Welcome back to the Automation
Podcast. My name is Sean Tierney, and today on the show I meet up
with David Peterson from control.com to talk about his experiences using
PLCs based on Arduino and Raspberry Pie. David, thank you for coming
on the Automation podcast. I'm so glad that you agreed
to come on and talk to us. I wanna let everybody know
that, uh, last year, uh, him and his company invited me
to be part of their automation, uh, day. And, uh, I did a presentation,
PLC's past, present, and Future, and I was just so appreciative. So I was
looking for a way to get somebody from, uh, their site on, and I would notice
that David was doing some stuff on, um, Arduinos and PLCs,
and, uh, I said, Hey, this is a great opportunity to get him on, just talk about everything
he's working on. So, before we go into any of these
topics, we're gonna talk about, could you first introduce
yourself to our audience? Sure. And thank you so much, Sean,
for having me on. This is great. Um, I love talking about all
of the, the new technology, and welcome to all of your listeners. It's great to be able to share
some information with you. Uh, so for just a little bit of background, let me introduce myself and then
kind of what I do and what our, our site is that, uh, that Sean mentioned that we had a
control automation day last year. So my name is David. I'm the director of engineering
Content for Control Automation, which is control.com. Uh, usually I
think of myself more as like a, uh, an engineer, but a
technical journalist. Um, I try to learn as much as I can
about both new technologies, but also the challenges that are, uh, always being faced by people who are
in the field. Um, so there's, uh, a lot of new emerging stuff that can be
applied to some of the more traditional troubleshooting scenarios. Uh, I
used to be a college instructor, uh, with a mechanical engineering degree, but I transitioned into control
systems engineering. Uh, so as past teacher, I can't
really get away from education. I just love knowing the,
the how's, the why's. We don't just plug
technology into a situation. We kind of need to understand why it
exists and why somebody developed it. Uh, so I love the, the teaching, the
training, the educating, but also the, the learning and the
sharing of information. And then just a really quick
introduction to what control.com is. Um, we publish news and information
and technical theory about the new products that are coming out
from big companies, from small companies, all those that are involved in
the control automation space. Uh, new products can also mean theory
and technical education on some of the existing products. You know, somebody comes out with a new
sensor for a process control, uh, what makes it better, what kind of
communication protocols does it have? And then diving a little bit deeper into
how does one use those communication protocols? How does Modbus work? How
does IO link work? Those kind of things. Um, so these days a lot of the conversations
are turning towards the industrial internet of things, industry
4.0, digital transformation, whatever words you use for it. So just trying to bring a lot of clarity
and definition to a lot of those. Terms. Now, for the audio-only audience, those people listen to iTunes
and Google Podcast, Spotify, although if we're looking at the,
uh, control.com website right now, and you'll see David's face in the
upper right hand corner where there's a link to, uh, PLC video lectures. So
that's the guy we're talking about. Yeah. I'm sorry, you have
to look at my face so much. . No, no, you do great videos. So I just want the audience
to know that you have that, that you're that guy
doing those, uh, lectures. Yes. Um, part of the reason why I love joining
this team is because of the educational focus. It is an editorial
team, which typically is, is just centered around journalism. But to have a department that's
dedicated to understanding engineering challenges, uh, I feel like is
really what sets us apart. We, we know what it feels like,
we know the challenges, haven't been in all of your
shoes, but, uh, been in a lot of. 'Em. That's awesome. Great. So that's
kind of your background in, uh, in, uh, your company. Now. I was,
you know, I was thinking to myself, I'm never gonna get a chance anytime
in the near future to play with any of these new or dino or pie based PLCs. And when I saw you were doing a series
on it, I'm like, oh, we gotta get you on, um, and, and have you talk about
what you've been playing with. So let me turn it back over to you. Sure. Um, and I guess I feel the same that that
same emotion a lot of times is I'd love to be able to play a lot more with some of
these devices that are coming out. Um, but the fact is there's so
many of 'em, and every time we, we go to a trade show, we,
uh, visit a vendor website, we see new products being launched. And
a lot of 'em, like you said, they're, they're based on these open source
platforms like the Arduino and Raspberry Pie, which typically
for, for the past number of years, we're kind of known as
the, the staple for makers. If you want to make some
little $20 project at home, you turn to an Arduino. And the,
the, the community behind those, the forums just incredible. You
find all kinds of information, um, and it's great for
building your own things. But then once we turn towards automation,
the more industrial applications, we really only saw those
showing up very rarely. Um, I do remember a couple of
times in past decade, uh, as, especially as the Raspberry Pie
was making an early entrance, there were a few industrial companies
that would reach out to us over at the college to find out, um, who had the
resources to help 'em troubleshoot this, this little computer device that they
had running a machine because nobody on the plant floor understood it. Um, so it's probably untrue to say that
it's never been used in industrial applications, but it's
certainly a lot less common, uh, than PLC platforms. Uh, it's,
it's, that's definitely true. I, I, I've, I've followed all these, uh,
all these, you know, back from the, uh, pick and I've just been
following 'em and I'm like, these would be great to do Home Project
Quest. But at the end of the day, you know, what turned me off is, you know, using JavaScript or C plus plus or
Python or it's like, ugh, you know, I spent all day in the I e C 6,
11 31 dash three space, right? And it's just like, I, I just like it,
you know, I'm just, that's what I use. That's the way I think
after 33 years, you know. And so when I started to see
the, you know, IDs for, uh, the, those type of languages like we
covered a couple weeks ago, open plc, that's what kind of drew me in. You know, we can actually call these
PLCs now cuz they use the IC 6 11 31 dash three. You can have normally open contacts
and coils and tons for your timers and to Fs and it's like,
okay, I can do this in my sleep. And that's kind of what
draw drew me to it. Yeah, absolutely agree. Um, because like the Arduino f
back from the very beginning, anybody could interface a 24 volt system.
