Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
2020 - The Year of Custom ERP Systems
199 Plays5 years ago
The year 2020 has not been exactly how we have imagined it. But it is in full swing for both Saunders' and Grimsmo, especially in the ERP department. Saunders' is cooking with LEX and Grimsmo is full GERP ahead.
Both John's and their teams have worked hard to make it happen this year. It is a great teaching lesson for anyone to make the best of what you have. Focus on the things you actually have control over and take a step back.
Grimsmo on the other hand is not happy with their Qualichen 450 and will be switching back to 251 for the time being. Also grinding rask blades on the Kern has been amazing! Saunders also touches on adding Shapeoko fixture plates to his lineup.
NYCCNC has a great article on advice for buying a machine tool HERE:
Recommended
Transcript
Introduction and Setting the Tone
00:00:00
Speaker
Good morning and welcome to the business of machining episode one hundred and eighty nine my name is john grimsmo.
Importance of Entrepreneurial Support
00:00:05
Speaker
And my name is john saunters one of those things that i think a lot of good entrepreneurs recommend is having a group of people or person you can talk to and john is that person for me for four years we've had a candid conversation about the successes and struggles about how to make decisions and how to get to pass to work.
00:00:27
Speaker
Yeah, exactly.
Technical Glitches and Solutions
00:00:28
Speaker
Sometimes we have to be honest with each other and sometimes we have to tell each other that your audio is on double audio right now. Right. Seriously, your audio is double from the video. Not fixed. Much better. I don't know if the audience can hear that, but good. How are you doing? Good.
00:00:47
Speaker
Good.
Streamlining Purchase Orders
00:00:48
Speaker
I was off Friday and it was great to take a day off, but I love working and we're cooking. We've got a lot of things change, which means ways to digest stuff. So we've had a couple, like I said, a couple of new people start.
00:01:08
Speaker
And Alex or Lex is pushing out POs. We finally flipped this trigger on that to where, for a while it would just email me the PDF or I could just download it and then I would forward it on. But now,
00:01:23
Speaker
So you scan things, it goes into an order queue and then when you're ready to order from McMaster or MSC or an aluminum supplier, you just say push to PO and a few seconds later, it is emailed directly to that vendor.
Automation vs. Manual Approvals
00:01:38
Speaker
I get a CC in my inbox, which I love because it confirms it went out, it gives me a timestamp or evidence that it was when and where it happened. So it still gives you the,
00:01:52
Speaker
It doesn't just be like, oh, you hit minimum. I'm emailing MSC to order these parts. It still gives you the impetus to hit send. Say that again.
00:02:03
Speaker
It'll compile a PO, but you still have to say send to vendor. We do.
Overcoming Vendor Queue Automation Challenges
00:02:08
Speaker
I think it's good. Well, we hit a technical glitch, which is fine, where what I had wanted to do was to take many of our vendors and set queues that were automatically pushed. So for example, every Wednesday at 3 o'clock, a McMaster would push, or every Wednesday at 3 o'clock.
00:02:26
Speaker
And that way, throughout the day or week, we could be scanning for aluminum and scanning for fasteners, and then those would just automatically go out. We're having a hard time because it's a server side change to queue those and create different batch trigger points for different vendors.
00:02:44
Speaker
I suspect we will figure it out, but what we realized is not a big deal. I can just go in every day or two and look through the order queue and just push as needed.
Order Management and Inventory Tracking
00:02:57
Speaker
What we also did, what this forced us to implement is a
00:03:02
Speaker
timestamp. When something is scanned, it gets pushed into the order queue. If it's in the order queue for more than a preset amount of time, I get a reminder. That's really what we want to do is make sure we didn't add something into McMaster and then six days later, nobody ordered it. All this is still me being involved. I'm still pushing PO. I'm still getting the CCs, but it would take two seconds to switch back to somebody else.
00:03:31
Speaker
It's a confidence level thing. Once you're happy with the system, you can simplify it further, automate it further. So are you inventorying everything like quarter 20 by half screw socket head cap screw from McMaster has a Lex number?
00:03:48
Speaker
If it's, we do use quarter inch socket cap screws for some of our products and those are stored on inventory bins and our, do you have a Lex ID? We also have quarter inch fasteners and a list drawer for general shop use. Those aren't. No, my point is for McMaster orders, it's got an inventory of everything you typically will order from McMaster.
00:04:11
Speaker
Yeah, it's interesting.
Unique Ordering Challenges with McMaster
00:04:12
Speaker
McMaster's the toughest vendor for us. They're the best vendor in the world period, but they're the toughest for us to deal with with Lex because we still order the lion's share of McMaster is one-off stuff. That will continue to be via the website. Really, the only issue is we would potentially do a double order on the same day. One would be via PO and one would be via the website.
00:04:35
Speaker
What's funny, this triggers a memory of I was up in Columbus, Ohio doing some hands-on training for a company that makes like museum exhibits, really cool company. I was there when UPS came and the UPS guy showed up. This is a big company, like 60,000 square foot facility, lots of people. The UPS guy showed up with like seven different McMaster boxes.
00:04:56
Speaker
And I was like, why don't you guys consolidate? And at the time, I was very much a bootstrap entrepreneur. That was the thought of $50 for shipping instead of seven was just incompatible in my brain. And the guy was like, oh, it would be so much more work for us to combine orders on the ordering side. And then when they came to try to figure out how to divvy them up. So everyone just has their own McMaster account. And Seth's package goes to Seth when it shows up. And Sean's goes to Sean.
