Episode 46: Benjamin and Stephen actually managed to keep an episode under a half hour for once! Steve jumps right into the global silicon shortage. Ben discusses how to assess the value of investing in robotics and automation. Stephen shares what he learned about laser-assisted tape winding. The boys close by rambling about alloys and how they’re getting old.
Free registration for all spark content on-demand until June https://www.imts.com/spark/
For the latest in Manufacturing Technology news https://www.amtonline.org/topic/technology
Transcript
Benjamin Moses: Hello, everyone. Welcome to the Tech Trends podcast. I am Benjamin Moses, the director of technology, and I'm here with-
Stephen LaMarca: Stephen LaMarca technology analyst.
Benjamin Moses: Steve, how are you doing?
Stephen LaMarca: What's up, Ben?
Benjamin Moses: I'm glad we're recording in this new format. This is going to be great.
Stephen LaMarca: This is really fun. It's so professional now.
Benjamin Moses: Yeah, absolutely. How do you feel at the end of Spark? We've been about two weeks out of Spark now.
Stephen LaMarca: Yeah, I really don't want this to be recorded, but it's nice that... It's like a breath of fresh air.
Benjamin Moses: Sure, [crosstalk 00:00:36].
Stephen LaMarca: I finally have times during the day where I'm not doing anything. I'm like, "This is nice."
Benjamin Moses: You have room to breathe just a bit.
Stephen LaMarca: Yeah. Just screw around.
Benjamin Moses: Yeah. So one thing that you can't screw around with is buying new graphics card, buying new GPUs.
Stephen LaMarca: Oh, man.
Benjamin Moses: You and I were chatting about the silicon shortage. Why don't you [crosstalk 00:01:00]?
Stephen LaMarca: Yeah, we've been blabbing about how hard it is to get a GPUs in previous episodes. And now, regular CPU manufacturers are having a tough time sourcing silicon for the aftermarket. And it's funny, I say that, after market meaning that for the DIY computer builders and people who want to do upgrades, it's nearly impossible to get your hands on the latest tech, at least at MSRP. It's out there, but the scalpers have it at like two, three times the price of what it's supposed to be sold at.
The cool thing is if you actually do want to get your hands on that stuff, this is the one time ever, the only time ever, that it's been possible to get a prebuilt PC, like an already assembled PC for you, with the latest hardware in it, like the latest 11th gen Intel processor, the AMD Ryzen, or the latest GPU, like the RTX 30 series, actually at MSRP, and it actually be cost less than building the computer yourself. It's a weird time we're in right now.
And it's funny how the silicon shortage is favoring the OEMs, as it's supposed to do. The OEMs get the new stuff first. That's the way it's supposed to be. And the aftermarket, for once in PC building, is actually suffering. I mean, I'm sure it's suffered before. But it's the first time ever that a prebuilt actually costs less to buy than building it yourself, which has always been the mantra of like the DIY elitists. Anyway, I didn't need to get into that.
But about the Silicon shortage, what I wanted to talk about was the silicon's also going to automobile manufacturers like Ford and stuff. So they need these to make these computer chips for their engine control units and their transmission control units because everything's run by computers these days, so auto manufacturers need silicon too.
But what's crazy is I saw a video on YouTube recently that was kind of like backpedaling on a video they put out exactly a year ago saying that if you need computer components, buy it now because we're not going to be able to get them from China in the future. Now they've got the video that came out just a few days ago saying that, "Okay, we were wrong, but we were still kind of right. Because you can't get the stuff now, but it's not because it's not coming out of China. It's coming out of China. It's just going to OEMs first. And yeah, we were right. You still can't get it right now. So aren't you glad you bought it back then?"
But anyway, it was making me think more about the silicon shortage and, sure, prices are going sky high on the aftermarket, way higher than MSRP for like GPUs and whatnot, but who else uses silicon? And being a huge watch snob that I am for high-end mechanical watches, a lot of watch manufacturers have shifted over the past few years from using specialty alloys in their main springs and hairsprings. These specialty alloys being relatively ferromagnetic, relatively anti-magnetic. Because magnets are really bad for watches in that electronics are all around us today. If it's electronic, there's a magnetic field associated with it.
Benjamin Moses: Sure, undoubtedly.
Stephen LaMarca: Almost most of the time. And mechanical watches don't like magnetic fields because of their metallic springs. If they get magnetically charged, particles get attracted to each other, when particles get attracted to each other, the springiness doesn't work as well. So your timing, your isocronism gets way off. I didn't mean to go in that deep.
