Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
#399 Too much on my plate
2.8k Plays5 days ago
TOPICS:
- Project management at Saunders
- Tolerance ranges and scrapping parts
- Python script deep dive
- Robots to change workpieces and pallets
- Willemin part release... magnetism?
- Too much on my plate
Recommended
Transcript
Introduction and Hosts
00:00:01
johngrimsmo
Good morning and welcome to the business of machining, episode 399. My name is John Grimsmo.
00:00:06
John S
My name is John Saunders.
Podcast Theme: Machining and Manufacturing
00:00:08
johngrimsmo
And this is your weekly dose of machining and manufacturing where you know two friends just kind of check in on each other, see how we're doing. See what's going good, see what's going bad, see what's stressing us out or making us happy.
00:00:18
johngrimsmo
And I think we we both have a lot of those today to share.
00:00:22
John S
Yeah, how are you doing?
00:00:24
johngrimsmo
I'm doing generally great.
Grimsmo's Fjell Machining Goals
00:00:26
johngrimsmo
um and Things are busy things are good. The team is crushing Handling so much. I still have critical things on my plate that are slowly coming off my plate, which is good Fjell production has been my focus for the past year, but especially now it's like crunch time, you know, I kind of made this Joking goal of finishing a hundred feels by the end of January and we're January 22nd and
00:00:57
johngrimsmo
ah We will machine a hundred fails for sure this month.
Team Motivation and Collaboration
00:01:00
John S
Good.
00:01:00
johngrimsmo
So it's, it's fantastic. It's, it's pushed everybody. It's kind of got that bug stuck in everybody's head. They're like, if you want to achieve that goal, like I'm going to need those parts, which is good.
00:01:05
John S
Yeah.
00:01:10
John S
Right.
00:01:11
johngrimsmo
It's a conversation that needed to be broached. You know, otherwise I'm just in my head like, yeah, it's coming.
00:01:14
John S
Right.
00:01:16
johngrimsmo
It's coming. But I've got the rest of the team going, dude, give me, give me, give me like I'm ready. I'm just, and it it's made timing ah more clear for things.
00:01:21
John S
Good.
00:01:26
johngrimsmo
because like the blade is so complicated. Well, we got to make soft blades. Let's make a ton of soft blades so that they can go through heat treating, grinding, and lapping, and and then the guys can finish them.
00:01:32
John S
Yeah. Right.
00:01:37
johngrimsmo
On top of that, I'm trying to carve out like fun, productive projects, you know improvements here and there, toolpath improvements, or tool holder improvements, things like that.
00:01:42
John S
Yeah.
Improving Grinding Consistency with Python
00:01:49
johngrimsmo
like one of them I mentioned last week how I wrote the Python script to create the grinding code.
00:01:54
John S
I have some questions about that later.
00:01:56
johngrimsmo
Yeah, please.
00:01:56
John S
Yes.
00:01:57
johngrimsmo
um It has been absolutely fantastic. I applied it. It was first to our Fjell, I applied it to the Rask, literally the same script. I just needed to like watch it and prove it, that it works on a different knife.
00:02:05
John S
OK.
00:02:08
johngrimsmo
um Works awesome, like huge benefit. um But the next step after that was to increase the diameter of the arbor that is holding onto this grinding wheel.
Reducing Vibrations During Grinding
00:02:19
johngrimsmo
Because I was using a half inch arbor sticking out maybe three inches. So your length diameter ratio is pretty long.
00:02:23
John S
kar
00:02:26
johngrimsmo
And in my head, I'm looking at the finish under the microscope and I'm like, it's kind of vibrating. Like that grind is a little chattery. How can I stiffen this up? So I went from a half inch arborate to a three quarter inch arbor, which math and Google tells me is five times stiffer, which is like massive, like absolutely massive.
00:02:43
John S
He done spiced me yeah
00:02:47
johngrimsmo
Even a two times stiffer should make a difference. And. The difference in the finish isn't massive, but I think it's enough to solve a couple of the problems that we were having. And so that was a fun little Wilhelmin slash Curran project to make those one night.
00:03:01
John S
Could you shorten them up too?
00:03:02
johngrimsmo
Now I need the length to to clear clamps and stuff.
00:03:05
John S
Oh, and okay, fair enough.
00:03:07
johngrimsmo
So the diameter helps. um But yeah.
Project Management and Accountability
00:03:12
John S
Good.
00:03:12
johngrimsmo
Oh, I got a good note for maybe later.
00:03:14
John S
We had a, you write that, I'll talk, we had a um similar thing this week with project management on the these this stupidest of silly examples that I think is a great example of we're redoing this opt-to clamp and I'm involved, Alex involved, Garrett's involved, the three of us mainly, but then Grant's involved because he's making the clamps on the will of it.
00:03:36
John S
So there's four of us involved. And it quickly became kind of the like, well, who's really like,
00:03:40
johngrimsmo
Yeah, I was in charge.
00:03:41
John S
Yeah. And it was a super, yeah I think what made it unique ah versus prior instances of this sort of thing is that we're now going to switch to using a ah previously non-critical diameter for these ID clamps.
00:03:55
John S
So now it's like, well, this this bore that we didn't used to care about. So when we don't care about something, I think it's probably plus or minus 5,000. It's way under that. But like we didn't.
00:04:04
johngrimsmo
You don't even measure it. Like you don't care.
00:04:05
John S
yes, and now we do care.
00:04:06
johngrimsmo
Yeah. Yeah.
00:04:07
John S
And so it's like, hey, revised drawings, how are we doing the cam on the production? um Just a bunch of questions. And so this ties into a bigger theme that Alex brought up that I think I probably need to give more.
00:04:21
John S
attention to of how does Saunders handle long-term project management, both from ah individual internal responsibilities, but obviously the cliche team examples of, you know, like, hey, let's let's scope out.
00:04:33
johngrimsmo
Sure.
00:04:35
John S
um Even the FIEL right now, or for us, the puck chuck, there's potentially months left of, for us, it's more like the sidecar development, like little accessories or other stuff, but laying that out, backing into timelines, like,
00:04:40
johngrimsmo
Yep.
00:04:44
johngrimsmo
so
00:04:47
John S
even if some simple task needs to be done for us, that may be, well, somebody needs to design it, some of the accessories for the um like the pneumatic valve. ah sheet metal covers that were we're selling with it are from Send Cut Send, which is works out great works not great.
00:05:02
johngrimsmo
Okay. Nice.
00:05:04
John S
We're actually using a turning shop to do the some ah covers, because it's like these are just integral covers.
00:05:10
johngrimsmo
Yeah, the red ones, right?
00:05:11
John S
yeah Exactly. They're super great.
00:05:12
johngrimsmo
Yeah.
00:05:13
John S
we I love making everything, and we might make it at some point, but the reality is it's a cover. It doesn't matter. Easy to outsource it.
00:05:20
johngrimsmo
Yeah.
00:05:21
John S
It's totally fine. But just who's handling that?
00:05:22
johngrimsmo
Hmm.
00:05:23
John S
So it's been top of mind. um And we switched so far using Google google Chat is OK. I don't feel like I like it as much as Slack.
00:05:35
John S
um I just don't want to deal with Slack at Saunders, both from a new architecture and it's not Google, so we don't have the benefit of integrated whatever Google stuff.
Communication Tools Challenges
00:05:45
johngrimsmo
Yeah.
00:05:45
John S
We're just not going to use Slack right now.
00:05:46
John S
So Google stuff is OK. There is a new thing in the spaces um So we have like R and&D, shop needs, um order fulfillment stuff where you can assign tasks. So instead of me DMing Garrett and saying, hey, will you look at that cam? I can create a quick task. I could put a deliver due date on it, which is OK. It kind of implies. But like if it's done today, it maybe I just want to take a quick look versus like, hey, this is hyper urgent. Drop everything else.
