting for a fixture plate, but we just threw a 5C column block on there because I had to cut a keyway shaft.
John S: And um it's, did talk about this already? I feel like, okay.
johngrimsmo: I think you talked about last week a little bit, yeah.
John S: ah Sorry, I don't want to repeat myself. But the that mean it was a 360 inch tool and a steel keyway shaft and the cut was awesome.
John S: Like if this is not the Tormach that you and I cut our teeth on.
johngrimsmo: Yup, yup.
John S: It was really nice. So I'm excited to get that machine commissioned.
johngrimsmo: That's beautiful.
John S: How about you?
johngrimsmo: um Been diving into heat treating more. So we probably talked about last week the the button on the fjall, the polished button that trying to get that colsterized.
John S: The one that's carburized or whatever?
John S: Thank you.
johngrimsmo: It's not. We tried that process. We used 17.4 and then we colsterized it, which does get the surface hardness way harder, but the substrate is still too soft, so it dents.
John S: Yes.
John S: Got it.
johngrimsmo: So we stopped using that. um Although there is an application for it. It's just not this application. so we're using 440c we're turning it soft and it's a really gross material to turn because it just smears and it streaks and it's just not like 17-4 um but it's the material we need because it's stainless it gets full hard and uh whatever the problem is when after we heat treat we tumble and then we go to polish the heads of the button um i put an instagram post on this there's like holes or porosity or
John S: Oh yeah!
johngrimsmo: right And they got like 120 comments of really smart people giving their advice, which was amazing because it was such good perspective.
John S: Yes.
johngrimsmo: you know So many people are like, it's the material, the material is garbage. Or like, it's the heat treat, the carbides are clumping together. Or it's the way you're polishing it, you're ripping out the carbides.
John S: Okay.
johngrimsmo: And it's like, all could be true. What is true for me?
John S: Okay.
johngrimsmo: the scientific answer is to take it to a microscopy lab and have them, you know, slice it, lap it, put it under a microscope and like see the grain boundaries. And this is a feature a service you can get for a couple hundred bucks yeah for any steel.
John S: Really?
johngrimsmo: So it validates your heat treat. If it's good heat treat, not just how hard it is, but how like, you yeah, exactly.
John S: Wow.
John S: The Martin Zeiss, yeah, yeah.
johngrimsmo: Your grain boundaries, your soak time. Is it cryo treated? Is it not? um This is how, metro not metrology microscopy met metallurgy labs um validate the quality of a steel the quality of a heat tree the quality of whatever um and i've done through the lapping research a lot of this stuff comes up um because those people might metallurgists slap specimens and uh make them mirror polish they don't have to be flat flat but they have to be mirror to be able to etch it and see the grain boundaries
John S: Yes.
John S: Yeah, sure, sure.
John S: You write.
John S: Write.
johngrimsmo: But you need like a thousand time zoom microscope to be able to properly see that and the right chemicals to etch it. And every steel has got a different chemical, whatever. So the other day, because we're having this problem and it's getting worse.
johngrimsmo: And this new batch of heat treated parts, we heat treated 120 buttons together in a little baggie.
John S: Mm-hmm.
johngrimsmo: And they have more of this pitting problem than ever before. The first batch, they just polished up and it was no, we care.
John S: do
johngrimsmo: The next batch, it's, ah holes are forming and he polished it some more and that hole goes away but another hole comes up and there's all these little tiny bumps throughout the whole surface and we're like what's going on so we I took myself, Angelo and Sky, who's doing the heat treat, um to our local heat treat shop.
johngrimsmo: We called him up. I was like, can we come over and bug your ear about 440C heat treat and stuff? And he's like, yeah, sure, come on over. And we chatted with him for 20 minutes and it was super helpful. And he's like, what's your heat treat temperature?
johngrimsmo: And Sky had his paperwork and he's like, 1950. And he's like, oh, that's too hot. Take it down to 1850. um How long are you soaking it for? ah are you cryoing it? What's your quench time? He's like, usually 440C is pretty broad. Like you can do a lot of different things and get away with it. But the result you're looking for is very specific polishability. So you want to reduce these features.
John S: Interesting.
johngrimsmo: And it's literally the carbide, the carbides of the steel are clumping together into little tiny bumps. And a lot of those tiny bumps are clumping together into bigger chunks.
johngrimsmo: And when we polish those big chunks get ripped out.
John S: Ripped out. Interesting. And they're are these visible to and to a...
johngrimsmo: And it's like, the the bigger ones are definitely visible and the little ones look like dimples, like, like, they're actually proud.