You could do use a relay in control, 24 volt discreet outputs. You
could use 24 volt sensor inputs. N P N P and P uh, making
hardware connections is pretty
easy. But like you said, it's, when does a PLC engineer wanna
have to learn Java or c plus plus basically become a computer science
major just to create a small PLC application. Um, so I think
that's one of two things that we're a, a big obstacle for
widespread adoption. Um, and then another one is the
ability to be able to connect and interface with other
things. Um, because the, the little microform factor
plc, like the old micro logics, great little devices, if you just wanted a few sensor inputs
and control a few relays or contactors on one small machine, but nowadays you
need to be able to connect in hmi, a few networked VFDs and a
small board with just digital and analog inputs and outputs.
Couldn't really do that. But people would come up
with, like, for the Arduino, they'd come up with shields and there
might be an RS 2 32 shield or an ethernet shield, and you could have a, a software
library that would allow you to use, um, Modbus TCP on your ethernet shield. But you are kind of relying on a community
of people to scrap together something that should be very reliable. So I think
as far as security and reliability, that programming language,
like those i e c languages, particularly ladder logic
and that interface ability, I think were two of the main reasons
why it was a, a, a cool idea, but maybe not a practical industrial idea. Yeah. Not having IO that's protected,
electrically protected, that's scary, right? Because you burn out that
point, it's dead, right? Yep. It's dead forever and maybe it takes out
the whole chip, right? The whole, uh, microprocessor. So, um, and I know on my pie I found a company
that makes these industrial shields or, or for the layman, you know, just add on boards that are
24 Voltz dc they're optically isolated and they're, you know,
rated for industrial environments, which is great. So I won't even use the
five volt TTL stuff that's on the board. Um, and so that to me,
that's cool too. And to me, I'm looking at more of this as
for a hobby home motivation for, for training because, you know,
trying, you know, it can be expensive, especially in other parts of the
world trying to get your hands on, you know, a $600 s
seven, 1200 starter pack, even though it has everything you'll
ever need that might be, you know, 10 times more than you make in a
month. Right? So this really, with, with these chips coming
in so inexpensively, um, it really can really help people
learn, especially overseas, where, but even for, you know, I remember
when I was going to college, all many, so many moon, many moons
ago in the eighties, how many capacitors and transistors
and microchips we blew up just in the lab trying to, you know,
put our own circuits together. So it's good for training too here,
here in the first world because, um, you know, kids make mistakes, right? Yep. And you wanna encourage that. Yeah. You certainly wanna make the mistake
when you're in a lab environment where nobody's gonna get hurt. And yeah, maybe these Arduino and Raspberry Pi
PLCs are a little bit more than the, the $20, $25 UNO boards
mm-hmm. , but still it's a whole lot
less expensive than, you know, blowing up an IO module for a much
more expensive chassis based system. Yeah. Oh yeah. Absolutely. So I love. Making mistakes. . I always tell people in, in training
when things are working perfectly, it's all right, but it's
not really real world. I, I love it when problems show up like, Ooh, the logic didn't work
the way you expect. Oh, the circuit didn't work the way you
expect. Okay, let's troubleshoot it. This is where the fun begins. Well, and I love doing that
and my courses, if I ever, if I'm ever teaching and then
something goes wrong, I like, I love it cuz it
typically, when it happens, it's something that's happened
many times in the field as well. So it's like I dig into
it with the students and I
just, I love going, I'm like, Ooh, look at this. This isn't working. And sometimes I'll do it on purpose
knowing that, um, knowing that, uh, it's a teachable lesson cuz the way
you think naturally and the way the, the these things work
sometimes is not the same. Now for our audio audience though, could you kind of just talk
about a couple of these, uh, pictures you have on the screen here,
cuz they won't be able to see the, uh, the finder in the revolution pie
and whatnot. Before we get to that, I have a special announcement. I've
just released my very first book. It's an e-book and it's entitled
Logic's Book of Knowledge. What I did is I combined all my control
logics articles from the last 10 years into an e-book. I updated
them, I added new pictures, and I made all available for just $9
