OTee Virtual PLCs (P252)

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Thank you for tuning back into the automation
podcast. Sean Terry here from Insights. And this
week on the show,
I meet up with Hendrik Peterson
and Jacob Abel to learn all about
virtual PLCs from OT. That's o t e
e. And, I just thought it was very
interesting. So if you guys have ever thought

(00:21):
about maybe running virtual PLCs to test some
processes out, I think you'll really enjoy this.
With that said, I wanna welcome to the
show for the very first time, Hendrik and
Jacob. Guys, before we jump into your presentation
and learn more about what you do, could
you first introduce yourself to our audience?
Yeah. Sweetly.
So my name is Hendrik. I am the

(00:42):
cofounder, COO, OT,
a new industrial automation company,
that, we're really glad to present here today.
I have a background from ABB. I worked
eleven years at ABB.
In terms of education, I have an engineering
degree and a master degree in industrial economics.
And, yeah, I'm I'm excited to be here.

(01:04):
Thanks, Rom. And I'll pass it over to
Jake.
I'm, Jacob Abel. I'm the principal automation engineer
at Edgnot.
EdgeNaught is a systems integrator
focusing on edge computing and virtual PLCs.
My background is in mechanical engineering, and I'm
a professional control systems engineer, and I have

(01:25):
thirteen years experience
in the machine building side of industrial automation,
specifically in oil and gas making flow separators.
And I'll hand it back to Henrik here.
K. Great. So OT, we are a a
new industrial automation company,
the new kid on the block, if you
will. We're a start up. So, we only

(01:46):
started,
about three years ago now.
And,
we focus solely on virtual PLCs
and and the data architectures
allow you to integrate virtual PLCs in in
operations.
And, you know, some of the listeners
will be very familiar with this first,

(02:06):
thing I'm gonna say, but I think it's
valuable to just take a take a little
bit step back and and remember what has
happened in
in history when when it comes to to
IT and OT and,
and and what really
what really happened with that split. Right?
So
it was probably around the '9 you know,

(02:27):
around nineteen nineties where
the the the domain computer science were really
split into these two domains here, the IT
and OT.
And, and that, that was, that was kind
of natural that that happened because we got
on the, on the IT side of things,
we got Internet,
we got open protocols
and, you know, we had the personal computers

(02:49):
and innovation could
truly flourish on the IT side. But whereas
on the OT side, we were we were
kind of stuck still
in the proprietary,
hardware software
lock in situation. And and that has that
has really not been solved. Right? That that
that is still kind of the the situation

(03:09):
today.
And it this is what this is obviously
what also,
brought me personally to
to really got really super motivated to solve
this problem and and really
dive deep into it. And I experienced this
firsthand with with my role in NAD and,
how how extremely locked we are at creating

(03:32):
new solutions and new innovation on the OT
side.
So so we're basically a company that wants
to to truly
open up the
the the innovation in this space and and
make it possible to adopt anything new and
new solutions,
that that sits above the PLC and and,
you know, that integrate effectively

(03:54):
to to the controller.
So I I have this this,
you know,
this slide that kind of illustrates this point
with with some some, you know, historical
events or or at least some some some
big shifts that has happened.
And,
Aurene mentioned a shift in nineteen nineties.

(04:15):
And it wasn't actually until
'20,
2006
that Gartner coined this term OT,
to explain the difference really what what has
happened.
And and, you know, as we know, IT
has just boomed with innovation
since since the nineties and OT is, is,
is slowly,

(04:35):
slowly incrementally
getting better, but it's still, it's still the
innovation pace is really not, not fast.
So, this is also, of course, illustrated with
all the new developments in in GenAI and
AgenTic
AI, MCP, and things like that that is
kinda booming on on the IT side of
things. And and and yeah. So,

(04:58):
but we do believe that
there is actually something happening right now. And
and we have data that they're gonna show
for for that. Like, the the large incumbents
are now working on this as well, like
virtual PLCs, software defined automation and all kinds
of exciting things going on on the OT
side. So we do believe that that we

(05:18):
will see, we will see a shift, a
true big shift on the OT side in
terms of innovation,
really the speed in which we can, we
can improve and adopt new solutions on the
OT side.
And this is kind of exemplified by, like,
what what is the endgame here? Like,
you could say that the endgame could be

(05:38):
that IT and OT once once again becomes
the same high paced innovation domain. Right. But
then we need to solve those underlying
problems, the infrastructural
problems that are still so persistent on the
OT side of things.
The fine point of this slide is to
just illustrate what's happening right now. It's like

