PI North America is a large trade organization dedicated to PROFINET, PROFIBUS, IO-Link and OMLOX technologies.
Join host John S. Rinaldi as he speaks with the Executive Director of PI North America, Michael Bowne, as they discuss:
Connect with John:
Email | LinkedIn | Newsletter | YouTube | Blog | Books
For all things industrial automation, visit rtautomation.com.
PI North America is a large trade organization dedicated to PROFINET, PROFIBUS, IO-Link and OMLOX technologies.
Join host John S. Rinaldi as he speaks with the Executive Director of PI North America, Michael Bowne, as they discuss:
Connect with John:
Email | LinkedIn | Newsletter | YouTube | Blog | Books
For all things industrial automation, visit rtautomation.com.
[ Intro music] Hello this is John Rinaldi with another networking connectivity podcast. Today, our honored guest is Michael Bowne, the executive director, wow, I don't think we've had anybody with a great title like that. PI North America. We're in awe of your title. Well, Michael,
Michael Bowne:Hey John, how are you today?
John Rinaldi:Doing well, so, um, tell us a little bit about, uh, what, uh, with the PA probably that's been around now for how many years, 20 years
Michael Bowne:It was the early two thousands. It was 2002, 2003?
John Rinaldi:20 years since you were in grade school at that time. So, so, you know, what does everybody know about PROFINET now? And what is it? What's the function of the PA what do you, what do you guys do? What's, you know, what are you trying to accomplish at this point in the life cycle of PROFINET?
Michael Bowne:Just a real quick it's it's PI for PROFIBUS International. Yeah PROFIBUS PA is the process automation.
John Rinaldi:Ah, sorry about that.
Michael Bowne:So yeah, I think, I think at this point, I'd like to think that everybody's pretty well aware of PROFINET, what it is as, as an industrial lease and that kind of the follow on to world famous PROFIBUS from the nineties. And it's done pretty well for us in the last 20 years it's growing and it's growing faster each year than I grew the previous year.
John Rinaldi:What, why is that? So tell, tell me why PROFINET it's growing as fast as it is.
Michael Bowne:It's it's, it's Ethernet. I mean, that's, that's kind of the direction that everything has headed in and continues to head. You know, when you, when you bet on Ethernet, which PROFINET is just Ethernet and then it's a protocol that lives on top of Ethernet, when you bet on Ethernet, you kind of get everything else for free. You know, you get all that other beautiful data that you can, any other protocols you want to run on, run on top of Ethernet, then you kind of can get that and you can kind of have your cake and eat it too.
John Rinaldi:So what are you guys trying to do now at the PI? So I guess everybody who's essentially familiar with, if you're in industrial automation and work around a machine as in a tech the technical way, you know of PROFINET. So what do you, what do you guys see as, as your function and purpose and what are you trying to achieve, uh, as a PI?
Michael Bowne:Yeah, we kind of have an all have always had before. I've been doing this for 10 years. It's, it's two goals and that's to educate and assist. So on the one hand we educate and that's targeted towards people who are using PROFIBUS, PROFINET, IO-Link all these technologies and they're helping with understanding how they work, any questions lately. I've been having this feeling, and it's kind of a good thing. You know, how you connect your iPhone to your WIFI hotspot and it just works well, usually there's actually a lot that's going on behind the scenes that we don't realize that's happening. I mean, all of this work has gone into make that connection very easy and seamless and very, very easy. And the same is true for PROFINET all these things. So, you know, one of the things we get to do is to educate people on kind of how really these things work behind the scenes. You can know they're pretty simple as you just plug it in, you don't have to worry about termination resistors or segments like you did with PROFIBUS. So that's one side of it is educating people on the other side is, is assisting. This is kind of where the Real Time Automation part from the assisting people in adding a PROFINET interface to their device. So on one hand you have the end users automotives the oil and gas the world. And then the other hand you have those because it's the marketplace. On the other hand, you have the device manufacturers that need to put a PROFINET interface, those are kinda our two...
