Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Building Up & Literally Building Out
235 Plays5 years ago
This podcast has always shown both sides of the business of machining. This week is no different. Saunders' & Grimsmo are both building out sweet central coolant systems. Saunders' is building out A.L.E.X which is their in house ERP system. Saunder's may have had a little bump with his swiss lathe, very veeeerrrrry slowly but everything is all good!
Recommended
Transcript
Introduction to Business of Machining Podcast
00:00:00
Speaker
Good morning and welcome to the business of machining episode 179. My name is John Grimsmo. And my name is John Saunders. TEG, SEAG, Tormach, Haas, DMG, Maury, Datron, Nakamura, Kern. These are the CNC machines that we all love running. And this is the story of running the businesses around running those machines. I'm sure I left some out, but that was off the top. Yeah, yeah. That's amazing. Right? It was a SEAG X2, right?
00:00:29
Speaker
Grizzly X2. Grizzly, I always say that. I owe you your apology.
CNC Machine Experiences and Maintenance
00:00:35
Speaker
Yeah, you started with the tag and yeah. Yeah. Awesome. I still remember watching your videos before I got the tormac and telling my wife about you.
00:00:49
Speaker
in like 2011. Hilarious. Yeah. Well, the, uh, I wonder if Ben, Ben still has your, he does. Yeah. He still has. Oh, it's the late. Okay. What happened to the mill? The mill I sold it to some guy here in Ontario for like, I think I got 3000 out of it or something. Um, it was very good. And I'd never heard from him since. Yeah. Fair enough. Yeah.
00:01:19
Speaker
Life goes on. Speaking of machines, we actually had our first kind of good, but simple maintenance thing. We had a, I think it was probably self-imposed damage to the spindle fan on our VF2.
00:01:39
Speaker
And when I saw a picture of the fan based on the part number, it looked like one of those computer, like CPU fans that are, say, 100 millimeters. And I saw the price tag. And, you know, look, service parts are usually a solid margin. And there's no reason to complain about it. It's just the fact of the matter.
00:01:56
Speaker
And so I thought, man, but that seems overpriced. It was like $245. Well, I got up on the machine and the thing weighs like six pounds. It's like a dinner plate size fan with metal blades. I thought, oh, okay, that's actually a much more completely acceptable. But I think when we were messing around with the, we did actually have a rotor reunion issue a while back. And so the fan had a nick on it, which was actually causing a fair amount of vibration. And so
00:02:28
Speaker
We just ordered a new one and it showed up two days later and it's funny because I feel like there's machine repairs that are incredibly skilled labor. If you've got to balance things that you're doing in the control, especially if it's like a service level that you have to get into, an access level that most of us can't even get into.
00:02:50
Speaker
compensating homing values and control values and servo values. And then there's this, which was four cap screws, pull out a fan with a Molex connector and brand new fan in there, which is now like it's refreshing because it's just perfectly smooth. Mm-hmm. That's awesome. And that was on the VF2, you
Machine Failures and Troubleshooting
00:03:12
Speaker
said? Yeah. Yeah. Yeah, like on our Moria, the only thing
00:03:17
Speaker
The only thing that's gone wrong is the door interlock switch broke about a year, year and a half ago. We were down for a couple of days because the machine just wouldn't do anything. You couldn't even bypass it and fake it. It was just bad. We ordered a new one in a couple of days later. It was a little tricky to install because you have to just get all the wiring pinouts in the right way, but relatively straightforward.
00:03:41
Speaker
But yeah, some of those things, it's not that hard to tackle yourself or at least look at it and like assess the situation and be like, Oh yeah, I could totally handle this. Um, or, you know, time versus money sort of thing, uh, get the expert in and just, you do other stuff while they, while they, they get it done. But I was thinking you were talking about spindle fans. Do you remember on, uh, like when I had my Tormach mill,
00:04:06
Speaker
It had the built-in plastic fan that was bolted on the top of the spindle itself. A lot of guys were finding that those were very out of balance. I remember I spent quite a lot of time balancing that fan and the assembly, just trying to get it perfect. Some people would even remove that fan and put one of those big fans you're talking about in its place and get better surface finishes because now you have less vibration on the spindle.
00:04:33
Speaker
Well, it's interesting. So, yeah, the Torwalk fan was attached to, I believe it's called the Stater, but the actual rotating shaft in the coil. The Haas fan is just mounted to the sheet metal at the top of the Z axis. But if it's vibrating, that ends up transferring it through, which I kind of thought was interesting because
00:04:54
Speaker
I guess I thought, why wouldn't you just install some dampers or springs that would mitigate how much? Because fans, there's no reason to have an incredibly high precise balanced fan. If it's not, it's simply to move air. It doesn't need to be rigidly attached to a part of the casting or motor.
00:05:15
Speaker
So I actually thought about installing some springs, but the new one was so silky smooth. And I'm pretty sure the reason it was damaged was was due to us. So not likely to happen again. Yeah.