(05:58):
cloud solutions
for control is actually happening.
Virtual PLC, software based automation,
AI is happening all at once.
And we see it with the big suppliers
and and also the exciting startups
that's coming into this space. So I think
there's there's lots of great excitement now that
we can we can expect from the OT

(06:20):
side, in in next few years.
Yeah. You know, I wanna just, just for
those listening, add a little,
context here. If we look at 1980, why
was that so important? Why is this on
the chart? And if you think about it,
right, we got networks like Modbus
and,
Data Highway in nineteen seventy eight, seventy nine,

(06:41):
eighty.
We also got
Ethernet at that time as well.
And so
we had on the plant floor field buses
for our controls,
but in the offices, people were going to
Ethernet.
And then when we started seeing the birth
of the public Internet, right, we're talking about
in the nineties, people who are working on

(07:01):
the plant floor, they were like, no.
Don't let the whole world access by plant
floor network. And so I think that's where
we saw the initial
the the divide, you know, was 1980. It
was a physical divide, just physically different topologies.
Right?
Different needs. Right?
And then and and as the Internet
came out in the early nineties,

(07:23):
it was it was now like, hey. We
need to keep us safe. We know there's
something called hackers on the Internet.
And and I think that's why, as you're
saying in 2006,
when Gartner,
you know, coined OT,
we were seeing that there was this hesitant
to bring the two together because of the
different viewpoints and the the different needs of
both systems. So I think it's very interesting.

(07:45):
I know you listeners, you can't see this,
but I kinda want to go back through
that and kinda give some context to those
early years. And
and, you know, like Henrik says, you know,
now that we're past all that, now that
we're using Ethernet on the plant floor everywhere,
right, almost everywhere, on all new systems, definitely,
that that becomes the right now on this

(08:06):
on the today on the, on the chart.
And I'll turn it back to you, Henrik.
Yeah. I'll search that. I just wanna echo
that as I think that there are really
good reasons for why this has happened. Like,
the there has
you could argue that innovation could flourish on
the IT side because there was less critical
systems, right, less,
more, you know, you can do to fail

(08:28):
fast and you can do,
you can test out things on a different
level. And so so there's really lots of
good reasons for why this has happened.
We do believe that right
now there is some really excitement around innovation,
the OT side of things and and this
pent up kind of, I wouldn't call it
frustration, but this pent up potential,

(08:49):
I think is the right word,
is is can be kind of unleashed in
our industry for for the next, next decade.
So so we are like this is really
one of the key motivators for me personally.
It's, like, I truly believe there's something truly
big going on right now. And and I
I do I do encourage everyone, everyone listening,

(09:10):
like, get in get in on this. Like,
this is happening.
And, you know, be an entrepreneur as well.
Like, build your company, build and, you know,
create something new and exciting in this space.
I think I think this is this is
a time that there hasn't been a better
time to create a new new technology company
or a
new service company in this space.

(09:30):
So this this, this is something at least
that motivates me personally a lot.
So let me move over to kind of
what we do. I mentioned I mentioned that
we focus solely on the virtual PLC.
This
this is now presented in the slide for
those that are listening as a as a
box inside
a open hardware.

(09:50):
We can deploy a virtual PLC on any,
ARM thirty two thirty two and and sixty
four bit processor and x eighty six sixty
four bit with the Linux kernel. So so
there are lots of great,
options
to choose from on the hardware side.
And and,
and yeah.
So you can obviously when you have a

(10:12):
Virtual PLC you can think of it new
in terms of your system architecture. You could
for instance,
you know deploy multiple Virtual PLCs on this
on the same hardware
and you can also, think about it like
you can use a virtual PLC in combination
with your existing PLCs and could work as
a master PLC or some kind of optimization

(10:34):
deterministic controller.
So it's it's really just opening up that,
you know, that architectural
aspect of things. Like you can think new
in terms of your system architecture,
and you have a wide range of hardware
to choose from.
And, and yeah, So the the flexibility
is really the key here, flexibility in how

(10:55):
you architect your system.
That CPU that you deploy on will will
obviously be need to be connected to to
the field somehow, and that's that's true, classical
remote IO,
connections.
So we currently support, Modbus TCP and Ethernet
IP, which is kind of deployed to

(11:16):
to,
our production environment, as it's called. So moving
on to the next slide. Like, this is
kind of the summary of our solution. We
have built a
cloud native IDE. So meaning anyone can can
basically go to our website and log in
to
into the solution and and give it a
spin. And, we'll show you that afterwards with

(11:38):
with Jake.
And the system interacts through a PubSub data
framework.
We use a specific technology called NUTS,
for the PubSub
communication
bus.
And you can add MQTT
or OPC UA to the PubSub framework,

(11:59):
according to your needs.
So, and from that, you can integrate with,
whatever whatever other,
software you might have,
in your system.
So we have these value points that we
always like to bring up. Like, this obviously
breaks some kind of vendor lock in in
terms of the hardware and the software.

