Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
#338 Bambu 3d printers!
370 Plays1 year ago
TOPICS:
- Lots of problems
- Building custom computers for CAD/CAM
- Willemin is sorta broken
- Bambu 3d printers!
- Calibrating printed parts
- Machining carbon fiber in the Speedio
- SMW puckchuck to market
Recommended
Transcript
Introduction to Episode 338
00:00:00
Speaker
Good morning. Welcome to the business of machining episode 338. My name is John Saunders. And my name is John Grimsmo. This is the podcast for John. He talks each week about what's happening in our private-ish lives as manufacturing folks. Each week, it hit me when you said 338. I'm like, that's a lot. It is. I like to think that you're sitting down for
00:00:29
Speaker
that many, well, what is that? 338 divided by 24, 14, almost even 14 days. 52. Well, okay. Days, yeah. That's crazy. Oh yeah. Like how many years have we been doing this? 338 divided by 52, six and a half years. Really? Yeah. That's nuts.
00:00:53
Speaker
Yeah. Good. Cool. I love it. Agree. Good. Okay.
PC Freezing Issues in the Shop
00:01:00
Speaker
I got lots of problems. I got lots of solutions. Good. I am the source of positivity and solutions. Good. Both of these are venting slash questions to you slash questions to our audience. Number one, we will get through this. It may actually already be solved, but we had a person who came well recommended build three PCs for our shop as kind of fusion.
00:01:22
Speaker
experts in life, I'm happy to admit, I don't know that much about computers anymore. It's a little bit hard because man, 20 years ago when I built them, it was so fun. You know, hearing about thread rippers and single cores and what's using the GPU CPU, I told you. Overclocking and all that stuff. Yeah.
00:01:37
Speaker
in brief background, I have a local guy who's excellent, but he doesn't know CAD game specific. So I thought I'll give this other guy a try to build three PCs for us that are, you know, that they're involved in a shop and they know this stuff. And Bill was, you know, great all that except two of the three are regularly freezing. And
00:02:01
Speaker
And I brought one of them into my shop just to make sure I wasn't missing something and have it set up on my computer. And sure enough, it can freeze intermittently but regularly. Just sitting there?
00:02:16
Speaker
Yeah, just sitting there like I was obviously trying to stress it so I was running long simulations. And like it one example, it didn't freeze then but then it I went to by closing the simulation that I went to close the file and just before my cursor hit the X, it froze. And so when it freezes the mouse
00:02:34
Speaker
cursor locks in place. There's no blue screen of death or any other visible acknowledgment of the freeze. It just completely is unresponsive. The controller delete doesn't do anything. It requires a full hard power cycle by hitting the power button. It's much different than fusion freezing on its own. Correct. It is not a Windows program. Yes.
00:02:57
Speaker
So we're in that situation that I think a lot of us have been in, where as business owners are buyers of stuff. Of course, I don't think the person that built these thought, ooh, I'll do a mediocre job. But I regret, I bought all the components and had them shipped to him. And he took a couple weeks longer than I thought he would, although I didn't discuss timing, so that's on me. So the components are outside of the return window.
00:03:27
Speaker
And so I'm sitting on these three computers that aren't functioning as well as I'd like them to. For sure. I believe one of them does work. So that's strange. So we had the SSDs replaced, and I've just continued to use it. It's actually been working OK on an insider build of Fusion and not on the core Fusion. There's an update coming out today. It feels fishy to me to say that it's, but what I've heard is like a Fusion MSI motherboard bug.
00:03:54
Speaker
Google search didn't yield anything super fruitful on it. And look, part of me is just regretting my local guy would have just taken these back and fixed them. And like, no stress on my like, no work on my work, which is what I'm doing now. I don't like and or buying like a Dell where it's just like, no, it's just they're just good. So you have any insights out on this? Or what's your end? Really? I mean, we've been buying several laptops for around the shop, but they're just kind of $1,000 off the shelf gaming laptops. Yeah.
00:04:25
Speaker
And no, I don't know. So like doing some research, like what's a good computer for fusion? And from what I've learned in my own experience, lots of RAM, everything else is decent. SSD, I don't know. Everything else doesn't matter that much, but I'd like more information on that because I don't know.
00:04:41
Speaker
Humbly said, I don't think that's the whole story. Like the HP folks did a little private session at DSI that you would have been to. I wish you were there. And they talked a lot about what aspects of fusion are single core versus what support multi-core. And like the idea that the problem with like adaptive is you can't distribute across multiple cores because adaptive has to know what's coming to it has to happen.
00:05:10
Speaker
in a straight line. You can't distribute it amongst literally eight different processors, if you will. But again, if you were calculating eight different toolpaths at the same time, could it distribute those better?
00:05:24
Speaker
I don't know. I think that depends on like a rest machine. There are some operations where yeah, I can like I'm making this up. It can handle seeping shallow while it's handling drilling while it's handling contour. So my layman understanding is that the CPU for sure matters a lot as well as as well as understanding if you need really good single core performance less ish on the GPU and yet for sure decent RAM and SSD my gosh for sure SSD.