00:05:27
Speaker
That's good too. Yeah. We also bumped our lathe again. So not a crash. Actually, a really good example of just a crappy outcome, which is my fault. And what happened was we were making some internal fixtures on it. And I had left the Fusion file open at work.
00:05:50
Speaker
went home, did more work on it, and then mucked up my revs. And so what happened was we have two different axial holders in the lathe, one that's driven, one that's static, and I had to do drilling and tapping. And I put the drill in the live holder, and then I put the tap in the static holder. So when Fusion, you have to check in the tool library, whether it's a live tool, I think is the phrase.
00:06:20
Speaker
Because basically when I drilled the hole, the turret was going to rotate the drill and the part in the chuck would stay stationary. And then right after that, when it would tap, obviously the tap is in a static holder. So the tap is just going to remain motionless other than moving a Z with the turret. And then the part was going to rotate. Well, when I had made those changes overnight, I lost. That was the only thing I lost. And I, of course, didn't realize it.
00:06:51
Speaker
The next morning when I ran the first part, this completely baffled me. I got to the drill hole drill fine, and then I went to tap it. And when the tapping was done and I was doing option stops, I was looking after each process, you know, what I generally do when I'm proving out a one off. Get this hole was not tapped. Machine did not alarm out and the tap was not broken.
00:07:20
Speaker
What were you machining? Bubblegum? Like... 41.40.
00:07:26
Speaker
Okay. I'm confused. Yeah. So was I. And I sat there for like 30 seconds and words were said that were not, that are not on this family friendly podcast. And I was like, what? And there was a little bit of a visible mark at the throat of the mouth of the hole. And I'm like, what the heck happened? Well, ends up what happened was neither the, well, the tap couldn't rotate. It was static. And the, um,
00:07:52
Speaker
part didn't rotate, so what happened was the machine pushed the tap into the hole without the part rotating, which did two things. Number one, it pushed the material back in the chuck about, I'm going to guess, 70% of the way or something like that. A little bit of the tap was able to make it into the hole.
00:08:12
Speaker
But unfortunately, it pushed our turret out by about fourth out. Not the end of the world. And I, you know, I'm not a machine tool designer, but my, I've heard it through the grapevine enough comments on the trade offs in machines. And you know, we had looked at the, I think it's the Samsung glaze, which are these box weighed kind of a poor man. They used to actually make the castings for, I believe, Morisiki. And those machines are tougher and stronger
00:08:41
Speaker
But apparently, if you do have a big crash, you've got a bigger issue. Whereas the Haasle than other linear design machines, they're a little bit less tolerant of bumps. But when you have a bump, you can fix it. So long story short, we have to realign the turret, which we had Haas do after our last crash, because that was a real crash.
00:09:03
Speaker
this time, we've got the PDF, we talked to our HFO on the phone about it. They're like, look, just to learn to do this yourselves. And I kind of get it like it's one of the things that where you actually have to do it maybe every year or two anyways. I know Lawrence has talked about doing it. And it's actually not a big deal, except there's a heck of a lot of sheet metal you got to get off. So I'm a little bit I was bummed it happened. But I'm trying to think like maybe fusion should have
00:09:26
Speaker
There may be, there should be some connectivity, uh, between the cam, the posted G code and the machine control to say, Hey, you're trying to tap, but nothing's rotating. This cycle shouldn't run. Um, but you know, we'll fix it. No big deal.
Fusion 360 Challenges and Solutions
00:09:41
Speaker
Yeah. That's a sucks, but that's why it like.
00:09:45
Speaker
I've really only had one computer one laptop that I bring with me everywhere right now so I haven't really run into that problem multiple instances yet but I have a second new laptop on order is coming in the next day or two and I'm gonna have to I figure I'll leave one of my computers here and open so I can
00:10:04
Speaker
What's that called? Online, see it? Remote. Remote PC kind of thing. Just if I have to go in and close something. But I do need to get into the habit of basically closing Fusion or closing all the project files when I leave, I think. Yeah. I don't think the software handles it as gracefully as it should now.
00:10:27
Speaker
which is potentially pretty catastrophic from either the integrity of your cat or cam, but then also just demoralizing. Because if you happen to do a lot of work on one or the other, I'm not really aware of a good way to reconcile the two without spending a lot of time trying to rethink, kind of trying to retype the paper that you lost in the power outage. Well, doesn't, I mean, autosave usually hits it pretty often. So the most recent version,
00:10:56
Speaker
will have been saved, don't you think? And then if you go home and open that most recent version, I don't think autosave pushes the version to the cloud. It could be wrong. I don't know. Pretty sure autosave is only the local cached version. Ooh, that's interesting.
00:11:14
Speaker
Because you can do that as kind of a cool little fusion hack. If you hit, I think, Control-Shift-S, it will save only a forced local version. So if you happen to be on a slower bandwidth situation or a huge file, and you just want to make sure you save something without worrying about pushing it to the cloud, which to be honest, most of us on an hourly basis don't really care about.