00:06:14
John S
um Anyway, I like the idea of a task because it doesn't get lost in the messaging of like, oh, oops. um But kind of goes back to what you were saying about like, OK, you have six people making these fails.
00:06:25
John S
Like, who's doing what?
00:06:26
johngrimsmo
Yeah, yeah. Yeah. um What I've read in a lot of books is like, who is ultimately accountable for that project or that division or that you know job or whatever?
00:06:32
John S
Mm-hmm.
00:06:37
johngrimsmo
um So you kind of need a a project leader. And this goes back to every every project that's ever been made for all time, from spaceship to you know a BIC pen.
00:06:44
John S
Yeah. Yeah.
00:06:48
johngrimsmo
um And it's funny how many parallels there are in all types of creation and manufacturing with multiple people that we are kind of stumbling and learning as we go, as most companies are.
00:06:59
johngrimsmo
But um I mean, on one hand, I'm super fortunate to have Angelo and Jeff on the team because Angelo comes from aerospace, big background. ah Jeff comes from a prototype automotive from big like like plants kind of thing, and but he was in the prototype division.
00:07:10
John S
Oh wow.
00:07:14
johngrimsmo
Um, the issue is they both have worked with a lot of these project management systems, but they just kind of used what was forced upon them.
00:07:21
John S
Right, right.
00:07:22
johngrimsmo
And to now have them kind of invent one here with me is, uh, confusing. Um, and we're not exactly sure what to do, but they're bringing all kinds of great suggestions and I'm like, okay, well, what are we, what are we doing?
00:07:33
johngrimsmo
You know, um, so they've been working on our drawing packages, which we've always been super weak at, um, both like.
00:07:38
John S
You're just top dimensional drawings for your parts.
00:07:41
johngrimsmo
Yep, yep.
00:07:41
John S
ye Yeah.
00:07:42
johngrimsmo
I have no interest in making dimensional drawings. I just look at the CAD and like measure parts, but that's not proper. But they've been crushing at the drawings. And and even coming up with a workflow project plan, and Angela keeps saying, like when you R and&D something, like it'd be nice if you had a structure that you followed that we could jump into.
00:07:59
John S
Yeah, yeah.
00:08:02
johngrimsmo
um So I don't have any solutions, but it's it's interesting. We're kind of both thinking about that. you know
00:08:07
John S
And that's where like our wiki has come in where that's now the go-to spot. Like if we're gonna create a policy, well, so like if we're gonna say,
00:08:11
johngrimsmo
No way. Explain that.
00:08:16
John S
what we want and don't want on a drawing, that that tribal opinion is gonna get dumped into a wiki. Or like here's a good example. we
Knowledge Documentation with a Wiki
00:08:24
John S
We don't want the wiki to have junk in it, meaning I did not create a wiki page for places that we can use to outsource machine parts, which is actually funny.
00:08:34
John S
We really don't do this, but we had, we did with the send cuts and brackets. We did with, we've actually used geometry for those covers as well.
00:08:39
johngrimsmo
Sure.
00:08:42
johngrimsmo
It's a great information to have around, right?
00:08:44
John S
Well, but I didn't, because the truth is, one Google search or one chat GPT search can say, tell me good options for outsource machine part services, and it'll tell you whatever all these vendors, this the zometries and the competitors.
00:08:51
johngrimsmo
Okay.
00:08:55
johngrimsmo
ah Except for the tribal knowledge of Saunders has used these before.
00:09:00
John S
So that's what flipped me on the head, was last week, um I said, no, we should put this as a wiki, because what we should say is, by the way, these parts we've been, ah we used to we used to make our own DIY Himer tips, and we had the,
00:09:02
johngrimsmo
Interesting.
00:09:14
John S
part of it SLA printed by, I think it was 3D hubs, if that sounds right. I don't know, it's been a while.
00:09:18
johngrimsmo
Yeah.
00:09:19
John S
um It's worth, from and that doesn't go in our ERP, but the tribal side of like, hey, we found this to be a really good solution for this part, or we don't like this, or or like even some of the parts, it's like, hey, the quote looks good, but you're gonna get nailed with tariffs if you actually, because the quote is actually overseas, it's not clear until you get through the process.
00:09:36
John S
like Um, yeah. And on drawings, I thought we've done, we've come so far in the past year and you guys will do fine. I don't doubt it, but our drawings, um, we continue to be really, I, someone just put this on my desk. Oh, it's that same project that ID clamp. So it's like, okay, what we're going to do is, um, what I wanted to show was I really like having the, um, target and scrap.
00:10:00
johngrimsmo
Okay. Target dimension, scrap, like your, your goal and your actual, like.
00:10:05
John S
Yes, and that's, I've said this before, I think as we're saying again, our target means this is what we want to hit. But the truth is, in this example, it's 343 to 338.
00:10:10
johngrimsmo
Yeah.
00:10:15
John S
So that's only 5,000. Oh, that's 5,000. This is a mile. This is not a big deal.
00:10:22
johngrimsmo
Yeah.
00:10:22
John S
um A lot of these it's tighter.
00:10:22
johngrimsmo
But still, conceptually.
00:10:25
John S
Yeah, we won't have a problem hitting that target there, but a lot of the mod vice parts are tighter, but the truth is if I want something to be 5002, so two tenths over half inch, if one of them is 5003, we're not throwing that away.
00:10:36
johngrimsmo
Yeah, what's what's the line, you know?
00:10:38
John S
So what is the line? And that the target means aim for that.
00:10:39
johngrimsmo
Yeah.
00:10:42
John S
So if you make one that's 5003, and that's within the scrap range, it's okay, but keep an eye on it because I don't want to make 100 more of that size and not care.
00:10:54
johngrimsmo
the
00:10:58
johngrimsmo
I think the problem is we're both too close to the products we're creating, not a problem, I like like it actually, I prefer it that way, but you go to a shop like our friend's shops who are making aerospace parts and the tolerance is plus or minus four tenths, no questions asked.
00:11:03
John S
Yeah.
00:11:12
johngrimsmo
And then as you're making those parts, you're like, well, scrap, good, scrap, good, like hold hold the line. um Whereas you and I are too close to the product, you'd be like, well, like it'll be fine if it's a tenth over my knowledge you know my tolerance, but That's too much variables.
00:11:28
johngrimsmo
Like, I don't know.
00:11:28
John S
Well, no, but that's what this solved was like if I
Scrap, Tolerances, and Accountability
00:11:31
John S
picked a part of it out of the scrap bin, which we now have a scrap ah bin that we look at why we scrap what we scrapped. um If I picked one up and measured it and it was 5006.
00:11:38
johngrimsmo
Good.
00:11:43
John S
And Garrett's like, yeah, the the threshold is 5005. That's the outside of the scrap range.
00:11:47
johngrimsmo
All right.
00:11:48
John S
It's a 10th over. Yeah, 10th is nothing. It probably would be fine. But there was no question about it was thrown away.
00:11:52
johngrimsmo
Mm hmm.
00:11:53
John S
um That's
00:11:55
johngrimsmo
Well, in in a lot of our stuff, a 10th will throw it out. A 10th over our max is not going to be good, but we're developing what the actual max is.
00:12:01
John S
Yeah, yeah.
00:12:03
johngrimsmo
And even still we have a goal, say it's plus zero minus two 10ths. But the reality is you could do plus one minus three 10ths or something like that.
00:12:12
John S
Yes.
00:12:13
johngrimsmo
And it'll be like, it'll work, but you don't want to get into that sort of like sideline. Like it'll be fine sometimes because that's too ambiguous for the team to like follow.
00:12:20
John S
you know
00:12:23
John S
yes But you can't, I mean, you can over tolerance things, but the reality is you're either going to be not profitable or not, or you're lying at some point.
00:12:36
John S
I mean, you can, you can attend. I can get something to read a 10th different easy.