and 99 cents. So if you appreciate what, what I'm doing, if you wanna support
the show and the channel and the blog, then please check out my brand new
ebook logics book of knowledge, which you can get
over@theautomationblog.com slash bk one. Yeah, absolutely. Um, this is, uh, a perfect segue because as we've been
talking about the Arduino platforms, a little bit about the raspberry
pie, there's a little bit of a, a difference in the way it started
versus what these platforms are now. Um, so there's a couple of examples
here from the Arduino professional line, the Arduino Pro series, um,
and one of 'em is a brand new, they actually labeled the device as a P
L C, it's called the Arduino Opta. Um, and I'd encourage you to go check
all these ones out just for research. Obviously not everybody's
gonna run out and buy one, but just to be able to
learn what's out there. The Arduino opta is kind of the
equivalent to some of those very small programmable logic
controllers. Um, that, um, I don't remember any of
specific part numbers, but Rockwell had one and then
Eaton had one as well. Um, but just a few tiny little IO
points, all 24 volt compatible, um, power supply input, a couple of
relay controlled outputs. Um, and then there's a much larger version
that they've called the Arduino Portent Machine Control, the P M C, uh, now that one's a little bit
different because it's based on first the typical Arduino chip
set. So we've got the, the ability to be able to run digital
inputs and outputs with your c plus plus programming. For those who have used
the Arduino I d e in years past, that's what you, you would
expect with any Arduino, like the Arduino UNO or
Mega. But alongside that, there's an actual operating system
that's based on Linux. Uh, and, and we'll get to that in just a second
because this is kind of where the, the conversation is going with
taking those abilities to interact with sensors and discreet hardware
and then elevating that to a more, what we'd consider a
computer or an IT level now, including the raspberry pie
in these, in this discussion, there's a few examples of some different
form factors of companies that have included a raspberry pie as
the driver for the devices. So it's still got a lot
of the inputs and outputs, digital analog with ad converters, um, but you notice a few more USB and ethernet
ports on those devices because the Raspberry Pie is legitimately a
small computer on a board with access to inputs and outputs. Being able to build these with
a little bit more functionality. Now they can network
with other devices. Um, sometimes we can even add modules
right onto the side, kind of like a, a compact logics Rockwell system for
those that are familiar with that. Um, and so with that expandability, suddenly we start building what
becomes a more familiar plc. But since the Raspberry Pie runs
that Linux operating system that I've found, myself included, that's where, where it starts becoming a little
intimidating. Like all of a sudden, oh, operating system, developing apps, that kind of language starts showing up
and it starts moving a little bit away from what we call a PLC and turning
more into a computer where I, and a lot of other controls engineers
probably don't understand a lot of that language. Um, so that's been a study
of mine over the past couple of months, um, including at a couple of recent trade
shows is seeking out engineers that are working in this space and learning a
little bit more about how does it look, how does it feel, uh, how does this actually interact with
these engineers at the shop floor? How does this development work? Do you, do you have to be a computer science
major to use these things? What, what kind of schooling
am I gonna need? Um, so that's something that I sought to
learn a little bit more about and have really enjoyed learning more
over the past couple of months. That's awesome. I just, just for the
audience, the opta, the first, uh, P l C we were talking about,
that's, uh, based on the Arduino, it looks like a, uh, maybe an Alan
Bradley Pico or a Siemens logo, you know, that type of very, uh, or might be a micro rate
10 without the display. So it's one of those small very,
uh, squareish, um, uh, you know, super nano PLCs. And then the Revolution
Pie that's kind of in the middle, that kind of reminds me of a micro logics
or an S seven 1200. Right. And, uh, just, just to kind of paint the picture
for those folks listening and, uh, if you guys think it looks like something
else, the guys who are watching, please chime in in the comments and tell
me what you think these guys remind you of. But I'll let you, uh, let me
turn it back over to you, David. Well, thank you. That Pico Logics, that was the one that was in my head
and I couldn't bring it to mind, so you just got it. Yeah.