John Rinaldi:I think that education is really important now because there's a lot of new people coming into the industrial automation space. Who've come out of college and you never heard of this kind of stuff. And you don't know what it is so that you need to have that educational material away available.
Michael Bowne:And w we've actually noticed that we've noticed that ourselves. So we look at all these Google analytics and all of these things and the questions we were getting more in the last three to four years on more of the basic questions. Hey, what is PROFINET? How does it work? And so we're, we've had to tailor actually we had to tailor our resources to some of the more generic introductory stuff, which is great because you can get through it pretty quickly. And then you can jump into the deeper stuff, which is what, so we've noticed that as well. Yeah,
John Rinaldi:Yeah. What I've heard is that a lot of control engineers are getting, are getting scooped up by the Amazons and Googles and Microsofts of the world who increasingly see automation as, as a future. And so they go, Hey, where can we get people that, that know how to make things work? And they go, let's grab them from manufacturers. So they're going to have to hire a new people just out of school and throwing them in the, in the situation on the machine. And they there's. I gotta, I gotta Modbus note over here. What the hell is that? We actually get a lot of hits on our website for Modbus still. And it just surprises me. Like Modbus is 40 years old now.
Michael Bowne:I mean, I don't have any data to support this or even anecdotes. I think it's a lot of what's taught in because it's different. I mean, it's so easy. It's how it's taught in schools. It's a great introduction into communication. Modbus is.
John Rinaldi:You know, that's, that's a really good point. And what I tell new people who join our staff is that once you learn one of these things, they're all fairly similar about that. Like, you know, when you get down to the nuts and bolts of EtherNet/IP and PROFINET, what do they do? They send inputs, go from a device, into a PLC outputs, go from the PLC out to the device. I mean, that's the core.
Michael Bowne:Yeah, by and large
John Rinaldi:It's, the formatting is different than how often they go out and all this other stuff varies, but that's pretty much the goal of all these industrial ethernets is just doing that. What do you see as the, uh, what do you try? And you got any, any big goals for this year for the PI that you're shooting for, um, you know, along along those lines, in terms of new programs, are you going to, you're going to go out and do you guys did road shows for years? Is that coming back or not?
Michael Bowne:Uh, God willing, God willing, we'll be back on the road. Um, we sit in the middle, this third or fourth waves of COVID-19 I, I'm not sure. And, and, and that's kinda, the problem is the uncertainty. How can you grind something like that? I plan to when, when, when this all comes to a nice finish, we, we planned to get back on the road with road shows for sure.
John Rinaldi:Are you guys going to be at Pack Expo is coming up here the next month? You guys can have a presence there?
Michael Bowne:We had talked about it, we had thought about it, but we, uh, we decided not to this year, perhaps, perhaps next year.
John Rinaldi:I think that, I think that their attendance is going to be a little, like, you know, Las Vegas seems to be a hotspot right now of, of COVID. And, uh, given the fact that you have to wear masks and at the show, I think that's going to hold a lot of people back from, from, from coming to that. Let's talk about some of the interesting things that are going on in the, and you know, I've always said this I've been around 30 years now in this, in this arena, industrial automation. And it used to be that thing, you knew a new something radically new would come out and you had about five years to digest it. And then some, something else radically new would come out. Well know we have 10 radically think new things coming out every month. It's hard for, you know, my job is pretty much to stay on top of technology issues and I have a hard time keeping track of everything, let alone a control engineer who's trying to keep a couple of machines running, trying to understand what's going on. And one of the things that, that, uh, the technologies that, that mean talked a lot about and has been for the last three or four years is, is TSN. TSN(Time Sensitive Networking). And I've heard people say that, you know, I I've, I've not been a proponent of it, but I've heard people tell me that, oh yeah, John, this is going to replace EtherNet/IP. And PROFINET, those things are going to be obsoleted by that. Where do you, where do you guys stand on that? And are you working, uh, you know, how does PROFINET, I mean, you've got IRT. How does, how does, how does PROFINET integrate with TSN? Do you see that as a, uh, how do you see that working out over the next year or two years or five years?