00:12:41
johngrimsmo
Sure, yeah.
00:12:41
John S
Um, so yeah.
00:12:44
johngrimsmo
But it all comes down to the actual assembly and what is required. um You know, I have, many times people have seen what we're doing and they have experience in aerospace or whatever, and they're like, you're working too hard.
00:12:49
John S
Yeah.
00:12:58
johngrimsmo
You're over tolerancing for no need. And just because you want it to be perfect, I'm like, well, of course I want it to be perfect. This is my this is my baby.
00:13:07
John S
Yeah.
00:13:07
johngrimsmo
However, I'm also seeing um As I have tighter and tighter parts, I can see what too big of a pin does to a too small of a hole. And I can learn what is a clearance fit and what is a sliding fit and what's a, what needs to be burnished or lapped or, you know, um I'm learning, which is fascinating.
00:13:17
John S
Yeah.
00:13:28
johngrimsmo
And you're literally developing those tolerance ranges based on actual stack up history.
00:13:33
John S
Yeah.
00:13:33
johngrimsmo
which is fun, but there's a lot of science and math behind it that I kind of don't care about, but it's like real world, you know?
00:13:38
John S
Yeah. Yeah.
00:13:43
John S
Yeah.
00:13:47
John S
Okay. Let's go back to that, uh, code. Cause I'd written a note. I was like, I asked you on to reexplain the chop grinding G coach has you BT explain like I'm five.
00:13:53
johngrimsmo
Sure, sure.
00:13:57
johngrimsmo
Okay, I'll start over.
Optimizing G-code with Python
00:14:00
johngrimsmo
What we're doing now is I'm having Fusion output a 2D contour toolpath.
00:14:07
John S
Okay.
00:14:07
johngrimsmo
And I got the tolerances and the, you know, when you hit post and it says like built in settings and that i that i never I never touched those.
00:14:13
John S
Okay, yes, when it's normally hidden hidden, yeah.
00:14:17
johngrimsmo
um I needed to because it broke a very large arc, like a hundred inch arc.
00:14:22
John S
Oh, sure.
00:14:22
johngrimsmo
into smaller sections or something. And I had to bump that up to to do that.
00:14:26
John S
Interesting.
00:14:27
johngrimsmo
And that was interesting. Anyway, so I'm having fusion output, a 2D contour, which posts as literally a lead in arc, a straight line, a big radius, and a lead out arc, four lines of code, more or less.
00:14:36
John S
Okay. Which is sort of the whole profile of the knife blade.
00:14:40
johngrimsmo
of what I'm trying to chop.
00:14:42
John S
or whatever.
00:14:42
johngrimsmo
Yeah, if you look at it, like if you're staring straight at the edge with the tip pointing at the sky kind of thing, and the fixture holds the blade at a four degree angle, so that bevel the side of the blade is vertical.
00:14:42
John S
Yeah. Yeah.
00:14:45
John S
Yes.
00:14:51
John S
Yeah.
00:14:53
John S
So you want to chop right up and down what a path that is normal to that contour.
00:14:58
johngrimsmo
Exactly. um And then in Fusion, I also have my Z heights like go down to my bottom chop depth, do the 2D contour, come back up to a clearance plane so that upper and lower clearance plane are now in the code as like clearance move to Z, move down to, you know, cutting zone and Z and then do the 2D contour and then come back up.
00:15:06
John S
Yes.
00:15:20
johngrimsmo
So. what I used in chat to create this Python script is like, here's my my upper and my lower of this top grinding code. And I want you to segment every arc and every straight line into variable, but in this case, 4,000 increments using math, pi, tangents, everything um so that it perfectly breaks an arc into 4,000 or arcular whatever around the circumference.
00:15:50
johngrimsmo
um You know what I mean? Like chunks, not straight lines, but actually like fourth out across the radius.
00:15:54
John S
Yes, along.
00:15:57
johngrimsmo
um and a couple back and forths to get it to work right. And then I got it great. And then when an arc ends and the straight line begins, if say you run out of fourth out and you're left with like two and a half, I don't want that either.
00:16:08
John S
Oh, sure.
00:16:09
johngrimsmo
So I'd have an average over the entire arc, average it less than four, but to make it even so that the next one is is an even distance.
00:16:11
John S
Good grief. No.
00:16:16
John S
Even.
00:16:19
John S
Yeah, sure.
00:16:20
johngrimsmo
And that works great.
00:16:21
John S
Okay. Can I see if I re-explained this again, like I'm five.
00:16:25
johngrimsmo
Sure.
00:16:25
John S
You ever seen the kids like gym toy thing or like jungle gym type thing where you walk along a rope, but you also have a rope above your head that you can grab onto?
00:16:35
johngrimsmo
Okay.
00:16:36
John S
you you That makes sense.
00:16:36
johngrimsmo
Yeah.
00:16:37
John S
so
00:16:37
johngrimsmo
Like a slack line, but with a top part.
00:16:40
John S
Yes, so I'm thinking, so the slack you have a bottom slack line, a bottom rope, and then you have a top rope.
00:16:41
johngrimsmo
Okay.
00:16:45
johngrimsmo
yeah Yeah.
00:16:46
John S
Those are your two rails.
00:16:47
johngrimsmo
Yes.
00:16:47
John S
You're not actually cutting when the tool is moving along those. You're cutting as you should move up, over, down, over.
00:16:51
johngrimsmo
Correct. Yep. Yep.
00:16:54
John S
So you chat GPT, put in the code, like does it modify your G code?
00:16:55
johngrimsmo
Exactly.
00:17:01
johngrimsmo
Yeah. It took, it took four arcs and lines and I added it.
00:17:02
John S
Dude.
00:17:06
johngrimsmo
I wanted, I told it, I was like, I want it to go, and it knows XYZ. It knows F for feed rate. It knows these things.
00:17:12
John S
you You pasted the code in, like you pasted the gcode in.
00:17:13
johngrimsmo
And I know my, the, I'm not using chat to generate my final code. I'm using chat to generate a Python script that modifies whatever code I send at the Python script.
00:17:25
John S
Stop.
00:17:25
johngrimsmo
Yeah, because I don't trust chat to give me like runnable code, but I trust it to build a Python script that is locked down.
00:17:29
John S
Interesting. So the Python trip, you feed the your gcode into the Python script.
00:17:36
johngrimsmo
Yep. Top line of code.
00:17:36
John S
Oh my god, that's so much smarter because
00:17:37
johngrimsmo
It says, it ah it says called the file name, this file name, output, this file name dash modified or whatever.
00:17:43
John S
Dude.
00:17:43
johngrimsmo
And then I can tweak. So when I changed between VL and Rask, I was just like, oh, 4001 and oh, 40021.
00:17:49
John S
Yeah. Yeah.
00:17:51
johngrimsmo
And it it just worked.
00:17:52
John S
Oh my God.
00:17:52
johngrimsmo
So. It was, it's, it's pretty awesome. And then I used and NC viewer to view what's happening in my output code.
00:17:58
John S
Huh. Yeah.
00:18:00
johngrimsmo
And I'm like, yep, down, over, up, over, down. And then the backlash compensation where it backs away and why by five thou moves over an X comes back in. And I made everything a variable. So my, my move feed rates are one inch per minute or four inch per minute, or whatever I want variable at the top, my, uh, fusions out in the Python program.
00:18:16
John S
Yeah. Variable energy code or just in the Python program. Okay.
00:18:22
johngrimsmo
And my, so my infusion, I basically say feed down in Z at 106 inches per minute and do a transition feed rate at five or whatever.
00:18:33
John S
Okay.
00:18:33
johngrimsmo
And I had the script realize those and be able to override them if I want.
00:18:38
John S
Yeah.
00:18:38
johngrimsmo
with another variable at the top. And yeah, so it's just going down, over, up, over, down, and across these arcs, and across a straight line with totally tangent, totally perfect, mathematically, no tolerance ranges, no rounding, no nothing.