00:05:27
Speaker
And what we do for other vendors, like for our aluminum vendor, that all needs to be through Lex. So when we need an oddball piece of aluminum, we have a Lex ID for a custom size and we just update it on the fly.
Tracking One-Off Inventory Items
00:05:42
Speaker
and add that in. That way, all that stuff still goes through Lex, but that would be way too much work for when I want a random fitting for McMaster or something. Yeah, so it's that balance between regular purchasing and the one-off stuff. If it's worth putting one off in the inventory system, and I'm looking at it as if I'm going to use more of it, if it's going to be a thing I want to track, then yes. If it's just literally a one-time thing, then who cares?
00:06:09
Speaker
The fact that I bought it once means it's over. I have it. Right. Very cool.
Purchase Order Mix-up Stories
00:06:16
Speaker
Yeah, I pushed my first PO out of GURP a week ago or so to Lakeshore Carbide. Nice. Huge order. And I emailed Carl and I was like, here's my PO. Does this work? Can I send this to you? Because normally we've had a website system before. And he's like, yeah, that'll work fine.
00:06:34
Speaker
And then, so we got the end mills a couple days ago and I was looking through it and we were totally out of this one end mill last night. So I went to grab it, I pull it out of the package and it's a regular length flute, not a stub length flute like I was expecting. So I'm like, oh darn, because I bought 33 of them and we had zero.
00:06:53
Speaker
So, I'm looking at their part number, and I'm looking at my part number in GURP, and there's like a one-digit discrepancy, like there's no L or something, L versus S for long versus short or whatever. And my system didn't have a complete part number.
00:07:09
Speaker
Oh no. So that makes me now question every Lakeshore in the inventory, and I'm just going to have to dig through and compare, basically, make sure that they are what they are. The good thing is I can definitely make use of them, and I'll place another order today. But it's like, oh man.
00:07:26
Speaker
I used to, the old ways to do this was a little better.
ProvenCut Data and User Base Discussion
00:07:31
Speaker
Growing things. Growing things, like not complaining, but yeah, it's like you throw an automated system and it makes a little mistake and you're like, oops, okay, okay, let's fix it.
00:07:42
Speaker
I almost had a full on panic attack maybe three weeks ago. Provencut member emailed us and one of the, I don't know if it was one of the summary displays, it wasn't necessarily the main feeds and speeds info. It happened to be a small end mill and the feed per tooth was say three tenths of an inch or six tenths of an inch. Well, Provencut was rounding it.
00:08:07
Speaker
to the nearest thousandths. As soon as he pointed it out to me, I said, hey, we're on it. We'll get this fixed. But my panic attack was figuring out, was it rounding it in the front end, which is easy to fix, or the back end meaning
00:08:24
Speaker
Oh, we have like 600 recipes. Is there the potential that anything that had, not even just a sub 1 thou feed per tooth, but anything that had a tenths value, like 1.3 thou per tooth, did that data get lost? Because that's going to be a lot of, actually, no, it wouldn't be a lot of work. We would build a script for Fusion that would scrape out the data and build a comparison chart and go through that way. But luckily, it was just rounding on the front end, so it was fixed in five minutes. Like on the display side. Yeah.
00:08:55
Speaker
But yeah, I get that. Total panic attack. Yeah. Especially in that scenario, because that doesn't just affect you. That affects all of your customers. You know, everybody using Proving Cut. Yeah. Hypothetical. Got it fixed. Awesome. Doing things, right? Speaking of that, Vince Fab has been in Zanesville now for about three weeks working. He'll probably be remote long term, but he's here for a while now, at least. And dude.
00:09:24
Speaker
It is awesome. He has a finish, an aluminum off of the Shape-O-Co that is almost, I mean, it's almost too good to be true. And as soon as I saw it, we started talking about it. And the one thing I said was, I want to make sure XYZ, Joe Smith in Denver buys a Shape-O-Co, can he replicate this? And Vince just looked at me and goes, yes. And I was like, oh, that is awesome. It is so cool.
00:09:53
Speaker
So he has a shape-oko he's doing recipes for right now, the little excess from Tormach, the new Bantam machine, and the pocket NC. And he's loving it. And it's cool. It's awesome. That's so much fun. Be curious long term to see what your customer base for Provencut ends up being, like hobbyists or industrial. I don't know if you even have that data.
00:10:19
Speaker
We have some of it. Yes, we actually have a bit of it. It's a classic mistake that folks make like me, which is I want it to be for everybody, all machinists. What's your real
Marketing Focus for ProvenCut
00:10:35
Speaker
market? I think we have a huge amount of value to offer to even seasoned machinists who all of a sudden need to do something like
00:10:43
Speaker
hard mill or tap stainless or whatever, but there's no question that we've seen a lot more enthusiasm and maybe that's just the nature of the customer, but the hobbyist users, they're going from nowhere to somewhere.
00:10:59
Speaker
They probably find the most value from it. Whereas I've used it before several times, but I'm not using it for every toolpath like a newbie would. But when I'm trying to drill hard stainless or something, it's something I don't do very often. And there's a recipe for that or super long drill or all kinds of things. It's just stuff I don't do, stuff I'm not comfortable with that you guys have already tested. Oh my gosh, that's the reason for it.