00:11:37
Speaker
Like, let's say your flight's about to take off and you've got to close your laptop or whatever. Control shift desk will force the save locally and then it can deal with the cloud push later. Cool. Anyway. Nice. Well, I'm sorry you went through that. But it'll be a good experience for you to learn how to align that turret and get a better feel for the whole machine and how it's put together. Yeah.
00:12:05
Speaker
Yeah, because I'm not really upset, you can actually still turn perfectly good parts with it right now because it's not that far out. And it's really only an issue for, I guess it would be an issue for like ID work maybe, but in the live tooling is a smidge off. But I think maybe I'll say turn alignment is something where it's never perfect. It's always just as good as you need to make it for the task at hand. So I'm not happy that it happened, but we're going to mix.
00:12:35
Speaker
Well, I guess if you had a long drill axially, that might not be super happy, right? Because it'd be kicked out a little bit. There you go. Yeah.
00:12:48
Speaker
Angela's been using the Nakamura to drill our pen tubes because we used to buy tubing before. Oh, yeah. But now, past few weeks, we've been machining it from beautiful centerless ground bar material that I buy for the Swiss. So it's like plus or minus two tenths diameter, beautiful surface finish.
00:13:08
Speaker
It's the right grade of titanium because the tubing is grade nine titanium, but everything else we do is grade five. And so like the tip is grade five, like on your pen, the tip is grade five, the tube is grade nine. They don't anodize the same color and it's been driving us crazy. Yeah, I can't tell.
Machining Innovations: Titanium and Part Ejection
00:13:28
Speaker
No, it's really most noticeable on purple. The other colors are similar enough that you don't really tell, but purple is like, oh, those are two different purples. Yeah, just little things been bugging us like crazy. So yeah, Angelo's got like a four inch drill in the lathe right now.
00:13:46
Speaker
And, uh, it was nerve wracking having that whole thing deep in the hole. It's like 15 times D or something. Come on, John, that's nothing. It's deepest I've ever drilled, but you've got, you've got through spindle, right? Yeah. We've got three, we have 300 PSI there. So.
00:14:02
Speaker
Yeah, it works. But yeah, Angela's been handling that and they're going really good now. It's like three minutes cycle time for a tube. Three minutes to drill or the whole tube? The whole tube with threading both sides and everything, transferring everything.
00:14:17
Speaker
Yeah, that's fine. Except right now we can't eject the part, so it's a manual pull and hit go again. Because the long part and it's clamped all the way deep on the sub spindle, we don't have an ejector with that much travel and we don't scratch it on the way out either.
00:14:36
Speaker
So I'm starting to wonder about those coolant fed grippers like used to pull bar material, but to use it to transfer like to eject the part. I mean, you know what I mean? Yeah. Like Royal makes the one it's like twelve hundred bucks. It's not cheap, but.
00:14:51
Speaker
It's cool empowered, so you'd have your through spindle or your regular pressure. It's got these three big fingers that with pressure, air or water will clamp down on the part. You can make little finger pads, not to grip or scratch the part, but to...
00:15:08
Speaker
hold it gently. And then that could use the turret to remove the part and dump it in the parts catcher. Sure. So that could work. What if can you see you can't use the part ejector even to push it out a little bit? Well, the part ejector is a the only one we have is a hardened spring loaded five C attachment thing. And it's only got like half an inch of travel.
00:15:37
Speaker
Well, but if you could push it back out at all, could you re-grab it with the main and then from the main... No, there's bar in the main. Yeah, you're right. We don't have a bar feeder on that machine to push bar forward at all.
00:15:52
Speaker
Right. Got it. I'm just trying to think of something on the turret or something on the parts basket itself that can grip and pull forward or the bee can move back or something. How far out is it sticking from the sub? It's like 3.25 inches long and in the sub it's sticking out under an inch.
00:16:13
Speaker
Oh, that's easy, John. Just take 3D print. Have you seen the spring-loaded fingers? They're kind of like chopsticks. Okay. Super easy. And then mount it to the backside of your turret and then either move the turret to the sub or the sub to the turret or both. And then you'll have a X-axis motion that'll come in and snap those chopstick fingers over the part. You open up the sub and then you're good. Now, oh, then you've got to strip it off of that.
00:16:43
Speaker
Oh, yeah. Unless you could re-grip it holding on to like 50 thou or something and then pull it out of the chopstick. There you go. No, without scratching it though. Re-grip, you're just going to scratch it?
00:17:01
Speaker
regripping or pulling it out of the chopsticks, depending on if they're plastic, it won't scratch it. Yeah, that won't be an issue. Well, I'm trying to think there's a way. Sorry, might be onto something though. Yeah. Boy, you have no other degrees. You have nothing else to trigger. You don't want to get it complicated with timers and electronics.
00:17:21
Speaker
Yeah, or other devices like little air cylinders mounted outside the bee or something like we have the turret that moves and you can use coolant to activate a piston or something if you need to or you just go spring-loaded chopsticks like you said.
