Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
#255 - 3 Axis Machines | What To Buy & Why?
224 Plays3 years ago
TOPCIS:
- Horizontal Machining centers for 3 axis machines.
- Renishaw touch probes. Double vs single touch.
- Grimsmo thinking hard about buying a Datron Neo.
- Saunders is still dreaming about a Hermle RS1 pallet changing 5 axis machine.
- Porsche update from Saunders on his 944.
Recommended
Transcript
Introduction and Podcast Overview
00:00:00
Speaker
Good morning. Welcome to the business of machining episode 255. My name is John Saunders. And my name is John Grimsmough.
Exploring Horizontal CNC Machining Centers
00:00:08
Speaker
And this is the podcast where John and John discuss their hesitancy but enthusiasm for horizontal CNC machining centers or otherwise forms of what is effectively three axis machining, but with some form of smart automation. Yes. Got something on your mind?
00:00:29
Speaker
Yeah, I think we both do. So let's talk about this and it's fun because you've got to ask, like I value going into a kind of a quiet place, getting yourself in the right mindset and thinking out why do you want this? What are you trying to do? So I'll share our side of this.
Mindset and Process Changes in Machining
00:00:49
Speaker
We have frankly had a great experience
00:00:54
Speaker
Getting our products made and built on mostly Haas machines that we have that have run really well and we've iterated in I think I'd say countless ways, but it's probably like any manufacturing entrepreneur does like we've changed fixturing we've changed the way we batch stuff up we've changed every tooling cam operations with change how many we've made at a time is it's that's the process we all I think love like that's the journey of it and
00:01:22
Speaker
And even as recently as the end of the year, we actually kind of got caught with our pants down. We basically overhauled, you know, we moved several of the DT to the VF2 with some of the VF2 to the VF3 YT and changed a bunch of things up. And I can come back to why that caught with our pants down later.
00:01:43
Speaker
And that process has been great, but we're ready for what I'll effectively say is the next step.
Benefits and Challenges of Horizontal Machining
00:01:49
Speaker
So the majority of the reason we would want a horizontal is to cliche kind of high-mix, low-volume-ish nature. I don't want to overproduce. I absolutely don't want to have to change setups, or at least as little as absolutely possible. And that's an interesting
00:02:12
Speaker
Well, I may be forced to compromise because we had been looking at something like a 10 pallet pool system, and it looks like it may have to be six, simply because 10 would push it to an overseas longer lead time. We're not even, that's next. It may just be the case. So, and frankly, the budgets or the quotes came in and I'm kind of like, okay, I got to get realistic about what we can swallow.
00:02:36
Speaker
But what a horizontal would let us do is it would let us keep mod vices, bases, the top jaws, hand soft jaws, and pallets. We don't really make money on the pallets. We make a few bucks, but I've always viewed it as a product that we need to sell because it's quite complimentary to our products. I like selling them, but it's just a very minor improved aluminum block, if you will.
00:03:02
Speaker
But the mod life spaces would benefit from having a B to index over, but not, it would benefit, but we've done it. We've got a great workflow. Of course, they'll be using our fixture plates now to make those. So it's not really unlocking anything like your current, where all of a sudden you had automation five axis, like a whole different level of, you know, we are
00:03:30
Speaker
incredibly happy with the finishes tolerances. Yeah, like, we know our C frame machines have a little bit of head nod, and we've actually figured out ways of probing reference datums to could cliche hold tents all day, it doesn't hold tents all day, it just comps work quite well. Yeah, I think some of that may be better. But that's not really why what it means is, when you want to make steel jaws, you call it pout five,
Machining Efficiency and Cultural Approaches
00:03:57
Speaker
and
00:03:57
Speaker
we're probably on the higher volume stuff, mirror tombstone so that there's two say jaw tombstones so that one can be loaded while the other's running. And the, so what I think of as the benefits is multi-sided tombstones for, you know, hopefully already four or six parts per tombstone, like sides per se. It's not really going to get us from two ops to one. We thought about doing some,
00:04:27
Speaker
window machining style work. It's a lot less efficient, meaning it's a lot lower cycle time because you're only making a few parts, but it's a one and done. Everything I've studied on our side and looked at other friends that have horizontal, it's still going to be a two op process. I think that's okay.
00:04:48
Speaker
Well, there's almost nothing that's actually one up. You're always going to leave a little bit of a tab or need some sort of secondary thing. If you just eat that and you go, okay, it's a two-off part, let me flip it and do this properly or re-clamp it or something like that, then you just accept that and figure out what is the best way to do that.
00:05:08
Speaker
Yeah. Yeah, that's a good point. And like the classic fixturing thing where it's like even on a three axis fixture you have on the left, you have up one and on the right you have up two and you'd run them sequentially like at the same time. Could you imagine doing that backward? The up on the right, up to in the left? Oh man. Oh, that'd be weird. Right? Yeah.
00:05:28
Speaker
I'll bet you they do in Japan where they read from left to right or from right to left. Do they? In Japan, yeah. They did not know that. Well, like Clara, my daughter's been getting into these manga comic books and they're Japanese based. Huh. And the other book you read from, well,
00:05:45
Speaker
You read from the back of the book to the front. Whoa. But I think on the page you still read, right? I forget. I don't know. Arabic or is it Hebrew? Like definitely a thing. You're right. I don't know. I don't know any Japanese.