00:05:55
Speaker
Yeah, a lot of the laptops are 16 gig of RAM, just how they came. Everything I've been running is 32 gig. Then these laptops, they drag a little bit, just using Fusion. I'm like, that feels a little bit slower than I'm used to. I don't know if I just need to upgrade them to 32 gig of RAM and call it good or what.
00:06:15
Speaker
For what it's worth, these three computers were identical. They are Intel, i7, 3.4 gigahertz, 32 gigs of RAM, and Windows 11.
00:06:30
Speaker
It's the G460 something. Good graphics cards for sure. I don't want to pull that up handy though, so I don't want to waste much time here. So that's the, oh yeah, Geforce RTX 3060. Yeah. All good stuff. I don't know what causes a computer to freeze. I don't know enough about computers.
00:06:56
Speaker
Yeah. And like, again, like back in the day, if I had a bad Ram, you usually would get a screen of death. Um, this is odd because it's just an absolute immediate hard freeze. Um, that's not always seemingly correlated with heat. Like when you're really ramping up simulation. So we'll figure it out. We'll figure it out.
Troubles with Wilhelmin Machine
00:07:16
Speaker
Yep. Um, the other thing that's broken is the Wilhelmin.
00:07:22
Speaker
So sort of broken. What Grant discovered is the Wilma has a U-axis. So that is an axis that moves in and out relative to the headstock, if you want to use some lathe terms, which I think are easier here.
00:07:41
Speaker
So if you think about the Wilhelmin being a lathe, this would be kind of like your carriage. And it holds the vise for up to, that vise pivots up, the pivot is, I think, just hydraulic. So it's not of concern or relevant here, but you have this vise that can move left to right as you're standing in front of the machine. On ours, about two inches, with the newer versions of our machine, it's almost double that. Which would be really handy. I'd rather have a larger bore, but yeah. Interesting, yeah.
00:08:09
Speaker
So Grant noticed that it was not repeating left to right, which matters because of how the vise would pick the part. So he wrote a program that ran a hundred left to right cycles of 1.9 inches, so just under the full stroke. And after those hundred cycles, sure enough, the vise is off by ten thousandths of an inch. And he said,
00:08:33
Speaker
Or we know it's not the case that it's an even amount off each stroke, like a tenth or whatever it may be. Sometimes it doesn't move at all, sometimes it moves up to like a half of a thou. So, and in fairness, I have it with...
00:08:48
Speaker
Grant has talked to Florian, one of the guys over 10. We haven't formally gotten super involved in this yet. I think it's the same issue, though, we're having with our tool changer arm, where the tool changer arm does seem to move.
00:09:04
Speaker
What I heard the guy say, I don't know this for sure, is that it seems to move about a tenth each tool change incrementally. I don't know if that's consistent in every one or if it's sometimes in some amount. But the solution there was the guy said, well, the tool change
00:09:23
Speaker
It only rehomes on a hard reboot of the machine. It doesn't rehome even if you hit home, I guess. They're like, just power the cycle, power cycle the machine once a week and fix that problem, which I was fine with because I didn't really want to spend the money if it's solved that way well. Coincidentally, power cycling the machine is what causes the U-axis issue to somehow doesn't get better, makes worse.
00:09:47
Speaker
We found the FANUC servo part to replace the whole FANUC servo which would be $2,500 plus probably a service call because it's pretty hard to get to that new access servo.
00:10:02
Speaker
Frankly, to be clear, all of this is very predictable in the sense that I didn't expect to not have to put some... It's a 20-year-old machine, yeah. For sure. Of course, now that the talk becomes reality, it's the idea of, well, no problem. This gets us another X amount of months or years of runtime, but if it's just every three months, we find a new
00:10:21
Speaker
With the service call, this will be a multi-thousand-dollar problem. But what I don't understand is, I guess I would love the nerd in me slash entrepreneur, owner, business owners, like how does an encoder-based servo motor inconsistently go bad? So off the top of my head, a ball screw could be loose, like the actual Lovejoy connector for them from the board of the ball screw.
00:10:48
Speaker
Because some of them on our Swiss, it's literally a slip fit clamp. If your ball screw has a round circle diameter and the Lovejoy connector clamps around it, it's slip fit. There's no drive. There's no spline. It's meant to fail on a super bad crash. So maybe that's too- Lovejoy is the three fingers on one hand, three fingers in the other hand, intermesh.
00:11:10
Speaker
with like a rubber in the middle or whatever. And so that's how the three fingers put together the connection, which allows some angular mismatch. But how those connect to the motor shaft and to the ball screw on our Swiss, it's literally a clamp fit. There's no key. There's no nothing. And on our Swiss, ours was loose. So we'd lose like exposition every now and then.
00:11:37
Speaker
It's like a friction clutch, I guess. Exactly. If we bump the machine too hard, it's meant to slip, which is handy instead of break something. But then our X is off by an inch. And then we just have to rehome the X for everything. And then that works. So I wonder if it's something like that. Maybe it's just gotten a little loose over time. It needs a little cinch. And I asked the tech when he came in. This was like two years ago. I was like, how tight do you go? And he goes, well, I'm not too tight. I was like, what the? Why, yes. I can give you four foot-pounds or something. I don't know. Yeah.