Uh, the Rockwell Pico Logics. . Yeah, I, uh, I
always avoided the Pico. I am an now a fan of the
microwave hundred, but, uh,
yep. Now that I teach it, I've really kind of really like it
now. But the Pico, I always, uh, managed to stay away from. It. . Yeah, I came into
the world of Rockwell PLCs and, and never got exposed to the Pico. My first dive was a bunch
of control logic systems, and now I have a couple of compact logic
systems and one of the 800 s and have a whole handful of the mic, the
old micro logics one thousands. But the Pico was kind of the one
I never, never got to play with. It's hard, it's hard to
learn, you know, because it's, it's hard to learn something new. It,
it's time consuming. It really is. So all of this talk about
Arduinos and raspberry pies and, and that time period where they,
they made a lot of fun projects, but I included a picture here. This
is just a, a picture that I got of, of a project that I was working
on a while ago. And, um, for those that are audio only, it's
just a, an Arduino board, little, little green chip with
some parts on it. Um, and it's connected to a breadboard with
some LEDs, some buttons, some resistors, bunch of wires. And that circuit was actually built
to replicate some ladder logic. But if you look at that circuit, there
is really nothing industrial about that. That hardware is not the kind of hardware
that you're gonna see running on a factory floor. So it's fun projects to build
and some fun things to learn, but you're always working
in an environment that's
never going to look and feel like a controls engineering environment. And that's a big obstacle for any
sort of application or training. It kind of limits you to just
the world of, yeah, this is fun, but how is this actually going
to apply to industry? So that is, that's the big hurdle that needs to
be overcome with these platforms. And so that kind of
brings me to what we've, what we've sort of
already talked about. Um, but I think it bears repeating
in that these little PLCs built on Arduino's, raspberry pies,
that hardware is only one element. The software being able to create
an environment of programming that has the look and feel of
not just the structured text, but all of those other i e
c languages function block programming, ladder logic, particularly
for those that are more in the, the hardware troubleshooting arena. If we don't have a way to be able
to program these boards using those languages, then there's really not
gonna be any industrial adoption. Um, and I know Jean, you just
mentioned that on a recent podcast, you'd had somebody who was able
to talk about open plc, uh, which is a great chance to be
able to bring a low-cost board to a programming environment that can teach
students about what does ladder logic look like, what do those standardized
symbols feel like? Great for high schools, it's free software. So
even a homeschool, that's, that's a perfect situation to
be able to first get a taste of ladder logic programming. Um, but the deeper you get into
most of these free software, you start running up against
the limits of education stops and industrial use begins. Uh, and so know that's, that's a real challenge there
is being able to get past that, that point of education
and industrial adoption. You know, one of the things
I think if you're not, if you're listening or watching and
you're not familiar with either Arduino or the, uh, the, um, raspberry Pie, you know, I think one of the things that we learned
in the open that I learned in the open plc uh, podcast was when
you look at the Arduino, you're really looking at a microprocessor. When you're looking at the
Raspberry pie, think of a small, inexpensive Lennox pc
because you can actually run a desktop and you can
browse, I mean, out of the box, you can ha you can do everything
and an extensive PC can do. And, uh, there's really that, that difference
between the two of them. There's a, there's a level of power differentiating
the two different systems. And, um, David, do you wanna add anything
to that? Do you think I got that wrong, or would you back that up? I would say let's devote the rest of
this discussion to talking about that. . Okay. Because actually that, that level of, of deeper use that a computer
can bring as opposed to simply A P L C is really the core of
where technology is advancing. Um, because if you just have the ability to
interface with some digital IO and some analog io, that's great,
but if you think about it, that's really what a computer
is already doing. Um, when you plug something into a USB port, it's recognizing discreet
signals coming in. They're just coming in
rapidly changing really fast, and it knows how to translate them.
So mouse, keyboard, things like that, the computer really already is set
up to be able to handle digital inputs and outputs. So if a PLC can do the same thing, then really a PLC is simply a computer
just with a very specific task, and it's designed to just do one thing
and to do it really well and really reliably. Now, as we talk about education, just really quickly before we
move to the, the I P C part of it, just a few examples and our
site control.com is just one place where you can find examples when
you're talking about Raspberry Pioneer. Do we know there are examples of how to
learn systems all over the place, um, including your, your previous
podcast about open plc? And that was exciting to hear
because we actually have a, um, an author who's a college professor
who wrote us a series of articles, um, how he uses open PLC
to teach ladder logic, both software and hardware
interfaces in his college courses. Um, but then also the Arduino PLC I D
e and recreating ladder logic if all you have access to is
structured text programming. So there's a lot of different avenues
that you can take with education, but I feel like we've gotten pretty
squared away with using Arduinos to teach and then moving it more
into industrial use. If we could start comparing how to
use these platforms in an industrial way, but compare them to the way
that we use traditional PLCs, I think that would help a lot of
engineers understand the context, because that's really what what engineers
are looking for is how do you use this the, not just the technical
specifications, not just what CHIPS sat, does this thing use that, that's not
really helpful information, but saying, here's how you use the, the
systems already in a facility. Now let's look at an equivalent use
scenario of like a raspberry pie or a non Arduino P L C and start
to draw the parallels between how those are used. And that brings us to that
interesting trend of IPCs. So as I was looking at some of
the specifications for these, and, and here this, this word that you, you
see on the screen for those of you, the, the basics basis of what that
Arduino Porta machine control, what that's built on is
their porta H seven core. And when you're looking
at the data sheet for it, you start seeing that that
processor is running in that case, an a R m or arm embed
operating system. Uh, there was the Raspberry
Pie systems, those ones, it says an operating system used on
those ones as a customized version of raspbian, which is an operating
system based on Linux. So that, that change is that we had
systems that were designed to create a task-based
environment like a plc, but they're built with operating systems. So PLC engineers could
understand the PLC task, the machine interaction part of it, but those that are more involved
with the IT and the OT end, they can understand the app development, installing and running
programs in more of a, an open source computer
system. But here again, we're still talking about
Arduino and Raspberry Pie. So in the back of my mind,
I'm still thinking, yeah, these are still kind of hobby boards.