(12:20):
But it's also, our virtual PLC is based
on on the six eleven thirty one. So
it's not a lock in to any kind
of proprietary programming language or anything like that.
There is,
there's obviously the cost,
element to this that you can potentially save
a lot of cost.
We have,
we have verified with with with some of

(12:41):
our customers that they estimate to save up
to 60%
in total cost of ownership.
This is
there is obviously one part is the capex
side and the other part is is the
opex.
And
and is this data framework, as I mentioned,
is in in in which itself is is
future proof to some extent. You can you

(13:03):
can integrate whatever comes comes in in a
year or or in a few years down
the line.
And,
there's environmental
footprint
argument for this as you can save a
lot on the on the infrastructure side. We
have one specific customer that estimates to save
a lot on and this this particular point
is really important for them.

(13:25):
And then final
two points is essentially that we have built
in a zero trust based security,
principle into this solution. So we have role
based access control.
Everything is encrypted end to end,
automatic certification,
and things like that.
The final point is, is that this is

(13:46):
the infrastructure that allows you to bring
AI and the classical,
DevOps, the the thing that we're very used
to in the IT side of things. Like,
you you commit and merge and release,
instead of,
instead of the traditional,
way of working with your automation systems.
So I know this is like, this is

(14:07):
pro pretty much, like, the boring,
sales pitch slide, but,
but, yeah, I just wanted to throw this
this out there for for the guys that
there is some there is some, intrinsic values
underneath here.
The way the system works, you will you
will see this very soon, through the demo,
but it's basically you just go to a
website, you log in, you create a project.

(14:29):
In there, you would
create your your PLC program,
test, you code, you simulate.
You would onboard a device. So onboard that
Linux device that you you want to deploy
on. This can be as simple as a
Raspberry Pi, or it can be something much
more industrial grade. This depends on on on
the use case.

(14:49):
And then you would deploy services like, as
I mentioned, MQTT and OPC UA, and then
you would manage your your your system from
from the interface.
And, I have this nice quote that we
got to use from one of the customers
we had. This is a global,
automotive manufacturer

(15:10):
that, basically
tells us that it's, they they highlighted
the speed in which you can set this
up,
as as one of the biggest values for
them, saving them a lot of hours and
setting setting up the system.
So I also wanted to show you a
real you know, this is a actual real

(15:30):
deployment. It was it was deployed about a
year ago,
and this is a pump station,
or a water and wastewater
operator with around 200 pump stations.
They had a mix of of Rockwell and
Schneider PLCs,
and they had a very high upkeep, and
they were losing a lot of data from

(15:51):
these stations because they were connected over four
g.
When the Internet was a bit poor, they
lost a bit of data in their SCADA
systems, so they had these data gaps and
things like that.
So pretty pretty, you know,
standard legacy setup to be to be honest.
Quite outdated PLCs as well.
So what they what they did for the

(16:12):
first,
pump station was they they, you know, removed
the PLC.
They put in a Raspberry Pi for for,
like, €60
or, like, $70,
connected it to to a to a remote
IO Ethernet IP module they had, in in
the storage,
and deploy this data framework as I'm showing

(16:33):
on the screen now.
So so they that was that was the
first station they put online, and they they
chose a Raspberry Pi because they thought, okay,
this is interesting, but will it work? And
then they chose a pump station, which was
was really just poor from before. So they
had very little to to to lose to
to deploy on this station.

(16:55):
So so, yeah, this has been running for
a year now without any any problems on
a Raspberry Pi. We have obviously advised against
using a Raspberry Pi in a critical environment,
but
they just insisted that that what that's what
they wanted to do for this first case.
And I'll back that up too. Your generic
off the shelf Raspberry Pi is just like

(17:15):
a generic off the shelf computer.
It's not rated for these type of environments.
Not that all pump houses are really bad,
but they're not air conditioned. And I think
we've all had that situation when it's a
120,
130 out that, you know, off the shelf
computer components can act wonky as well as
when they get below freezing. So just wanted
to chime in there and agree with you

(17:36):
on that.
For testing, it's great. But if you're gonna
leave it in there, if you were in
my town and you say you're gonna leave
that in there permanently, I would ask to
have you,
assigned somewhere else for the town.
Yeah. Yeah. Exactly. No. So and and that
point is also illustrated with the second station
they brought online. So there they chose a
much more industrial grade CPU,

(17:59):
that,
that, was much, you know, cost cost a
bit more, but it's more suited for the
environment. And,
and yeah. So this was, I can disclose
it was a Bayer Electronics,
CPU.
So so yeah.
And, and they reported,
some good,
good metrics in terms of, like, the results.