Aligning Machinery with Production Goals
00:05:59
Speaker
So it's weird because like we're not fitting the, I think normal checkbox or horizontal, you know, tens of thousands of parts where it's going to run for dozens of hours unattended. Um,
00:06:10
Speaker
But we, and there's some new products that we want to roll out, which will be a good fit for it. Um, so it's kind of like, look, it's all there. So here, um, the big question is not if, but what. So, uh, and the two biggest questions are 400 millimeter or 500 millimeter. Um, we can make a 400 work. Most of the people that I've talked to, uh, have been like, Oh my gosh, go for the 500. If you can. Um, a 400 is only 22 by 22 inches or so. It depends on which brand.
00:06:41
Speaker
It's not a ton of travel. 500 is about 29. And you have some better spindle reach abilities if I recall.
00:06:53
Speaker
So that's the big first question and then the second question is do I stick with the much less expensive max ATC on the machine so like 64 or whatever number it is that stays on the machine tool or do you go matrix where it's you know hundreds
00:07:10
Speaker
The difference between 164 and 218 is trivial, actually, but it's almost six figures. It's like its own machine to get to that matrix. It's another 50 grand in tool holders to fill that up. But yes, having that capacity is real nice if you need it.
Machine Workflow Preferences and Automation Potential
00:07:30
Speaker
Yeah. I remember buying the VM3 how I thought it would be the only machine we'd ever have or have in years and I needed it for everything and probably bought an extra $20,000 in options that in hindsight would have been better off just putting towards the next machine. I'm in that same boat now. It feels
00:07:53
Speaker
impossible to prudently think about the second horizontal because I can't even... Exactly. Because you want to spec out the first one as like, this is the last and only one I'm going to need, right? Yeah. It's going to do everything. Absolutely. And I'm with you on that thought process.
00:08:12
Speaker
Yeah. I struggle because I want to love horizontals, but there's something about it that just doesn't click with me yet. I can't quite put my finger on it in order to explain it. One thing I've noticed between loading up the table on the mori and having it run for 10, 16 hours with the door closed, I don't like having the parts locked in the machine finished and it's moving on to the next one. I don't like that.
00:08:37
Speaker
There's ways to pause the machine and pull them out. I'm not complaining too much about that. Whereas on the current, having smaller fixtures with parts that cycle in and out often, everything on the pallet changer that's done is done. You can access it, you can reload it, and you can reschedule it and just put it back on the machine. The individual pallet cycle times are not that crazy for a lot of the parts.
00:09:02
Speaker
which is super nice to have that flow through. Whereas a horizontal is this big like rotating three axis machine where you could put, you know, 10 hours in one pallet if you want. Um, and then the parts are locked in there. I don't know. But they're not. I mean, the cycle, I think if I'm reading the brochure correctly, the APC swap time is like seven seconds. Oh yeah. That's not it though. When it's, um,
00:09:28
Speaker
Like if you have, say, 50 mod vices on one pallet, the 49 of them are finished and waiting and not doing anything while the 50th one is being machined, right? Right, totally. And we have such a tight, almost hand-to-mouth flow through of parts in our shop that I don't like parts being locked in the machine when they're done.
00:09:57
Speaker
depending on how you tool it up, a horizontal of it. Oh, you're saying it could still be tied up because the tombstone's in there going a bunch of work. Got it. Yeah, you almost want like a mini tombstone, like a horizontal that had like 40 mini pallets just shuffled them in and out. It's basically, it's kind of coming, what we were talking about last weekend, like you want to have a machine tool that can load in pallets, but you don't want just three axes.
00:10:28
Speaker
A lot of the work we do is three axis would be totally fine. Like if you put, say you put a robot pallet loader on a speedo or robot drill or something, like that could do a lot of work for us. But the 21 tools on those machines is killing me. Yeah. I mean, I think for the sake of progress, you just got to strike those period. Yeah.
00:10:54
Speaker
I don't know. No, but you do. I know you like I know we were talking offline and you were listening like hey I do I use 17 tools for this part alone and next for this and sure you could try to consolidate that you're fighting the wrong battle and it just
00:11:14
Speaker
Yeah, it's funny. Phil said the funniest thing on our WhatsApp. He's like, don't buy a horizontal because what was it like? Buy the machine that will spark joy for you. Like a good Mary Kondo quote there. Yeah. To which the next comment was YOLO. Yeah, exactly. But it's funny because either I get like I could go out and buy a speedio tomorrow if it was the right decision.
00:11:42
Speaker
And I'm so on the fence with it, like it would do some work and it would be great. But unless we put a robot on it to automate it, um, it will only run like 40 hours a week maximum. Whereas. Unacceptable.
00:11:58
Speaker
I've been tracking it on the current. For the past four weeks, we've averaged 125 hours a week of runtime. It's incredible. And we could probably do 140 once we really get up and running. That's fantastic. So the machine's a lot more expensive, but we're running it three times more per week.
Control Systems and Investment Justification
00:12:16
Speaker
Yeah, sure. Hugely productive. And a horizontal, you could make the argument that it would do the same thing for way less than a five-axis current.