00:18:45
John S
Yeah.
00:18:53
johngrimsmo
And it's freaking sweet.
00:18:55
John S
Dude, tip of the hat. I give you a lot of credit too, because it's like, you're put you're exactly right.
00:18:57
johngrimsmo
Thank you.
00:19:00
John S
I would never trust chat once, let alone every time.
00:19:03
johngrimsmo
Yeah, 100%.
00:19:03
John S
It's just not hallucinate, and that's bad news bears.
00:19:05
johngrimsmo
Yep.
00:19:06
John S
But a Python script, you can read and edit and look at it, and then you can kind of lock it down.
00:19:09
johngrimsmo
Mm hmm. Yeah, there's not.
00:19:11
John S
Ho, ho, ho. And the Python script is not 20 pages long.
00:19:15
johngrimsmo
It's 200 lines.
00:19:17
John S
Eh, it's pretty legit, yeah.
00:19:18
johngrimsmo
Yeah.
00:19:19
John S
You wouldn't have written that from, you or I wouldn't have written that from scratch.
00:19:20
johngrimsmo
No, impossible, like people, smart people can, but I, I can copy and paste and tweak and modify, but
00:19:24
John S
Yeah.
00:19:28
John S
Can you read through it and kind of see what's followed along what's going on?
00:19:30
johngrimsmo
Yeah, the but I just enough like, I don't need to.
00:19:34
John S
Yeah.
00:19:35
johngrimsmo
It's basically go back and forth with chat and be like, I needed to do this feature and and that's wrong and that's wrong. And I actually got to a character limit in the chat.
00:19:42
John S
Yeah.
00:19:45
johngrimsmo
There's like in chat, you use a breakout thing where it edits code and
00:19:50
John S
Okay.
00:19:53
johngrimsmo
It, chat told me it was 25,000 characters, but my code was only 10,000 characters and it was still saying truncated file too long and it wouldn't finish generating the file.
00:20:01
John S
Make it be counting spaces or something weird.
00:20:03
johngrimsmo
That could be, um, but anyway, I was able to get around that and, uh, just weird little limits.
00:20:08
John S
you You're not even paying for 4.0,
Learning from Robotics Integration
00:20:11
John S
you're just using, okay, okay.
00:20:11
johngrimsmo
I think I am. Yeah. I have a subscription. I don't know if it's required for this, but I'm like, I use it enough.
00:20:13
John S
You do. Yeah, yeah, yeah.
00:20:16
johngrimsmo
Why not? Just support these.
00:20:18
John S
Okay, can I segue that to a new point?
00:20:20
johngrimsmo
Hmm.
00:20:23
John S
Okay, very quick recap, thinking about buying automated five-axis with robot, mixed loading solutions, Hermla is the one that makes this, it's great.
00:20:30
johngrimsmo
Yeah.
00:20:32
John S
Lots of questions, uncertainty, major capital expenditure. So you do you date do digging, and get smart. Shout out to some viewers who have reached out. One guy has an incredible ah
00:20:44
johngrimsmo
No
00:20:44
John S
sell, build around brothers. um He's in the firearms industry. Yamazin started it, but he ended up taking it over for reasons I won't get into that, you know, kind of sour grapes, but he also, it's kind of like, look, at the end of the day, he was going to have to learn the system because that's the point.
00:20:52
johngrimsmo
way.
00:21:01
John S
If you're going to do good, not cheap because of Yamazin, I think.
00:21:03
johngrimsmo
Brothers with what robots just tie together? Okay, so but like industrial like fanic style, okay Sick
00:21:10
John S
Yes, yes, um but he's doing what I want to do, which is he's using these robots to load and unload raw material and load and unload different work holding fixtures. Yes, and like super dialed stuff like using, grabbing castings and using a laser to back plot or like ah adjust how it's grabbed the casting so it can load it into a fixture better. um And I bring all that up to just sort of say there's a thesis around if you're going to do this, you might as well learn robots full stop.
00:21:39
johngrimsmo
Yeah.
00:21:39
John S
That's not a small undertaking.
00:21:41
John S
But um then I thought we have a UR on our floor.
00:21:41
johngrimsmo
No.
00:21:44
John S
It's getting installed next week on the Wilman. But that's going to be, is ah that doesn't I don't want that to be my sandbox test robot um for a whole bunch of reasons. So then I was like, well, I could just go buy a used UR. And that might be still what happens.
00:21:56
John S
but then It occurred to me, William here, hold on. I still have a very good relationship with Tormach.
00:22:06
John S
We are not like active members of the Tormach world anymore.
00:22:10
johngrimsmo
Yeah.
00:22:10
John S
um We've for be blunt kind of grown beyond them. um Although total random part, we are getting a 1500 because ah Everybody here wants an R and&D machine that doesn't get turned into a production machine, which has happened with our last two hostas.
00:22:22
johngrimsmo
yeah Every other, yeah.
00:22:25
John S
But anyway, I reached out and they are probably going to be willing to lend us z a ZA6 robot. um And what I love about that is
00:22:32
johngrimsmo
Okay.
00:22:36
John S
to really two things. Number one, um it is fed off Python and PathPilot and ROS, and it's very like it's a robot operating system.
00:22:41
johngrimsmo
Really?
00:22:44
johngrimsmo
What's Ross?
00:22:46
John S
It's like Linux CNC for robots. It's a huge ecosystem in the world.
00:22:48
johngrimsmo
Interesting.
00:22:50
John S
um That's number one, that's a good point. And the second point, which is related, is that a year or two or three ago, me diving into Python with this is different than it is today, where I can use use cursor or chat.
00:23:00
johngrimsmo
Yeah.
00:23:02
John S
Actually, I got to give Justin at PDX a laughing point. Shout out.
00:23:06
johngrimsmo
Yeah.
00:23:06
John S
He DM'd both of us and was like, oh, if you're using chat, you got to use cursor. And then he's like, oh my God, 20 seconds later, sandra Saunders mentioned cursor. But PSA, there's some really incredible coding tools out there, even beyond chat.
00:23:19
johngrimsmo
Yeah Yeah
00:23:19
John S
and So to bring this story home here, what I'm going to do is not build the system with a CNC machine, because I don't care about the machine right now.
00:23:31
John S
I care about, can we build a robot to do what we want? So offline in a closet or whatever, I'm going to build a put a robot down with material, vices, puck chucks, and just see, can I get the robot to first just unload and load material?
00:23:45
John S
That will not be difficult. But then can I get it to switch grippers and load different material?
00:23:47
johngrimsmo
Yep.
00:23:49
John S
Then can I get it to lift the part off the puck chuck, put a different part on, and just kind of build from there and gain that confidence around. There's a lot I've got to learn on robots.
00:23:59
John S
But at the end of the day, this is it's checking a box for me personally, if I'm being honest, you know you know I talk about this.
00:24:00
johngrimsmo
For sure. Yep.
00:24:06
John S
It's going to help us make a better decision if we end up going with an integrator or an OEM solution. And I suspect in many respects like things like Lex or other systems, we're going to end up just probably crushing it.
00:24:17
John S
and you know whether we buy a big UR or ah another industrial robot that probably has more reach than the Tormach one, I'm grateful that we have the chance to kind of sandbox this and and learn.
00:24:22
johngrimsmo
Yeah.
00:24:29
johngrimsmo
Yeah, and play.
00:24:31
John S
Yeah, I think that's the plan at least.
00:24:32
johngrimsmo
My like, other than the aroa palette change we have, which is PLC driven, and I've i've hacked into it enough to like, change
00:24:41
John S
Mm-hmm.
00:24:44
johngrimsmo
forget what the term is, but like where it perks the palette in the current, basically, you can tweak the in out up down and kind of thing.