00:17:37
Speaker
I'm trying to think about those drawer pools that everyone has in their kitchen these days that soft close. Okay. Do you know what those things look like or are? I know what they are, but I don't know how they do, I guess. Yeah. So I did the same thing.
00:17:55
Speaker
we have like old cabinets in our house. And so I was at some house and they had the drawers and coverage that you can't slam. And I'm like, how do these work? And so I was looking at them and it's interesting if you open the drawer, you actually are resetting the either hydraulic or pneumatic device. And then it's just kind of like a passing fluid through a small orifice. So when you go to close it, it bleeds it out. So I'm trying to think if there's a simple way of
00:18:21
Speaker
On those chopsticks, when you push the chopstick over the tube, that could charge a cylinder that then starts bleeding out in a controlled and repeatable time interval. And by the time it re-relaxes, you've already got the parts catcher up there. So then when it relaxes, it falls. Just falls into the parts catcher, which that's your goal, right? Yeah.
00:18:47
Speaker
or is there a way for when the parts catcher flips up that it could grab the part, but that's potentially still an issue because it's still then. No, no, no. The parts catcher will hit a tab on the bottom of the thing that'll release it. There you go. There you go. That's it. Oh, that's it. Something like that.
00:19:06
Speaker
Because then you need it. Where does the parts catcher dump it into a bin outside the machine? Yeah, it shows it to the door and then it flies it forward. Right. So when it goes to catch part number two, part number one is already outside of the parts catcher. OK, yeah, there you go. Little bit of a tight tolerance to have the parts catcher trip a thing, but doable 3D print it that way. If it crashes, it's just breaking a piece of plastic. Exactly. Yeah, the parts catcher comes out pretty aggressively.
00:19:36
Speaker
You can slow it down, I think. Because there's no reason. Well, I don't know. I'll play with it. But yeah, there's some cool ideas here. Well, here's what you do, John. Sorry. This is even better. OK. Chopsticks grab the part in the sub. Sub spindle opens. Turret moves the or beam moves out so that the turret chopsticks are holding the part
00:20:04
Speaker
with nothing else around it. Parts catcher comes up, is in place, then the turret moves over to the parts catcher, and the x-axis motion of the turret pushes against part of the parts catcher, and that causes the finger to open. Yeah. Because that's something you can control in G-code. That's safe and controlled, and nothing slamming, and yeah.
00:20:29
Speaker
In Royal, I think, or somebody makes these sort of style fingers, but they're usually meant to be a puller, but just print your own. This is easy. For sure. I'm endorsing a DIY project. Yes. I'm loving this idea. This is really cool. I think I said a couple weeks ago, somebody gave me that Prusa Mark II printer, and it's phenomenal. It's so good.
00:20:59
Speaker
Yeah, I've been having a lot of fun with that. The hot end is leaking from somewhere. It'll drip outside and mess up my prints. I might just get a new hot end and throw it in and call it good. Yeah, we love our... I think we have the rev above yours, like a three. Yeah, the newest one. Yep. Cool.
00:21:25
Speaker
I wasn't going to say about that. Oh, I was just, that one made me think of the 3D printing video, which is in a couple of weeks, but yes, or today's video, which is two days, sorry, if you're listening, it's Friday.
Major Tool's Shop Tour and Hiring at Saunders
00:21:38
Speaker
So the video that came out two days ago was the shop tour of major tool in Indianapolis, which is
00:21:45
Speaker
I mean, we saw the space shuttle, the new SLS thruster machine. I mean, there were parts in there that were by far cooler, larger, crazier, more impressive. I mean, 600,000 square foot job shop with machines that were measured in meters or had stairs or the one that had a separate gantry, like controllable gantry system just for the operator relative to the rest of it. What? Yeah.
00:22:15
Speaker
so big that just unbelievable. Oh, I can't wait to watch that. Yeah, it was a really good tour. How long is this one? 50 minutes maybe. Perfect. Yeah, really cool. I mean, just one of those reminds me very much of
00:22:32
Speaker
when we were kind of doing the research and reading up on the Apollo program for the 50th anniversary last year. And you start to realize like you just can't throw money at certain things. Like you have to have a deep seated roster of multi-talented companies and people across all sorts of industries who know material and metallurgy and welding and prep and sourcing and QC and have incredible machine capabilities and just, just
00:23:00
Speaker
It's what life's all about. It was really cool. That's awesome.
00:23:07
Speaker
Speaking of that, we are looking to hire two people. You ready for this? Yeah, bring it. Number one is the job will be machinist slash SMW employees. Helping us do everything we do here in terms of running the shop, making our products, that type of work. The other position is, I think a really interesting one, if we can find a person that's a good fit for it,
00:23:34
Speaker
which is PocketNC, the Tormach XS router, the ShapeOco, even the Datron and the smaller Tormach machines. We have another benchtop router that I'm not recollecting off the top of my head. PocketNC was the one I mentioned. So with all these...