Benjamin Moses: I like the technical term, springiness.
Stephen LaMarca: [crosstalk 00:05:15]. Yeah, the spring constant, that's it. But over the past few years, mechanical watch manufacturers have pivoted from these specialty alloys to silicon because it's anti-magnetic. And you can wear a mechanical watch with a silicon hairspring and put it right next to like a 1 tesla permanent magnet, and it won't affect the timing at all, which is really cool.
And a couple of years ago when it came out, it was slowly catching on. I think Omega was one of the first people to adopt it, and then over the years, eventually Rolex bought into it. And if Rolex buys into it, usually the market's going that way in the watch market.
But nobody ever thought of silicon as being a precious metal because when you think high end watches, you think, "Oh, where's the old man flexing with a gold Rolex?" But it's crazy now that this pandemic bringing on this huge silicon shortage has actually made the value of watches... Which, the pandemic also came with an economic downturn, some people may have been affected by it, and yet a luxury good is skyrocketing and price. The demand is higher than ever. The prices are higher than ever. And it's probably due to the silicon being used in hairsprings now, and effectively, taking something that was not a luxury good or a precious metal and turning it into a precious material.
Benjamin Moses: Yeah. That's fascinating that as big as the electronics market is, broad spectrum, so just computer components to industry goods, to other consumer goods, the trickle on effect from the raw material of silicon being chewed up and the demand going way, way past anything anyone forecasted. That was one of the interesting things that the video brought up is, yeah, these computer electronic manufacturers had a forecast. I would say it looked at some level of escalation because every other generation of cards, there's more demand. And, of course, automobile demand had a certain level forecast. When you far exceed that level of forecast and the demand is this high, it's very fascinating.
And the idea of why the demand was so high, sure, the disruption to the supply chain definitely there, but there's also just more people using those goods and wanting those goods. The [inaudible 00:08:00] also mentioned the cryptomining. Yeah, that is a factor, but that's not the only factor. There's all just significant demand across the board on all the sectors.
Stephen LaMarca: Oh, yeah. Yeah, it's like a lot of gamers point the finger at cryptominers are the reason why you can't get a GPU. It's like, no, we're all the reason why you can't get the latest GPU right now. The fact that so many people around the world, not just in the US, are working from home, and a year ago, they had to dust off the 10-year-old eMachine computer to open up some Excel spreadsheets off of your job's VPN, and you find out, "Wow, this hardware is not up to snuff. I need a new computer, or I need to upgrade this computer."
Everybody, literally everybody, is buying electronics. So yeah, there's going to be a materials shortage. And that's what it is, it's a material shortage. The industry's making the stuff, have not had a problem pivoting from one product to another. We've watched over the last year how beautiful all industries have been pivoting their product lines. But at the end of the day, there's only so much matter.
Benjamin Moses: The only fond memory of the early pandemic I have is when my daughter, who was five at the time, in kindergarten, got shipped home and said, "We're going to work from home the rest of the kindergarten year." Great, let's figure this out. She's using my desktop to log into her class remotely using this giant mouse and giant monitor, I was like, "Well, this is going to suck."
Stephen LaMarca: Yeah, it's just another thing just proving the point. Your daughter's how old?
Benjamin Moses: She's six now, but five.
Stephen LaMarca: She's six-years-old, and she was five at the beginning of the pandemic.
Benjamin Moses: Right. And now she needed a computer.
Stephen LaMarca: She had a full-blown computer.
Benjamin Moses: Oh, she was using mine, yeah.
Stephen LaMarca: She had her very first... No, but you guys bought her her first computer at like five or six years old. I didn't have my first computer until I was like 15.
Benjamin Moses: That's right. Over the summer, I was concerned that-
Stephen LaMarca: Probably not even, it was probably like 17.
Benjamin Moses: I was worried that the school wouldn't hand out Chromebooks in our county, so I bought a laptop for her, which she does use for church activities. But yeah, at six years old, she has her own personal laptop. Which I found out the hard way about setting family permissions and basically have my own IT infrastructure for all these new logins I had to create.
Stephen LaMarca: We don't need to get into the demand for materials because that makes you think of additive, and we don't need to get into the need that their first investment before investing in any of the transformative technologies in the manufacturing industry, get an IT department. Then there's going to be demand for IT people because now every home is going to need an IT department.