This is cool I guess for home projects, but realistically, are we gonna start sticking a board
with an operating system like this into an industrial process? So I think that at, at this point, just talking
about these open source systems, that's still a holdup that I think
engineers would, would have as a concern. But then in some recent trade shows, um, I was able to go to ATX West
down in Anaheim recently, which was an awesome show. If you haven't
gotten a chance to go to trade shows, they're super fun just to see all the
new kind of stuff that's out there. Uh, but I got a chance to set up some
meetings with a couple of companies, and I started seeing a lot of parallels. What these little hobby companies are
doing is not something unique to the hobby industry like Arduinos and Pies. Um, I've got a couple of
examples up here that again, I didn't encourage you guys to check out
just for research sake if you haven't already. Um, one of 'em is Bosch Rex Roth, and they have a platform
called Control X Automation. And it's designed to be a
system that can include and work with not just the, um, the
plant floor level devices, but also be able to work with
cloud computing and edge computing, because there's a lot of applications
that are needing that kind of data access these days. And I pull out
of the data sheet there, it says Control X core is
completely based on Linux, which to me that sounds a lot like
these raspberry pie things, huh? Um, beckoff, they have their CX line
of embedded PCs operating system, windows embedded compact, seven
a Windows operating system, um, Rockwell Automation, of course, most automation users know
about their offerings, but they have an industrial pc, an I P C, and it says Microsoft Windows 10
i o t. That's an operating system. So this is not just something
that's showing up in the Arduino and raspberry pie environments. This is something that's being driven
by a lot of the large automation companies. So that again, is why I wanna to turn a lot of my
attention towards understanding what are they trying to do with
this? How does this work? What's different between PLCs
that I used and was familiar with versus the industrial
PCs? And I know Sean primarily, a lot of your audience is more in
the, the PLC programming environment. Would that be a, a correct assumption? Yeah, yeah. Matter of fact, on the morning
show is sharing the beckoff, uh, uh, IPCs industrial PCs, which look
just like a plc, like an L one or a, you know, maybe a WAGO
or a Phoenix plc, right? And I was sharing with them
because I thought that was, it was pretty amazing that, that
they're running these as PLCs, but really at their heart they're
an IPC and industrial pc. So, um, and you know, I think there's
a lot of pros to that, but there's also some cons to that
because you're not gonna hack an embedded operating system, right? But
if it has the word windows or Lennox, then yeah, it's, it's hackable. So I, there's pros and openness and
there's cons with openness, right? Yep. Absolutely. And, and cybersecurity
is a huge thing. I mean, we, we look all the time at, at some new,
somebody gained access to some system and, you know, usually it's, it's kind of a scary thing when it's
like a bank or a financial system, but certainly somebody having
unauthorized access to, to your data or even worse
control over your system, that would be catastrophic. So
absolutely, security is a big thing. But, you know, even before we get to the point of
talking about security in a discussion, the first question is, can I even use
this? What does this even mean to me? Um, and I read a lot of data
sheets. What, what is an I P C? What does an I P C do? What's
different? Um, but I, I'll tell you, I did not really understand the
difference until recently when I actually got to meet with some of these
companies and get to see, uh, a hands-on walkthrough where I got
to ask some of the questions that me as a, as a skeptic for the most
part with some of this technology, what am I curious about? What,
what would concern me about it? Uh, and what kind of answers
could they offer? So I, I've got just a few of
those in here that, um, even though I can't in the, in
a, in a podcast kind of environment, I can't share with you the same
hands-on experience that I got, which would be great. But so far
audio technology doesn't allow us to, to do that. But I can at
least share a few of the, the insights and inputs that I got from
some of these industry leaders. Um, so you mentioned that you had, you
talked about it, um, during your, your other morning show. Um, so maybe it would be fair at this point
to, just as a survey for those who, who aren't familiar yet,
what is an industrial pc, we've covered little elements of it, um, but I thought it would be a nice to be
able to sort of compare how would a PLC programmer approach a programming
process versus how would an I P C programmer approach the same
kind of programming process? Um, and all PLC softwares are a
little bit different. So I, I guess I can only generalize
this process so much, uh, because everybody's gonna have a
little bit of a unique approach. But usually with PLC programming,
um, you're going to use a, a laptop or a desktop that might be in a
shop somewhere that's connected to your PLC network, and you need
to open the software for it, t i a portal logics or Studio 5,000,
whatever it is that you're using. Um, and sometimes you might connect
to the PLC beforehand, but, uh, at some point you're gonna be
making the edits in the program. Uh, but it's a separate step to connect
the PLC and download the program. Because you're programming computer, the PC is different than the plc. You can't program the
PLC on the PLC itself, and your PLC is just one system. All
the external devices, maybe you've got, um, you're, you're trying to
log data into a local database. And I've got a couple other examples
here that might not be as common, but like sending data up to a cloud for
a data dashboard using like an M Q T T gateway. Not everybody is gonna
have all of those applications, but it's becoming a whole lot more common
to have things connected to the edge or to the cloud for storing and logging
data, looking at machine anomalies, uh, being able to just see when faults are
occurring so that you can quickly respond and clear those faults.