00:11:27
Speaker
Yeah, we are succeeding and we will do fine with that, but what we need to do, what will really change the trajectory of Provencut and what we want this to be is when you, as an example, when you bring a new guy on, the first thing you say is like, hey, if you're struggling when you make those soft shots, go just Provencut, like boom. Absolutely. I look, you're my friend and we talk to this a bunch, but like,
00:11:53
Speaker
We're not there yet. Like we have not done enough of a job and it's funny because I'm just not a We were talking on whatsapp about people that succeed on Kickstarter and it's like some of it is just you got to push push push you got to shove this up in people's face and you got to promote it and I like succeeding for technical reasons like the oh man, the data sets impressive what they've done is impressive, but like I need to do a little bit more of the Push marketing. Yeah, I get that
00:12:22
Speaker
You guys are doing a great job with like the Insta stuff. I assume you get a good response on just so many stories and feed things about on your side, but also the customer side. Yeah, absolutely.
Delegating Tasks for Efficiency
00:12:32
Speaker
Customers are extraordinarily happy. And now we have me doing a little bit of it and Fraser doing most of the marketing. He's got the Grimsmo official account and is handling the website and a lot of the emails as well. And it's like, oh, that's such a weight been lifted off my shoulder the past few months.
00:12:48
Speaker
It's getting handled and much better and deeper than I was able to do, just time-wise. And I mean, he's like slammed now with busy work. But yeah, especially when something big is going on, like we're doing a drop on the website or something, he's like, DMs just don't stop coming in. Just on my phone nonstop. That's the point I was trying, and I hope succeeding at making last week. It's that once you see,
00:13:16
Speaker
what a person like Frasier can do or what Vince is doing with the shape Oco or the, I never, I would have liked to think I could have done it or maybe I was doing some of it, but no, that is not to the level that's possible, right? Yeah. But you don't go into it with that much necessarily conviction. For sure. So that's, I think what's interesting about the decision is when you don't necessarily understand or appreciate what the outcome could be.
00:13:41
Speaker
Yeah, you might have theories, but you don't feel it yet. And not till it's done. We're very tactile people. You want to make the thing and hold it in your hand and be like, yeah, OK, now I know I can do it. But it's the same with hiring and distributing work and things like that. They can be all theories, and you think it'll work. But then when it does work, you're actually blown away. And it's like, I should have done that six months ago.
00:14:04
Speaker
on that note, I am contemplating some sort of business things, decisions and so forth. And what hit me yesterday when I had a few minutes to just sit down and think was there's a difference when you're making a decision, there's a difference in
00:14:23
Speaker
Let's say there's anxiety or stress or just concern or risk. There's a difference in what that all is relative to the change that represents versus what the new kind of stabilized look would be.
00:14:36
Speaker
So let's say you're thinking about adding a product or stopping a product or hiring somebody or letting somebody go, it's so easy to focus on, oh man, there's so many things that could go bad with that hiring the wrong person, firing somebody, it's not fun.
Decision-Making Stress and Business Clarity
00:14:51
Speaker
Or let's say making something crazy up. Let's say you're going to stop selling
00:14:55
Speaker
the Norseman because you need to focus on flashlight or something crazy like that. Well, there's a lot of concern around the change that represents, but if you just fast forward and said, no, this company could be this if we were selling, and let's say flashlights, it's weird. I guess that's a bad example because flashlights aren't who I think of as Grimsmode knives, but if it ends up that that
00:15:18
Speaker
what that new kind of pro forma, what that new run rate looks like is good, then that makes you realize, okay, so there's some change stress, but set that aside, because if it gets you to where you want to be, hey, rock and roll. Yeah, exactly. And I've actually been thinking a lot about your flashlight example, because I mean, I have been wanting to make a flashlight for like four years now. That's one of the excuses I used to buy the Nakamura, was like, oh, I can make flashlights on it too.
00:15:45
Speaker
four years ago. And yeah, you said last week, like imagine if you were making 40 flashlights a day, just pull a number out of the air. And yeah, I think about that as a
00:15:56
Speaker
What would that look like to our bottom line? What would it take to get from here to there? Yes, there'll be stress. Yes, there'll be change. Yes, there'll be relocation of time and resources and money and things like that. But what is the outcome six months down the road? What's the smooth running look like? And is that worth the pursuit between now and then to do that? And I've got several examples of ideas we want to pursue. And it's a good way to analyze it. What's it actually going to look like once it's working?
00:16:22
Speaker
And what are the steps between now and then to make it work? And if it's worth pursuing, then go. Go, John, go. Yeah, I know. I'm in slow motion for a lot of things. Yeah. Like take my time. Yeah. No, no, no, I know. But I'm also like running around doing all kinds of great stuff. So. Right, right.
00:16:43
Speaker
It's good though. I, you know, 2020 has been such a strange year and I am a positive person and that I love a lot of what I do, but I'm not always a, I'm a realist. So I will be happy to call out things that are sour or bad or whatever. I don't know what that puts me middle of the road, but you know, 2019 was,
00:17:07
Speaker
was frankly a horrible year for some personal reasons, including Yvonne getting lung cancer in 2020.