Benjamin Moses: Absolutely. Yeah, we'll definitely touch on that, maybe the next episode. Let's jump into the article, Steve. I found a really good one about how to assess the value of a robot investment. So the idea, we've talked about how important automation is and it's the full spectrum of automation, right?
Stephen LaMarca: Oh, yeah.
Benjamin Moses: So you could use conveyors, you could use robotic arms, even robot process automation for software and back office type environment. But the idea that that this article from Automation World talks about is avoiding some of the say misnomers of going super cheap. So the idea of getting a quick return on investment, that's the foundation of running a good business. Make sure you get profitable return on something that you invest on.
But the idea of going super, super cheap could be a problem. So they look at the total cost of ownership, and I'll read a couple of quotes here. "So the less expensive robots may not provide the necessary capabilities in terms of throughput, payload, or reach, where higher quality ones will." So the idea that buying the consumer-grade robot that you and I have for the test bed may not fit in the industrial space. We are moving pieces of [inaudible 00:12:06] from one side of the table to the other, checking positional accuracy. In the end, to be honest, making sure we can get the data off those pieces of assets and trying to simulate those in the future.
But if I'm running that 24/7 or, okay, maybe eight hours a day for five days a week, that's going to put a lot of wear and tear. And that's why you look at different say tiers of automation equipment.
Stephen LaMarca: That's what you're paying for when you buy an industrial robot. Because like I remember showing our robot to [Monir Hailu 00:12:38] from NIST. And I was like, "Dude, look at this thing, man. We have a spec sheet. It matches the same robot that you could get from UR, FANUC, KUKA, name and industrial brand." And he was just like, "Yeah, it does right now."
Benjamin Moses: [crosstalk 00:12:57].
Stephen LaMarca: "But put it in a manufacturing environment for a week, it's going to break." He wasn't like, "Wait for it to break," it's going to go break.
Benjamin Moses: It's going to break, yeah.
Stephen LaMarca: It's going to break.
Benjamin Moses: Between the usage and the different environment. So I mean, most factories I go in, there's always, I wouldn't say contaminants there, but you got the smell of cutting fluid. If it's assembly, you could have some welding going on somewhere else. All that migrates into the joints somehow, and it's just going to keep wearing it down.
Stephen LaMarca: Yeah. I mean, think about the difference in a Corvette, racing Corvette versus the Corvette down the road from you that's driven by the 70-year-old man who only takes it out on Sunday and never goes over 40 miles an hour and always keeps it under 2000 RPM. That Corvette's going to last forever. The one that went to the 24 hours of Daytona, will be lucky to make it 24 hours.
Benjamin Moses: It's going to limp home.
Stephen LaMarca: You know?
Benjamin Moses: Yeah.
Stephen LaMarca: Because it's under more load.
Benjamin Moses: One of the other core elements that the article talks about is, "But sacrificing core capability is a bad way to save money. Any type of robot that fails to meet all the core [inaudible 00:14:06] will always end up costing more in the long run." And that's a really interesting idea of, hey, I have this use case, let me get this robot into this use case versus I have this factory that has a bunch of potential use cases. And the idea of understanding the life of a robot and what to do when, say, the small project or end-of-life of this project, and then transitioning to the robot to do other things.
I know this is something we talked about in the past episodes about, flipping the robot into doing multiple things in the life of a robot. Once you've paid the $100,000 or $200,000 that the robot costs, including the end of arm tooling and some of the the controller packaging, how do you continue harvesting that value if the machine is still working and the controllers are still up-to-date and things like that?
So I think this a really good look at don't just think of one use case, think of a series of use cases that this might be applicable for. And you could either replicate that scenario or think of transitioning that robot to other scenarios.
And one other side note, I'm the liaison for the Automation and Manufacturing Committee that AMT hosts, and we talked about mobility in the factory at Ford specifically. And one interesting use case that's growing in popularity is adding a robotic arm to a mobile platform. So I found that very, very interesting where IMTS 2018, we saw a lot of robotic arm on linear slides, sliding back and forth along the booth, picking up different things, and sliding back. We've seen a bunch of implementations where it's on the ceiling, so it's not tying up floor space. The next iteration is put it on a mobile platform, and you don't need rails. It just goes wherever it wants. It does the task and goes back to charge.