Michael Bowne:I'm going to put it bluntly. Um, anybody who says that might have, uh, a little bit of a misunderstanding of the seven layer, ISO OSI model, which is what I always, always come back to. And it always helps me understand things in questions like this. So if you know the ISOs layer model, you know, that ethernet, do, you know, layer one you have your physical layer, layer two you have your data link layer and that's, that's where TSN lives, right? Protocols like PROFINET and OPC UA, and HTTP. They live at layer seven. There is absolutely no way or effect of, of a TSN replacing before you still need the protocol. You still need a way to format the data and do all that stuff that the protocol layer 7 does. All TSN is a transport. Is i t a great transport? I mean, you're taking deterministic, y ou're t aking u ser and you're making it deterministic by design. That's incredible. Well, it's not cause we've been doing it with PROFINET for a long time, but the point is t hat y ou're having it for free in all these in it. And t hat's, that's a beautiful thing. So, no, I absolutely, it does not replace these technologies. I t's j ust a n ew t ransport.
John Rinaldi:What about IRT? What explain, explain to the people, listening to it may not be familiar with what IRT is and, and how that is if that's impacted by TSN.
Michael Bowne:Yeah. So, so IRT or isochronous or isochronous real time. It's, uh, it's some of the things that we, that we, we have a flavor of, of PROFINET for your audience that is not aware of flavor PROFINET for super, super high speed synchronized PROFINET. This is where you want to synchronize in sub-millisecond timeframes. And so we do a couple things, synchronization, bandwidth, reservation, scheduling, fast-forwarding fragmentation, all these different things to, to allow these super high-speed sub-millisecond timescales. And so does it coexist with all the other ethernet traffic on a, on a network? Absolutely. That was the whole point. You know, you can run your normal PROFINET traffic, your normal TCP IP traffic. So a lot of the mechanisms in IRT, that's what I was kind of getting out before. A lot of the mechanisms in IRT only made sense for factory automation for a long, long time. Then all of a sudden other industries want to get deterministic ethernet, like in car, in car automotive infotainment systems and all these sensors on cars now. So they want, or the audio visual bridging folks, you know, they wanted to turn to ethernet. And so a lot of the mechanisms that we are heading in an IRT, you were adopted by the eye chip lead for, TSN not all of them, some of them, and are there differences? Sure, but it's something that's very familiar to PROFINET. So what's going to happen is now all of those things that were just PROFINET IRT now become TSN again, deterministic ethernet for the masses. And we'll, we'll take that. We'll, we'll do that. If TSN had existed, you know, 15 years ago, what we would've done, it would've been great, but instead it didn't exist. So we had to kind of, we had to do it ourselves with IRT.
John Rinaldi:Uh, well, how many people are using IRT? Do you guys have a feeling for the percentage of people in the, in North America that have implemented PROFINET? Is it a small amount, a large amount of, of those customers that are using IRT? I have no feeling for that at all, actually.
Michael Bowne:You know, I, I get this question a lot in that I actually don't the insect that I had is on a global level. So the way, the way it works is, um, every year we hire a independent auditing firm. It's like a, kind of like one of these KPMG guys, they go out to each and every single PROFINET device manufacturer and ask them to give them a count of the PROFINET nodes that they sold last year. They then take those numbers, put them all together, add them up. And they give us the total because it has to be anonymous. We shouldn't know who's selling how many and where. I have a global number. And we feel very confident that number, because it's all outside of our hands in a way. I m ean, i t's all independently handled. We don't have insight into where they're being sold. I mean, I have my own ideas, but that's not an official number, let's say.
John Rinaldi:So on a global scale, you know, how much, how much IRT is there compared to how much PROFINET?
Michael Bowne:Well, we sold, we've sold 40 million PROFINET nodes.
John Rinaldi:Really? 40 million?