Business Adaptations in 2020
00:17:15
Speaker
I mean, yeah, COVID is a curve ball beyond a curve ball, but 2020 has been an awesome year. Yeah, it's been my best year of my life. Yes, yes.
00:17:24
Speaker
And I miss seeing people. It would have been awesome to be at a trade show and some other things that I miss traveling. But darn it, we have completely changed our business. We've got new machines and I've learned so much about 5axis, about Laze, about the Datron. I just did a blog post.
00:17:44
Speaker
talking about, and this hit me like a ton of bricks, I bought my 1100 Series 2 almost 10 years ago this week. What? That was a huge moment in my life, moving out of Manhattan to the suburbs to get that tormach to help with strike market prototyping.
00:18:05
Speaker
Boy, what made me think of it is the, I think it's a Tom Lipton kind of quote, or something he had said that resonated with me, which is that the trades have been good to me. Yeah. I'm grateful for that. Man.
Early Experiences with Machine Tools
00:18:21
Speaker
Yeah, I got my 1100 probably eight years ago, eight and a half, 2012. Okay.
00:18:28
Speaker
But I remember when I got it, we didn't know each other yet, but I was watching your videos. I was like, honey, there's this guy in New York. It's got the same machine I'm looking at.
00:18:39
Speaker
Yeah, but I had a tag and you had a modded X. Modded X, yeah, the mini mill. Right, which I think was a bit more capable than the tag. Possibly, yeah. The tag was when you would crash the tag or bump the tag. Literally, the Z column was just a piece of tubing with a single bolt at the bottom, so it would just bend over
00:19:01
Speaker
Which is like made it super easy to tram because you just swept the whole column left to right Kind of like how how how lays crash you just realign the turret ends up. There you go Make it fast and easy
00:19:17
Speaker
No, so that was, I had, I cashed in my half a day of lay the training. Side note, anyone buying a machine tool, go reread the article on NYC CNC and on advice on buying and negotiating for a machine tool because I fell victim to my own advice, which was when I bought the lathe, I focus on, it was very intimidating with all the tool holders and options and spindles and sub spindle and details.
00:19:44
Speaker
I should have because I'm 100% sure that I would have gotten more training had I asked for it for free. And all I got was a half day. Now, oddly, I'd never even used it. We just figured it out on our own. But I think you have to use it within one year and it's coming up on a year. And so I cashed that in. And luckily, we got
00:20:03
Speaker
I got to know what the hfo is probably one of the best apps guys out there period he is you would love him he's just sharp as a whip and do we spent the whole morning deep diving on. Dual spindle complicated parts dialing in you know i'm tweaking i got my y axis just tweet ever so slightly rewriting the master over offset so that those are that's the default.
00:20:28
Speaker
We redid custom probing templates so that it correctly probes the spindle rotation for Capito, which turned reverse relative to the Eppinger, which turned forward. Haas has an M109 command, which is a super easy way of doing a user input. So you can say, on the lathe now, it pulls up a picture of the setup of a complicated part.
00:20:48
Speaker
And we already had that done. That's easy with an M 130. Then what we're going to implement is an M 109. Yes. No, which will make the user acknowledge. Hey, does the setup match the picture? That way you can't just say, Oh, I saw a picture, but I just hit.
00:21:02
Speaker
go. What you can also do is say like a return value, like one, two, three. And if it's one, it runs this program, if it's two, it runs that program. And this idea of something that we never even thought about, which is like, you know, interactive interaction between the machine is pretty cool. Yeah. So
00:21:24
Speaker
There's value in getting upfront training, because especially if you're not super comfortable with the machine, take your training days on day one. You took the opposite approach. You took training almost a year in. So you're quite comfortable with the machine already. I mean, it's good because you could deep dive on super advanced stuff right away. But I assume you learn things that you didn't know about, period, in that training.
00:21:49
Speaker
Yes, more the latter. I'm very glad I, you know, there were some awkward moments in the beginning, but look, to be honest, that's one reason why we were joking with Trevor, because I was like, I never intended to become like a Haas
Insights from Haas Training
00:22:02
Speaker
shop. That was not like a goal or intentional, but we
00:22:06
Speaker
We like them we know them we keep making money with them and one of the things is also the user base and so we kind of struggle through in the beginning and figured stuff out on our own and now i'm glad i had that training to tackle.
00:22:21
Speaker
I'd say complicated or advanced stuff, but stuff that I really care about now about, um, you know, I'm going to do a video on how we're producing. Parts are diamond pins and we're holding a few 10ths across our batch cycle. And I say that as somebody again, who loves trolling false claims, that machine tool, but a lot of that has to do with the thermal stability and what we've done and sharing that. Um, and.
00:22:46
Speaker
I want to figure out a way in the video of like time lapsing it to show that we're as we're pulling them off They're coming off that way because I want to yeah, I want to prove it kind of like I want to show folks this is working Absolutely. So right now you have one Real lathe right in the shop like once did nice CNC lathe Tormax
00:23:06
Speaker
We had a tormac lathe, right? We sold it. Sold it? Okay. Yeah, sold this lathe. Normally, my real question is, do you see yourself getting more lathes over time, or are you good with the one for now, for a while? We're not at... We make two parts in production, the mod vice washer and the diamond pins, and those are, I'd be 20%. I mean, not even of the capacity of that machine, so no need right now. The one who wants awesome. Yeah. Sweet.