00:23:53
Speaker
We don't have a random, but I'm sure we can. They would probably be happy to get us one for ProvenCut, which is basically part of this job, which is we want to find somebody who wants to spend their full time figuring out how to run these machines as best as possible. What are the tips and tricks? What are the recipes for ProvenCut? And there obviously can be YouTube or Instagram type of content as well.
00:24:19
Speaker
We've got a good selection of these sort of hobby grade or benchtop CNC machines, and we want to be creating more content and information around them. And so if anyone's listening and is interested in a job here in Central Ohio, you can email jobs, J-O-B-S, at SaundersMachineWorks.com.
00:24:39
Speaker
That would be such a cool job for someone. Like, it's definitely not for everybody, but it is absolutely for a few people who would just totally nerd out and go fun and have fun with that kind of stuff. Yeah.
00:24:53
Speaker
And it's something that you figure out in life what's important to you. And it's cool to talk Kern and Okuma or DMG and automation and high end stuff. But I will never venture too far away from the passion and love for the folks that are learning or getting started or only have the $1,000 machine or trying to figure out how the heck do I actually figure out how to cut aluminum with this 50 pound machine? Exactly.
00:25:20
Speaker
I've seen it lately playing with the Prusa. This one was given to me as a gift, but you buy a new one for $800. It's fantastic. It's so much fun. It's affordable for anybody. It's really good. It's not perfect. There's always issues.
00:25:40
Speaker
It's just fun. It's just so fun. And with this upgrade, I gave my old mono price mini to a 13 year old. And he like he almost cried like, you know, getting it and he's been having so much fun. He's been printing Minecraft figures and little things and he designed something in fusion 360 already. And that's so cool. It's so rewarding for me to you know, this was passed on to me. So I passed mine on to somebody else. And it just it works so well.
Developing In-house ERP System: Alex
00:26:07
Speaker
Yeah, awesome.
00:26:09
Speaker
We are also, I'm going to steal the phrase from that book, we are going from good to great here. And it all surrounds Alex, which is what we're calling the in-house ERP system. We've made huge strides on it in the last week. Yesterday, I tried to put my first part into it.
00:26:33
Speaker
which was a planned process, and we hit some roadblocks. And as expected, we're fixing those right now. Not even fixing them, but just kind of we're thinking about little things like, hey, when we add a supplier, do we want to have a default? So that initial inquiries to this supplier go out in the form of a PO or in the form of an RFQ. POs would be for McMaster, where it's just the same thing every time, whereas RFQs are for material, where the pricing changes. How are we tracking the history of those orders and understanding
00:27:03
Speaker
the data set that will give us information over time that will be quite valuable. And then how does Alex tie into the real world of, hey, where is stuff on the rack? How is it labeled? Are we doing color coding? The big thing yesterday was
00:27:17
Speaker
all the internal part numbers. Alex will assign a unique internal number to everything that's added to it, but we don't want those numbers to be totally arbitrary because it will be helpful to have some meaning to them. They'll be like a categorization. Yeah. The current thinking is
00:27:38
Speaker
I feel pretty good about, but obviously keeping things flexible is always good because you never know what you don't know. But we're doing A, B, C, and S. A is an individual item. B is an improved product. C is a subassembly. S is kind of a finished or customer-facing product. A would be a screw. You literally just buy it.
00:28:01
Speaker
B would be, A would also be then a piece of raw material, assuming that it's saw cut to size. So when it comes to our door, it's ready to go into a machine. B is an improved product. So a machine's product or a
00:28:16
Speaker
Well, it would mostly just be machined products. I'm trying to think what else we would do. So B's are often made from A's and then sub-assemblies see like when we ship a fixture plate. A fixture plate is actually not a single product even if they just buy a fixture plate because it would also include a sub-assembly which is the correct T-nuts and fasteners so long as they're included with that model plate for that machine. And so,
00:28:42
Speaker
ABC will help us with how we label inventory and just how you think about stuff. And then, you know, if you run out of a C when you scan the work order for it, it's going to tell you, hey, this C was made up of these A numbers. They're located here. Here's a photo or video or a QR code or instructions or comments, et cetera. Yes. Yes. So much. Yes. Yeah. And in fact, I just thought of something, which is do we want to have a different number
00:29:11
Speaker
A was individual items, but I had been thinking yesterday about those all being product items, screws, fasteners, et cetera. But we also will be putting a ton of office supplies into Alex.
00:29:28
Speaker
poly bags or stuff that we use internally. I don't know that I want those to be the same A because they're ... If you need ... It's still an inventory item. A screw or a plastic bag are still sort of ... I don't know if you need to categorize them separately.
00:29:47
Speaker
I'm trying to think. Because what some systems do is they'll have cots items or consumer off the shelf or commercial off the shelf. Then it's anything you buy, basically.
00:30:00
Speaker
And there could be subcategories within that. You have hardware, you have office supplies, toilet paper, food, like categories within that category. But how deep do you want to go? I mean, if you just need some broad strokes, your ABCS system, then make that work for you for sure.