And there's obviously a ton of problems that could occur. You got to charge it. You got payload capability and balancing all that stuff. But it's a very fascinating look at a more flexible environment and being able to arrange your facility to how you want. So I thought that was a really good article from Automation World.
Stephen LaMarca: For sure.
Benjamin Moses: Steve, take it away, man. Will you talk about laser-assisted tape winding?
Stephen LaMarca: Laser-assisted tape winding, so I thought this video, or not video, this article was really fun, and I learned a lot from it, and I probably need to reread it because I'm sure I missed a lot. But one of my favorite technologies in the manufacturing industry, and sadly, I'll admit, it is not a transformative technology, but one of my favorite technologies is fiber tape and tow placement. It's fascinating. It's incredible. It's really fancy, and because I'm attracted to all things expensive, it is very expensive. Those machines are not cheap, and there's very few of them, even in the US.
But its less fancy little brother, or maybe older brother actually, is tape-winding technology. And this article was really cool and eyeopening because this article was essentially... It was titled The Challenges of Laser-Assisted Tape Winding with Thermoplastic Composites. But it's a really good article in just everything you need to know on tape winding.
Benjamin Moses: Okay, cool.
Stephen LaMarca: And because I talk about in the Tech Report that back at the office, I have this, from my middle school years of paint balling, I have a carbon fiber high pressure air tank. And I always assumed, I knew it wasn't made using fiber placement or else it would be way more expensive than what I paid for it, which was expensive, but I always thought it was made on a carbon loom, a carbon fiber loom. And reading this article made me realize, and I remember what the tank looks like, though I haven't seen it in a while, but it is back at the office... It was cool because I was like, "Oh, man, this wasn't made on a loom. This was made using tape winding."
And it was just a fun article. It was a very informative article. And it was a new technology that I learned about that isn't necessarily new. It's been around, I just didn't know... It's new to me. And it's fun and just as exciting as a new technology to everybody.
Benjamin Moses: Yeah. And I do like the idea of being an explorer of new materials, or we have talked about new materials in the past also, and I'm really a big fan of winding and your application of this high pressure vessel that's made out of that material versus like the one I have is probably made out of steel. And running around the field, obviously we're just paint balling, but the amount of weight reduction is incredible, and it's so much easier and so much user-friendly to go to something that lightweight. And it's a good experiment of the advantages of going to something more progressive on materials like that.
Stephen LaMarca: Composite light-weighting, and it's crazy to think that I bought that thing in middle school, which was a long time ago.
Benjamin Moses: A thousand years ago.
Stephen LaMarca: It seems like a thousand years ago. And we're still talking about carbon fiber manufacturing today. Now more of like additive manufacturing with carbon fiber, but carbon fiber is still, is actually really old. As relevant as it is and as popular as it is and sexy as it is, seeing it on a super car, whether it's your standard woven prepreg carbon fiber, or it's on a McLaren, which uses a lot of forged carbon fiber, it's still a really cool material. And it's amazing to me how old it is.
But then again, if you want to dunk on carbon fiber, another really old material that is only now becoming more popular, inconel. I can't believe inconel came out in the '60s and only now we're figuring out, oh, we can actually work with this now. Manufacturing technology has come so far that you don't have to be Sandvik or Boeing to actually work with this stuff anymore.
Benjamin Moses: Yeah. There's two layers to that. One is the composites, the carbon fiber and the different composites that's used in aerospace. My first kind of introduction to that was learning about the B2 bomber that had a lot of carbon fiber and composites into the skin. And then, back at Eden, our last project with Rolls-Royce had a very, very specific cost for adding anchor points in their composite bypass shell, bypass tube on the engine.
So I found it very interesting that they fully understood the cost of if I run this duct system, and I have to anchor it to the bypass, I need to embed this piece of metal that you could attach a screw through instead of screwing right into the composite, right?
Stephen LaMarca: Yeah.
Benjamin Moses: So they fully understand the cost of drilling it out, adding and gluing this basically stud that you can attach to. And it was every time you added a new anchor point, the cost just keep rolling and rolling up. So I found that very fascinating that they fully understood this at that point, that in the design phase, to minimize cost, it was a canned cost every time you added that point.
Stephen LaMarca: Is there a specific term for that anchor point in the composite or in the composite?
Benjamin Moses: I'm sure there is. It's a, not stud... So the issue is distributing the load into the composite, the screwing right into it doesn't work, so you have to put some kind of load bearing attachment point. So it's a load bearing stud or something like that. There's a bunch of different references.