00:23:30
Speaker
We have a, we have a, I don't even know if I'm allowed to say this, but we have a new small lathe from a brand that I just mentioned. Small lathe. Yeah. And it's kind of sort of not a secret, but.
00:23:43
Speaker
I actually am legit excited for that because there's some things like Ed or Jared wanted a quick threaded adapter thing. And I don't want to run that on the ST20 because it's just too, it's not so much risk. It's just that I just don't want to crash that machine for no reason. Like if you're making a $2, $2 part, don't risk a hundred. Don't do that, you know?
00:24:05
Speaker
Yeah. You're turning a production machine into a prototyping machine, which involves different setups and collard liners and things like that. I noticed that too. We've got the Nakamura set up and when I want to play with something and just make a bolt or something, I'm like, I'd rather just pay somebody to do it because I don't want to. That's a good answer too. But we've got two high school interns, one of whom is in a machine program. I want to have him go say, hey, on the
00:24:30
Speaker
smaller lathe, go program or hand write a code. He'd hand write because they do it in school a lot and go thread this adapter for this. It's fine. No one's going to check him. He's just going to go do it on his own. That's not happening on the ST20Y. No, exactly. Yeah. The other thing that we did, this was actually Jared's idea, was we do a blow off
00:24:52
Speaker
Now we've been using a lot of air for finishing parts. So not only is every edge chamfered, but we're blowing parts off in the machine with just through spindle air, no need for the chip fans. And Jared was like, Hey, I keep posting this program, but we have all these different fixture plate files. Is there a way to not have to do that?
00:25:11
Speaker
What we did is we just wrote one program, threw it on the Haas control, and then Haas has an alias function. So like in the 9,000 programs, you just assign an M code to a program. So now M156 calls program 9,002. Program 9,002 is this blow-off program. And that means all we have to do is do a quick manual pass-through of M156, which we could do
00:25:35
Speaker
infusion manually with a pastor or in the post. And then what I'm thinking about doing is going back to that M109 and having three different blow off programs, which would sort of represent the three different sizes of fixture plates. Because there's no reason to blow off a big plate when it's a little plate.
00:25:53
Speaker
Yep. So nice. Yeah. Yeah, we've got we've got the Lang fan in our Maury. And at the end of every Norseman cycle, the fan comes down and it sweeps over one palette moves over sweeps over to the other moves over sweeps over to the Pearson palette. And then it kind of traverses back and
00:26:11
Speaker
Our guy Steven just asked me the other day, he's like, can you do a second pass on the first pallet again? Because it's still like, it gets blown on basically from the second one. And he comes in in the morning and unloads the machine after it's been sitting all night and dried and sticky and gooey and stuff. And he's like, a little bit more cleaner would be fine. I'm like, yeah, I could do that, no problem. Don't be afraid to ask for stuff like that because that's
00:26:33
Speaker
That's easy. I don't know. I'm not the one doing it. It's like Jocko says, you're the boots on the ground. You're the guy doing it and seeing it. I can help with that. Easy. Yeah. Two blowoffs is the way to go, we found as well. Yep. The first one just blows off a lot, but a good portion gets moved back. Yeah. Yeah. Speaking of waves. Good. We're not happy with our coolant right now. Ooh.
00:27:03
Speaker
Yeah. Okay. My interruption was going to be a derivative of that. You go first. I've been a fan of Quality Chem. I haven't tried too many other brands, but we've used Quality Chem 251 for the past five years. It's been great.
00:27:18
Speaker
Since the move to the new shop and the Kern and everything else, we switched to a quality Chem 450 Eco Pure, which is supposed to be like the premium.
Coolant Dissatisfaction and Solutions
00:27:26
Speaker
It's more expensive. It's like awesome for titanium and aerospace. And Lockheed did their test of a billion different coolants, and they said this one's the best. And that's all the marketing stuff that they sold us on. And it may be amazing, but it stinks, and it's sticky, and it's yellow, and it's gross. I don't want to like it. Switch.
00:27:47
Speaker
Yeah, so we're going to switch back to 251 because we were happy with that. Yeah. Yeah. We did the very similar story. I mean, I think the very honest answer is other people that we looked up to, largely through social media, were using QualiChem. So I tried it years ago. And it was great. Don't fix it. And then I switched and tried Blosser. And I think Blosser makes some really good coolant. We tried the Synergy. We did not like it.
00:28:13
Speaker
for a variety of reasons, some of which are the stinks and smells and sticky. I think there's some technical issues with that synergy with how it reacts to zinc and the water. Well, actually, the real criticism, in my opinion, is synthetic coolants have incredibly stringent water requirements to the point. It's just not worth it, period. Anyway, I don't
00:28:35
Speaker
There's a different, I balanced sort of pointing out technical shortcomings versus throwing shade. And anyways, I was like, let's just switch back to quality cam. Here's our issue. We had our anodizing process down pretty darn well. And what we're now seeing is
Anodizing Process Challenges
00:28:55
Speaker
Sorry, third-party anodizing. I just think in house, I'm like, you guys are doing anodizing? Yeah, we have a whole separate building, John, for anodizing the plate line. I don't want to use the wrong terminology, but they say etching, but it's clearly more than just a stain before
00:29:19
Speaker
they're anodized. Now, I don't know where that falls or whether they're doing like an etch or a clean or a rinse or whatever. What's odd is that, quote unquote, nothing has changed on our end. Now, what has changed is we have two new machines, but those machines are being run with the same 250C that we've been using for a while. Now, nevertheless, as an example, learned that experience that the Blaser taught me that different
00:29:46
Speaker
types of machines have different exposed metals, like the blazer reacted with the zinc or the copper. And so I'm thinking, well, hey, Haas does change the sheet metal material, the casting, the paint. So maybe there's something there. But the problem to solve from a much more practical solution standpoint is we need to get our plates really clean and dry before we freight pack ship them to the anodizer.