00:12:07
Speaker
But it could be something like that. It could be chips accumulating on the ways. Oh, good call. So it's just like a chunk. It's, I don't know, something like that. Yeah, that's horrible. That's what comes to mind. I don't think it's like losing steps like software or encoder. Maybe the encoder is bad. I don't know.
00:12:28
Speaker
Well, that's kind of what I was wondering. I have not never taken apart like Haas or FANUC servo motor, but my understanding is the backside of the motor has an etched
00:12:43
Speaker
glass or a circular thing that's etched. So basically, it knows that on rotation, 472, it should get me to the home position, but then within that rotation, go to this encoder position to, I guess that's home, sorry. But it seems like that shouldn't be something that has a physical failure to it.
00:13:05
Speaker
Like, look, I get it, over, it's kind of like that saying I remember here in 20 years ago, over all time, all mechanical things will fail. Like, it's 100% absolute certainty that at some point, your current in Myo Kuma will be in a dump. Weird to say that, but I get it.
Maintaining Aging Machinery: Is It Worth It?
00:13:20
Speaker
But what would, like, I don't want to spend $200 bucks and do all this work if that's not the issue. And I was just trying to think, what would be, in a servo motor like that, what would
00:13:31
Speaker
cause it to lose steps like that? Ironically, like stepper-ish loose steps, you know what I mean? Yeah, like early tormac days. Yeah. And I should do. I didn't even think to walk up to it and try to move it. Could it somehow be slopped? Is it preloaded? I don't know if you can, even if the machine's off, it doesn't slide very easily, very difficulty. If you wanted to actually slide the axis and feel the waves or whatever, you'd probably have to disconnect something to do that.
00:13:59
Speaker
But there should be some, I can't remember the correct pillow block bearing-ish terms, but there's somehow that the ball screw is preloaded into its axial position to avoid floating the whole screw left to right, which is the behavior we're seeing. Yeah. But yeah, it sounds like an intermittent mechanical issue to me.
00:14:24
Speaker
Talk to Florian, talk to Marcus, get a feel for it. If I do this, I'll probably end up doing the tool change arm as well. At the end of the day, if I'm being an open book here, I'd rather openly fail and regret
00:14:45
Speaker
I'd rather fail over investing in a machine. I know it sounds weird, but rather than be a bootstrapper, spend the money, keep treating this machine well because it is freaking awesome. I love it.
00:14:56
Speaker
And, you know, that's kind of what I've tried to brace myself for is like, okay, if in, gosh, I hate to say it, but in another X amount of time, we crash or we wear out the main spindle or the B-axis pivot head. And that's a five-figure replacement. I want it to be a position where I realize, hey, nope, write that check. Say, thank you, guys. We love this machine. We're looking forward to a new one. Exactly. Because the bones of the machine are nothing wrong with that.
00:15:25
Speaker
Exactly. It's keeping an old car on the road is what it is. And it's such a good machine, so productive.
00:15:33
Speaker
Yeah, an old car on the road where we don't have mid Rust Belt salt corrosion on the frame driving in the winter. It's like an Arizona car. Yeah, exactly. Okay, well, that's the top of my mind. Yeah, I hear you. Yeah, I've had some UX issues on the Wilhelmin 2.
00:15:56
Speaker
a lot of which was the firmware or the software settings, which Mark has helped me kind of update and fix and works great now.
New Equipment and Personal Projects Update
00:16:05
Speaker
Okay. Yeah, that's a good thought. You have not had any in your overhaul work, repair, testing, et cetera. You've not had anything with these servo system doing what we're doing on any axis that you know of.
00:16:21
Speaker
Okay, so yesterday something similar happened. I did lose U-axis position. So I've got this set up where the vise comes up and it pinches between centers a pen tube so that I can put a pattern on it. Pinches between centers, you lost me.
00:16:39
Speaker
So the vise comes up and there's a bearing, a live center in the vise. Sweet. Right? So I can, I can actually pretend that it's a lathe with a tail stock. Okay. And I'm putting a tube between centers and now I can mill a pattern around the outside of the tube. So I've been playing with that twice a few days. It's super cool. I got two patterns done yesterday and they,
00:17:00
Speaker
They look super sick. That's what we drove on Instagram. Yeah, exactly. One of them is using a 10,000 tapered ball from Lakeshore Carbide running at 20,000, 25,000 RPM. Spindle goes up to 30. I was like, well, let's not run it at 30. Let's cut it down to 25.
00:17:19
Speaker
And because the tool bed takes a long time and some of the a axis rotational moves weren't doing polar interpolation properly. So they were like way slower than they're supposed to be. And I was like, whatever, just let it run. So it ran at 25,000 RPM for like many, many minutes and the spindle overheated. Oh, no, not a bad thing. It just hit the max temperature and it just kind of stopped itself. Okay.
00:17:44
Speaker
When my machine stops or he stops or you open the door, whatever the Z drops a good quarter inch, half inch. Does yours do that? Negative. It's the Z axis break is totally worn out on our machine. So normally not a problem, but you can like watch it. If you hit the stop button, the Z drops.
00:18:04
Speaker
Like a bit. Unsubscribe. Yeah, exactly. So there's a local place that can rebuild the Z-brake. We just got to pull the motor and drive it over and we've just been ignoring it. Anyway, so machine kicks an alarm, which basically is an E-stop position. So the spindle dropped when the tool's in the part.