Um, a lot of advantages, but it usually requires a whole
bunch of external equipment. So your P L C is just handling
the tasks at the device level and some other computer
somewhere is shipping and
receiving that information off to other onsite or off-premises storage systems, servers, clouds,
things like that. Um, so some people obviously
have more, more of a, a need for those kind of
applications than others. But if we change that around
and look at what does the I P C, this industrial PC
programming process look like, we kind of cut out the step of
connecting to the PLC because we're creating and editing the program on
an operating system that's already on the plc. So our PLC functionality
is still there in an I P C, but we're opening an operating system
and instead of connecting your laptop and opening the software, you're opening
the programming software, right? On the I P C itself. When you make
the edits and verify the edits, those edits and changes are already
taking place on the p so that you can monitor, monitor tags,
look for faults, uh, and that doesn't require the
use of an external computer, which removes a lot of the, some
of those security issues. Um, what kind of computer is
somebody connecting to? It is this unauthorized access. Uh, there's a lot of security issues. The more pieces are involved
that make connections. So that was kinda one of
the original design intents, is somebody had to bring in a more
sophisticated computer to program the plc. What if we could just do that
programming right on the PLC itself? So that to me was kinda the
first eye-opening difference, is that programming, the PLC itself is not a different system, it's just how we access
and connect to the plc, which is on the PLC itself. So now you're programming
environment is actually a, um, you connect your keyboard and
mouse right to the I P C, um, you connect your monitor
right to the I P C, which is why very often you'll see
H D M I or D V I connections on the plc, the I P C itself, you'll see a lot of u USB ports so that
you can connect your peripheral devices. And those are things that
traditional PLCs didn't have. The only USB port that they would
have was one that was for accepting a download of a program, not for connecting a keyboard and
mouse to treat it as a computer. So seeing how somebody would approach
the PLC programming environment for me was kind of a first eye-opening
step in how approachable this is to somebody with a PLC background. Now I'm gonna take the other side of that. I think some of the listeners
may be on this side as well. You never want to program the p you
don't wanna plug your keyboard at mouse and monitor into the plc. There is never, in 99% of the applications are
maybe 90% of the applications. The PLC is someplace where you do not, when you're not gonna
do your best thinking, where you're not gonna be the most
comfortable. And, and in many cases, you're gonna be just up in
full OC flash gear . So, so there's that side of it as
well. Industrial applications, typically the PLC is in
with all the control system. And it is, it is, uh, it is not someplace
where I would do my best programming. I would just say that. And nor is it a
very comfortable place. So, you know, again, if you're in a clean area,
you're working in a very, very, um, you know, like a pharmaceutical,
you know, they're very quiet. They're very clean.
Everybody's already dressed up. You can't even get into the building,
building without putting a hand head on, right? So yeah, maybe that's not so bad, but I think in most of the applications,
maybe that's not a big selling point. That's true, that's true. Although it also is a little bit of a
change in the architecture of the system as well. Um, you mentioned having the control system
being located at like the machine control center, um, inside the panel
close to the center of the machine. And, and that's certainly the case for a
lot of PLC applications. Um, in fact, one question that I had that, that I was posing to the
engineers on site was, is, is this something that's gonna replace
PLCs and really not for the foreseeable future? Cuz you're right
in a lot of systems, yeah, you're not gonna drag around your keyboard
and mouse and monitor so that you can set up a little table right next to the
control cabinet and plug all this stuff into it. Um, and plus, just
for the security sake of it, if you've got USB plugs
on the front of your plc, does that mean somebody could
plug in and unauthorized, uh, little thumb drive with a virus on
it and all of those? Absolutely, yes. But now a lot of the, the I P C systems are
designed for the control system that uses on machine io, so like a remote IO system
that's attached to each machine, which means now that the network can be
controlled from that more comfortable workroom environment, all of the
motion control devices, the VFDs, the motion controllers, those are what's contained in the machine
control centers where the central I P C, or it could even be a plc,
uh, now that it's more of a, a process automation control
system can be stored in the more comfortable control system room
where somebody can sit down in a, in a much easier environment
to think and process. So that monitor and that keyboard and
that mouse connected to the computer are still in the office environment, but the computer is now
connected to the remote io, all of the, the, the machine centric IO collection points
that are distributed to all of the machines. So it's, it's a bit of
a shift in architecture as well, which is why I think that it's, it's certainly not gonna be something
that suddenly replaces all of the PLC applications. I, I think that's
certainly a fair statement to say. And I, and I'm sure somebody's thinking, well then we can just remote
connect to the p c from our desk, and it's just like, we're there
vnc? And I'm like, yes, and you can, and guess what? If you can VNC
into it, soak can a hacker. So we're back to losing the
advantage of having it, you know, of having it out there in the
plant floor in a lock box. If I can remotely do everything without
even having to buy a expensive license of software from a, uh, reputable, uh,
you know, then it's kind of like, well, if all I need is vnc, I can
access your entire plant floor. There's a security issue
there too. So yeah, I could definitely see
the pros and the cons. And it seems like the endless battle
of that endless tug of war between security and physical security and
network security and openness and closed systems and just, uh, just, I
