Hello and good day to you. From episode 22. of our podcast series Project Breakaway. A metaphorical and literal time in the day when we here at Predator cycling take some time away from working in the back shop to come and share with our listeners what we're doing, how we're doing it. What it takes to do it, our ideas, our innovative success stories and even our missteps and failures. If you find yourself with an interest in bicycles, composite manufacturing, out of the box design or even curiosities beyond, I encourage you to stick with us, settle in and learn a little. I'm Courtney B, co-owner and project manager of Predator cycling. I'm here with my partner Aram Goganian, the other co-owner, CEO, lead designer and engineer and night owl machinist of Predator cycling. How's it going, Aram? Uh, it's going pretty good. It's uh, is it night time, is it day time? I don't even know anymore. I know, well, before we get into why I called you that. Um, I just wanted to quickly hit on last week's episode. Yep. When we had a discussion with Mike Guyer from Nvidia about Omniverse. Yep. Um, so pat yourself on the back because I think it was one of our most popular listens. It it was, it was definitely up there. Definitely got a lot of good feedback. And I think that we'll definitely be circling back. Um, with them as they say in the corporate world. We'll be circling back. Um, uh with Nvidia to further discuss Omniverse in the future. For sure. I mean, I think we're, you know, I think we kind of touched on a bunch of different aspects of it that we're probably going to be using or I mean, we're already currently using it for some stuff, but. Right, but I think we peaked some interest maybe into further future projects or how it could be used or. Yeah, and it it got me thinking some more too. Like I have some other ideas on how we could maybe implement it in our manufacturing workflows. Yeah. So, no, it's cool. It's cool stuff. It's and it's exciting because like I when I log in into um my nucleus uh server. It I like literally every couple days there's an update and like, what did they add? What's new now? Like it's exciting. Yeah. Yeah, it's cool. It's cool. I I don't know because the computers are listening to me. I'm getting like ads on like Instagram and Facebook for Omniverse. Oh, really? Yeah. Like they're legit listening. Like I'll say something like. I'm not even looking it up. I'm not Googling it. That's for certain. Yeah. But uh. Uh. Yeah, I don't know how that works. I don't know, or maybe because you're Googling it. It shows up on my ads. Maybe it's the same IP. Maybe it's the same IP. I don't know. It's weird. Anyway. Uh, so yeah. So that was cool. So let's move on. So. Yes. I referenced you as our night owl machinist. Because you keep working through the night. I do. I I do keep working through the night. Yeah. You're diligently cutting out RF 20 aluminum molds. Yep. Yeah, we got through most of the custom orders and stuff that we had. And we're starting to make all of the full-size production molds. Mhm. So, um. And it's it's tedious work. Yeah. It's there's metal flying everywhere. Yes. It's probably my least favorite part of the process. Yes. Because it's so gross and messy. Uh yeah, well especially because we have a, you know, our what's the CNC is open that we have. So it it kind of shoots aluminum chips everywhere. Yeah. It's hard to control. Yeah. We literally go home and there's like metal like little tiny little slivers of metal everywhere. Especially when you're doing the finishing ops, it's just it's not taking off much material. So it's just like aluminum dust and. Yeah. Yeah. So, as you said, we have like a open air gantry gantry style system. So ideally, we would love to invest in maybe a more superior machine. Because you definitely have the know-how on how to operate one. And you'd probably max out every capability that it offers. Um, but we just don't, you know. Yeah, we we have to pick and choose. They're an investment. For sure. And we definitely invest in other things like technology and stuff instead of like maybe machinery. But you've had even me researching the machines, which is not usually something that I sit there and do. Yes. So, um, they seem really cool. They're costly, but they're cool. Yeah. Um, so it's definitely maybe on a future need list. It's a want list. Definitely a want. Yeah, it's need and a want list. Yeah, I would say it's both. Uh, so yeah, that might be a, you know, we got, you know, just got the mini vans, so maybe we'll get the machine. So a CNC machine next. Um, because you really like the company Datron. Yeah, I'm a big I've been a Datron fan for oh, I don't know, five years now. I've been eyeing one. Which are like the smaller like size smaller end of like. Because I've gone to some of these conferences with RM and he'll be like, they have all their machines lined up like for showing people things. Right. Some of them are freaking huge. Yeah, no, they are. And they all serve different purposes. Um, I mean, maybe like a little overview of different types of machines. So like we're the the so what Courtney referred to is is we have a a gantry