00:18:24
Speaker
So it chipped the tool and the parts held between centers so it moved them down a little bit. So once I undid everything, I brought the U-axis up to the same position that it was and now the pen tube which had some tension before is now loose in that position. Interesting. So I'm like something moved in this scenario because
00:18:45
Speaker
Now bringing the U to the same position, your tail stock, same position, I need to go another 40,000 more for it to actually like tension this part. So that's as far
Excitement Over New Bamboo Printers
00:18:55
Speaker
as I got. I don't know if anything shifted or got weird or whatever, but everything looks normal to me and I'm like, I don't know.
00:19:03
Speaker
Literally, this is just like your vices acting like a literally like an old manual blade tail stock. Yep a live center you would think that a hundred pound or 500 pound part falling not even fine like Sliding down into the pen to would put very little Radial load that would cause your u-axis to yes a lot, right? I
00:19:25
Speaker
I mean, the hydraulic lifting of the U-axis is not super strong. Sure. You can actually overcome it. So that to a degree, but I don't know. Or maybe it lost position in the E-stop and it needs to home or something. I didn't get that far.
00:19:40
Speaker
That's true. If he stops and that somehow relaxes the U, like it could be U then, I don't know, seems interesting. The U doesn't home during a homing reference. It doesn't. I don't think so. They're not absolute encoders. There has to be some home to it. I don't know.
00:19:58
Speaker
That's a good question for me. How does you use it? You watch it. You hold X, Y, Z, B, and then it's just a little rotation. Like your headstock. I don't think it moves the U. That's your sixth. Grant said that it's not consistent anymore. When they first mentioned the problem, I thought, well, if it's moving a tent each time, maybe one of the encoder etchings is somehow compromised. And we can just do like, what is it? A G10L, whatever, and like comp.
00:20:28
Speaker
Every time. Which I readily admit is something I've vehemently been against because you're moving machine positions systematically without regard to what's happening in reality. Yeah, exactly. In and out my head in the same moment. But anyway, overall, my system of holding between centers is giving a lot more predictability than I expected, which is good.
00:20:56
Speaker
You have to put the soft jaws in that hold that center though. Yeah. Okay. There's no, I'm just trying to think like, I mean, you could have a little UR robot that moves it into soft jaws that also fit the center, but that's not what you say or the squeeze there. Yeah. And I still have to manually put in each pen tube clamp between centers and then run it. And you could have a UR robot do that if this were a high production thing, but for now, whatever. Load one, run a five minute program or whatever, and then just do it again. I wonder if.
00:21:26
Speaker
Not with saying coolant issues, splashing out of the machine or onto the robot. I wonder if you would ever be allowed to have a robot come in and have the end effector include the live center. Wouldn't be strong enough, I don't think. Yeah, right. Yeah. Unless it was some big fanic robot, but yeah, too many variables. Yeah. But yeah. How you doing? Doing good.
00:21:54
Speaker
So after the podcast last week, I ordered two bamboo printers and 36 rolls of filament. Holy nuts. Yeah. And first one came the other day on Monday. Second one came yesterday. And I was like, Eric, you run with this. I don't want to touch them. You just give it to me working and tell me how to do it. And he's been having a lot of fun.
00:22:15
Speaker
Awesome. Yeah, super cool. You got that first printer set up within an hour and printing parts right away. I checked on him as I was leaving and I'm like, I see the printer's out of the box. He's like, oh yeah, I've already printed two things. What? Yeah. Yeah, it's cool. He's like, it's game-changing. It's just a whole different world than the Prusa. He's like, it just works. It's just easy. Everything just does what you tell it to do. You don't have to worry so much. It's all self-comping and really, really cool.
00:22:45
Speaker
I have finally found one thing I don't like, I think is a genuine criticism and one thing that I just think is...
Bamboo Slicer and Prototyping Challenges
00:22:53
Speaker
against the ethos of bamboo. The latter being, if you go into the slicer, the device page has a, I mean, it's as close to hidden as you can say, hidden option where you can, I'm in it right now, I don't even see where I choose it, which is kind of my point, where you can change it to ludicrous mode, like from, here we go. Okay, on the device page, there is a gauge, like a speedometer gauge.
00:23:20
Speaker
happens to be next to the lamp icon. And that's currently shown at 100%. That is standard mode, you can go up to sport or ludicrous mode. But to me, it's not that I'm against having that separate from the slicer that way, like normal users don't even have the risk of doing this. But it seems odd that they're separating out what I would consider kind of
00:23:44
Speaker
general print slicing features and putting it in a separate hidden area. Because part of it is like, well, I remember doing some comparison data with Spencer Webb, and perhaps you could argue that the 100% mode is the fair comparison because that may be the most reliable. But the reality is for a guy like Spencer, when he's doing
00:24:04
Speaker
production printing, it probably would be worth his time to look at what he needs to do to get ludicrous mode working if it's... Yeah, every time. That much faster. What do you normally print in? Well, I only found this recently because it goes back to my first issue, which is my general legit criticism. We're printing some mod vice and fixture plate prototype stuff.
00:24:25
Speaker
And I care about some tolerances here. Now, caveat, it's a $1,500 filament printer, and I'm wearing the hat of a, you know, Akuma machinist nowadays.