00:24:48
John S
Yes. Yes.
00:24:51
johngrimsmo
Because when we move the current, we reintegrated that ourselves, we retaught it the endpoint kind of thing.
00:24:55
John S
Oh yeah.
00:24:57
johngrimsmo
And talking to Tina at current, she's like, most customers don't do that.
00:25:01
John S
but or Right.
00:25:01
johngrimsmo
I'm like, I'm not most customers.
00:25:03
John S
Right.
00:25:03
johngrimsmo
um So that was cool to be able to do. But otherwise, I've never touched a robot. But my one of my concerns with your plan is, Say you put all this work into the tormac robot and then you choose to go with a notchy or whatever.
00:25:16
johngrimsmo
And the language is completely different in the, like you got to rewrite it all. But maybe there's super value in you proving the process and the work holding and the g gripper and the changes and the quick, all that stuff.
00:25:27
John S
Yes.
00:25:29
johngrimsmo
And then you could just reapply all learned knowledge onto whatever programming platform is required. Like they can.
00:25:34
John S
No, so it's a great question. I had the same question myself. The the specific answer is that Tormach uses Python, URs use URScript, which I believe is Python-esque, or it wouldn't be from...
00:25:51
John S
You're right. There may be some redundant overdue work, but it's absolutely still the path I'm going to go. It's like the kind of fail fast, fail cheap. and the Right before you and I hit record, I was commiserating how I do feel like I've got a lot on my plate and anyone who's listening that runs a small business probably ebbs and flows feeling the same way.
00:25:59
johngrimsmo
I like it, yeah.
00:26:08
John S
and Part of me even thought about putting the robot at home and treating this as a side personal project um because
00:26:13
johngrimsmo
yeah
00:26:15
John S
I don't want to. i don't if this If two months in, I realize this is a mistake, then no big deal. It's not going to be that way. But um I will gain so much from the confidence of setting up a robot and then understanding the frames, the kinematics, whether you're driving it with with yeah angular specific specificities or or coordinate systems.
00:26:29
johngrimsmo
I agree.
00:26:37
John S
And you know how are you doing sensors to detect material present or not present to fail? like i'm I'm totally confident. And again, like Lex in our ERP system, if the whole point of Lex was to show really our P-System company what we wanted, then we ended up never going there.
00:26:53
johngrimsmo
yeah
00:26:54
John S
Here, if we build this working functional, then it and it makes sense to hire an integrator or a carmela, be like, hey, we built this as a replica. um Here's everything it does. Here's what it needs to do that it can't do.
00:27:05
John S
And that would be a great start-up point.
00:27:07
johngrimsmo
Sure. Or as I see it, you'll just be rolling your own and you'll just have to ask her really like, how do I make it push cycle start?
00:27:12
John S
ah Yeah.
00:27:15
johngrimsmo
Because that's what I need from you right now.
00:27:17
John S
no
00:27:19
John S
Well, and you said there's a UR30s and there's, um yeah, that that'll be the last, and I was, you know, a person who was a mentor here was like, that will be the hardest part, will be um not only on the robot side, but having the machine to recognize different workholding, different material, and now loading a different NC file because, you know, the going to be the hardest part.
00:27:43
johngrimsmo
Yep.
00:27:47
John S
but And you're going to laugh. This is probably not the way to go. But fun fact, a UR robot can go up to a machine tool control panel and start pushing the buttons.
00:27:57
johngrimsmo
Yeah.
00:27:58
John S
So if it's as dumb and brutal as switch to this palette and when it switches to that palette, it's going to pick up the soft point pen and it's going to go up to the, it's not going to be a Haas, but a Haas control panel and punch in, load this program, you can do that.
00:28:01
johngrimsmo
three Yeah, absolutely.
00:28:11
johngrimsmo
That's a good point.
00:28:13
John S
I'm having fun. It will figure it out.
00:28:16
johngrimsmo
That's good. Yeah, and tos so many people have done it. There's solutions for most of it, you know? You just kind of kind of put it together into your perfect plan.
00:28:24
John S
Yeah. Yeah.
00:28:26
johngrimsmo
Well, that's super exciting.
00:28:27
John S
Thank you.
00:28:28
johngrimsmo
I'm here for it.
00:28:29
John S
No, I appreciate it. The code, um I really am fired up just because I um but blown away what chat GPT does with code. It's nuts.
00:28:38
johngrimsmo
Yeah, yeah.
Challenges with Willaman Machines
00:28:40
John S
Yeah.
00:28:42
John S
yeah
00:28:45
John S
Oh, I had a Wilhelmin question. Can I ask or talk about?
00:28:47
johngrimsmo
Okay, and I have a Wilhelmin point, you first.
00:28:49
John S
Okay, this has actually been on my list for like a month. I told Grant I'd ask you and we just haven't gotten to it. um He was having a problem with some parts either getting scratched or not releasing from the Wilhelmin sub-axis vise and the Jaws that I believe were the culprit were, yeah, they were 41-40 that we hardened to about 50 HRC. um Side note, were we're not really using 41-40 anymore. but We now prefer the air hardening tool steels that we're now buying and just building workflows around. but
00:29:25
John S
These happened to be 4140. Had some light scratches on them, um and they could have been stoned out better. But what we think happened was the opening and closing of the automatic vise on the parts turned those jaws slightly magnetic, which is Like if you bang material together, but but I wish I could tell you more, it becomes magnetic or lightly magnetic.
00:29:46
johngrimsmo
I don't know.
00:29:53
John S
I know this because years ago we fought with our suppliers about material arriving in different different levels of magnetism. We bought a Gaussian meter and one of the culprits is when materials on a truck bouncing around and slamming into each other.
00:30:06
johngrimsmo
No way.
00:30:08
John S
Oh yeah. And some of the saws that they use are so, you know these guys are just running these saws full out. So it's just slamming the parts against the backstop, slamming an automatic vice closed on it. um That also magnetizes or c creates some form of magnetism in the parts, which attracts chips and that's not fun.
00:30:26
johngrimsmo
That is wild. Do you have a demagnetizer?
00:30:29
John S
We have a demagnetizer. We use it on a lot of material, but I don't want to use it. I mean, Grant didn't want to take it. it's a It can be picked up and held.
00:30:41
John S
It's like a shoebox thing, but we didn't want to hit, it's a giant transformer.
00:30:42
johngrimsmo
Yeah.
00:30:45
John S
So it's not like, I'm sure um putting an iPhone on it or something could probably be a bad, ah actually it could be a great way to have a nice days off in the shop.
00:30:52
johngrimsmo
yeah
00:30:55
John S
No distractions for hours. um Anyway, i just I guess the question is, have you seen this in your turning centers and your Wilhelmin?
00:31:04
johngrimsmo
I can't say I've seen that anywhere. um There's something tickling my brain about hearing about that, creating magnetism from something, but I barely remember it.
00:31:12
John S
yeah Okay.
00:31:14
johngrimsmo
And that's super fascinating.
00:31:14
John S
Okay.
00:31:16
johngrimsmo
Most of my jaws are not steel.
00:31:20
John S
What?
00:31:20
johngrimsmo
They're just the aluminum ones I buy from CJ.
00:31:22
John S
Oh! Okay, noted.
00:31:24
johngrimsmo
I just buy them like 10 pairs at a time and I'm like, CJ, I need more.
00:31:28
John S
Yeah, sure, sure.
00:31:28
johngrimsmo
Um, I made some steel ones, but I had them fixed in the Kern and it's just took up too much time, too much current time to like worry about them.
00:31:35
John S
Yeah.
00:31:36
johngrimsmo
And when I can just buy them from CJ and put them right on all day long.
00:31:39
John S
Okay.
00:31:40
johngrimsmo
Um, so yeah, I don't have that problem, but they don't close that fast. Do they? I mean, they're, they're hydraulic the way they close and mine just kind of go.