style machine. So basically what that means is we have a work envelope, which is approximately a four foot square. The size of a bike frame. The size of yeah, a little bit bigger than a bike frame. Um, and basically there's a spindle and the spindle moves across the table in both X, Y and Z coordinates. So it's it's basically the the cutting tool is moving around and cutting the material. Um, and that that's a that's a gantry style router. Um, and a Datron is is the same except that um it it's much more accurate. It's much faster, it cuts smoother. It it's much higher resolution. It's much more like plug and play like like. Yeah. One button screen. Ours has been um very much modified so that we can make our molds. Um and make them accurately and make them well. Because when you see these machines at conferences, they're they're cutting tiny little pieces. Yeah, and so like the the machines that Courtney's referring to that are like the more of the the uh the large machines. They they have a spindle head that's stationary and the bed moves. Or even the arm like move, like isn't it just kind of like go around. Like this. You can't see. Yeah. But yeah. Yes, so like the the It's like all access. It's not five access. It's five access, all access. Uh, you can't have more, but usually a five access is is is a style that you can hit all the sides of a part pretty much. Yeah. Um, but the point being that usually that when you have a CNC where the the the spindle head is stationary, um, in order to get the the X and Y coordinate, the machine has to be twice as take up twice as much space because it has to travel the bed the whole length. So if you had four feet on either way, it'd have to travel eight feet in order to go left to right. Um, and so and the problem that we have is is that our molds, our largest molds are are just about 30 inches, um, in the shortest distance, in in the width. Um, so that means that in order to have a a CNC machine, you would need a very, very, very large like 60,000 pound, 70,000 pound machine in order to Well, when I look, you know, when I'm when I'm feeling low for you and I'm looking for used machines, these things are literally the size of our building. Yeah. It wouldn't fit. Well, I mean, I'd have to get rid of like half of everything else we have. Yes, they're they're massive machines because you need that that coordinate. So like things like um a Datron and like the machine that we have, which is a router style gantry system, um, work because the spindle moves. And so they're basically essentially about a five-foot cube. Instead of a uh like a seven-foot, 10-foot by 30-foot. They're enormous. Yeah, they're huge machines. They look like trash compactors. Bigger than trash compactors. They're they're very big. So yeah, we have a a a gantry style machine, so it means that we have to be pretty um uh smart about how we implement our strategies for machining aluminum molds. So let's talk about that. Yeah. Smarty, being smarty. Being smarty. Being smarty. Uh, uh, so we have the open gantry style router cutting table and um, let's talk about your ups and downs lately because there's been some. There has been. Um, so let's talk about um, basically, let's first talk about how the software that we use, the design software works with this style machine. Because you can't just like make what you want in the software and then hit a button and then this amazing five axis machine cuts every nook and cranny and corner of it. You have to actually think about like, okay, I have to cut the top and then I got to cut the side and then I got to cut this one angle. And then I like I have to physically like move the aluminum. Right, so there's a lot of different things to think about. So whenever you're designing for a mold, um, you know, the the real So there's a couple things. One, you have to design the part so that it can be molded. Depending on the molding methodology that you're using. So that's like the And that's what you always talk about the difference between like a design engineer and a machinist. Yeah. Because they are two usually two separate jobs. And the designer makes this amazing design. Okay, go make it. And the machine says like, Yeah, dude, I can't. Well, so there's you're actually missing a person in between. There's usually an engineer, like an application, there's a designer typically, and then there's the applications engineer, uh, mold expert, composite tech, whoever is that person. And then there's the machinist. Okay, and you're a three-in-one. We have Yeah, I I do all three of those. You're Um, but uh, yeah, so like it there's a benefit to that because I know how the machine works. I know how to cut it and I know how uh, I know how the mold works and how the process works. Um, we also simulate the composite work ourselves. And so I do that as well. So you're a four-in-one. I guess, sure. Yeah, 101. Oh, thank you. Um, but the idea is that you have to think about how you're going to mold the part. And how the part's going to be made. Um, so the mold line is really important. There's a lot of inserts that go inside of it for like water bottle cage, um, bosses. Um, um, the