00:24:38
Speaker
but one inch features in smaller were approximately five thou undersized consistently, like very repeatable, but undersized. So a one inch, I printed a one by two block with a half inch hole in the middle. Hole was 495. The one inch dimension was 995. And the two inch dimension was about nine thou under.
00:25:01
Speaker
And I did some poking around, and there doesn't appear to be a way to change that. Meaning that there's a lot of buzzwords around calibration, but that's not what I'm talking about. And the best way I saw to do it, and shout out to the folks like the WhatsApp people that were trying to basically to do a scale change, like print your model at 1.006,
00:25:28
Speaker
or 100.06% or whatever. And Spencer was talking to me about this. And he was kind of like, dude, you're within whatever percent, 5,000 across one inch. You're good. It's not a printed part. But it bothers me because the machine seems to be quite repeatable. Yeah. Yeah, yeah.
00:25:48
Speaker
Hmm, you almost need like servo mapping. Like you would do on a big fancy machine. Oh no, I just need to comp it. Like just say, hey, what you think is one inch is actually only moving to 995.
00:26:00
Speaker
There's got to be that table somewhere in the machine. Lawrence or somebody in the WhatsApp was saying there's a consequence with regard to layer stuff, admittedly, over my head. One of the reasons I bought the bamboo was to not need to become a firmware in the weeds expert. Sure. That's my only. Only grape? Yeah.
00:26:25
Speaker
I would assume that's consistent across most all bamboos like Spencer's or mine or whatever. Print a one inch cube, see what happens. Yeah, I think he did it and just got similar. I mean, if you want to do it, would love to see what you get. Oh, but then that's what came up was making sure you're not in ludicrous mode or I was pretty normal settings and using their filament and so forth.
00:26:53
Speaker
Have you printed anything to it? I have not touched it yet. Eric showed me a couple things and he's printing this fixed blade knife that he designed. And the one was still printing on the Prusa. It was about a three hour print. And he's like, there's an hour left. Let me see what I can do on the bamboo. Oh, 45 minutes. So I could start it and finish it before the printer Prusa even finishes. Hilarious. So he did. That's great.
00:27:18
Speaker
But yeah, yeah, he hasn't tried multi. We're still waiting on the rest of the filament to come in, like just the one printer and then the second printer came in. But he hasn't tried multi colors yet or anything fancy, but yeah. Cool. It's great. Super happy with it so far from what I've seen. Yeah, awesome. We sort of forced a cliffhanger on the last bomb. On polishing parts?
00:27:47
Speaker
Well, that, I'd love to come to that. No, are you going to convert? Did we end up out? We talked about it briefly. Yeah. Convert my Amazon to electric. I got a couple of DMs from people being like, do not do that. How dare you? Like sacrilegious or too much work? I think so. I don't know. They didn't go into details.
00:28:07
Speaker
I don't know. I think it'd be really cool.
Electric Conversion of Classic Cars: Why Consider It?
00:28:10
Speaker
As someone who's driven an electric car for the past five years, I love it. And to do that to an older car, breathe new life into it. I mean, if it was quiet and non-stinky, it would be a totally different car and fast. It'd be amazing. Although I think you'd hear a lot of noise vibration harmonics in a quiet car that's like 55 years old. But whatever, you just deal with it.
00:28:34
Speaker
Are there practical concerns around the frame strength for both the weight and the speed? Sure. Okay. Yeah. There's a bit of frame rust on the car. I almost said on the machine. I don't know. Just playing with it. Just been driving it around, enjoying the heck out of it. Awesome. I need a lot of stage zero stuff. I need to keep working on the brake system.
00:28:59
Speaker
Keep tweaking little things here and there. But like, wake up in the morning, no issue, hopping out, rolling the ignition, driving to work, good to go. Yep, yep. Crank's a little hard, but it does start. And then once it's warm, it starts to first crank. Got it. No problem. It's got a choke. I've never had a car with a choke. I'm like, how does this thing work? It's a carburetor, so I've never had a carburetor. Yeah. Yeah, it's super fun, though. Yeah, that is cool. Yeah. Well, you had mentioned
00:29:27
Speaker
that this was kind of your side hustle, like finding a sense of identity and doing this. Yeah, like self-purpose, an excuse to spend fun money kind of thing. Yes. And yeah, it is nice. It's just really nice. Go out for a drive. Kids kind of like it. Yeah. And it's just my fun thing. It's nice.
00:29:56
Speaker
version of that is I'm trying to force it to be a similar thing, but with a skid steer and a mini excavator. Okay. Of which I own neither, but I have been binging for probably nine months. Mostly a gentleman on YouTube named Diesel Creek, awesome channel. There's others as well, but I really like his style of videos and
00:30:18
Speaker
The problem is that he makes everything look so easy. He buys a Richie Brothers crane truck and figures out what's wrong and fix it. He's got 15, 20 years of experience. I have zero, but it's that confidence-inspiring goal. I would use these here and down at our farm and elsewhere, but it's ultimately for fun. There's no business justification to going out and buying brand new ones, blah, blah, blah.