Michael Bowne:Yeah, 40 million at the end of 2020. I mean, that's over 15, almost 20 years, or i s, it's taken 30 years to get to 65 million PROFIBUS nodes. So pretty quickly w e're going to overtake that. Yeah. I wish I wish I could give you a harder number on how many of them are really companies adopting IRT. We see them, y ou k now, creating devices with them. So there must be some market for m ean, p eople d on't j ust b uild t he product for, for p roducts s ake, r ight? They have a customer.
John Rinaldi:We don't see that. And you know, the applications that our customers have and discrete automation is that it doesn't seem maybe things like paper-making, they use that maybe more pro you know, where you can have processes going on, but pulling the discreet market, this very few people come to us and say that they need IRT.
Michael Bowne:Yeah. I mean, it also helps that PROFINET RT satisfies 90% of applications.
John Rinaldi:Sure. That was one of my arguments against TSN is I don't, you know, ethernet continues to go up in speed, you know, w w when we got to one gig ethernet, I thought, wow, that's amazing. I'm very, I mean, the problem in industrial automation is we at the end of the day, you still have a physical world at the end of the link. And then the physical world does not outbreak that fast.
Michael Bowne:You are exactly right. And so the, the, so somebody says, well, you already have IRT. Why are you doing PSN? And if it only satisfies 15%. Why are you doing this? And here's the answer why do TSN it's because down the line, you can get it or whatever. The, the, the thought is that there's going to be so much other traffic on the networks. And as, you know, OPC UA frames are Mack trucks compared to a little tiny PROFINET frame, right. And so the bandwidth is going to get utilized more and more and more and more, and you can't have your automation, traffic at the expense of a video camera, or a web server or something.
John Rinaldi:So, so, so, you know, but as ethernet continues to advance, you get 10 gig ethernet, a hundred gig ethernet. I mean, geez, that's, you know, we can, you can push a lot of stuff through pretty fast at a hundred, a hundred gig. I still wonder if we're, if we're ever in the, and the thing is, there's not that many applications that really need a lot of speed. Oh. A lot of things are like injection molding. We're going to make a part every three seconds, you know? So I don't know. We'll see how that plays out. What about, um, you know, another, another area of, uh, of some interest right now is single pair ethernets is, uh, is a PI involved in, in the, in the ethernet standard, what are you, what do you guys do in there?
Michael Bowne:Quite, quite heavily? So this is the 802.3CG thing. I also say something a little controversial, but I'm still unsure of, of the prospects of single pair ethernet in factory automation is my feeling, although it's certainly possible that could change, most people call us new use cases, but it seems to me that four wire ethernet or two pair ethernet seems to work pretty well. In fact, the exciting part for STE is when you take it and then you make it explosion proof or intrinsically safe, which is what we've done. And that's APL, that's pretty exciting. This IEC version of,
John Rinaldi:I agree. I agree with you completely. Um, when I take a look at discrete automation, single pair ethernets I keep thinking, well, do I really want every single node to be connected on that ethernet link or IO-Link seems to work perfectly fine? I mean, what is the advantage of doing that over IO? It seems like it's harder to manage ethernet nodes, a bunch of 500 ethernet nodes then it is to manage lots of IO-Link nodes that that can be that kind of get compacted together and aggregated and, and commit over PROFINET.
Michael Bowne:Yeah, totally agree. I lead this. I mean, if you want to talk about gross, IO-Link, is going is going nuts there. They're growing really really fast and here in North America too. That one we know for sure. That's a really, really nice technology. It's expensive. It's really great. And then you just pump it up the other side.
John Rinaldi:Yeah, I-O Link, I think is I-O Link about 10 years old or is it less than that?
Michael Bowne:It is. It is. Um, there was a name way back called PROFI points and that they, they, they kind of didn't have a home within the, they fell back into the, into the PI umbrella, I would say about 15, but only in the last half decade or so, has it really started to ramp up very, very quickly.