cable guide systems, bottom bracket inserts, chain stay plug systems. All those have to be thought of. Right. How they're going to be implemented into the mold. So do we we can't because we manipulate. I mean, we use the aluminum mold, but you're smart. And we also use like 3D printed inserts. Which is that something new or creative or is that just Well, it gets so. Okay, well, that's let me just back up to the mold. And then I'll touch on that. It's part of the mold. It is absolutely part of the mold. But it's there there's another issue of why that's complicated. Um, but the mold itself, one of the things that we have to think about is the strategy. And how we remove material. Um, so when you're designing for a mold, it's surface finish and tolerances that matter. So we have to take a very a very systematic approach to doing. Methodical approach. Methodical approach. Um, and what we're doing is we're basically removing very small layers at a time. And getting this makes it look kind of like a a topology map when we're when we get our first roughing pass done. And then once we get to there, we do multiple semi-finishing, finishing passes. And then we have um, all of our precision geometries, which we're very cautious to make. Perpendicular to the spindle, so we can make really nice accurate cuts. And then those are actually done very um, there's a reason why we make the parts where we make them so that we can actually index those holes and make check our tolerances. So we're actually able to hold about a thousandth of an inch tolerance across the part. Um, obviously on some of the 3D surfaces, it's harder to measure, but on the critical tolerances, we're holding a thou. Um, which is really good. Um, and then that kind of goes on to your 3D printed section. We do use 3D printed inserts, um, quite a bit. And that gets really complicated because um, you have to account for thermal expansion. Um, across it and you have to account. So composites, carbon fiber has a negative thermal expansion, so when it heats up, it contracts. Um, aluminum expands. Um, most 3D printed materials also expand. Um, and they they expand at different rates. So you have to be very cautious on how you make everything and the rates of expansion. So if you look at some of our inserts, there's some very creative shapes on the insides of them. Um and that's to to compensate for rates of thermal expansion because of. Well, I mean, separate from all your material science jargon. Uh it's to help assist the cuts that we just can't simply do on a gantry machine. Absolutely. Yes. So there's there's things that like, yes. There's cuts we can't make. Like the curves on the RF 20. Yes, there is. I'm looking at it right now. Uh the top thingy, the chain stays. No. Seat stays. Seat stays. Seat stays, yeah. The top of the seat stays. There's like a weird curve that hits the top tube. That's that's if you look at it carefully, it's also it's impossible to mold. It's like almost an impossible to mold part. It's it's. Because I'm looking at that and then I'm looking at like a normal bike next to it, which is like a tube to tube construction. Yep. And that just kind of just comes off as like a degree. Well, and you don't even notice. No one would actually even see the hardest part because the seat stay. If you look at the RF 20, it's a seat stay that goes up. And Courtney's right, it's a ridiculously difficult part to do. And it goes basically up and then it turns almost 90 degrees and goes straight into the frame. But what you won't see and you actually will never see is that there's actually the way it it it bonds into the frame is this. It's this plug. But it's not just a plug, there's actually a flat section and then there's a taper. It's a it's a it's a concave taper on it. And so that is really difficult to do. Now that I'm looking at it. Why did you make it, why is that for clearance or it's much wider than like. Yeah, so it's it's. Well, that's a. Oh, that's a track frame. You're looking, oh, that's that's an old track frame that's a tube to tube and that's narrow for other clearance issues. Um but the RF 20 is done like that because of CFD results. So because of the way the wheel spins in the back of the frame. The amount we wanted clearance so that the air could actually go between the frame and the seat stay. So that we don't have a a pressure spot right there. We could actually relieve that pressure and get that air out of there. Um and make a little cleaner approach. But anyways, the the point is is that. Um we have to and the advantage that we have is is that we have software or this disposal. We have a lot of um powerful cam software to design and and model stuff and um. Get draft analysis done so we can get really accurate um mold constructions. But also we have a mass amount of computational power. So we can make really complex machining operations and do them very efficiently. We would just like to match that complex software with some. Yeah. For sure. If any Datron is listening. If anyone from Datron is listening. Listen to me very closely. RM loves your machines. If you want to give us like a trial run. Oh my gosh. Like