00:30:10
Speaker
And we can easily dry them off and blow them off, but the problem is we're getting coolant that just gets trapped in the threads. And there's 476 threads on a Tormach 1100 plate. So the two thoughts that we have right now are number one, using an air knife to blow them off. And the second thought is buying a pizza oven.
00:30:36
Speaker
to bake the coolant off? Like one of the conveyor. They're not expensive. I don't like this idea, but you either need to mechanically remove it, like forced air, like blowing with force, or a combination of that plus if you moved it through a conveyor oven at some relatively low temperature, 175 degrees, it would almost certainly massively aid in. Well, it's going to evaporate the water, but the oil is not going to leave, I think. Yeah, that's a good question.
00:31:05
Speaker
Yeah, I would worry about baking the oil more into the metal somehow. And I think you need to like, I would say get a big ultrasonic and like clean them that way. But they're big plates. Yeah, I don't think that's going to happen. What's your biggest bloomer plate? 24 by 56. That's huge for an ultrasonic.
00:31:28
Speaker
I mean, set up like a... We have a rinse station. But a boiling water soap, Dawn dish soap, rinse station, that would make it clean. And just like a basket that goes up and down, you can agitate it through the boiling hot soap water. It'd be clean. So we've done that sort of not with boiling water, but with a
00:31:50
Speaker
pure clean water and you still end up with a plate that has liquid on it. And over time that liquid will become further contaminated with residual from the prior. And so my thought is it's not so much, I'm not necessarily trying to solve the problem of what liquid is left. I'm trying to get rid of any liquid period, whether it's coolant or water or Dawn soap, detergent, whatever.
00:32:15
Speaker
Which is why I'm leaning toward an air knife, which would give us a solid blade to blow versus the gun nozzle yesterday. We'll see how it should come next week.
00:32:27
Speaker
Yeah, I know Sylvain makes them like the air guns that I have, the quiet ones. Yeah. They make air nozzles, air knives, and a bunch of companies do as well. But yeah, I did a bunch of research a while ago on cleaning and drying parts automatically, like let's say our little screws coming off
Enhancing Cleaning with Crest Ultrasonics
00:32:45
Speaker
the Swiss. They're full of oil. Oil gets trapped in the torques or in threads. And we have, we just got, we had the Amazon like $400 ultrasonics and they weren't okay.
00:32:57
Speaker
We thought they were great, and then we just bought like $1,800 Crest Ultrasonics, and they are sick. The guys are so, so happy with those. Everything I've read is like Crest is just the brand. Maybe they're not the best, but they're certainly big and strong.
00:33:16
Speaker
And they range from, I don't know, probably low 1,000 to 2,000 or 3,000, depending on how big they are. Now, one for the size of your plate. I'm sure they could make it, but it'd be like $6,000. Probably I'd make the numbers up. But I mean, it might be worth it. It might be worth it.
00:33:34
Speaker
Because ultrasonic, it'll vibrate the oils out of every single thread. And as long as you rinse it and agitate it properly, Angelo's done so much research on how to ultrasonic properly. And he found this whole PhD article on agitation for ultrasonics. Oh, yeah? I want to build a, what did he call it?
00:33:56
Speaker
Like a merry-go-round, you know, a thing that just goes, that's the wrong way. Oh, like a carousel. It goes up and down in a circle, like a carousel wheel kind of thing. And the parts go in and out and in and out. Yeah, yeah, yeah. Apparently the, even just up and down motion, slow, it doesn't have to be hard, in and out of the water, it helps hugely. And like there's scientific studies like proving that it's good.
00:34:20
Speaker
And we've been trying different cleaning methods and Simple Green or ZEP as a pre-clean for some of the parts, like lapping machine stuff. And Simple Green's winning. And it's safer. Apparently, ZEP is pretty gross as far as toxicity. We're starting to read MSDS sheets and everything. And it's like, OK, let's not use that one, especially when you heat it up. It's bad.
00:34:42
Speaker
But yeah, so we're coming to the conclusion that even just rinsing knife parts off the mill in the sink, hot, hot tap water and don dish soap works really well. And then if you want a spot free shine, rinse it with like super clean water distilled or deionized or RO should be clean enough. And the funny thing about the RO system that we installed about a month ago,
00:35:10
Speaker
We filled the IBC tank full of RO. It's all good. Everybody's happy. Ultrasonic guys are like, yeah, it's cleaner. We're running it in all kinds of machines and everything. And then the quality chem guy comes in and we measure the TDS of the tank.
RO Water System Setup Mistake
00:35:25
Speaker
And it's not RO water. What do you mean? We didn't put the membrane in the RO system when we installed it. Shut up.