00:31:50
John S
yeah Sorry, that is true. It is not a slapping, pneumatic-y slam action, but um
00:31:55
johngrimsmo
Yeah.
00:31:57
John S
I assume,
00:31:58
johngrimsmo
We just think that they're not magnetic and then they get, they are magnetic.
00:32:02
John S
that's what I'm saying. I'm not saying that with like proven conviction, but um somehow they became slightly magnetic over time and that's probably, even like, I don't even, if the material is magnetic, I assume that running, we have not run 10,000 this part, but it running thousands of the parts could itself transfer pass along some.
00:32:06
johngrimsmo
Yeah.
00:32:11
johngrimsmo
Okay.
00:32:22
johngrimsmo
Possibly. Yeah.
00:32:23
John S
um
00:32:24
johngrimsmo
So something I've been thinking about a lot lately is trying to find out what the real issue is. Like, you know, we all kind of complain about a thing and you're like, but but what's, what do we really complain?
00:32:31
John S
yeah
00:32:34
John S
Yes.
00:32:35
johngrimsmo
What do you really want here? You really, you're having chips stick to your vice jaws and those chips are causing marks in your part, right? Like if you break it down.
00:32:45
John S
Worse, it's causing unreliable failure to reject.
00:32:49
johngrimsmo
The part is sticking in the vice, which also has like stickiness of oil and issues like that.
00:32:51
John S
Yes.
00:32:54
John S
Mm-hmm.
00:32:58
johngrimsmo
Is it a lightweight part or kind of a heavier part?
00:33:01
John S
Yeah, this is a lot of ice washer. So it probably weighs an ounce.
00:33:05
johngrimsmo
Okay. Have you done?
00:33:06
John S
Sorry. Sorry. Huge for you.
00:33:09
johngrimsmo
Yeah, yeah, yeah.
00:33:10
John S
Um, but small for us.
00:33:10
johngrimsmo
Well, it like the the clip the clip for our Fjell, I don't know what it weighs, but it's the size of a large paperclip and heavier, but not very heavy.
00:33:17
John S
Yeah.
00:33:19
johngrimsmo
And so I'm using CJ's hack where you put a pipette, a little plastic dropper, eyedropper thingies in an ER-16 collet, and you just push it.
00:33:24
John S
Yes.
00:33:30
johngrimsmo
So for the clip, I have that routine where it opens the jaws, pipette goes down, shoves it out, make sure it's out of the way, and then you kind of don't care.
00:33:30
John S
Yes.
00:33:39
John S
Yes. um
00:33:40
johngrimsmo
Magnetism might be weird in that case, but.
00:33:43
John S
No, and I'm not against that. i I, when I first looked, I don't want to be up-grant here by any means, but when I first looked at the jaws, I'm like, oh, that was one of our first sets we made.
00:33:55
John S
Throw those away, get a set of A2, we'll hard mill, you know, ma machine them, we'll hard mill them or we'll grind them and they'll be 60 rock on that 50. They won't have scratches on them, which all of that is, that could all be done better at this point.
00:34:08
John S
And that's one step.
00:34:08
johngrimsmo
Yeah, yeah.
00:34:09
John S
but Um, I just wanted to ask if you had seen, oh, and then it's also like, well, what if we did like 17 four? I don't know if there's higher hardness, samosas avoid, you know, generally they are.
00:34:19
johngrimsmo
they're still magnetic.
00:34:21
John S
Yeah.
00:34:21
johngrimsmo
Yeah, I mean, there are some that are not, like make them out of titanium or something, but then it won't be as hard.
00:34:22
John S
Okay. Yeah.
00:34:27
John S
No. Yeah. These need to be, it's crazy when you run production, how, how you see where and tear on fixtures and,
Thermal Impact on Machining Precision
00:34:37
John S
and stuff.
00:34:37
johngrimsmo
Yeah, yeah.
00:34:37
John S
And it really is.
00:34:39
johngrimsmo
Yeah. So my Wilhelmin point is, um, I definitely on my older 20 year old Wilhelmin, I definitely noticed thermal growth in Z, like up and down.
00:34:50
John S
Okay.
00:34:50
johngrimsmo
um Like when turning specifically, if you turn the the diameter half inch diameter on a cold machine, and then you run a bunch of parts and you turn it again, it's going to measure out
00:34:53
John S
Okay.
00:34:56
John S
Yeah.
00:34:59
John S
Hmm.
00:35:00
johngrimsmo
It's growing, but almost a thou, let's say, which I guess is half a thou of real diameter radius kind of thing. Um, with milling, I don't see that. So for our field clip, I have a feature that is a critical feature has to be plus one 10th minus one 10th in a perfect world.
00:35:17
johngrimsmo
And holy cow, the machine just hits it cold hits it hot because it's interpolating it.
00:35:22
John S
Right.
00:35:23
johngrimsmo
It's it's not referencing a milled feature to centerline.
00:35:27
John S
Bingo. Yep.
00:35:28
johngrimsmo
It's just making its own feature.
00:35:30
John S
To itself. Yep.
00:35:30
johngrimsmo
And that's you know backlash tool way or things like that. and And hot or cold, it doesn't care. So it got me thinking, like is it better to machine machine critical features with milling as opposed to turning if you can get away with it?
00:35:44
johngrimsmo
And you know what I mean?
00:35:44
John S
Yes. There's probably a change in like the circularity depending on how your machine ball bars at B90, but I believe, I don't know, I say this with hesitation because I don't want to put words in their mouth.
00:35:51
johngrimsmo
Yep.
00:35:56
John S
I believe one of the Willam and Apic guys told me when you need stuff that's critical, use turning to rough it down, leave a couple foul, and then finish it with an end mill interpolation, which just warms my heart to speak, yeah, to talk about how bad lids are.
00:36:05
johngrimsmo
Yeah.
00:36:08
johngrimsmo
That's pretty cool. I could see if, uh, Z growth, like up and down growth could change your. Cylindricity from the actual turned center line to the milled hole in the end or whatever, they might not be concentric to each other.
00:36:21
John S
who
00:36:25
johngrimsmo
Um, but as there's ways to do it, I don't know.
00:36:26
John S
hu
00:36:29
John S
yeah Yeah, the best thing we figured out on stuff like that is also take the machine temp out of it.
00:36:31
johngrimsmo
Yeah.
00:36:36
John S
So doing critical features with the same tool or doing them right back to back so that it's like pinch milling with separate tools where I don't care where the Z casting is.
00:36:46
John S
It's gonna grab tool one and do feature A and then tool two and do feature B. And they will be relative to each other quite accurately, much more so than, um, not do it that way.
00:36:53
johngrimsmo
Mm.
00:36:57
johngrimsmo
I see what you mean, yeah. That's a good point, actually.
00:36:59
John S
That's awesome though.
00:37:00
johngrimsmo
Yeah.
00:37:01
John S
I'm good at it.
00:37:01
johngrimsmo
And then i i I've wondered about and permanently installing a little, ah not a ball bar, but a little spherical ball, like one inch ball or whatever in the corner of a machine that I could probe and like auto-com, you know, it's like, cause that's a fixed place and in space.
00:37:01
John S
So it was okay.
00:37:11
John S
Okay. Oh, competent. Yeah.
00:37:20
johngrimsmo
And as thermal grows the machine, uh, it's just auto-compet. So that's a theory, but, um, otherwise.
00:37:26
John S
Just warming up. we do We have to do it with ours too, um which kind of stinks.
00:37:29
johngrimsmo
Yeah.
00:37:32
John S
You kind of wish, with because ours has all the fancy thermal stuff on it.
00:37:34
johngrimsmo
Yeah.
00:37:34
John S
But grant Grant was just talking about this yesterday because it's been so cold here um that it takes a half an hour to get that machine running.