00:12:26
Speaker
But it's like, you know, you look at the $100,000 robotrills or speedios and that's one thing versus the, you know, $700,000 currents or whatever, 800. Is it, although, because even though you're sharing the aroa, would it still be that? Yeah, it would be cheaper because you wouldn't have to buy the aroa again.
00:12:44
Speaker
Um, but, but my point is like, either you get some cheaper machines, but anything in the middle ground just gets rounded up to, I just want another curtain. I understand. You know, it's like, I know it so well. It does such a good job. It has all the tools, all the pallets, all my programs, all my fixtures. I've done a ton of custom programming on it to make the interface and the palette changing work and tool life management is sick on it now. Like so many things. What control would you want?
00:13:14
Speaker
on, on what I love the high nine. Yeah. I mean, I don't, I mean, I'm sure there's a high nine or no somewhere, but like that's generally not what you tend to run into. Yeah. It's like, I can deal with Fannock. It's fine. Um, I'm pretty good at it. I don't love it, but it's very usable and I know it very well. I love the high nine control though. I would want that on everything if I could. The, so literally the only drawback
00:13:41
Speaker
The current is the absolute perfect solution as a machine tool, as an integration with your workflows, an existing current, an umbrella, and knowledge. It checks every single box except capital. It's just really expensive, but that's also really productive. A horizontal could be too. Do you need the five axis nature of it?
00:14:06
Speaker
I don't know stuff. I've never been a one to shy away from buying the right machine, buying the expensive machine if it solves the problem that we're facing. It's an interesting capital dilemma. You're certainly checking the box of, hey, I've got one. I've proven that it works and we're running it 125 hours a week. Maybe there's some
00:14:34
Speaker
room, meet on the bone to actually, so this is the question, could you increase your efficiency by doing a little bit of a more, find the long still paths, shorten them up on the current somehow to buy yourself more free current time and or take just some of those off
00:14:51
Speaker
toss them onto a speedio. Look, you can always sell a speedio. Even if it's not automated, run a speedio 40 hours a week for the next year to make some money, sell it when you're done or even keep it whatever. I still just can't get behind spending another 100 grand on some, frankly, difficult to resell, not tried and true automation on a three axis.
00:15:17
Speaker
Yeah, like a robot or a cell or something. Yeah. If you DIY it, then it's very custom. If you buy some Triton cell or whatever they have. Trinity. Yeah, Trinity. It's very specific to your setup, I guess. And I want to be fluid. So like you said, at the moment, we don't need to double our production.
00:15:44
Speaker
another current would do that, but we don't necessarily need that right now, but will we need it in a year when the current arrives or in eight months? Anyway, I like your plan. We can absolutely shave some cycle times on the current.
00:16:00
Speaker
I'm hoping for like a few hours a day of cycle just through looking at efficiencies. Like I've got so many programs and toolpaths running every single day that yeah, I know I can shave some time. And then if I put a few of those parts on even more on the Maury and then if we also got a speedio, put some on that, I think we'd hit our goals for the immediate future. And then is the question of when do you place that order for the second current?
00:16:30
Speaker
Yeah. And you'll have a conviction later on that. Exactly. So that's kind of exactly where I'm at right now. Ask
Machine Selection: RoboDrill vs. Speedio
00:16:40
Speaker
him. Ask. It's fun to think about.
00:16:42
Speaker
It's Elliot Metzger that's now the... Elliot Metzger sells Robodrill right now. Oh, okay. They used to sell Brother, but they dropped it and they picked up Robodrill instead. So I'm going to quote both Robodrill and Speedio. I like them both. I've always liked Speedio better, not having run either one of them. So I really don't know what I'm talking about, but I know like 20 people that have Speedio is off the top of my head and I know like three that have Robodrill is off the top of my head.
00:17:10
Speaker
Yeah, I know Robodrells are everywhere. You know more than that. Yeah, I probably do, but I'm just that tip of your tongue kind of thing. I'm like, everybody's got speedios. Yeah. You know, Fannock. I mean, that's the knock against- I know, right? It's worth asking Elliot. I mean, they do sell, Mitiko is the name that comes to mind. I don't think this is going to make sense, but it's worth asking about the ... Because I think they're also like half the cost of the machine, but- Like a two-pallet changer on the Maury?
00:17:39
Speaker
Yeah, no, no, no, no, no, on the new Robo. Oh, okay. Brother. I think you just keep it simple. Yeah, basically find a machine that gets some work off the curve. I mean, heck, could you even
00:17:55
Speaker
So the weird thing about this is the horizontal, about sharing, workholding between it and another machine. Could you put a row of bases on the Robo or Spedio so that you could, quote unquote, rough on it and then just move the pallet over to the current? Well, it's the finishing that takes the time. So it depends on the part. Some parts take a lot of time roughing.
00:18:21
Speaker
But yeah, keeping the machines kind of modular between each other, where you can move fixtures and pallets, or I could make the pallets on the Kern. Like for doing the Rask blade, the blade has to be tilted at four degrees. So on the Kern, I just rotate five axis and it's done. But on a three axis machine, I'd have to make the fixture at four degrees.