00:30:21
Speaker
no need to over complicate it too much, but you want to build the system now so that it's set up for the future. Right. Yeah, I need to think about this critically. I can't do that live on a podcast, but I can. It actually is helpful to talk about with you because I saw this thought that A has to do with
00:30:41
Speaker
what has to do with value, products that have value added to them or become more of something down the road. Like A is stuff that we do kind of care about and that's a great way to describe it. Whereas B is, like you said, if we end up doing, I don't think we're going to put toilet paper into Alex, but the poly bags or just the little stuff
00:31:04
Speaker
The cardboard boxes, I don't think of them as the same as A's. But from an inventory purchasing perspective, I don't know if it matters. I see what you're trying to say. You're trying to say one is kind of a value added product and the other is just necessary.
00:31:26
Speaker
you know, like, yes, paper towels in the shop or sharpies or shallower bins or something. But yeah, you know, like, like a quarter 20 screw is for a thing.
00:31:41
Speaker
Yeah, simpler is better because- Simpler is better. You have to decide whether it's an A or D, then that's no good. Think about it from everybody on your team's perspective and how they're going to categorize a new thing that they have to add to the system. You don't want to add too much pain or too many questions your way. Hey, is this new thing a thing or a thing? No, you're right. We'll just keep it as A. Yeah. You could have subcategories within A if you really want.
00:32:10
Speaker
Yeah. Yeah. Okay. So you're doing, is this through air table or all through WordPress? It's a SQL database that has a right now effectively a WordPress front end. Okay. So it is a SQL database. Oh, interesting. And it's amazing.
00:32:26
Speaker
Yeah, everything you know, we have will have three different live monitors. One will be outstanding purchase orders, which will be waiting to come stuff that has been ordered to come in. And then we'll have an internal work order. And then there was one more maintenance, I think was going to be the third screen. And then you will have triggers like
00:32:47
Speaker
If a, when you receive an item, you're at that computer with that station to receive stuff in, and if something isn't received by the scheduled date, like the automatic, like we assume MSC stuff shows up in four days or some comfortable lead time, that triggers the follow-up to say, hey, where is this or what happened? Or maybe we just forgot to mark it as received.
00:33:08
Speaker
Yeah, I feel like those custom dashboards, little tablets around the shop with the critical information you need in various places, I think that's where this is going to feel like it's winning, like it's amazing. You have your screen with your outstanding orders, and you're like, oh yeah, that should be here any day now, or the work orders to be made, or inventory levels to order next, or things like that.
00:33:33
Speaker
Yeah, it's a good idea. I've been thinking TV screens and that's what I've often seen in some of the other shops that I've been impressed by. But I think having iPads as well, it's just all going to be web hosted so that iPad would let you, who could build a little front end webpage, which would let you quickly toggle between any three of those screens. That way you don't necessarily feel like you have to go over to the receiving computer to look at current outstanding
00:33:59
Speaker
Yep. Yeah. And some stations will be just monitoring. You just want to display some information and some will be interactive. So you might have a keyboard and mouse attached to it, like a little tiny computer setup or something.
00:34:13
Speaker
Yeah. Are you going to have any interaction with, like obviously the receiving side. So when you receive product, it needs to be entered into the system. But as far as work orders or job tracking or things like that, is that going to be kind of computer-based project, you know, updates? What do you mean? Like as you're making a fixture plate, are you going to be like, oh, I just did this up. Click. I just did this up. Click now. Unlikely.
00:34:39
Speaker
I don't think we're going to have, right now at least, we're not going to have details within work in progress, nor actually are we going to have much with, and I recognize, well, so a big part of Alex is execution on it. We know there's going to be improvements and changes down the road, but basically we're not going to complicate Alex with a lot of inventory quantities right now.
00:35:07
Speaker
We're going to instead rely on the Kanban style system, both with things like screws. So I don't actually have a problem over inventorying things like a 30 cent nut. You know, we have an extra 500 in the box below. When you, when you tip in, when you get to that box, you trigger it. Um, that's fine. And then most of our.
00:35:27
Speaker
supply chain with the exception of the steel for the fixture plates. Most of the rest of that supply chain is generally a quick turnaround and reliably in stock. So it'll be the same thing where Alex doesn't know how many we have in stock. It's just as soon as we get to say the third to last plate, you hit the card, it generates the PO and it comes in. And then the same thing, once we get caught up here in the next month and the VF6s are making plates, then we're going to start inventorying many of the plates that we have
00:35:57
Speaker
that we're currently basically doing just in time. You have a random precision Matthews or a machine that we don't sell a lot of, especially on the aluminum ones. I'm willing to inventory just at least one of them. And then over time, we'll start looking at actual sales data to know, OK, if we sell one every three months, we may want to change that to having two in stock or something. Right, right. Yeah, it's a good system because setup and lead time are always factors to consider with things like little one-offs.