don't know if it'll ever end. I think it'll just keep on going. Well, yeah, because the
problem is, it goes, the, the further something goes in capability, it's the more open it's going
to be to security issues. So the only way to make something truly
secure is to put it in a little locked box with no network connection. But I
think most people would kind of feel like, well that's, that's not really a
direction we can go. Um, so you're right, it's always gonna be a battle of
the more capable we wanna make it, the more risky it's going to be. Um, and I think IPCs have more
of an appeal to people who are looking at using their
data across larger systems. So when we're talking about using
like an AWS cloud database, um, or, um, sharing the, sharing some
of the tag temperature data, process data to a, um, to a database to be able to look at
long-term trends and data log much longer than what the PLC itself can do, then we already have a network
connection and we're gonna have external equipment that the PLC will have to
talk to so that we can ship that data off to another system. So I think that's really where most of
the companies who are developing these IPCs, that's the market
that they're looking for, is those ones where people are
looking to use their data in a more widespread optimization environment
for looking at preventive maintenance situations. Um, whereas the folks who are
looking to keep a system very secure for a very small application, this I P C is probably a lot more
than what they're looking for. And the risks far outweigh the benefits. Yeah, I can definitely see in
two in a r and d situation, having your P L C on A I P C where you
could do run a bunch of other stuff would be, you know, extremely
helpful. And, and you know, now we get into digital twins and we start
talking about all that. So, you know, it really is application
specific. I think, you know, most people are gonna make the decision
based on the application they're using it in. But let me lemme
turn it back over to you. Well, yeah, I think the application specific is really
important because if you look at most of the big companies, you'll notice
that they sell PLCs and IPCs mm-hmm. . So if the
company was saying, well, the I P C is better in all situations,
they wouldn't sell the PLC anymore. So I think that kind of speaks for itself
when they recognize that an I P C as interesting and intriguing as it is, it's not going to be an
application for all situations. Um, but I also have learned that it's not
really safe to predict technology. So if we were to say, well, what's it
gonna look like 10, 15 years from now? Oof, that's risky. I don't know, I
don't know what it's gonna look like, but it's still interesting
to look at how this feels to somebody who has these same kind of
concerns. And I put together here, just like I said, a few of the questions
that I had as I was getting kind of a, a walkthrough of, of a sample of
creating a P L C program and accessing a, a couple of access controllers, um, and some distributed IO systems. Um, so just thought maybe a, a few
of these questions that I had, that might be some of the same
questions that people might ask. Um, and since my background
is in ladder logic, that was kinda one of the first things
that I wanted to know is you're talking about this like a computer, is
this still run ladder logic? Like, is somebody who understands more of
the electromechanical versions of these things still gonna understand
it? And instead of creating, creating a program and
downloading it to the computer, you simply double click on the software
and you're opening your PLC I d e directly on the machine. So yes,
the PLC software is still there. You can still see it and interact
with it just like you would've felt comfortable in a typical plc.
So aside from all of the, the security issues, that was
my very first question is, Hey, if I can't use this thing,
it's not the right fit for me. If I don't understand how
the programming works, then all the security in the world doesn't
matter because I can't really use it. So that was a very nice
thing to learn at first, is that all of my familiarity with ladder
logic and how to use the laptop and open the machine interface with the
plc, all of that is still the same. It's just how you get there. So that was a really good
thing to know just immediately. Now another question is remote io, um, or other types of peripheral equipment. So a lot of 'em will be running on
mod bus or like an RS 2 32 4 85, and most computers that
I've ever used, my laptop, although it has an ethernet port, it doesn't provide support for
any of these other protocols. So that's something that is different, but well thought out in an I P C is
that almost all of them that you see will have the ports that are required
to connect to all these different types of networks. Um, and they support all of the different
protocol that somebody might need so that it can be universally connected to
pretty much anybody's remote IO or motion motor control type situation. And then another question
is local io, um, your, your concern earlier is certainly valid
in that if this is gonna be something that's stuck in a control cabinet
where the local io the, the, the IO connections are
supposed to take place, then maybe an I P C isn't the right fit, maybe remotely connecting to your
PLC at that machine is the right fit. So very often these IPCs are not designed to have local
IO attached to them, but in many cases they are, they'll have either built in
embedded io kind of like a, um, some of the compact logics series
have the embedded io, um, or the, the Siemens S seven s, the 1200 s, but a lot of them are just the computer
processor and the connection to the computer processor and
the network ports from there, it distributes out to remote io. So these are definitely machines that are
intended for a scenario where you have a central control center with
the distributed connection to all the IO and the
equipment around the facility. And certain people have that, that kind of architecture to their
facility and others don't. They have the, the control right on the machine itself. And I had some other questions. We've, we've talked about the security because
that was certainly something where once I felt like maybe this would be
approachable for me as an engineer, what about the security?
Anytime you connect something, as soon as you start saying
network or cloud, especially, there's a lot of confusion there.