00:18:43
Speaker
So that the blade, like, so I can make the fixtures on the Kern and then run it on the, on the three axis machine. Um, not a huge deal. No, not a huge deal, but, but having interchangeable fixtures between machines becomes huge. Yes. Look, if, if this is where the conversation is heading, I think it is don't buy a horizontal because it's, it's, well, it's probably more than half with it to where the cost of the current is actually used with the current. Yeah.
00:19:13
Speaker
Um, so what can you do to get the most out of the three axis for now? Uh, and in most part, what does, what do you need to decide when you're working? Like, do you need, uh, yeah, do you need through spin or cooling options? Yep. Do you need chip conveyor? Probably. But yeah, to look through all the things and figure out what, uh, what's efficient and beneficial. And, uh, yeah. I mean, look, I, I, uh,
00:19:44
Speaker
I wonder what a used one would sell for. I'll make up some numbers though. A used which? Well, something like the argument of buy two, say there are 100 each, buy two for 200 and then sell them for 10, 15, 20% off in a year or two. That actually does seem expensive to lose that much money. But how much have you made in that time?
00:20:10
Speaker
Well, and I guess that you'll grow into keeping at least one of them. That's where we're really struggling now is we've missed having a machine that is generally idle for R&D, for prototyping, for testing. Like I even have some Renishaw stuff that I'm working on where I just need to go to a machine and play. We're trying to find cycle time by stopping the double touch. So look, it's the reason I love Haas. They've done a great job of just ready to use turnkey Renishaw routines.
00:20:38
Speaker
That work very well, but when you do a web like a pocket or measurement it comes in Relatively slowly with two touches and there's actually significant error Renishaw has a white paper on this about if you only do one touch the higher potential higher feed rate or inconsistency in a potential lag like how the interrupt works if you will between when the probe trips and when the machine recognizes the trip can cause a
00:21:06
Speaker
a real amount of delay. I think the math was like over 10 thou, but a lot of times we're just probing for stock location. We have an extra 50 thou. So I don't care. So I want to, I want to take an 18 minute program that has two minutes of probing in the beginning and take that two minutes of probing down to 15 seconds. Yeah. So you're saying on the Renishaw, if you go to the single touch, it's significantly less accurate than the double touch.
00:21:33
Speaker
It has the potential to be less repeatable, I think is the better word. Okay. Because on the bloom probes and touch cycles that I have on the current, it's single touch and it's super accurate. And I've known about, you could switch the render stuff to single touch as well, but I haven't done it yet. And like the Haas VPS, like 9812 is a single axis. No, that's a web.
00:21:59
Speaker
You can't modify that. That's a pre-programmed thing. You can if you go in the macros. What I mean is you can't add a, what do they call it, like a parameter to 9812, like a K GQRS value to append it to change the number of touches. You can change the over travel or other things, but I've been going through that and just looking at the sub program that it calls to generate that code, and then that's what we're going to strip out.
00:22:30
Speaker
Interesting. Yeah. Machine purchases have been big on my mind lately. I would guess you can get a drill mill too in weeks versus a horizontal or
00:22:50
Speaker
current is months plus months. Yeah, if I'm going to custom order something, I'm just going to custom order a current. Yeah. But that's the thing with the RoboDrill Earth video. It's candy. They're available. They're everywhere. They're fairly easy to sell, I would assume, even if you're willing to take a hit. If you just need it gone, somebody's going to buy it. Yeah. Probably say, is Haas the same way? Oh, for a resale? Yeah. Oh, yeah. And just purchasing, right?
00:23:18
Speaker
I think it's also just the market's been strong. The weirdest way possible, the economy is strong. Things just seem so weird. And I think there's real inflation out there.
00:23:30
Speaker
I regret we sold the VM3. I don't regret it because it wasn't the right machine for us anymore, but I wish we had kept the spindle. You know what I mean? You could use the X-ray. Right now, yeah. My mistake. The other one I wanted to ask you about, we texted about a little bit, but the Datron Neo. You guys had one on loan for like six months, right?
00:23:54
Speaker
I like it on paper. The work we want to put on it is the foam for our cases. Right now we do that on the UMAC, as you've alluded to for the past year or so, I think it's time we get rid of the UMACs. I want to love them so freaking much, but I don't need a project.
Considering Datron Neo for Composite Work
00:24:13
Speaker
I need a machine that just works. There's a lot of bugs on them. Even the one that works is not solid? It takes me 20 times to start it before it'll say drives not operational.
00:24:26
Speaker
Yeah, okay. So actually I texted one buddy and I'm like, hey, you've been looking for a machine? And he's like, yes. Yeah, that's great. And he's looking for a project. So like that could be perfect for him. But yeah, so the foam for our cases, our foam volume is going to probably triple this year because we're going to do it for the pens as well. So we're going to have a lot to go through. And then also I want to do more carbon fiber inlays and G10 work and things like that.
00:24:55
Speaker
So, looking through the Datron with the vacuum table and just it's a small enclosed machine that's totally dry for dusty stuff. We hook it up to a HEPA dust filter, dust filtration system and like profit. I don't know. Yeah.
00:25:13
Speaker
When you hear about some of these startup companies that are flush with cash and need fast iteration and product development without nearly as much constraint as you and I have in the Datatron platform to be totally make sense.