(18:20):
They they said around 50 on the hardware,
75%
on the management of the PLC system. So
this relates more to that they have very
a lot of, you know, driving out with
the car to these stations and doing
changes to their systems and,
and updates.
They no longer have any, any data loss.

(18:40):
It's local buffer on the data framework.
They've increased tag capacity with 15 x, resulting
in in four
fifty five x better data resolution
and a faster scan frequency. And this is
actually on the Raspberry Pi. So so just
just think of it as as the the
even the even the,

(19:01):
kind of the lowest quality
IT off the shelf,
computers,
are are able to to,
to execute really fast in in in, or
fast enough for for,
for these cases.
So,
Sean, that was actually what I wanted to
say. And, and also,

(19:22):
you know,
yeah, we are we are a start up,
but we do have,
fifth users
now in 57 different countries across the world.
And it's it's really cool to see our
our our,
our technology being deployed
around the world. And, and yeah.
I'm really,
really excited to to,

(19:45):
to get more,
users in and and hear what they what
they, think of the solution. So so yeah.
I'll I'll with that, I don't know if,
Sean, you wanna you shoot any questions or
if we should hand it over to Jake
for for for a demo.
Yeah. Just before we go to Jake, if
somebody who's listening is interested, this might be
a good time.

(20:05):
It said that, you already talked about being
cloud based. It's, o t e e.
So Oscar Tom, Edward Edward for the the
name of the company. Where would they go
if if they like what Jake's gonna show
us next? Where will they go to find
out more? Yeah. So I would honestly propose
that they just, reach out to to
me or Jake, on on one of the

(20:27):
QR codes that we have on the presentation.
But they can also obviously go to our
website, 0t.io,0tee.io,
and just,
either just,
log in and test the product, or they
could reach out to us, through our website,
through the contact form. So yeah.
Perfect. Perfect.

(20:48):
Alright, Jake. I'll turn it over to you.
Thanks, Sean.
Fantastic stuff, Henrik. I wanna take a second
too to kinda emphasize
some of the technical points that you,
presented on.
Now first, the the fact that you have
the built in zero trust cybersecurity
is so huge.
So, I mean, the OT cybersecurity is blowing

(21:10):
up right now.
So many certifications,
you know, lots of,
consulting and buzz on LinkedIn.
I mean, it's a very real concern. It's
for a good reason. Right?
But with
this,
zero trust built in to the system, I
I mean, you can completely close-up the firewall

(21:30):
except for one outgoing port.
And you have all the virtual PLCs connected
together
and it's all done. You know, there's no
incoming ports to open up on the firewall
to worry about,
you know, that security concern.
You know, it's basically like,
you know, you've already set up a VPN

(21:52):
server, if you will. It's it's not the
same, but similar
and,
you know, taking care of that connection already.
So there's an immense value in that, I
think.
And I wanted to add to the zero
trust. We've covered it on the show. And
just for people, maybe you've missed it.
You know, with zero trust is you're not
trusting anyone.
You authorize connections.

(22:14):
Okay? So by default, nobody's laptop or cell
phone or tablet can talk to anything.
You authorize, hey. I want this SCADA system
to talk to this PLC.
I want this PLC to talk to this
IO. I want this historian to talk to
this PLC. Every connection
has to be implicitly
I'm sorry. Explicitly,

(22:34):
enabled and trusted.
And so by default, you know, an an
integrator comes into the plant, he can't do
anything because in a zero trust system, somebody
has to give him and his laptop access
and access to specific things. Maybe he only
gets access to
the PLC, and that makes sense. Think about
it. Who knows whether his laptop has been?

(22:55):
I mean, we've heard about people plug in
to the USB ports of the airport and
getting viruses. So it's important that person's device
or a SCADA system or a historian only
has access to exactly what it needs access
to. Just like you don't let the secretary
walk on the plant floor and start running
the machine. Right? So it's a it's an
important concept. We've covered it a lot. And

(23:15):
and, Jake, I really appreciate you bringing that
up because zero trust is so huge, and
I think it's huge for OT to have
it built into their system.
Yeah. Absolutely. Absolutely.
I wanted to highlight too the Henrik mentioned
that the the backbone of the system is
running on a technology called NATS. That's spelled
n a t s.
And

(23:36):
why that's important is this is a a
lightweight messaging,
service,
and it's designed to send millions of messages
per second. You know, that's opposed to,
you know, probably the best Modbus TCP device
that you can find. You might get a
couple 100 messages through per second. It's millions
of messages per second. It's,

(23:57):
you know, especially
with,
you know, we're dealing with AI machine learning,
you know, training models.
I mean, we're data hungry. Right? So this
gives you the backbone too. You know, it's
like it can push
an immense amount of tag data, you know,
with ease. I think that's another really important
point.