What does that mean? How secure is it? Can a hacker gain access to it? Um, and something that's been pointed out
repeatedly is if somebody is designing their own system to be able
to have access to the data, then the, the IT team at that facility
is gonna be 100% responsible for building all of the security support. But if a company has designed a system, an operating system
with built-in security, kind of like a Microsoft
operating system, um, then they have the ability to keep
up to date on all of the latest security patches, uh, I don't think that ever means that
it's a hundred percent secure. I think there is no computer system
that's connected to the internet that is a hundred percent secure. I, I just,
I don't believe that can happen. But if a company is providing
continual up-to-date support, then that can be a whole lot more reliable
than a lot of companies that really lack the IT support to
build it in-house. Um, so as far as security goes,
everybody has that as a concern, but it's nice to know that the
big companies can support that. And the more I was exposed
to some of these systems, the more this this question of do
you need to have a background in programming? Do I need to be an
expert in Python because I'm not, I've used some Arduino stuff before, so I know understand c plus
plus a little bit, but, but as a controls engineer thought, I, I don't wanna have to
learn to be a programmer, but if you can simply download
programs, all of the apps are available, sort of like a computer has programs
that you can go to the internet, you can download a program, you can install that program on your
computer and you can run that program that doesn't take programming skills. And that's primarily what these
industrial PC environments are built around. There's a set of applications
that are supported by the company, be that Rockwell or
Beckoff, Bosch, Rex Roth, they have a central storage of
apps that work with this pc. You purchase and download the app and
potentially a license that goes along with it, not unlike a, a license that you'd have on a
laptop for a programming software. And if you can download the program, then you can download
the program on an I P C. But most of these environments
also allow a team with knowledge to be able to create their own custom apps
to interface with things and run it on the same machine. They have
the ability to do that too. So if you don't have a
programming background, it's fine. If you do have a programming background, you've just unlocked more
capability of the system, uh, and you can even submit some of those
applications to the company in hopes that it might help somebody else
who's in a similar situation. Um, and just one final question
that I posed to them, um, is whether the companies who are
developing these feel like this is going to replace PLC architecture and the kinda resounding consensus that
I got is in the short term, no, but in the long term, these
computers really are a plc. They're just a PLC with more capabilities. So there will always be cases where a
small system is going to rely on a small dedicated logic controller that
has no connection to the outside world because it doesn't need that,
it doesn't need the extra complexity. But a lot of the systems that have three
or four computers all connected to a plc, each performing their own
task to share data can easily be consolidated into one I P C system. But I think as it is with all technology, the more functions that
something provides, the more confusing it is and
therefore the slower adoption that it's going to be,
but for good reason. Because these are all very valid
concerns that people have. And, um, I would share some of the same
concerns and would never say, oh, here's a new technology, let's
immediately replace our system with this. Yeah. You know, it's, it's, it's so hard
to predict the future in technology. I remember back when there
was a report saying, you know, a new co a company came out with a
new micro camera to go on a phone, and this is before Facebook,
this is before the iPhone. And I'm kind of thinking like,
who wants that? And then , here we are today. I think the number one camera almost
everybody uses in their life is the camera on their phone. I mean, in their, they've
just grown so fast and gotten so good, you know, that um, that, you know,
it's hard to predict cuz you know, if you didn't know about
social media exploding, if you didn't know about the popularity
of iPhone and Android, you'd be like, well, what are you gonna
do with that? You know, text messages are already an
nominal leg. I'm gonna send a, a picture over a text message. But, um, and so those are the things we don't
know what the future holds. Um, but, um, you know, and, and so we'll just,
and I don't think anybody does, we just have to, we have to go, we have to go down there on this journey
and we have to watch and and and see how these things turn out. But I
think you did give us a lot, you know, sharing your experience, talking to
these vendors, playing with it yourself. I really think you did give us a good
idea of where the industry is today and, um, how these systems are developing
and I really appreciate that. I really appreciate you coming on this
show and sharing all that with us. I, I love sharing, I don't
know the answers to things
I don't, like you said we, it's impossible to know. Probably the first time somebody explored
putting an ethernet connector on a PLC and connecting it to the internet,
people probably just had a fit. You can't connect this kind
of stuff to the internet, but now we do it all the time. So things
have slower adoption. But like I said, it's for very good reason. Um,
people's concerns are not unfounded, they're not invalid. Um, but
they do need to be addressed. And if they can be addressed
securely and effectively, then the efficiency of systems, the
ability to keep things running longer, the ability to flex your system
if something major happens, can you turn around your process and do
something better in a short amount of time? Having access to your
data is really important. And I don't think anybody
would argue with that. It's just how it's done and what's the
right context and what is this new thing you're developing, what
are you trying to sell me? Those are all very valid concerns and
it's fun for me to just be able to learn more about 'em and share more as I attend
trade shows and meet people with more experience than I have. Well, David, I really appreciate coming on the show
and I hope I see you at some of those trade shows this year. But, uh, with that, I just really want to thank you for coming
on and, uh, sharing all that with us. Really appreciate it. Thank you, Sean. I really
appreciate it. And thanks to your, your audience and your
viewers for joining us today. Well, I hope you enjoyed that episode and I
wanna thank David for taking time out of his busy schedule, the come on the show and share his
experiences using PLCs based on Arduino and Raspberry Pie. Now
if you did enjoy this episode, please consider giving
us a like and a sub. And if you wanna follow us and keep
track of everything we're doing, you can do so over@automation.locals.com. You'll also find all of my training
courses over@theautomationschool.com. With that said, I wanna wish issue all
a very happy, safe, and healthy week. And until next time, my friends peace.