John Rinaldi:Right. I, I just don't see it displaced by single pair ethernet making as much sense to me because you have to look at it from the sensor manufacturers point of view, if you're making a photo wire proximity switch or any of that, I mean, the I-O Link is so you don't have to do anything to support IO-Link where you have to have a, uh, a diff a whole different kind of technology to support SPE. And I don't know that they're going to get pushed to do that.
Michael Bowne:It's unclear how much a little SPE interface will cost you, but that's one of the reasons it's become so popular is that it's so inexpensive to put on an interface on a little, put a$15 internet chip or a$5 proximity sensor. It makes no sense.
John Rinaldi:Right. Even adding a dollar to the case, to the sensor. You know, I don't know, you know, you get into that market, that's approximate it from this vendor versus a practice switch from that vendor pennies matter because they're buying hundreds of thousands. Right. And so you start adding another dollar for every one of them you've run into lots of money. We'll see. That's another thing that's going to play out. And it's interesting. I mean, there's so many issues. Like another one is the information modeling I've lately been in touch with a number of, well, you know, this there's lots of trade organizations that are all in the, in the information modeling kind of business. I'll call it right now. And it's, and that seems to be, to be kind of a mess. If you look at it from the outside as an independent user, trying to figure out what's going on with information modeling, lots of, lots of standards being developed, which is kind of scary.
Michael Bowne:We're at least for now, we're trying to stick with. And, you know, cause you're a big, pretty big proponent of OPC UA. I mean, from our perspective that it does what it does and it does it really, really well, great information models. And so that's where we've got our efforts.
John Rinaldi:And I agree with that completely OPC UA. And I remember our, our good friend, Carl Henney used to kid me about that, that I was drinking the OPC UA Kool-Aid, but at one time in my life, I was actually onto something early that turned out to be pretty big. I missed a lot of buses in my life but I didn't miss the OPC UA bus. That was certainly, it seems to be the home for all of these, you know, for definitely there's no other communications technology that can support an information model like OPC UA. So I I'm very bullish about OPC UA. I'm somewhat less bullish about information modeling just because in the United States, we've got so many different trade organizations that are all ready to just do. They just do their own thing. They don't want to cooperate. And that makes it difficult to come up with a common standard for these things.
Michael Bowne:Agreed, agreed. That's kind of, I mean, it would be nice, but the nice thing about the OPC UA information models, they're pretty flexible. And if you've got a trade organization that has a very niche market, whether it's packaging machine or printing, I mean, whatever, you can take an OPC UA information model, right. Your own companions. And there you go. That's, that's a great way to do it.
John Rinaldi:Yeah. Right. Because it's true. And one of the, you know, I w I've been in, um, uh, looking at, uh, in getting involved with the Sesame organization for listeners out there that CESMII, you can find out about them, certainly on the web, uh, Sesame has this idea that they would like to make integration of machines as simple as you integrate a printer driver, when you buy a new printer for your, for your windows PC, you just go to a cloud and you download the information model for that machine and away you go, boom. Now all your applications know how to talk to that machine. That's kind of their, their, their vision and a couple of sentences. I mean, it's, it's incredibly, it couldn't be more expansive to try to change, try to change the whole of industrial automation to work that way.
Michael Bowne:Well, not a new idea. Let's, let's be frank, I mean...
John Rinaldi:They are getting, but they have lots of, they're getting a lot of funding for the department of energy. That's the different part differences. When, you know, with your, you know, other trade organizations, they have to get funded by their members and that's difficult. And they get, they get time. People, you know, donate labor to them to work on things, but they don't donate labor for years when you're a partner with the U S government. There's a right. Especially right now, there's an open tap there for money. And the government recognizes that, that the United States of America has a problem in that our manufacturing technologies and has moved overseas and that we can't build the kinds of things that we need to build for our defense, for our security and for other things. So they're rightly funding technologies like that. So we'll see how that goes.
Michael Bowne:Yeah. Let's see. Let's see.
John Rinaldi:Uh, so what do you guys got on the roadmap is PROFINET going to change more, you know, every once in a w hile there's a new rev new revs come out a nd are we still g oing t o be on the, y ou k now, I think it was about every year, every 18 months, there was a new rev of PROFINET. Are we still on that kind of track? A re we s till see more of that in t he f uture? In t he next f ew y ears.