(24:19):
With that, though, I'll I'll get on to
the demo. Oh, that's great. We do we
do see that, Jay, that most of our
customers report on that, you know, 400
or 700 x better data resolution. And so
it's it's a step change for for for
the data resolution there. Yeah.
Excellent.
So one of the things that I personally
love about OT is

(24:40):
how quickly you can get into the PLC
once everything's set up. So this is OT's
website, obviously, ot.io.
So once you're here, you just go to
log in.
And that brings in the login screen. Now
I'm are I'm using my Google account for
single sign on, so I can just click

(25:00):
continue with Google.
And this brings me into the main interface.
And
another thing that I love is that,
you know, it is very simple and straightforward,
you know, and simple is not a bad
thing. Simple is a good thing. I mean,
the way that things should be is that
it should be,
it should be easy and the finer details

(25:22):
are taken care of for you.
So right here, we have our main project
list. I just have this one benchmarking program
that I've imported in here.
And you also have device lists, just a
a test device that I've installed the runtime
on.
Just real quick. You know, you have a

(25:43):
Martha, the AI assistant in the corner here.
And,
the documentation
guides is up here.
So you can get
help or look into reference material
very easily. It's all right there for you.
So I'm gonna open up this program here.

(26:05):
So just a quick tour here.
Right up here in the top left is
basically
where where most everything's done.
So
if you click on this little down arrow,
you can choose what virtual PLC runtime to
attach it to. I've already attached it to
the device. I installed the runtime on.

(26:28):
You can add,
you know, a new program,
driver, function blocks,
custom data types real quick here.
Compile your program, download it to the device.
Check the release history, which is really, really
great.
As you can, you can go into release
history and you can revert to a prior

(26:49):
version very easily.
We got built in,
version control, which is another,
great feature.
I can also just comment on that, Jake,
that
we do have we do have,
in the quite short term roadmap to also
expand on that with Git integration,

(27:10):
that, a lot of our customers are are
asking for. So yeah.
Awesome. Yeah. I mean, that's that's another,
very
hot topic right now. It's,
you know, getting getting the revision control systems,
as part of,
you know, at least the textual,

(27:32):
programming languages.
See, so,
you know, we have a few, like, housekeeping
things here. I mean, you can delete the
program,
export it.
It's a good good point here is that,
OT complies with the PLC open,
XML specification.
So you can import or export

(27:53):
programs,
in this XML format, and it should work
with
solid majority of other automation software out there.
You know, if you need to,
you want to transition over to OT, you
know, you can export it from your other
software and import it
rather easily.
Got your program list here and,
you know, just the basic configuration

(28:15):
of,
you know, you can add global variables that
you wanna share between the different programs
and POUs
or, you know, change
the,
cycle rate of the periodic tasks,
add more tasks.
Let's just get jump into this program here.

(28:39):
Both the system uses the IEC
sixty one one thirty one dash three standard
structured text.
So here's just a little,
quick benchmark program that I've been using to
do some performance testing.
Like you, you have
the, the code right here, obviously. And on
our, our right,

(29:00):
the variable list,
very easy to
add a new variable and pick out the
type.
You can set a set of default value,
add some notes to it.
Super easy.
So
let's go online. So if you have these

(29:22):
little glasses up here in the top, right,
you display live tag values.
And so it's grabbing from the runtime
that's running and
plopping it right in here in the editor,
which I I love the way it's displayed.
It makes it.
And, you know, it's one of the question
marks is if you're doing structured text instead
of letter logic, like how it's gonna show

(29:43):
up and how readable is it gonna be.
I think the, the text, like the color
contrast here helps a lot. It's
very,
very readable and intuitive. And we also have
the tag browser
on the right hand side.
Everything
is,
organized into, you know, different groups. There's the

(30:07):
the resources
and instances that you've set up in the
configuration tab.
So the by default, the tag the tags
are all listed under there.
And here too, you know, you can set
tag values
doing some performance testing, as I said. So
this is,

(30:27):
recording some some jitter and task time metrics.
And that's that's really it. That's the
that's the cloud IV in a nutshell.
Super easy, very intuitive. I mean, it's there
there's zero learning curve here. For the, audio
audience,
just a little comment here. First of all,