Stephen LaMarca: They actually do a similar process in watchmaking. No joke. So in a mechanical watch, all of your springs and gears which power the watch, and, well, the springs power the watch and the gears just deliver the power to tell you the time. But all of that stuff is held together with these plates, and the main plate is called the base plate.
And usually those plates are made out of brass, very soft material. Or, in some cases, like the Germans, they like to use German silver, which fun fact, German silver is neither German or silver. It's actually, another Tech Report article gets into that, and that's the copper nickel alloy is German silver. But anyway, those are typically soft metals. Brass is definitely a soft metal.
And when a lot of watch manufacturers... They love thinking ahead in terms of serviceability and maintenance. What's funny about Rolex is, I swear to God, Rolex hates their customers, but they love their service people. When a customer, when their watch stops working and they take it back to a watch store and be like, "Oh, the battery's dead," because they think it has a computer chip and a battery in it, and that's the typical Rolex buyer for you... The Rolex engineers their movements to be an absolute dream come true to work on. And in those soft metal plates that things are bolted to, literally screwed down into, they actually embed a anchor point or a stud that you're talking about. And the French word, well, the French language word that these Swiss people use, I'm probably butchering it, it's called a [foreign language 00:23:49], I think, or [foreign language 00:23:51].
Benjamin Moses: Sure.
Stephen LaMarca: And it's basically, like you know how on some AR platform rifle hand guard systems, the QD sling swivel mounting point-
Benjamin Moses: Yeah, [crosstalk 00:24:10].
Stephen LaMarca: The hand guard is usually made out of aluminum. But aluminum is soft, and by entering this thing in and out, and then, removing a sling and entering the sling back in, you can wear down that aluminum. So instead, they'll use an anchor point made out of steel embedded in that aluminum. It's a similar concept.
Benjamin Moses: Yeah, so the idea-
Stephen LaMarca: I just wanted to know if there was a term for it other than what the French watchmakers use.
Benjamin Moses: I'm sure there is. We just, attachment point.
Stephen LaMarca: We'll call it an anchor point.
Benjamin Moses: Anchor point. Yeah, and also your comment about using nickel-based alloys, so aerospace, of course, they've been using it for a long time because of their high temperature applications. It's always used on jet engines a lot. So we use inconel 625, inconel 718, if you want something high strength at high temperature that you can age. And then, you got the other alloys where you have like cobalt, so you get some abrasion resistance.
And it is fascinating that the evolution of some of those materials in terms of more market acceptance... We started experimenting with L605, I'd say like 15 years ago. By experimenting, I mean our first integration to a product that we sent to a customer, and then that we started learning how to machine.
So the idea of these mature materials getting more acceptance into other industries, and then they're learning all this tribal knowledge again about how to machine it. And it is fairly difficult to machine. Do you want to machine the anneal condition, or do you want an age condition to be on the surface finish? Are you going to get chatter? What type of cutters do you want? There's a lot of nuances that are learned, and the manufacturer's learning this on the fly. So it was great.
Stephen LaMarca: For sure.
Benjamin Moses: It's a very challenging material.
Stephen LaMarca: Yeah, and it's awesome that it's trickling, it's made it to way to the consumer aftermarket. If that's your thing, if you're a Honda Civic owner, you can actually buy an inconel exhaust system for your Civic.
Benjamin Moses: Yeah. I remember-
Stephen LaMarca: It's absurd, but it exists.
Benjamin Moses: One of the stage three modifications for my Golf was a inconel 625 manifold.
Stephen LaMarca: No way.
Benjamin Moses: Yeah, this was back in 2005.
Stephen LaMarca: What?
Benjamin Moses: I didn't buy that. It was really pricey. I went cheap and-
Stephen LaMarca: That had to be five figures.
Benjamin Moses: And the motor didn't last too long after I modified, but this is a cheaper version. So rest in peace, Golf. RIP.
Stephen LaMarca: Right.
Benjamin Moses: Steve, where can they find more info about us?
Stephen LaMarca: They can't find us at our old news website anymore. We're phasing that one out, just like we phased out Zoom. But they can find us at amtonline.org. I don't know the actual subscribe link yet, but I'm sure if you go to amtonline.org, there's somewhere you can click for news, and then you'll be able to find us there. We'll throw it in the link below.
Benjamin Moses: Awesome. Thanks, everyone. Take care.
Stephen LaMarca: Bye, everybody.