Machinery Projects Inspired by YouTube
00:30:45
Speaker
like a buddy of mine rented one last weekend, an excavator mini one, we were doing work and like, there's just something, I don't know if it's all guys or just me, but like, man, it is the most just wonderful thing to just sit there and like rework, move dirt, all that. Yeah.
00:31:01
Speaker
Yeah, so I'm debating, look, the problem is that the only deals, like true deals I found on non-working units are usually, they're usually skitziers that have burned, like a fuel line broke or something. And I don't think that's a smart- Yeah, it's a bit far. First project, but- Yeah, replacing the wiring harness on a skitz gear or something like- Oh, yeah.
00:31:26
Speaker
So, um, and then you quickly move into like competing with people that are really writing businesses using these construction landscaping. And you know, a new skid steer is, this blows me away. It's like 80 to 90 grand. Um, and so, you know, there's no reason to rush, but it's also, there's a YOLO element to it of like, no, I really enjoy this. And you know, William and I had a lot of fun on that Porsche 944 project and this could be, I was just going to ask, do you still have it?
00:31:52
Speaker
No, it was absolutely perfect story. Bought it, got it running, did the water pump issue, the timing, the belts, a couple other fuel line system things. Did a whole brake job on it, rotors, pads, calipers, all that. And then the seat frame was rusted to the body, the rails. So you were going to have to saws all them out. And that opened up a whole other can of ones that I wasn't interested in. The seat was too far back, frankly, for me to drive it, but especially for my wife to learn to drive stick on it. And so I thought, you know what, John?
00:32:22
Speaker
In six months, I took a car that wasn't running and I got it running. It was super fun. Time to book the win and sold it to a young gentleman in the reserves who always wanted a Porsche. Yes. Great. Oh, that's amazing. That's cool. That's good. Yeah. Yeah. Hmm.
Challenges in Machining Carbon Fiber
00:32:48
Speaker
Over the weekend, I went on a deep binge, YouTube binge on carbon fiber. Oh, yeah? And just kind of learning more about it, how it's made, how to make it, how to cut it. I was trying to look for videos or information on, tell me what end mills to use to cut carbon fiber. And there's nothing out there. I was actually extremely shocked. There's like some information, but very little because
00:33:14
Speaker
At the end of the day, any end mill will cut carbon fiber. You can cut carbon fiber with a HSS drill bit. But it's going to suck. And it's going to get dull after three holes. So you can, but not for production. And I mean, we've machined dozens, possibly hundreds of carbon fiber inlays over the years. And I'm getting back into it now, because it's been a while since I've done it. And when I cut a sheet, a 12 by 6 inch sheet of eighth inch carbon fiber, my end mills were toast by the end.
00:33:42
Speaker
Absolutely hilarious under the microscope. They're just rounded. It's gone. I couldn't believe it. Same thing with cutting rich light, which is a paper-based phenolic material. The binder and just the abrasiveness of paper or carbon fibers is massive.
00:34:06
Speaker
So I was looking for some production YouTube videos from anybody, being like, oh, you got to use these end mills. This is the way to do it. And there's not much at all. Harvey Tools has quite good information, and Corehog is one of their new subsidiaries that specializes in making end mills for composites. So I spent a lot of time on their website, reading their articles, things like that.
00:34:30
Speaker
DLC coding is a great option. And then there's CVD coding, chemical vapor deposition coding, which is like the next level. And apparently it's a lot thicker than DLC. And it's also like a lot, a lot more durable and operation resistant. So I bought a bunch of end mills.
00:34:49
Speaker
Well, I was going to say, I would think diamond, either diamond, not just coating, but like MCD-PCD. Yes. Because carbon fiber, I mean, I don't know much about it, but I assume some majority part of it is carbon. And diamonds, fun fact, carbon, but harder. Like when? Exactly. So Corehog does make PCD end mills, like a tiny little face mill. But they only go down to, they do make eighth inch PCD end mills, but they're only one flute. So there's a quarter inch two flute.
00:35:19
Speaker
that I did buy for $216 but I'm going to use that for facing the carbon fiber which I think will make it look insane and lasts for a very long time because I buy eighth inch thickness and I have to face it down to 70 thou or something.
00:35:34
Speaker
They're grinded, huh? No, they're too messy. So far I've been using just off the shelf, eighth inch, four flute, Alton coated end mills. They're like eight bucks each and they work so long, you know? Yeah. And even for eight bucks each before you replace them. But if they die sooner than a sheet is done, that is no good to me. And if I have to replace them every sheet, like that's no good to my team because it's like work and time and stuff.
00:36:01
Speaker
The fancy core hog ones, they were only about 40 something dollars for the CVD code at Edmill. And even if it doesn't last four times longer, or it does, or I don't know, we'll see. Yeah. Waiting for those to come in. Which machine is this in? It's video. Okay.
00:36:21
Speaker
Big win on stuff like that is get a hydraulic holder because it's like the same thing with those ceramic tooling and inking out. You got like 19 minutes of tool life. So it's like, you know, run your timer and then get a wrench in there, replace the tool. Just like don't even probably don't even take out a spindle. Yeah.
00:36:39
Speaker
Cuz yeah exactly we've got them currently in er call it and then a regular er call it that all that carbon fiber dust finds its way into the car or so i've got regular er call it with through spindle coolant in the operation to keep them clean.