John S: just Okay.
johngrimsmo: Um, they're like little tiny ripples in the surface, but the bigger ones absolutely like would never ship apart with the bigger ones.
John S: Interesting.
John S: Yeah, yeah.
John S: Okay.
johngrimsmo: Even the little bumps, um, are like, they're just there. And, I think it all comes down to, don't want to say a poor heat treat, but an improper heat treat for the results we're trying to get.
John S: Sure.
johngrimsmo: Not just the hardness, but the actual carbide dissolution dissolved into the whatever, whatever. um
John S: so what's the answer?
johngrimsmo: So the answer is we're making new buttons.
John S: Just, just cook less lower. Yeah.
johngrimsmo: So we're turning new buttons now. And then don't do about a batch 120, 10.
John S: Yeah. Right.
johngrimsmo: and then try different temperatures, polish immediately and get that validation. So that's what we're doing right now. And pretty excited, like many hours of research and thinking and phone calls and to figure this out.
John S: Cool.
johngrimsmo: But I think we're on the right track.
John S: Do you try to stop the parts from touching each other in the foil baggie?
johngrimsmo: No, they're all clumped together.
John S: And doesn' that doesn't... There's no correlation of where the part's touching to be more...
johngrimsmo: My hands are up. I don't know.
John S: Yeah, okay. Because I've noticed you'll get like... Hotspots may not be the right word, but I kind of don't love it.
johngrimsmo: Yeah.
John S: Even where a part may be tipped up against a ceramic post. um
johngrimsmo: Yeah.
John S: So...
John S: Interesting.
johngrimsmo: Yeah, it is interesting. and And one of the variables is this batch was 120, previous batch was 40 or something. So like, you know, this guy's like, well, more parts fit in the bag and the first batch worked.
John S: Yeah.
johngrimsmo: So this should probably, why, why wouldn't it, you know, I don't know.
John S: Right. In this, night your oven has no problem handling that higher. Sure.
johngrimsmo: That's one of the other questions, because you start to look at every variable and you're like, what could possibly change?
John S: Sure.
johngrimsmo: Last time we did it, we just put a new thermocouple in the machine, in the heat treat oven.
John S: Interesting.
johngrimsmo: And now it's been running for six months every single day at 1950 degrees. Like they get old, they have to be replaced every now and then.
John S: Huh.
johngrimsmo: So now technically it's a more used thermocouple, which may drift more. It may be the wrong temperature. Like, i don't know.
John S: Well, that's crazy because, yeah, if your if your oven's actually at 2020 degrees and the in the guy was saying 1950's too hot and you're inadvertently...
johngrimsmo: Yep. Yep.
John S: way like Sure, that could be... Hmm.
johngrimsmo: yeah So then I also go and I asked ChatGPT, I'm like, just, this is what's happening.
John S: Thank you.
johngrimsmo: 440, like, amass all this information you know. i'm not going to trust it 100%, but give me perspective, you know?
John S: Yeah.
johngrimsmo: And they said pretty much the same thing, like nineteen fifty s kind of your upper range. If it's too hot, the carbines will clump together. um The way you're polishing it, if it's a, say it's a red stick of rouge on a buffing wheel, that's iron oxide, um which is not a very sharp abrasive like diamond is.
John S: Yeah.
John S: Got it.
johngrimsmo: So that could grip and right out grab and pull as opposed to using a diamond paste on your buffing wheel, which will be sharper and more aggressive.
John S: Rip it out.
johngrimsmo: And, you know, it'll actually cut through the carbides instead of just smearing the material below it and ripping out the carbides.
John S: Shear it off. Yeah, sure.
johngrimsmo: So it's kind of super nerdy fun to go through all this, even if it's frustrating and causing everybody.
John S: Yeah.
johngrimsmo: Like as a company, we've probably put 20, 30, 40 hours into polishing these stupid buttons and talking about heat treat and things like that.
John S: Yeah.
johngrimsmo: And that's a lot.
John S: That's why you are who you are though. you know, yeah, but you gotta figure out something to keep moving product too.
johngrimsmo: I know, right?
johngrimsmo: Yeah, so, and I told him, I was like, I'm happy to throw these away and make more buttons and, you know, try again, but we need to know what we're doing.
John S: Yeah.
John S: Yeah.
johngrimsmo: And i think I think we're on that path now. So by next week, we should have the new ones heat-treated and answers.
John S: Yeah.
John S: Yeah, good. I made a note to that and I want to hear more about the brother thing this week.
johngrimsmo: Yeah.