00:37:42
johngrimsmo
yeah Yeah, we put a therm temperature gauge in the coolant tank, so at least I have a quick visual of like of like cold, is this hot, is this really hot, is this?
00:37:48
John S
Oh, yeah.
00:37:53
johngrimsmo
And we have that in our Swiss too, and the Swiss gets pretty hot.
00:37:55
John S
Grant brought that up, I'm glad you said that, when I was, we were in a meeting or something and it wasn't appropriate to discuss it further, but he was saying it's something to do with getting the chiller fluid level up, and then it made me think, well, if it was just the coolant or the oil, we could put an aquarium heater in it, the' like a super common trick, yeah.
00:38:12
johngrimsmo
Yeah, keep it warm.
00:38:14
John S
um Or even the ability to run that system overnight without having the machine on, um just keep that stuff stable. um I need to look into that.
00:38:24
johngrimsmo
Yeah. I mean, the coolant being temperature is one thing, but you want, what you really want is the machine casting to be at a temperature because that's where the growth comes from. But the coolant kind of delivers that temperature, you know, it's all of it.
00:38:35
John S
I think it's creating a, yeah. e Yes, I know. um I think if you had, I guess what I'm thinking in my head is if you had,
00:38:40
johngrimsmo
Yep. It's interesting.
00:38:45
John S
a very, very accurate or very, very robust thermal comp management system that's using probes and temperatures and you know grids to adjust things for. You're not doing yourself any five favors by then pumping 15 degree different coolant through the machine that's gonna warm up over the course of 45 minutes.
00:39:03
John S
Like you you're asking and the system to do a lot more work than it should have to if you could just get...
00:39:03
johngrimsmo
Yeah, yep.
00:39:09
johngrimsmo
I feel like I've heard it. I've heard some shops keep a machine in a warm up program and just keep it keep the cool and flowing and keep the machine just slow XYZ moves like the Wilhelmin warm up program, but just kind of all the time.
00:39:13
John S
Yeah.
00:39:20
John S
Yes. I've heard, I mean, none of the aquarium pumps I've heard about shops with, uh, I believe it was grinders putting those Mylar space blankets ah around the grinder column.
00:39:29
johngrimsmo
No way.
00:39:30
John S
So yeah.
00:39:31
johngrimsmo
Well, that's kind of cool.
00:39:32
John S
Yeah. All right.
00:39:35
John S
Yeah.
00:39:35
johngrimsmo
But yeah, otherwise ah clips are going great. I'm in one toolpath tweak this morning, so I'm getting some chatter on one feature. Hopefully this gets rid of the chatter, but I've made a handful. And if this gets rid of the chatter, then I'm just making parts and they're, they're tolerant.
00:39:50
johngrimsmo
They're good. Surface finishes are fine. And then I'll i'll keep making them. And then I'm teaching Jeff how to like, how to Wilhelmin.
00:39:58
John S
Good.
00:39:58
johngrimsmo
And he's excited, but he can't just like jump on and start poking buttons because the Wilhelmin's
00:40:03
John S
Yeah, you can.
00:40:04
johngrimsmo
intricate. There's a lot. Yeah. Like I have my I sent you my chicken scratches of these are all my Wilhelmin notes, right?
00:40:10
John S
No, yeah. this case yeah um Because when we were talking last week, you had said.
00:40:12
johngrimsmo
I still have that I haven't formalized it. But
00:40:20
John S
We were talking about, I think focusing in like just hyper, like just do one thing. Don't worry about everything else. Um, wasn't that the Wilhelmin clip or the field click on on the Wilhelmin and you did great dude.
00:40:29
johngrimsmo
Yeah. That was part of it. And I did that. Yeah.
00:40:33
johngrimsmo
And then I did a bunch of other side projects too.
00:40:33
John S
Awesome.
00:40:35
John S
The same guy.
00:40:36
johngrimsmo
That same week anyway.
00:40:38
John S
Okay. Yeah.
00:40:38
johngrimsmo
Yeah.
00:40:39
John S
Okay.
00:40:39
johngrimsmo
No, actually last week when we talked and I was like, today, all I could do is clip. I didn't get to that till like Sunday, to be honest.
00:40:45
John S
Come on.
00:40:45
johngrimsmo
Yeah. No, just didn't.
00:40:48
John S
You want to talk about it?
00:40:49
johngrimsmo
No.
00:40:50
John S
OK, fair enough.
00:40:51
johngrimsmo
Yeah.
00:40:52
John S
OK.
00:40:53
johngrimsmo
But I did other stuff that was important and needed to get done and some of which was maybe not as important, but I wanted to get done.
00:40:58
John S
No, look, yeah, God bless you for that.
00:40:59
johngrimsmo
And here we are.
00:41:02
John S
I i really like. Um, it's why I was so, I mean, I don't, I think that the point of this podcast is keeping things candid right before we hit record.
00:41:11
johngrimsmo
Yep.
00:41:16
John S
I sort of told John, I was like, Hey, uh, I got too much going on both today. Um, and like, I'm trying to do, so do too much right now.
00:41:25
johngrimsmo
Yep.
00:41:25
John S
Period. Like, and that's some of this bingo.
00:41:26
johngrimsmo
Family work all the same time. Yep.
00:41:29
John S
Yeah. And I know folks listening are are there so it's not I'm not looking for sympathy. I'm not trying to complain It's more like what do you do and? um but For me, I love piano. I picked up adult piano three years ago and four years ago. and I think because It's almost like one of those naysayer, like because everybody quits, like because people don't stick with it, I'm i'm like just going out of my way to not be that person.
00:41:54
johngrimsmo
yeah yeah
00:41:55
John S
And I truly love it, but it's hard for me to love it when I feel like, oh, my kid wants to build a Lego together or you know what, I need to do work stuff or this morning it's like negative seven and I got a pipe that burst.
00:42:08
John S
It's like, I, that is more, that has to be, like I'm the person that has to take care of that period or at least start figuring out who to call, et cetera, et cetera.
00:42:16
johngrimsmo
Yeah, yeah.
00:42:18
John S
So that was why I was like, well, maybe you just quit for six months. This sucks to say it, but might as well say it. Like maybe you quit for six months. And then, um you know, like I was saying, do you like build Johnny five, get him finished, which he could be done in that short of timeframe.
00:42:30
johngrimsmo
Yeah.
00:42:32
John S
Yeah.
00:42:33
johngrimsmo
Yeah. We can't do it all, but we have to choose what, what fills it up, what fills us up, you know, what, what fills our cup kind of thing.
00:42:37
John S
Yes. Yeah.
00:42:40
johngrimsmo
And, uh, and those people around us too, your wife, your kids, your employees, your team, um, your fans, us out here in the internet. Um, you know, you're trying to keep everybody's cup full, but you can't always do that.
00:42:48
John S
Yeah.
00:42:52
johngrimsmo
And, uh, I do too, a hundred percent.
00:42:54
John S
I really want to, John. Sorry. like
00:42:57
johngrimsmo
I'm with you. I want to do it all.
00:42:58
John S
I know sometimes we sit here and I i feel like I tell you, oh, John, you're doing too much. you're Like I will tell you right now, i'm I'm proudly taking over that seat of like, I want to have it all.
00:43:03
johngrimsmo
Yeah.
00:43:07
johngrimsmo
yeah
00:43:08
John S
Yeah.
00:43:09
johngrimsmo
and that's That's ambitious. That's okay. um I mean, your piano lessons are are one thing you cannot delegate. to but is That is a personal thing.
00:43:16
John S
Yes. Yeah. So the lessons are the, the lessons are the easy part. It's like what I enjoy is spending 45 minutes in the morning. No one else is awake. Like kids are sleeping and I just go play by myself.
00:43:25
johngrimsmo
Yeah.
00:43:27
John S
Like that's great.
00:43:28
johngrimsmo
Yeah.
00:43:30
John S
Yeah, be it'll be okay.