00:25:32
Speaker
You do have to learn it. It's not like it's just a walk up and anybody can run it. But it's intuitive control and I like that. You could build a format to teach somebody pretty quickly how to make parts on it. And they're great. I mean, I remember the first time we machined one of our pallets on it and we set it on. And this machine is tiny. It's like a phone booth. And the travel area relative to the machine size is such a win.
00:26:00
Speaker
And they're just, they look cool, they're great. It's just incredible what it can do with that size footprint. And we took this plate that we just machined and we set it down on our granite fixture plate and it almost fell off because air hockey skated across the pallet. I mean, that's a level of flatness that is, you know, incredible. So in many respects, I have nothing but good things to say about what we were able to do with it, what the machine's capable of. The two things that,
00:26:29
Speaker
our noteworthy in this instance, one is against the machine and one is, is, is criminalized. The machine, I think maybe the heavy filter would, would help, but I didn't love from the long-term process reliability standpoint, the fact that there's no tool holder. It just holds the six millimeter shank. Eight, I think. Thank you. Eight.
00:26:53
Speaker
The eight millimeter tools are just held in a wine rack at the back of the machine, which is covered with a garage door, but we found under relatively light use that that became contaminated. And so every time it changes tools, it's just picking them up itself and it's retouching them off. But I don't think, and maybe I'm wrong, but I don't think that's the
00:27:14
Speaker
best unattended or lights out or, you know, we've had a couple times where pickup tools and they got more run out. I suspect that could be long-term. I just don't know, I'm not a bad. A strong dust collection thing could help. The second thing is just, look, if you're just cutting foam, John, and you need to look at what's gonna help, you know, you have so many chips to play with right now in the next 12 months.
00:27:43
Speaker
buying that cost of the machine for phone is not where you should be spending your money. Yeah. Not just phone though. Like we want to do a lot more composite work too and handles things like that. So it's almost the cost of a speedio. Yeah. You know, it's like a hundred grand for that machine. Yeah. So I totally understand that. It's more than I think. Yeah. Yeah. Be a bit more with optioned out. Um,
00:28:09
Speaker
So it's probably getting close to 80, 90% of a decked out speedio. So I don't know. It's a tough call. The floor space is a concern. Yeah. I'm not against it. I'm just saying. I know. I know. It's a tough one. And like our buddy said, it's current money for a router.
Exploring Affordable Machines for Specific Tasks
00:28:33
Speaker
Here's what I would do. The ShapeboCo HDM is a linear rail, high RPM. What Vince has been doing with it here is, frankly, amazing. It is, I think it's five grand. It's literally 5% of the price. It is not a day trunk. But for foam, it's going to be... I need a tool changer. Touche. Okay, that's gonna be it.
00:29:00
Speaker
by three HDMs and just move the pallets between the machines. Yeah. You move the entire vacuum pallet, still vacuums. Yeah, no, seriously. Just been unlocking things. Okay, that's a fair point. But it also kind of makes my point. A $5,000 machine frankly is capable of doing most of what you need. You shouldn't have a tool change.
00:29:26
Speaker
I mean, I want it enclosed. I want suction on it. I want at least a dozen tools for various other works. Like I've been making these aluminum plates for our heat treating setup, trying to make them on the U-Mac. The U-Mac keeps failing, E-stopping in the middle of the program, yada, yada. And then yesterday I just finished it on the Kern. But now I'm taking Kern time to make this stupid aluminum plate. Like if I had the data on that bit done,
00:29:55
Speaker
like six hours ago. That's John, send it out to a local shopper. We'll make them for you. Like that's not what you should be doing or buying machines. But again, it's, as you said, it's having that spindle available sometimes, you know, like, like the day. So we thought about this too, like in a day, trying to actually would be an incredible,
00:30:15
Speaker
R&D machine because of how good the build quality is, how intuitive the control is, the probing, the footprint. I give it all pluses there with one caveat. You should also have a more versatile, powerful, capable travel envelope speedio as well. Yes. The current is the... Sorry, the Datron is a phenomenal number two R&D machine. Or if you're at Graemer in his garage, it's a great machine for him as his only machine.
00:30:45
Speaker
But for us, that's not the point. So yeah, this whole conversation in my head started with, should I just sell the two UMAX and buy two Robo drills? And then talking to Angelo here, I was like, oh yeah, the Datron, I totally forgot about that. Okay, let's think about that. Buy one of each, the one Datron, one Robo drill or Speedy or whatever. And then
00:31:07
Speaker
Now I'm like, okay, the cost of the speedio and the cost of the Datron is almost the same. Should we just buy two speedios? I don't know, back and forth. I do want one machine that stays dry. If I bought two speedios, one of them would stay dry. The versatility of it, or you could throw anything on it if you needed to, you'd need to add coolant and make a mess.