00:35:34
Speaker
Well, hold on. The obvious reason is that if you have the RO working properly, it's a drip feed. It'll take three, four, five days to fill up an IBC tow. So if it fills up quickly, you know it's going fast. It still takes two days to fill up. Really?
00:35:50
Speaker
One of the guys thought that the membrane filter was an extra that I had bought a spare, and he never thought to check the windows. Because there were charcoal filters in the canisters. Silly mistake, it's fine. There's a- Go ahead. We fixed it. There's a, we have that like water page on NYC CNC, but if you just go there or Amazon, inline PPM or inline TDS meter, it's like $7.
00:36:18
Speaker
And then that just gives you a live feed of what your output is. Because that's all I care about. If there's a problem, I'll go figure it out. I just want to know if I'm putting out bad water. Yeah. I bought that with the system, but they didn't install it right away. Oh my God. It's installed now. But now we're getting about 130 TDS in and like two out. Yeah. That's amazing. And you check that weekly or whatever and we'll keep an eye on it.
00:36:43
Speaker
Yeah, that's totally fine. In fact, honestly, we got so focused on getting to zero when we were trying to play with the synthetic coolants that I'm kind of remembering now that going back to a semi-synthetic, you don't need or even want zero. Our water is 230 out of the tap and our RO membrane, this morning I checked it and it was a month old and it's back to like 30. When we started out, it's like 15 and that's totally fine.
00:37:09
Speaker
Yeah, the cleaner you go, once you go to zero, I don't know. You'll pull another contaminant. Yeah, it's hungry. It's pulling metal from everything, from every fitting, from every- I believe the word is leech. Yeah, I guess, yeah. Yeah, I had a two-hour conversation with one of the water system guys on the phone, and I learned a lot. It was great, but it's like, wow.
00:37:33
Speaker
I've seen YouTube videos like, what happens if you drink deionized water? It's really bad for you. We'll go look at the guy who puts a functioning computer in a true DI water with zero TDS. There's nothing conductive in it. So water is not conductive. H2O is not conductive. Super cool. Awesome. Yeah. Yeah.
00:37:55
Speaker
Good. What's up today? How, dude, how's, uh, can we, can we talk about the grinding wheel or it's too soon? I didn't explode one. Um, super amazing results. Yeah. I guess last week, uh, I had test cut, but I hadn't actually done anything. So. And the last week I was able to do some amazing grinding. Um, so it's like a, like a disc.
00:38:17
Speaker
No, you would cut a blade, because I gave you a hard time about not tramping it or dressing it, but you would cut a blade. Yeah, and I figured out how to step it down properly at the end. The wheel is 100 thou thick, and I'm stepping down. And this whole time, I thought I could get away with like a 90 thou step down, 90%, and have it kind of overlap, like a step over kind of thing. But you can absolutely see every single
00:38:46
Speaker
step down every streak. And I'm like, it's not acceptable. So Angela was like, just try a 1,000 step down. And who cares if it takes a long time? You've got to go pick up your kids anyway. Just do 1,000, and then leave, and then come back. And oh my goodness, it was ridiculous. So freaking nice. Like, perfect. Like, I could actually scale it back from here to a certain level and speed cycle time. It was like 36 minutes for that finish pass. But I'm like, don't care. You're not as crazy as I thought. Yeah, right.
00:39:15
Speaker
Go to 2018 minutes. Exactly. Done all day long. So happy with this. And the crispness of the plunge lines and just everything, it's so good. And then Eric threw it in the tumbler for his typical quick cycle, two hours in the media and then like 20 minutes in the finer media. It's perfect. John, well done.
00:39:37
Speaker
There's one section at the very bottom of the blade or at the top by the spine, but when it's fixed, let's call it the bottom. There's a clamp there holding the blade still, and I can't go too close to the clamp. Otherwise, obviously, I'm going to crash the clamp. I'm like five thou away from it as it is.
00:39:55
Speaker
But the wheel is still on the blade. And I need to get away to feed the wheel off of the blade completely. Take my 1,000 steps all the way off the blade. Because now there is a band at the bottom with like one stripe left. And that looks kind of gross compared to the rest of it. So other than that issue. Trim tool path.
00:40:15
Speaker
A trim toolpath. Fusion has a quote unquote new feature where you can create a toolpath and then go and manually create a sketch and just trims away the section that you sketch over. Oh, I have heard of such things, but I haven't done it. Yeah, it's amazing. I mean, it's not amazing. It has some, I think some bugs or quirks about how it handles the linking move. Like what you trimmed away, how does it lead in and out of that? But then also, but you have a current, so I don't care.
00:40:39
Speaker
And it is not parametric or it wasn't parametric. So like you had to create a sketch, but then that sketch was gone. And obviously we want that to be, I think that was just a, it may have already fixed that to where you have an actual parametrically driven sketch, which you could literally do as a project from the clamp with a 2000 offset and then use that as the trim tool path. Okay, there you go.
00:41:03
Speaker
Yeah, what I really need is more clamp clearance. So today I'm making another clamp that's got more room for the grinding wheel to go over. So I'll be making that today and then make another blade today. Had a half hour conversation with the company that makes the grinding wheels yesterday. And I was like, so here's where I started. Here's my speeds and feeds and recipe and everything. How close am I? And he's like, yeah, you're actually pretty close. We'd say 5,000 to 6,000 SFM for a grinding wheel of that size. 5,000 SFM.