Michael Bowne:Most of them, I think it's been stable for at least two big version numbers. Um, most of the stuff has maintenance update kind of things. They, they moved to 2.4 when TSN came along. But again, it's you, TSN as you know, it's not ready. It's not out in the market yet. I mean, it still has to be finalized. And IC608.2 then it has to get into the chip makers and the chip makers have to get it into the devices and the device, you know, so there's still some, it's still gonna be some time before this stuff. And that was the last major update. Um, this stuff is pretty stable. So a lot of the things we're working on in the future now are kind of so PROFINET is doing well. And, you know, the TSN OPC UA specs are doing well on ATL and all this SPU stuff. We're looking at diversifying a little bit, not just data communication side of things. That's been the bread and butter for 30 years, packets, ones, and zeros over PROFIBUS. So now we're kind of diversifying a little bit with IO-Link stuff. Well, let's still get the communication down to go down to the central level, but we got a new technology in our portfolio last year. It's called OMLOX and that has to do with r eal t ime location systems.
John Rinaldi:How do you spell that?
Michael Bowne:O M O L O X, the Hermes award, a prestigious award in 2020 has to do with locating things in a factory. Customers notice that the t hroughput isn't limited by the machines. I t's actually t he, if I find the materials a nd w arehouse to get them to the machines to be processed. So, u h, those companies got together and they wrote this second, they were looking for a nice home and was able to provide that to them. So t hat's kind of,
John Rinaldi:Yeah, there's a lot of need for that incredible need for the factories, get bigger and bigger and bigger. Someone told me that at Tesla, they have these$50,000 tool cribs or something that move around and they lose them. So they just have, they just ordered more. So being able to locate stuff like that is incredibly important. But especially with high-priced equipment like that.
Michael Bowne:The ROI is one day. If it's$50,000.
John Rinaldi:Yeah, absolutely. Well now you know who to call when we hang up here.
Michael Bowne:When we hang up here, yeah.
John Rinaldi:Trade them for a Tesla. So we're pretty much at the, at the end of our call here. Um, Michael, you have anything you'd like to before, before I let you go, though, I want to talk a little bit more about developers. I mean, that's who we support. We've been supporting PROFINET developers now for, well, I don't know, forever, as long as I can remember. And we, we, we do a lot of it. Are you guys doing anything? You know, what are you, what are you doing for new developers who want to incorporate PROFINET technology inside their devices? What kind of resources do you at the PI have available for them?
Michael Bowne:Two main ones. One of them has to do with, if you're just starting out on your journey, we've got this really, really nice document, and it's been around for a long, long time. And we update it every couple of years. It's called the The Easy Way to PROFINET, it just has these are your options when developing for your PROFINET device, and it also has social providers and there's of course, Real Tme Automation is in there. And then a couple of years ago, or maybe last year, I guess it was, we came out with a second really, really nice document. It has to do just certification and testing because we've got a lot of questions. How do I get my post tested? How do I get my device certified? Because PROFINET license to be certified, frankly, you guys do the hard work. We, you know, we just, we just tested bugs. So you gotta make it really easy. Real Time Automation makes it really easy for somebody to put a PROFINET interface on their advice. And then they come to the lab, they get it tested, and checked, they're done.
John Rinaldi:That's what we, you know, that's our goal is to, is for, for device vendors to come to us and say, I need PROFINET and then we help you on that when we're that you understand your technology, your device, and we understand PROFINET and put our heads together and we come up with a solution and get to go in and get you integrated. So you can tap into that huge market for us factory systems that are using the PROFINET architecture. So thank you very much, Michael. It was a pleasure as always to talk to you and, uh, hopefully we'll, I can see that with some trade show or some road show at some time in the near future.
Michael Bowne:Yeah, it would be nice. Thanks for, thanks for having me on. It was my pleasure.[ End music].