00:43:31
johngrimsmo
I get that.
00:43:34
John S
um hey Another one I want to check off my share list.
00:43:35
johngrimsmo
Cool.
00:43:37
John S
A couple weeks ago, we were like, oh my God, the Okuma's leaking coolant. What's going on? And that sometimes that machine can get weird. Like if the conveyor gets turned off by accident or you get a chip built up in a weird way, it can start flooding over.
00:43:50
John S
there's only yeah hasn't It doesn't happen, but it could happen. And so I was like, what's going on? And Garrett took a look and sure enough, um well, what had happened is after three years of that machine running a lot, there was one hose, coolant hose that was up against a piece of sheet metal corner and it vibrated up and down and it wore through that.
00:44:10
johngrimsmo
Okay.
00:44:12
John S
Now this hose, if you told me this hose was like the main hydraulic hose to lift up Mount Everest, I would believe you. it's like
00:44:19
johngrimsmo
Massive.
00:44:19
John S
ah inch and a inch and a half OD, you can barely move the hose. It's so stiff and thick. um It's just one of those crazy hoses.
00:44:26
johngrimsmo
Yeah.
00:44:26
John S
So I'm kind of like, you know, see, you know, I'm thinking to myself, Oh, and then at first I was like, Okay, what do we do? I Tried we have that um roofing patch Material stuff like kind of like flex seal, but whatever put that on there immediately blew right through that It's it's not the high pressure but even the flood going this is probably 100 plus PSI.
00:44:42
johngrimsmo
Okay.
00:44:51
John S
It's a lot and so I Looked at it.
00:44:52
johngrimsmo
Yeah.
00:44:55
John S
and I'm like waiting here Well, I don't have to call Gossager and spend two grand or probably more to have hoes shipped in, service tech come in, all that. I was like, wait a minute, we got plenty of, there's actually plenty of slack in this whole line.
00:45:07
John S
And so we just figured out the fitting size, bandsaw through the hose, put a male to male coupler with with with two high quality codes close hose clamps fixed.
00:45:15
johngrimsmo
Okay.
00:45:20
johngrimsmo
Done.
00:45:21
John S
Yeah, couldn't have been easier.
00:45:22
johngrimsmo
It's not like a braided, super rigid, um, you know, hydraulically pressed together kind of hose or crimp fittings or anything.
00:45:29
John S
Sorry, great point. No, it's not a true hydraulic hose.
00:45:33
johngrimsmo
Okay.
00:45:33
John S
It's like 5,000 PSI. And i knew I knew we could do this because the way it connects into the pump and machine terminals is the same thing with a barbed, one inch, a one inch barbed hose with a high quality closed clamp done.
00:45:34
johngrimsmo
Yeah.
00:45:42
johngrimsmo
Got it. Yeah.
00:45:47
johngrimsmo
Perfect.
00:45:47
John S
Yeah, that felt good.
00:45:48
johngrimsmo
Good is new. Yeah.
00:45:49
John S
Yeah.
00:45:50
johngrimsmo
Love it.
Shop Maintenance with Concrete Slab System
00:45:52
John S
And the last thing, this is the start of this conversation for us, not the end, but We blade out a shop zone system and the zones are based on the concrete slab cuts.
00:46:05
johngrimsmo
Okay.
00:46:06
John S
So we have four by three, so we have 12 zones in the shop and that will serve as the reference guide for weekly cleaning tasks by person like clean the toolboxes in zone four or clean the ceiling in zone six or look at the Whatever in that zone and then also um Probably in the spring, you know, we've got some stuff to do ah right now Otherwise, but the spring I'm going to start having everybody spend five minutes or ten minutes together Like hey everybody let's go to zone 11 and let's just all look around like why is there that?
00:46:23
johngrimsmo
Okay.
00:46:39
John S
piece of cord laying on the ground or what does this do here and breaking it into those little
00:46:41
johngrimsmo
Yeah.
00:46:44
johngrimsmo
Kind of like that.
00:46:45
John S
zones um I think will help and the concrete slab makes it easy to know what zone you're in.
00:46:51
johngrimsmo
Easy. Yeah. Yeah.
00:46:54
John S
Yeah, with the, they're on a, yeah, I don't have it in front of me, I'll show you, but it's, we we um
00:46:54
johngrimsmo
Are you going to physically, physically label them like at, at each zone?
00:47:04
John S
Actually, sorry that's how we did it. ah we have When we created the 3D model of our shop, we lasered in the slab cuts so that we knew where to place machines relative to expansion joints or the slab joints.
00:47:09
johngrimsmo
Yeah.
00:47:13
johngrimsmo
Right.
00:47:17
John S
so And so that's accurate enough. And so I just took an aerial photo of our 3D print and then labeled that.
00:47:21
johngrimsmo
Nice.
00:47:24
johngrimsmo
That's perfect.
00:47:25
John S
Yeah, that felt good.
00:47:27
johngrimsmo
Awesome. All right.
00:47:32
John S
Anything up to today?
00:47:33
johngrimsmo
I have two tasks today, which I will accomplish, um because they're pretty easy.
00:47:36
John S
Good.
00:47:38
johngrimsmo
One is keep making clips. If I have time, I'll teach Jeff some stuff on the Willyman.
00:47:41
John S
Yep.
00:47:42
johngrimsmo
but um We've made so many handles, fail handles, and so many blades. And I've test fit various things, but I need to do a random fit up test between handle 16 and blade 37 or whatever.
00:47:52
John S
Oh. Mm hmm.
00:47:56
johngrimsmo
And just, I need to like that mental closure of like, yes, they've passed all of our gauge checks, things like that. But I just need to grab anyone and anyone and just, you know, see what the reality is.
00:48:08
John S
Sure. Proof is. Yeah.
00:48:11
johngrimsmo
So I'll be doing that today.
00:48:12
John S
Good. Good.
00:48:13
johngrimsmo
But otherwise, I mean, what's today, Wednesday? um We're gonna finish some PLs this week, for sure.
00:48:20
John S
Awesome.
00:48:21
johngrimsmo
Which is gonna feel amazing.
00:48:24
John S
Yeah. Good for you, dude. That's awesome.
00:48:25
johngrimsmo
Yeah, it's exciting.
00:48:26
John S
Even if you don't do a hundred this month, you're like, it doesn't, you're there.
00:48:28
johngrimsmo
Oh, for sure.
00:48:30
John S
Like you're.
00:48:30
johngrimsmo
Honestly, that goal, like, I've always viewed goals as just made up targets. I was like, yeah, whatever. like But but it's it's changed how we've done this month.
00:48:34
John S
Yeah.
00:48:41
John S
Yes, that's the point.
00:48:41
johngrimsmo
Like it it actually, it changed every decision we made this month by having this audacious goal and it's working.
00:48:50
John S
Good. Does it help you? You don't have to answer this if you don't want to, but does it help you also say no to stuff? Cause you're like, no, right now?
00:48:56
johngrimsmo
Kinda, yeah.
00:48:56
John S
Yeah. Good.
00:48:58
johngrimsmo
100%, which I need.
00:48:58
John S
Good.
00:48:59
johngrimsmo
Cause otherwise I'll say yes to everything and then get nothing done.
00:48:59
John S
Yes. ah Yes. Yeah. Yeah.
00:49:03
johngrimsmo
So that's kinda cool. i'm I'm chewing on that.
00:49:04
John S
Yeah.
00:49:05
johngrimsmo
It's like, I gotta do more of this, like.
00:49:07
John S
I need you to find you, I need to find you a t-shirt that just says no means no.
00:49:11
johngrimsmo
I like that.
00:49:14
John S
Awesome.
00:49:15
johngrimsmo
Cool, man. All right.
00:49:16
John S
I'll see you next week.
00:49:17
johngrimsmo
See you next week. Have a good day.
00:49:18
John S
Bye.