(30:48):
structured text to me
seems to be, like, the
most compatible between all PLCs. So, you know,
everybody does ladder a little bit differently. Everybody
does function blocks a little bit differently.
But structured text and, again, I could be
wrong if you guys think out there in
the in listening, think I'm wrong about that.
But when I've seen structured text and compared
it between multiple different vendors, it always seems

(31:11):
to be the closest
from vendor to vendor to vendor. So I
can see this makes a great a great
place to start for OT to have a
virtual PLC that supports that because you're gonna
be able to import or export
to your maybe your physical PLCs. The other
thing is I wanted to comment on what
we're seeing here. So, many of you who
are familiar with structured text, you know, you

(31:31):
may have an if then else,
or an if then.
And and you may have, like, tag
x,
equals, you know, either some kind of calculation,
you know, maybe, you know, z times y
or just maybe a a constant.
But what we're seeing here
is as we're running,
they have inserted

(31:51):
at a in a different color the actual
value of, let's say, tag x.
So in between you know, right next to
tag x, we see the actual value changing
and updating a few times a second. And
so it makes it very easy to kinda
monitor this thing while it's running and see
how everything's working, and I know that's that's
huge.
And I know a lot of vendors also

(32:12):
do this as well, but I love the
integration here, how it's so easy to see
what the current values are for each of
these variables.
And, I'll turn it over to you, Hendrick.
I think I interrupted you. Go ahead. Yeah.
No. I was just gonna comment on that.
Jake said, like, this is the this is
the POC editor,
and the next the next big feature that
we're releasing very soon is essentially the service,

(32:36):
manager, which is the,
which is the feature that will allow our
users to deploy
any kind of service very efficiently,
like another runtime or OPC UA server or
an entity server or
or or whatever other, software components that that,
you want to deploy, like a Knox server
or things like that. So and that's that's,

(32:58):
we were really excited about that because, that
will kind of allow for a step change
in how you kind of orchestrate and manage
your system and your,
your system and your, your, you have a
very good overview of what's going on with
versions of, of the different software components running
in your, your infrastructure and your devices and
things like that. So we're really excited about
that, that it's coming out. And it might

(33:19):
be that actually when when this,
episode airs,
who knows if it's if it's done or
or not, but we're very close to release
the first version of that. So excited about
that.
Now I have a question for you guys,
and maybe this is off topic a little
bit. So let's say I'm up here in
the cloud. I'm working on a program,
and I have some IO on my desk

(33:40):
I wanna connect it to. Is that something
I can do? Is there a connector I
can download and install my PC to allow
the cloud to talk to my IO? Or
is that something where I have to get
a a, you know, a local,
you know, like we talked about those industrial
Linux boxes and and test it here with
that?
Yeah. So I think you what you what
you're you're after is, like,

(34:00):
the IO configuration
of, if you wanna deploy a driver, right,
or, like, a modbus
driver and how you figure out the system.
Right? Yeah. Because this is in the cloud.
It's not on my desk. The IO is
on my desk. So how would I connect
the two of them? How would I is
is that something that can be done?
Yep. Yeah. Exactly. That's that's actually the

(34:22):
you know, I I think, Jake, you might
just wanna show why you deploy a driver.
Right?
Sure. Sure.
And I just wanna take a second to,
clarify. You know, it's something that kinda comes
up often,
and I I don't I don't think it
gets it's it's cleared up enough is that
so, you know, we have this cloud ID

(34:42):
here. So, you know, you can open this
from anywhere in the world.
But the virtual PLC run times get installed
on computers
preferably
very locally,
you know, on the machine, on the factory
floor, something like that. I I've got,
an edge computer right here.
Just as an example. I mean, this is
something you would just pop in the

(35:04):
control panel and you can
install OT on this.
So
to answer your question better, Sean, you know,
to get to,
you know, the remote IO that you need
essentially, or actually in the, in the case
of this, this has onboard IO.
You know, you're looking at connecting with MOBAs,
PCP, Ethernet IP.

(35:26):
I I know that a lot more protocols
are coming. Profinet.
So how you would do that is
that you have
that plus sign up here
and add a driver config.
We're just gonna do,
Modbus real quick.
Mhmm.

(35:48):
And we wanna add a TCP client. So
you can name the client,
tell it how fast to pull, you know,
any delays,
put in the IP address.
Just an example. Do the port number if
you need and then add your requests.