John S: Oh, you're the heat treat thing made me laugh. Did I tell you about putting the aluminum parts in the dishwasher?
johngrimsmo: You told me on WhatsApp, yeah, not on the podcast.
John S: Oh my God. I should find a picture. Maybe I'll throw up on Instagram, but we do those shimper valve covers and we wanted to put them dishwasher. And I said, well, before I buy a dishwasher for the shop, I'll take them home.
John S: My wife didn't care.
johngrimsmo: You're trying to clean the coolant off.
John S: They were
johngrimsmo: Like...
John S: Like try to clean and cool off and just like really clean them, dry them so that they're not, cause you know, like if you blow them off, rinse them off, they're not dry. Like it's dry. Dishwasher does a great job at this. And I took two of them home.
John S: They were good covers. I didn't think. I didn't think to try to find blems or scrap. And I certainly, I almost put like whole 12 of them in there. But i let me just do, let me just put two in there. And I put them on a quick wrench cycle, which is still like 50 minutes or an hour.
John S: And didn't really think anything of it. And I come back in an hour and they're done. Dishwasher dings. I open it up. Oh my God. it These are 6061 domestic aluminum that we machine, have machined for years here.
John S: It looks, it,
johngrimsmo: And they were like white silver going in, right? Like beautiful.
John S: They look beautiful. They look like a perfectly high-end machine part. um And we I opened the dishwasher and they look like somebody took a fire torch to them.
johngrimsmo: Yeah.
John S: The aluminum apparently got so hot from our dishwasher at home, like this regular dishwasher, that it, i mean, i I don't know. What else it could have been, whether there's a chemical thing going on.
johngrimsmo: Did oxidize the surface?
John S: i assume it's heat.
johngrimsmo: Like...
John S: The surface feels the fine, but it literally looks like dark blotchy heat. Like I've never even done this to aluminum with a, tried a blowtorch.
johngrimsmo: Yeah.
John S: Dimensionally didn't change, but it looks terrible.
johngrimsmo: Yeah, yeah.
John S: The whole thing looks black and blue and all that.
johngrimsmo: It's got to be oxide corrosion, but just a thin, super thin layer kind of thing.
John S: Man, nothing, nothing at all.
johngrimsmo: And what was your detergent? Like, what did you use? Anything? Just hot water.
John S: Yeah, hot water.
johngrimsmo: Maybe residual detergent from last load of laundry or dishes.
John S: Could have been that. mean, we had rinsed the quality cam off them, so I don't think there was that much. I'd be shocked if that little bit could have caused it. Now, I don't know how much residual detergent is in that whole system from the dishwasher.
johngrimsmo: Yeah.
John S: um What I also learned, which is one of those awesome like little factoids, ah tip of the hat to Rob Lockwood, was that um commercial dishwashers are far less um abrasive or a less ah harsh on your parts on your department dishes because commercial dishwashers assume that a professional dishwasher has first pre-rinsed the dishes like you know use the like the um sprayer that's suspended from the top so you like rinse most of it so the dishwasher can do 10 minutes it's not nearly as harsh these commercial as you mean these residential ones are made to assume that you're the laziest person in the world and you like got gravy and potatoes all over your plate.
johngrimsmo: Yeah, that top pose thing, yeah.
John S: And so ah a rare instance where the commercial one is actually less yeah aggressive and less industrial, if you will.
johngrimsmo: Aggressive. and Yeah.
John S: um
johngrimsmo: That's crazy.
John S: So that's all I got for now.
johngrimsmo: Yeah, I'm leaning more towards the detergent than the heat. But maybe it's the moisture, the vapor, the the drying cycle, all of it. I don't know.
John S: now now Now that we're talking out loud, I should, ah and I got to run, I think you do too, but I should just take a couple of scrap pieces of aluminum, like face them for their machine surfaces and go put them in the dishwasher and see if ah if it, I don't know how you could, I would assume if I ran a cycle empty, that that would flush most of the detergent out of the system, right?
johngrimsmo: Okay.
johngrimsmo: Keeps doing it, or...?
John S: Because whatever, anyway, actually,
johngrimsmo: Yeah. or take take five five pieces and run five loads and see if it gets worse or better.
John S: touch no, John, I'm not going to do it because it doesn't make the beer taste better.
johngrimsmo: Who cares? There you go. I like that.
John S: um Sorry, this is not something I should do. But at some point I want to do it for fun.
johngrimsmo: Sounds good.
John S: just
johngrimsmo: Okay, so don't go do that, and I'll see you next week. All right, bye.
John S: Sounds good. See you