00:31:32
Speaker
Yeah, I'm with you. When we got our Datron, it was right when COVID started getting real. So some of the weirdest times are all lives. And we ended up making hundreds, I don't even know the count, yeah, hundreds of polycarbonate, Lexan, is that right? I think, basis for a local architecture company, or whatever you call it, contractor company, they got the,
00:32:01
Speaker
Bids in requested from hospitals and banks and all these retail places and they were Canadian giant sheets of Whatever plastic they could get at the time. There's a big run-on supplies like this and then but they needed a machine to base that could intersect by 3d printed teas, you know where they mesh in and
00:32:19
Speaker
They had all these drops from jobs, like pallets of these drops that were about 12 inches long and quite long. So we had them chop them down to, I think it was 20 inches or something to fit on the Datron. And we said, hey, get us, they rough saw those to get them to cut to length so they would deliver us pallets of these parts. And we just sat there on the Datron. And it felt good. It wasn't the original reason we had the machine, but it was also like, this is what it's all about.
00:32:49
Speaker
I'll tell you that made a great example of like just dropped on there. The vacuum with the Datron vacuum card was incredible. And it just went correct. Like the foam is awesome. It just can't get over it. Not having a 40 taper or 30 taper as well. Yeah. Well, and that would be the plan maybe, you know, because like I wouldn't put knife work on the Datron unless it's like light engraving or some sort of super, you know,
00:33:19
Speaker
Yeah. It's fun to think about, but it's kind of, you don't want to make the wrong decision. Well, in a weird way, I almost be more interested to see what you could do with a, you are robot automated, a daytron too. Yeah. Because the door lifts open, the vacuum could work well and you could suction the, uh, and the factor to grab the phone. So you could potentially.
00:33:48
Speaker
That just makes more sense. Like the robots, you don't have to, you'll have to reach all the way in. Yeah. Yeah. I did see one YouTube video where they were auto-loading these little aluminum blanks into an air vice on the Datron. And, uh, yeah, that's, that's a cool point. Yeah. But it's also like, man, if you're buying a speedio and a Datron optioned up, and then you are, you're 250. It's like, you're not.
00:34:15
Speaker
It's terribly, I mean, it's probably from Kern to Leich. But it's, the Datron has nothing to do with the Kern at this point, like in my head. It's like a speedio would match the Kern or would supplement the Datron something else. That's foam inlays, handles, carbon fiber, things like that. Okay, go ahead. So I can't not get a machine to solve those problems, especially if I get rid of the UMAX.
00:34:44
Speaker
Well, that's a good point then. And maybe if you kind of need the Datron as is, then can it help also solve any of the current time issues right now as well? Yeah, I thought about maybe a couple of things, but nothing huge. Yeah. Looking at the horizontal kind of led me down the path of looking at what we might want out of the next machine.
00:35:13
Speaker
This happened two years ago when I worked at Horizontal's and thought, well, maybe we just go with the five axis because all the reasons that we all know about it, it's just the price may not even be that much different and it gives you more of a facility capability. The lights started going off or clicking on the Hermla RS1.
00:35:39
Speaker
It just seems like it is beyond our needs and abilities. So it's their version of an automation system that mixes multiple size palettes and raw material in one automation platform.
00:35:57
Speaker
so it can change out. The 5-axis machine has a coupler-style table, so it can take whatever you want to put on it, directly put work piece of the tile pallets on there, or put a pallet that has a hydraulic vise on there. And they show the demo of it having an automatic vise on there, where it can load raw material, where you could have these sheet metal racks that have scissor nests, so the nesting patterns change size by just pulling it on one corner.
00:36:27
Speaker
you could go from holding one inch round slugs to four by four inch saw cut drops in two seconds. So I just slide it around. So raw material worth holding, super inexpensive. The problem with us on that idea is it's like the same issue with the Wilhelmin of if you're loading a mod ice one at a time
00:36:51
Speaker
you're spending, whatever it is, 12 or 20 tool changes on one part. And tool changes ultimately cost money and time and weight. So that's where the horizontal kind of weighs out of like, no, you want to have a tombstone that's got two strips on each of the four sides, eight strips of mod vices. So that face mills going across 30 parts, not one. But I still like that idea of even the mix of the two, like, hey, do you take this product idea that would be a lot of roughing
00:37:22
Speaker
You could rough out a majority of it on the horizontal and then have it move over to the five axis to do all the finish work. Yeah. Yeah. It's something I've realized the past few months thinking so much about machines and fixturing and sharing work between them and stuff. Fixturing is absolutely everything. Like I commend you guys for making fixture plates because you're solving a real problem in the world right now.
00:37:50
Speaker
I'm stressed about what pallets systems to put on my machines, what vices, what interchangeable, what's the most repeatable, whether you go the Lang, four post, whatever, fifth axis, or Shunk makes a really nice VRO system. I bought one to test it out. It's fantastic. We've got the Aroa clamping, like what's repeatable, what's easy, what's fast, what's cheap enough, what's scalable.
00:38:15
Speaker
Yeah, it's everything. Fixturing is probably one of the most important parts of manufacturing and it goes unnoticed. When it works well, it's something you take it for granted, but it's always interesting when you come back. For us, we're looking at
00:38:34
Speaker
this potential improvements of instead of buying saw cut for size, buying saw cut strips, because we've got some good roughing, slotting recipes where you can slot away the material. If you're only loading one longer bar in there at a time, in that weird ways, I think it would be a potentially better option. We just talked for 40 minutes about one topic. I love it. We've never done that. Yeah. So what's next?