00:36:56
Speaker
which so far is working well. But it's like that whole, you can fan through an ER collet, but it literally creates a fan of coolant. And I'm like, how much of that fan is reaching the end mill? Or is it just creating? None of it. Maybe none of it. I don't know, it's last thing, it's working. And then we've got the sealed collets with the four coolant jets in them that kind of direct. I haven't tried that yet, but I think that's the answer.
Ultrasonic Cleaners and Tool Maintenance
00:37:19
Speaker
We use some of those collets in the horizontal and they do aim it because otherwise, considerably, you're blowing the coolant further away. Yep. I'm going to try that next. We just randomly reminded me that we're trying to do a better job at often using the ultrasonic to clean out holders with- Right. We just bought another one for that.
00:37:42
Speaker
ERs and the ER collets and the ER nuts because you just get nasty. Why isn't it positive pressure and option? I don't want through spindle air like at 100 PSI, but just something that creates outward pressure from the spindle so that you're not encouraging debris to get soaked up there. 10 PSI of through spindle cooling, please. Have you been using the ultrasonic to clean those parts?
00:38:10
Speaker
We need to move it. It's literally on the ground next to Garrett's. We just need to improve it. But we have, yeah. Ours, we bought one for that purpose. It's on the bench next to the sink, hasn't been plugged in yet.
00:38:29
Speaker
But I think it would work. I mean, it adds time, like minutes, many minutes in a tool change. And I think the only way to do it effectively is to literally have a duplicate tool holder set up. Like these are my ER16s, same gauge length, everything. Backup one's clean, ready to go. And then dirty one goes in the dirty pile. It gets washed eventually.
00:38:48
Speaker
That's how you see those big shops that have the tool crib, right? Where it's like, use tools, go here, new tools, come out, never reload, use tool holder kind of thing. I don't know. That's a whole other level, but we're kind of starting to eke there.
00:39:03
Speaker
two minutes in an ultrasonic though does wonders. Yeah, I can imagine. Yeah. Two minutes across a bunch of tools is a lot of time, I know, but some of the tool holders I've seen, I'm like, there's zero chance I'm wiping this with a rag and reassembling it like it needs. Right. But like you said, even the collet and the nut alone, and I can have duplicates of that, like dozens of them. Oh, that's actually a great point because- And then you just wipe the holder.
00:39:26
Speaker
Yeah, the holder is the easiest to clean, man. That's a really good point because nuts are the same. Collars are easy. The holders are real sensitive about switching the diameters and gauge links because they need to say the same. Yeah, you need a rigid process there. Good.
00:39:46
Speaker
Yeah, that's my, I meant to bring this up last week. You know, Alex has been working on the puck chuck, it's going great. But between that and then the Gen 3 Mod by stuff, it's a lot of work bringing a product to market, like just all the processes around it.
Effort Behind Launching Gen 3 Mod Vice
00:40:01
Speaker
And that realistically is going to take up the next, you could say a few months or really this year, but then
00:40:08
Speaker
My sort of plan is, for the sake of making a plan, is as that wraps up or winds down toward the end of the year and I start transitioning some of those things off to the team, is really some quality of life overhaul. We really cleaned up the back of the shop around the Okamoto, which felt great. And then things like, hey, where the Spironi is with our tool tools,
00:40:35
Speaker
like Torx wrenches for all of our inserts and end mills. And where's the ultrasonic? Where do we have rags? Do we have them on magnetic clips? It's kind of what I've said before that it's easier to talk about, harder to do, which is I can do things like that really well if I spend the time on it. And right now the truth is I'm going to hang up this podcast and I'm going to go troubleshoot a castle grip fixture that I'm not happy with the obtuse on. That's what I need to do.
00:41:02
Speaker
Yeah, and I'm in the same boat where my wish list, my to-do list, my plans and ideas and new products and upgrades and fixes around the shop and tweaks to code and all that stuff is literally way longer than I could imagine having time to do. And I can distribute some of it towards the team and probably need to start hiring some more people to be able to realistically accomplish any of it.
00:41:31
Speaker
And that's okay. That's cool. Well, look, I mean, again, I kind of, this is one of those moments I wish this was private conversation, but like the thing I first do is just like, you know, my name is John and I'm here because I'm an addict. Like, you know, you have to at least acknowledge like this ain't happening. Yes. Like I'm not going to, so yesterday morning was a sort of fake days off in the shop for the first 90 minutes. All I did was palette Jack.
00:41:58
Speaker
material around, we got rid of some old boxes, we consolidated some inventory that had been stacked on different pallets. And you know, it's amazing what you get done in 90 minutes. And I those so it doesn't have to be okay in January 7, I'll start doing this. But man, you've really got a discipline to carve out time to do that stuff.
00:42:22
Speaker
Yeah, you make it a priority. And, uh, you know, I definitely fall into the camp of, but everything's a priority and I want to do all of it. Yeah. And I, I, I'm just, it's just hitting me so hard that I can't, and I feel, feel bad about it. I feel like guilty and like, like I'm not enough or whatever, but it's like, no, that's okay. Like you have grand visions and so much skill and like, you don't have time to execute all of it. You need to distribute that time amongst your team and more team and that, that makes it okay. You know,
00:42:50
Speaker
Yes. Yep. Um, yeah, it's interesting of all of the like mentors and books I've read and all that. I don't know that I've had one that's made me break down the hurdle of like, you've just got to learn to enable the team. Like I know it, but it's, um, I have a lot of respect for those that have done that. Yeah.