00:41:33
Speaker
Exactly. Which is, on my machine, it'd be about 20,000, what do you say, 19,000 to 23,000 RPM. And I was running at a 20. Perfect.
00:41:43
Speaker
And I'm running it at 100 inches per minute. And he's like, yeah, you could probably go faster. If you go too fast, the wheel will start to hop. And you'll see a bad finish. So push it. You've got a kern. Go as fast as you want. And then I'll assess the finish and the time and the cycle time and balance all of the step overs and everything to find a recipe that's just super bang on consistent. And then I'm going to have the dressing stick on one of the small pallets.
00:42:12
Speaker
and write a routine so it automatically comes in. It grinds five blades and then it comes in and dresses. I'm going to probe the dressing stick so that I know how much meat to go. Then I'm going to laser probe the diameter of the grinding wheel afterwards, do cutter comp for all of the toolpaths. This is my plan. There's a lot of pieces in motion, but it's going to be so sick.
00:42:38
Speaker
This is, I mean, maybe it's not the first Aroa, but I want to say it's groundbreaking.
Automating Grinding Precision
00:42:43
Speaker
That's freaking awesome. I have Aroa Compact 80 and I palette change in my dresser to run a routine to touch up the wheel. John, 2020 is a great year. I know. I was having trouble measuring the diameter of the grinding wheel because it's just the design of it. I have to measure
00:43:05
Speaker
quarter inch up the shank. That's where the wheel is. There's a set screw at the bottom, basically. Sure. Catching off on the cap screw, which I don't love because it's not super precision lengthwise. But I have to, yeah, I'm gonna have to make a custom solution. But so I have to probe the diameter like quarter inch up. And there's a safety built into our probing cycle on the current that only lets me go five millimeters, which is 0.19 inches.
00:43:31
Speaker
So it would just alarm out every time I tried to probe diameter. So I reached out to my guy at Bloom, and I'm like, what's wrong? And he's like, OK, send me this parameter file. And then he sent it back to me, and he's like, change this one number from 5 mil to 10 mil, and you're golden. And now it works great. Oh, that's awesome.
00:43:46
Speaker
Then the other thing I'm going to do is after I finish rough milling and or grinding, I'm going to come in with the probe and measure the surface, use a surface inspect, and compare theoretical surface with what it actually did and then offset tool wear based on that. Not only will the grinds look amazing, but they'll be dimensionally currenty. I love it. Great. Well done.
00:44:10
Speaker
The process that you went through there is when I always continue to remind myself, this is like t-shirt worthy, which is find the solution that works first, then you can improve it. When you're dialing in a tool, instead of being like, I know I need it to go faster. Well, I don't care what you want or need, you got to start with what works, and then we can figure out where to go from here.
00:44:31
Speaker
within the limits of don't cut too many corners. Get it done, but don't slap it together kind of thing, because I find myself rushing sometimes too, grinding without dressing the wheel kind of thing. Make sure you're doing it well, but you don't have to have the whole system in place before you hit go, which is something I struggle with, because I want the whole system in place. I'm like, I've got this theory. I need it all to work. But I try to pick it away piece at a time and then put it all together at the end.
00:44:59
Speaker
We had a hiccup yesterday, which is we started selling shapeboco plates and the larger two sizes are relatively large and even thinner plates than normal. Then a shapeboco doesn't have any frame to bolt it down to like a normal plate where you're sucking the plate down. We need to do a better job on a more difficult aspect ratio of keeping these plates flat in an unconstrained state.
00:45:25
Speaker
That was a lot to say there. And so we have a problem. And so there's a lot of different ways you can tackle this. But what we realize for now is we're going to solve it the right way, which is we have a different way. We're going to hold them. And then I just went ahead and bought more stock yesterday that's thicker. That gives us potentially better flatness from the mill, from the vendor, I should say. And then we have the chance to deck more material off. Now, I'm sitting on a lot of material that I want to make work, but it's like,
00:45:53
Speaker
put that aside. Like we need to figure this problem out today the right way, and then we can improve it later as we, as we learn and get stuff out there and all that. Yep. Move. Not always, not always easy to, it's very easy to, to judge yourself in hindsight. It's, it's difficult in the moment. In the moment or even future planning, like, like what is the next step? I want to do everything. I want to do all of these things, but what is I'm trying to learn how to do,
00:46:23
Speaker
bunch of terms for it, but small wins, you know, so that you move forward, you make progress, you gain confidence. Minimum viable product is another one that you talk about in software. Like, like, what is the minimum amount of work for me to achieve the next goal, the next best goal or
00:46:39
Speaker
for me to sell a rask. What is the minimum amount of things that I have to do for me to be happy to sell a rask? I don't need to have 20 pallets machined tombstones ready for massive production. I just need to have my test fixture like making parts, you know? And as much as I want any more fixtures, and I will, but I don't need it right now. That has nothing to do with me selling my first rask.
00:47:03
Speaker
So, leave it alone, drop it and move on. So, I've got my list, I've got my to-do list, I've got it sorted into different categories and I'm just picking them off every day. Good. Good. Well, Phil's doing laps. Sounds good. Sorry, Phil. Hey, good week to you. I'll see you next week. Sounds good, man. All right. Have a great day. See you. Bye. Bye.