00:39:04
Speaker
talked about it until it's still kind of fresh. What's the next shoe to drop? We've got a couple more products we want to come out with over the next year, probably. So I'm trying to figure out what machinery will fit in that realm, right? You do too. So it's like, okay, we could keep doing what we're doing. We could scale that, we could grow that, but I want to do a little bit different stuff too. So I want to make sure we can scale into that as well.
00:39:34
Speaker
And I mean, at some point, for a business to scale, it's not just by one machine, double your production. That includes twice the lapping, twice the surface grinding, twice the heat treating, twice the cleaning of photography, twice the assembly, twice the sharpening. Our business would have to grow everywhere in order to sustain that kind of volume. So a second current has its upsides and downsides as well, right? Sure.
00:40:00
Speaker
It's not the only solution to
Scaling Operations with New Machinery
00:40:02
Speaker
the problem. It's like, okay, we'd have to hire like five more people and probably get another lapping machine or something, you know, all solvable problems, but it's not a simple thing, right? Yeah. Is it what you want? Yeah. Yeah. I do want to grow, but sustainably and at a pace that feels right. I don't know.
00:40:28
Speaker
You want the current, you'll probably end up with a second current cycle. I agree. But it's not happening today, so write that off. I think automating a three axis is not something you should do right now because it's too much money time, risk hassle. It complicates the machine itself about having some other bottlenecks of tool stuff. You make bad parts, just no. Absent the idea of like, hey, could a UR help on a Datron? Maybe that's what you're looking for.
00:40:58
Speaker
So that's a good point too. It's like, I don't know. I see the cycle times on the current and it's like ridiculous. If I had parts that only took, you know, 20 tools and I could do that all day long, then I needed all day long production. Then a robot on a three axis makes a lot of sense, but I don't. And I don't know. I agree. Yeah.
00:41:26
Speaker
Yeah. Anyway, so I'm going to call the speedo dealer today. It's on my list and, uh, just get some information. I've been there a couple of times for open houses, um, but I don't know them very well. So I need to see that. Are you guys still the dealer for brother? I don't know. Yeah. Like, or just to see, I mean, what automation, you know, gossegers are a Kuma rep. They have an in-house automation team now. Like it's, see if they.
00:41:52
Speaker
see what your dealers want to show or push and make you think about, because look, I have to say I'm wrong. Like maybe there's a bunch of people out there, automated three axes and they're crushing it and you're just quietly just like you mentioned a guy, just killing it. Yeah. I've seen a few examples. Like I've got a buddy here, um, modus who makes fire firefighter stuff. So he's got a, uh, palette changing brother, like the R six 50 or 450 or something. And also, uh,
00:42:20
Speaker
older speedio or older robot drill. And he said he bought a KUKA robot and is part loading slugs of aluminum into the robot drill, I think. And then they come out, they get, I think they get cleaned and then they get put in the tumbler or on a conveyor belt that go into the tumbler. And then when the conveyor belt's full, then it turns on the tumbler automatically and he came up with this whole like system. And I'm like, that is so cool. He seems happy with it.
00:42:48
Speaker
What the, is there a single part that you would most like to automate on the three axis, like knife handle or blade or something? Blade grinding, like the way we grind the rask bevels, um, except never really showed it that much, like how we do it. So I don't know if you have a good visual, but it's a time consuming process and it's current time right now. It takes over an hour to grind a rask blade. If I could get that, like for each blade, if I could get that off the current onto a speedio, I'd be happy.
00:43:18
Speaker
Dude, buy a junk. Well, I guess you want to get rid of it. If the u-mox don't turn on, like reliably get rid of that. But like, like thinking back to like AB tooling, who would have these Haas machines where they would proactively replace ball screws every few years because they were grinding on them in these wheel packs. Get a used machine and set it up to only grind, rough grind, brass blades. Can you do that? Can you rough grind and leave? It's the time it's in the finished grinding. Yeah, yeah.
00:43:48
Speaker
Improve your grinding. Yeah, this is where I've landed to get the right finish we've need. But an hour of time on a Kern, I can't imagine it should be less than $100 in value. And I don't think the rest justifies that. Sorry, I love it. It's beautiful. I'm just saying, it's not a $3,000 knife. It's what is the rest? $100,000 for? Yeah, $900. Yeah.
00:44:18
Speaker
Yup. That's a lot of- Yeah, I've been comparing the Rask cycle time is all current and it's a lot longer than it takes to make a Norseman and the Norseman is half current, half Maury. Yeah. So I'm playing with that and we're retailing them for the same price because it's a Grimsmough knife. But I guess technically we are losing because it takes so much longer.
00:44:44
Speaker
Yeah. We should charge more, but it's just like, look, it's just the way we're going to do it. Exactly. We need to be smart to be competitive. Exactly. Yeah. Yeah. Look at the, I would think that's the better meat on the bone. Even if you could shave that path. Yeah. Cycles wheel. It's tough because you don't have the time to play because you're making, you're making parts. But there's definitely cycle time in the whole run.
00:45:14
Speaker
to, to pick away at. Yeah. So I'm going to start digging into that. That'll help buy a, buy a Walter. Yeah. No, I'm serious. Yeah. I've thought about it, but a different, it's a different grinding strategies. Okay. Yep. Cool. Very cool. How's the Wilhelmin? The Wilhelmin is all put together. We have all the air system.