00:36:28
Speaker
Well, sorry, but to answer your question, like when you scan for like a mod vice top jaw, when you scan the work order for that, it'll send an email. It'll say the standard quantity that we make is 100. Here's where the material is. Alex isn't going to then assume that 100 or 98 or whatever were added into inventory because we don't have the resource or manpower to do inventory audits. And basically, that system is going to then immediately be wrong on day two.
00:36:56
Speaker
So, and I don't think it's necessary right now at least. Well, what some places do is if you're going full accurate inventory, you'd have barcode scanners. So as you take out two washers, you go click, click, and then it's tied to a computer and then it's updated. Yeah. It's a system you put in place.
00:37:19
Speaker
I understand the theory of that. Execution-wise, I can't get there. Honestly, it reminds me of when I toured Harvey and Helical Tool, and they talk about how they're 99% accurate on their order fulfillment, and they don't use barcode scanning for positive
00:37:38
Speaker
checks on how they pull and fill inventory, but rather they do a two-man system with color-coded bins and a spot check. Basically, they looked at all ... They process enough orders and they have enough resources to do proper studies of this. They were like, a scanning barcode system would have only gotten us ... It wouldn't have gotten them better, basically, and it would have cost a lot of money. It would have resulted in less training and knowledge and more burdensome this and that. It was like, okay,
00:38:08
Speaker
It's not better, right? Aside from needing some inventory levels for accounting purposes, I'm trying to get away from needing to know that. That makes it interesting. Do you know if that made it to the video with Harvey? I know I watched it a long time ago. Yeah, and I'm pretty sure it's toward the end of the video when we're walking around. Well, there's all these list of cabinets that are waist height or something.
00:38:37
Speaker
talking to a female person, I don't recall her name. She was like, she was the best order fulfillment guru I could have ever met. She had so much pride and passion in the system that they had developed. She knew how to defend it and knew it was right. But clearly, she was also smart enough to realize, hey, there's times where we need to reevaluate or think about, you know, wasn't she wasn't boxed in, if you will, I was it was phenomenal. I'm gonna go back and watch that. Yeah, I actually should too. Yes, that's awesome.
00:39:06
Speaker
Yeah, because I think about that a lot too. We've kind of been stalled on our ERP system for quite a while, but I want to dig back into it. How complicated do I want to make it? Do I just want to set up something and get started so that we see what actually works versus theorizing and putting a lot of time and money into implementing?
00:39:25
Speaker
and then testing and realizing you don't like it. So obviously fail fast, fail cheap is the better scenario here so that you can move forward and know what you like and what you want. And that's what you guys are doing too. And it's really cool to see.
Coolant Systems and Water Quality in Machining
00:39:39
Speaker
What's going on today? We've been building out our central cooling system. Oh. And I thought I heard you mentioned something about that last week too. You have a new thing coming in. We do. It's here.
00:39:52
Speaker
Yeah. So what did you get? So we partnered up, let me actually pull up their name because they did a lot of really excellent work on this. And it deserves its own video and longer discussion, but it's a fellow named Steven from the Valley Christian Schools that have a group of students that really put in an amazing effort to this Raspberry Pi and kind of Arduino system that will
00:40:18
Speaker
has a digital touch screen, it has all of our machines. This simple explanation is you walk up to the touch screen, it has a list of all the CNC machines at Sonos Machine Works. You hit the machine that you want and the amount of coolant that you want, and it then starts pumping water to that machine. And then at each machine, we have bypass valves, which will let you either add the straight water or
00:40:38
Speaker
we can quickly quick disconnect in the piston mixer that will then instead of pumping straight water, it'll pump water through that piston mixer with a drum coolant drum barrel and that'll add actually mixed concentrate or mixed coolant.
00:40:57
Speaker
Love it. That's very similar to what I'm building out here, except I'm going to have a pressure. So I'm going to have treated water, RO, then DI, and then to a, to an IBC container. And then I'm going to pump fresh water, keep it pressurized in the lines through the shop with a pressure holding pump. And then I think I'm just going to buy a mixture on for each machine and then have like a little five gallon drum at each machine. And, but this can all be automated and it would just float valves and it would just work.
00:41:28
Speaker
And then you can monitor the coolant level concentrations at each machine, because I know the Nakamura and the Ory take totally different coolant levels to top up. So right now we only have the one mixture on, so we're going to figure it out. But yeah. That's awesome. Yeah. That is a better system. I can't justify the cost of a mixture on at every machine right now. Yeah. Yeah, that's sort of, I just occurred to that. I kind of came up with that yesterday or a couple of days ago.
00:41:58
Speaker
I was like, well, in for a penny, in for a pound kind of thing. I think that's the better solution. But for now, we're probably just going to have the one at the head, and we're going to feed mixed coolant to the machines and kind of see how that works. I do think all of those are better than having hundreds of gallons of premix already mixed. I don't think that's a good solution. I know some people do that, but I don't think that's the best. Even I just had an hour-long conversation yesterday with,
00:42:25
Speaker
U.S. water systems, they sell DI systems and RO systems, and super knowledgeable guys. So that's where I'm getting my RO system from.