a year would be nice. Uh. We could do lots of things. On your machine. Yeah, they trying to be pretty fun. Um. Yeah, I mean that that's the goal. Because that's that's. Anyway, thetron's really cool because it gives us the surface finishes that we are after. It also gives us some of the tolerance. And it'd just be nice for me to see like you like not yelling and screaming when something happens. Oh. Well, let's talk about our power supply issues. Oh. Power. Guess what? The electric goes out sometimes. And we our operation, your CNC operations are sometimes hours, hours, almost half a day. Uh. 14 hours. Uh, so the last operation we ran the finishing pass on the RF 20 on one side of the mold was 19 hours. Okay, and you can only stay asleep at work so many days in a row before you have to come home. Shower. And eat. So, uh, you left you left the machine running as we always do. And then, you know, of course, it's that one day and there's like a one minute power outage at like 5:00 in the morning. And, you know, our machine doesn't have any sort of backup. No, we don't have backup power. It also doesn't remember the location where it was. Right. And so that ended up causing a really big problem. But you found a solution. Yeah, so we were able to um re re-probe the machine to the original home location before the. Anyways, it gets kind of complicated because the the origin that we had specified for the finishing pass was referencing one of the areas that hadn't been machined yet. So it was this like. Anyways, it was like a three prong problem. Of why one of them was my fault that I shouldn't have made that my home location. Um, but it was just easier to do it that way to set up some of my tool paths. Um, I didn't account for the machine turning off. Stopping in the middle. Right. I didn't I didn't think about that. thing. Which sucks too, I mean, not just for that, but like with 3D printers, that happens. I mean, it's 2021, machine should know where they left off. Yeah, so I mean, well. Yes, yes. Uh, some of our 3D printers can actually remember the location they have they have power failure systems. And others don't. Well, there's also issues with how they're printing. Plus materials and like. Yeah, it's it's it's bad when the power goes out. Um, and also like a power backup power supply on the CNC, I don't know how realistic it is to do. If it would actually work. But on the printers it does work. Um, and we need to set that up better for some of our production printers. But basically, you're out here with like your calipers for like three hours. I was here for about four hours. You should be Mr. Caliper, not Mr. Night Al machinist. Yeah, Mr. Calipers. Um, yeah, we got it, but we got it back probed all back in and we were able to finish the pass. And it looked really good. Um, so it came out nice. Um, and now we're just doing the final polishing and finishing. We're going to lay the sucker up this weekend. Ready for Monday here. I don't know. Yeah, I'm just getting the all of the mandrels finished and then um we'll be able to lay up that RF 20. And we'll have another production frame production mold ready. Is there anything else you want to talk about um CNC machining before I move on to the next thing? Um, no, I think that's. Okay, I mean, I think I don't know. So I just wanted to quickly address another issue. Um, it's about. components versus the parts and accessories that we use in the components. So basically we launched our 3D printed cleat adapters last week and we've gotten great feedback. Um and then I posted on Amazon. Yeah. And in Amazon universe, we got a poor review. Yep. Um and I know it's uh Amazon and that's what happens. Um and I know people will never be satisfied, but reading the product description also seems like a vital part of being an informed customer. So, basically in our kit, we sell a two degree or two two degree, two one degree uh adapters for cleats. And we provide screws. Now, I'm going to explain this much better than I can. Um but we decided to include a torque type of screw. Which are a little uh star-shaped head type shoes. Um rather than I guess a normal screw. Yeah, an Allen key or a flat head. Um so this customer gave a poor review, um not based on the success or failure of the product uh in use. But um because he didn't have a compatible screwdriver for the screws that are listed in the description of the product. Yeah. So I'm a little frustrated. But um it's also, you know, that's me. So, um I'll digress for a moment and let arm explain the decision that we made and why we feel that this screw is a superior. I don't know. Well, it's so I mean we had this debate before we even launched it if we were going to go with the path of a torque screw versus a Allen key. Um and if you notice on our RF 20 and a lot of our products that are coming out with, um, um we're actually switching over entirely to torque screws. across all of our products, um, uh and torque plus for the most part. Um but a torque screw basically, like Courtney said, it looks like a star. Um or it's uh referred to as a uh a oblong uh uh a six uh what is it? Six star, six star. I don't