(36:09):
You know, you have support for,
all the main function codes and mod bus
right here,
you know, read holding, read input,
you know, write multiple coils, all that good
stuff,
you know, tell address
how many registers you wanna do,
timeouts, slave ID.
And then, you know, once you've done that,

(36:30):
so let's say, you know, I'm gonna read,
and holding registers here,
the table on the right auto updates.
You can do aliases for each one of
these.
You can just do

(36:51):
register one
Mhmm. As an example
It's showing just for the audio audience, it's
showing the absolute address for all these modbus,
variables and then, has the symbols, and he's
putting in his own symbol name. It has
a default symbol name of symbol dash something,
and he's putting his own in, like, register
one, which makes it easier. Yeah.

(37:15):
Good point. Yeah. Good point. Thanks, Sean.
So, yeah, once once you put in your
request
and you can throw in some aliases,
for the different registers,
you know, you can go back to your
program
and
here's this,
sample variable that I just added from earlier.

(37:36):
You know, you can
the registers are 16 bits. I'm gonna select,
an int.
And what you can do here now is
select those modbus requests that you just set
up.
So it automatically maps these to those variables
for you.
So that that way you don't have to
do anything

(37:58):
anything manual, like have a separate program to
say, you know, this tag equals,
you know, register 40,001.
You know, it's already mapped for you. So
that's that's essentially how you would connect to
remote IO
is,
just add a client in the driver configs
and,
fill in all your info and

(38:18):
be off and running.
That's excellent. I really liked how you were
able to easily map the register to the
modbus value you're reading in or writing to
to your, variable so you can use that
in your program. That was very easy to
do.
Oh, yeah. Yeah. It's
that it's like I said, that's one of

(38:38):
the things that I love about this interface
is that everything is just very straightforward.
You know, it's it's super easy to just
stumble upon
whatever it is you need and
figure it out.
And just just, to add to to kinda
your your processors, like, once you have created
that connection between the IO and

(39:00):
and and the program, you basically just, compile
it and download it to the to the
runtime again, and and it executes locally the
based on the yeah.
Nice. Oh, right. Good point. Yeah. Of of
course, after we add something, we do have
to redownload. So
Very interesting. Well, that answers my question.

(39:21):
I think that's that's about it for the
the demo. I mean, unless,
Sean, you have any more questions about
the interface here.
No. It looked pretty straightforward to me, Hendrik.
I don't know. Did you have anything else
you wanted to discuss while we have the
demo up?
Nope. Not nothing related to this except for
that, you know,

(39:43):
this is probably something that's quite new in
the OT space is that
this is a software
service, meaning that there are continuous
development going on and releases,
and improvements to the software all the time.
Like literally every week we deploy new improvements.
And,
and
what,
I typically say is that like,

(40:05):
the, you know, if you if you if
you sign up with OT, what you what
you will experience is that the actual software
keeps on becoming better over time and not
is not going to become
outdated. It's going to be just better over
time. And I think that's
part of what I really loved about the
innovation space, innovation
happening around IT is that that, that has

(40:26):
become the new de facto standard in how
you develop software and great software.
And I think we in, in, in the
OT space, we need to adopt that same
methodology
of developing software, something that continuously
becomes better over time. Yeah. And I would
just say, you know, if you're if you're
on the OT side of things,
you wanna be in six eleven thirty one

(40:46):
dash three languages,
because these are things that your staff, you
know, what you know, your electricians and technicians
and even engineers,
you know, should know, should be getting up
to speed. I don't know. We're at the
automation school. We're teaching, structured text.
And so,
easier. I look at this, and I'm like,
this is a lot easier than trying to
learn
c plus or or JavaScript.

(41:08):
So in any case, I think, you know,
if it's an OT side real IO control,
real control system or data collection,
you know, you know, very important, you know,
mission critical data collection,
then,
you know, I'd rather have this than somebody
trying to write some custom code for me
and, you know, use some kind of computer
language who doesn't understand,

(41:29):
you know, the OT side of things. So,
I could definitely see the advantage of your
system, Henrik.
Yep.
I I I also wanted to say to
that, Stike, the I I do not believe
the EIC
standards in general will disappear.
They exist for a very good reason. Right.
Exists to standardise to to ensure safety and

(41:49):
determinists,
determinism in this.
So I don't think they will disappear. But
there are obviously advances now with AI and
things like that that can can help us
create these things much faster and much more
efficient and things like that. So, so
but,
but the EIC standards, I think, will be
there for a very long time. Obviously, the
06/4099

(42:10):
standard is is really exciting,
and and we believe that that can be,
yeah, that that can clearly be there, but
it's still a new EIC standard. So,
it's not
think what we're gonna see is we're gonna
see a lot more libraries fleshed out. There'll
be a lot less
writing from scratch. We've interviewed on the History
of Automation podcast. We've interviewed some big integrators,