00:42:37
Speaker
I learned that as you strip all the minerals and stuff out of the water and gets your parts per million down to almost nothing, then the water is hungry. It'll eat away at metal. It'll leech. Yeah, it's spongy and it's hungry and it just wants to fill itself back up with garbage because you've taken all the garbage out.
00:43:00
Speaker
So you have to think about things like metal fittings and pump fans and motors and all that. I didn't know that. And I was like, oh man, okay, so I have to replan my system of how we're storing that water and what it's pumping through and what fittings and things and attachments. And I was like, oh man, that's the last thing you want is a leak like a year down the road because your water was so clean it ate through the garden hose.
00:43:24
Speaker
It's like that high school experiment of pure, pure water is actually not conductive. You can run electronics in it. Oh, yeah. I think we were talking about that.
00:43:35
Speaker
Go watch our water video that we did with Qualicam. It's awesome when you... Well, it's a little bit overwhelming, but things like zinc and brass or copper are common occurrences in machines and plumbing systems, and those can be problems, but a lot of our problems were frankly stemmed from some temps with synthetic coolant, which is much more sensitive to water quality.
00:43:59
Speaker
Basically, you're better off having water that basically isn't perfectly clean, if you will, because then you don't have that water act as aggressively towards leaching. I think leaching is the right word, but like you said, being hungry.
00:44:16
Speaker
Mm-hmm. Well, they were saying you can drink RO water. It's not too clean. But you don't drink the water because it's too clean. It's like less than five or something. And it's bad for you. But he said, you can store the RO water. It's OK. So we bought an IBC tank, which makes me feel like big boy, like, oh, I've got a big coolant tank now, stepping up in the world, not these five gallon buckets.
00:44:46
Speaker
So we're going to feed RO water to the IBC tank. We're going to store that. And then it's going to be pumped through the pump and then through the DI. So the only DI stored water will be what's in the pipes, the PEX lines going through the shop. Got it. And even that, even it sounds like that is less than ideal.
00:45:05
Speaker
Ideally, this guy wants the DI system four inches before you're spout your outlet. No, DI's are cheap if you just do the desiccant. The concern with DI's is you have to, if your RO membrane, not if, when your RO membrane starts to fail, it just doesn't filter or act as a membrane that well anymore. Your water will go back to the input quality, which
00:45:30
Speaker
Depends on what your initial hardness is. The problem with a DI system is that once the DI system becomes saturated, which again will happen, it effectively starts outputting salt water.
00:45:41
Speaker
So it's an actual incredibly bad consequence. And so having a bunch of DI's around the shop or even just one is kind of that ticking time bomb unless you've got a really good maintenance or checking system. So we have an inline PPM sensor. I have a weekly calendar reminder and we're pretty proactive about checking and replacing it. Does it show you a visual? Like you see the number on a daily basis? Weekly, but yeah.
00:46:11
Speaker
Yeah, like whenever you look at it. Yeah, the one that they have a little kit with just a little light bulb that lights up after a certain conductivity. So like when the light goes off, it's time to replace your filters even with a safe margin.
00:46:25
Speaker
John, for you at your level, I would just do the service. So that's the thing. If you do the DI system only, not the RO, you can use the pressurization from the city of Toronto or wherever. So you don't need to deal with the IBC and the tote, which would be really nice, and the well pump. Yeah. That's the better way to do it. With the service? Well, running DI only.
00:46:54
Speaker
It's putting so much work into the DI system that it needs to be replaced. The filters need to be replaced a lot more often. But if you feed it RO water, then the filters on the DI system, which are expensive, last for a much, much, much longer time. RO filters are way more expensive than DI. And it depends on how hard your water is. So we're trying to soften our water with a salt water softer before the RO memories to make them last longer.
00:47:21
Speaker
But do what you need. But the big shops that we've seen are using DI systems, but also because you don't have to store. The problem with storing water is it takes up a ton of space and you have to repressurize it to work in the system. And the pressure matters because you want the proportional valves or the mixers to mix correctly, which means it needs to have the same amount of pressure. You're better off just using the city water pressure for that in DI.
00:47:47
Speaker
Well, and I also want cooling guns at each machine that basically pump premix into the machine so that we can wash down with premix. So I need pressure. I want good pressure coming out of it. So there's lots of ways to skin this cat. Yeah, I haven't fully looked into a rental system, whatever you call it, a service.
00:48:08
Speaker
Before I buy something, maybe I should look into that locally. It's not that expensive at all when you're a real company. Because it can get expensive to buy what I need, the good quality, high flow, all the filters and DIs and RO's and stuff. The service is scalable and sustainable. The homemade system, which I love, which what we're using is not the same.
00:48:39
Speaker
Cool. Awesome. I will see you next week. Sounds good, buddy. Take care. All right. Have a great day. Bye. Bye.