know. Yeah. When you ask me for a normal screwdriver, I still ask you if you want the. Yeah, six star. The line or the X. Um it's uh it's so basically it's a star. And it is got rounded corners on either end and it has an increased amount of surface area compared to an Allen key. So you can actually tighten a torque screw much tighter than you can an Allen key. And it's less prone to stripping the threads. Or stripping the head of the screw. So a torque, you get a much more consistent tightening. You get a better amount of torque. Um it lasts longer. Um for the cleat adapters, it's really important. Because those cleat adapters. I mean, you you ride your bike. You get on and off. You get in the dirt. You get grow, you know, road grum all inside of that. Um. Grime. Grime. Road grime. Grum. Grum. Grime. Excuse me. Road grime all inside of that screw head. Um on an Allen key head, it's it's much more difficult to clean it out correctly. And if you get a little bit of dirt still left in there. movie doesn't work. On a uh a torque key, you can actually clean that out way better and if you don't get the screw the the head all the way inside to set perfectly, you can still get enough torque to actually tighten and loosen the part. So it's just it's a far superior way to do it. Also, there's a whole thing about how actually manufacturing them, you can hold them at a tighter tolerance. You can also actually mill the head. There's there's a lot of things. Like if you notice on some of our products, we actually don't um um broach the screw. It's actually a milled finished screw. So that's kind of cool for anyone that's really into manufacturing. So basically it's to compliment the product. It's to compliment the product and also for it to work better. It's it's a better performing screw. Cuz I know all the Allen screws that I get that like come in like all the baby product I've put together. I pretty much stripped all of them. Like good luck try to get that crib apart. Yeah, probably. So I yeah, I understand. So it's it's it was an issue of that. And so we and we knew that when we did it, we were going to probably get some push back. It just sucks that the first review was like, oh, seriously, like Yeah, so I mean, Yeah, we knew it. We knew it coming. But my problem is reviews not based on the use of a product. And in fact that it says it. Like if product sucks, tell me the product sucks. Don't say like I just don't have a screwdriver. Well, also the the part that kind of frustrated me is that we made sure that throughout the description we kept saying torque screws. Yes. So if you ever visit the listing on Amazon after today, I will be posting a picture of a screw. Arrows and a diagram. Yeah. And a song and a dance. Yes, it was a it was an internal debate if we were going to take the easier for consumers or for trying to make a better product. And so we chose better product. Because that's who we are. Yeah, we tried we tried to do it. So it was a very thoughtful approach on how we what we pick and all the screws and hardware. If you look at the RF 20 or if you look at the cleat adapters or some of the other new products coming out soon, um all of them are specified for a reason. There's no oh, that's just easier. We're going to do that. We don't usually do things easy. No. Never. I wish we did. Anyway, okay, that's just my little side note there. Things to mention coming coming soon. Our friends at Ansis are flying in next week. They'll be getting a little shop tour here and we're collaborating collaborating on a little piece. So I've got a lot of cleaning to do so that we look pro. Not messy. Yeah, and just get some. We have to also put some some some things away so that they're not shot. And some things put them out. And all these metal chips. Got to clean up all these metal chips. Yeah. Anyway, um so we'll have an episode next week probably. I'm not sure what it will be talking about yet. And then after that, we'll be taking a little break during Memorial Day week because we will be headed to Sunny LA. Yep. To visit some family that we haven't seen since pre-pandemic. Yeah. So that's what's coming up. Anything else you'd like to say before we wrap her up? No, I think that's it. Yeah, that that covers it all. Okay, well, we thank you for choosing to take some time with us. And we look forward to future breakaways. Look for us on Instagram and LinkedIn, Facebook, Twitter, and in person here in Tennessee. We ask our listeners to share, like, and subscribe. We're available on all major streaming platforms. Thanks for listening, have a good one and find some time to break away.
EpisodeMay 21, 2021 · 24:35
← Podcast
Project Breakaway with Predator Cycling
22: The Night Owl Machinist, Ep. 22
In this episode, hosts Courtney B and Aram Goganian of Predator Cycling discuss Aram's late-night work as a machinist, diligently cutting RF 20 aluminum molds for their bicycle components. They delve into the tedious, messy process of using their current open-air gantry CNC machine, and express their desire for a more advanced, enclosed Datron CNC machine to improve efficiency and precision in their manufacturing.
Loading player…