00:43:22
Speaker
Yeah, tricky. Tricky sides of business. What do you, uh, we have today? Today. I don't know if I got that for you. Okay. Yep. Yep. Got to check in with a vendor. Uh, and the other thing is I made a new pattern on a Norseman, um, that I tested one yesterday, but the whole pattern is shifted off about 80 thou or something. And I'm like, what the heck?
00:43:50
Speaker
Who set that fixture origin? So I got to debug that. It looks great. It's just wrong. And it's missing a bunch of features because I pasted the code improperly. John.
00:44:03
Speaker
Yeah, like the way we make our Norseman on the Maury is literally 60 year old code that I'm limping along. And there's going to have to be a time which is on my massive to do list of like, I need to redo this new fixtures, new pallets, make it simple, make it updated. But until then, I'm managing the way it works right now.
00:44:24
Speaker
You don't, so you never repost that code out of Fusion. Like it's just, that code's there. It's almost too far gone to repost out of Fusion. There was a time when I could have reposted it with all the tool life macros and all that stuff. It just worked. I don't trust that anymore. And Fusion has changed over the past couple of years and like, I don't know if I, if I would just post it and go. No, I hear you. That was my other, I don't think I mentioned this.
00:44:49
Speaker
in line with that list I just said was probably to delete all the programs off the horizontal and then we post them one by one. That was zero risk there. That'll be fine. But then like watch each operation and each tool path because they so often happen overnight or it works great so we don't worry about it. But there's probably a chamfer operation that's plunging at six inches a minute. Yes.
00:45:14
Speaker
haven't caught. Now that is absolutely something where I will sit down probably well not probably with Garrett and maybe we'll do the first few together and then you know whether he does it or he does it with a camera and scrubs through the footage. And we'll probably wait to do that only on Gen 3 because Gen 2 actually another kind of question of like okay Gen 2 and Gen 3 devices are different
00:45:36
Speaker
enough to where we'll want to continue to have some stock of Gen 2 for a period of time, but then it'll be phased out. Just if somebody has eight of them, then they want two more. The Gen 3 Adjustment would be fine, but the Gen 3 Fix will be different enough for them to have a Gen 2. So we obviously want to do our best to accommodate that. But I also want to eventually rip those fixtures off the machine to free up that space. Yeah.
00:46:08
Speaker
Excellent. What are you up to today other than, uh, debugging a castle nut thing, castle grip. That's our new one inch. Like we used to make a reversible insert. Now it's a one inch version of that, which bites in very similar to something like a talent or there's other similar products. Um, and, uh, learning how to actually balance a grinding wheel.
00:46:29
Speaker
Oh, yeah. Yeah. I've done it before, but I wouldn't. I'm a student, not a teacher. Exactly. Yeah. Take your time. Make sure it makes sense. Understand it and all that. I don't know how to do it either, but I know technically how it should be done.
00:46:42
Speaker
What I liked at the risk of inciting riots amongst the tool makers on the listening, like Diebolt, who I look up to a lot, Paul Diebolt, he doesn't change the wheels on his grinder. Paul is, to me, a Robin Renzetti-like person. He makes very good parts. He put his wheel on his Chevrolet 1224 years ago. He doesn't change it. He doesn't grind that much, so it gets dressed down, but he doesn't rebalance it or worry about it.
00:47:13
Speaker
Not saying that's how we're gonna have to run the shop, because we'll use it more than he will, and so it'll wear down, but I am not gonna go uber nutso. What I am excited to try is I did grind a sample part with the factory wheel that was not balanced, and you can see the bouncing after you PFG it, or stoned it, so I'm gonna grind half of that with the new balanced wheel, and try to see if I did. There, yeah, totally.
00:47:43
Speaker
Because you need that feedback loop. You need to know if what you did worked, and why it worked, and what happened. Yeah, I know what Angelo says is he'll rough balance it first, and then he'll put it on the machine and dress it clean, and then balance it again. Because once you dress the initial out of round, out of it, you've got to balance it again for a new wheel. What I did.
00:48:06
Speaker
think is right is mounted the wheel in the arbor and just put it on the machine and initially dressed it because it's going to be so far out before you dress it. There's no point trying to balance it. Yeah, maybe.
00:48:17
Speaker
But then I didn't want to use marker to mark the outside of the wheel because marker will soak in. So I had to find an actual number two graphite pencil. Mechanical pencils don't have enough lead thickness. And so I wanted to mark the OD of the wheel to make sure when I was dressing it, that it was dressing around the full circumference. And it got there. It's on the balancing stand. It actually doesn't seem terribly out of balance, but go play with it.
00:48:44
Speaker
It's cool. When you're balancing, put your surgeon hat on and do it well. That's a very precise precision operation where you get to like, oh yeah, this is good. I got this nailed kind of thing. Yes, exactly. Cool. Awesome, man. I'll see you next week. All right, dude. Later. Cheers. Bye-bye.