00:45:41
Speaker
I'm having an error that says, 1309 spindle servo not okay.
Troubleshooting Wilhelmin and Personal Projects
00:45:48
Speaker
Sometimes it'll work and sometimes it won't. I've been chatting with the Wilhelmin techs and they're like, okay, first thing you got to do is you got to find a computer that has Windows XP on it. Then we'll give you the software that's not supported anymore and it only works in XP.
00:46:04
Speaker
We're going to load that and we'll be able to plug into the parvex spindle drive and see the error that's on the drive, not just what shows up in Fannock. Um, so luckily Hansel has an old laptop that's still on XP and he brought it in and maybe today we'll find some time and, uh, do that to find the error. I love this. I'm also scared. I know. Like originally scared. Cause I asked the way one of the guys I'm like,
00:46:29
Speaker
Can this machine be reliable? Because we're going to start trusting it and relying on it. And he's like, yeah, yeah, we just got to find all the bugs and fix them. They're all, they're all fixable. I think about the work we're putting into the Willeman, it's a 2004 and the work we're putting into the UMAX, they're 2002s. And I'm like, I would rather spend all my time on the Willeman. Like that's going to have an ROI, whereas the Willemans need to become a speedio, you know? Yeah. And I'm okay with that now.
00:46:57
Speaker
I know, I think I said that wrong, but, but yeah, you know what I mean? And I know I, for the best of reasons, gave you a hard time on the U box. Um, because, what, because the alternative was a much more reliable road road gorillas, four years old, regardless of the bridge. I wholeheartedly want to see this willingness. Exactly. And I'm super scared though, that it's like one of these, like it's not, it's just like,
00:47:26
Speaker
spending reverse analyzing doesn't get doing good, but it's like, wait a minute, this was supposed to be a working condition when it left a year ago or whatever. It seems like it's maybe more gremlins. That's a good point. I didn't think about that. It was in operation when they unplugged it and why am I having a spindle drive error now? I didn't put those two together.
00:47:46
Speaker
So the audience that has run the Willams and Chiron's has reached out to us and they're like, look, they are awesome machines. Make sure they have like no downtime because they just machines what to stay on period. Like don't let them go out for two weeks. And we've had ours on our shop floor for seven months, I think. Yeah. So it kind of bugs me, but we're so close. Oh, so close. Yeah. But I mean, the alternative is spending
00:48:15
Speaker
seven times more than I spent on this one on a brand new one. Oh my God. I don't want to do that right now. I mean, put it this way. If it comes to the point where it's like we run it hard, we run it into the ground, it's unreliable because it's old. And then it's like, yeah, we love it so much. It's time to buy a new one. Then we'll have that conversation. But until then, like we got to make this work. Yeah. I will say the new ones look sick. Yeah.
00:48:44
Speaker
I got to share, it's like the most heartwarming thing. I've had quite a few people reach out that are either fellow like Porsche enthusiasts or hobby racers. I don't have any racing desire per se, but I guess inevitably the two are interrelated. But over the past week, I've had most of the engine front end off, sorry, not the engine, but the timing chain, the pulleys,
00:49:10
Speaker
the last night got the water pump off, which was the culprit. Uh, why the car sat for eight years. And this is the 9 24 9 44 9 44. Yes. 1985 actually 19 five dot five. So it's like the half year rev up. Um, and unfortunately the two M six bolts at the bottom of the water pump, uh, which where the gasket was leaking
00:49:35
Speaker
were so far rusted and corroded that they both snapped off. So I've got to figure out, I will get them out and I've got to do all the old like tricks of thinking through like, okay, you know, heat and penetrating oil. Yeah, you can drill a hole down the center and use like a reverse out left hand drill bit. Yeah, there's not that much clearance in front of them though, because it's kind of the front of the engine, which is in behind the radiator. Oh, I guess I could pull the radiator.
00:50:04
Speaker
if I had to. Yeah. We'll figure it out. Like also consider just drilling it out and putting a helicoil in there if there's room. I just don't know how you picture the face of an engine block. It's facing the radiator. Yeah, you only have six inches or more. How do you drill? Well, one of them is protruding. So that one, I took a M six not saw I haven't done this. I cut the nut. I haven't tried yet. Yeah, I'm letting it soak it well. But
00:50:34
Speaker
cut the nut so that I can then clamp on the nut with vice grip signal around the screw. And I'm hoping that one will back out because the rest of them, frankly, came out relatively easy. It's not like they're seized inside of the aluminum. I've definitely welded a nut to the stud before too. And then pull it out that way. That's the other option because the other one is like
00:50:55
Speaker
one millimeter proud of the casting, but that's a gasketed circuit. So I don't want to go like a Dremel in a slit for a slotting screwdriver. I don't want to touch the aluminum. So I'm not totally sure my game plan just yet. Yeah, it's fun. Yeah. Oh man. I remember doing this stuff on the Volvos. Kind of glad I don't anymore.
00:51:20
Speaker
But if I can get that back together and it runs, I'll be incredibly proud. And are you doing this in your home garage? Yeah, just in the garage. Yeah, that's awesome. I'll see you next week when we have all the answers on our automated three-axis. Yeah, horizontal debate. All right, man. Take care. Have a good day. Bye.