(42:33):
and they're at a point now, you know,
twenty, thirty years on that they have libraries
for everything. And I think that's where we'll
see, you know, much like the DCS, I
think, vendors went two years ago. But I
still think that the there's a reason for
these languages. There's a reason to be able
to edit things while they run. There's a
reason for different languages for different applications and
different,

(42:54):
people maintaining them. So I agree with you
on that. I don't I don't think we're
we're gonna see the end of these, these
standard languages that have done us very well
since the, you know, nineteen seventies.
I just wanna add a bit on there
about,
Sean, you mentioned, you know, doing less code.
I I did show earlier in the bottom

(43:16):
right hand corner here, we have our our
little AI assistant,
Martha.
I don't believe the feature, it has been
released yet. You know, Henrik, correct me if
I'm wrong, but I know
one of the things that's coming is,
AI code generation,
you know, similar to that of cloud or
chat GPT.
So it's going to,

(43:38):
you know,
you can open this guy up here.
You know, right right now, I think it's
just for, help topics, but
you'll be able to talk to Martha
and she's gonna generate code for you in
your program there
all built in.
Yeah. Yeah. That's that's coming really fast now.
So,
it's it's not been implemented yet, but it's,

(43:59):
it's right around the corner.
Yeah. And it's it's not gonna be able
to it's you're not gonna be able to
hook a camera up to it and, like,
take pictures of your machine and say, okay.
Write the control code for this. But, you
know, if you had a, you know, process
that had 12 steps in it,
the AI could definitely help you generate that
code and and other code. And we'll have
to have Henrik and Jake back on to

(44:21):
talk about that when it comes out, but,
you know, it's gonna be able to save
you, reduce the tedious part of the the
coding. You know, if you need an array
of so many tags and so many dimensions
or, you know, the stuff that, you know,
it would just be the typing intensive,
it's gonna be able to help you with
that, and then you can actually put the
context in there. Just like,
you can pull up a template in Word

(44:43):
for a letter, and then you can fill
in the blanks. You know? And and, of
course, AI is helping make that easier too.
But, in any case, Henrik, maybe you can
come back on when that feature launches.
Yeah. Absolutely. And I'm also excited about just
a simple
a use case of of translating something. Right?
Translating your existing let's say if it's a

(45:04):
proprietary
code or something like that, like, getting it
getting it standardized and translating it to the
ESE six eleven thirty one standard, for instance,
or,
so so the obviously AI is, like, perfect
for this space. It's there is no doubt,
And and it's,
like, that's also why I'm so excited about,

(45:24):
like, what's going on at the moment. It's
like there's so much innovation
potential,
in the on the OT side now that,
they are with all these new technologies.
Yeah. Absolutely. Absolutely.
Well, gentlemen, was there anything else you wanted
to cover?
I think just just one final thing from
from me is, like, we thought a lot

(45:45):
about it, like, before this this episode, and
we thought, like, let's offer let's offer the
listeners something something of of true value. So
so we thought,
the,
you know, after this after this episode launched,
we want to
want to offer anyone
out there that's listening
a free, completely

(46:05):
hands on trial of our technology,
in their in their in their environment or
on their Raspberry Pi or whatever.
So just just reach out to us if
you wanna do that.
And,
and I yeah. We'll get you set up
for for for testing this, and it's not
gonna cost you anything.
Well, that's great. And, guys, if you're listening,

(46:25):
if you do take advantage of that free
trial,
please let me know what you thought about
it. But, Henrik, thank you so much for,
that offer to our listening audience. Guys, don't
be bashful. Reach out to him. Reach out
to Jake. Jake, thank you for doing the
demo as well. Really appreciate it.
My pleasure.
Any final words, Henrik, before we close out?
No. It's been great. Great, being here, Sean,

(46:47):
and thanks for for helping us.
Well, I hope you enjoyed that episode. I
wanna thank Hendrik and Jacob for coming on
the show, telling us all about OT virtual
PLCs, and then giving us a demo. I
thought it was really cool. Now if any
of you guys take them up on their
free trial, please let me know what you
think. I'd love to hear from you. And,
with that, I do wanna thank OT for
sponsoring this episode so we could release it

(47:08):
completely ad free. And I also wanna thank
you for tuning back in this week.
We have another podcast coming out next week.
It'll be early because I will be traveling
and doing an event with a vendor.
And so expect that instead of coming out
on Wednesday to come out on Monday if
all goes as planned.
And then we will be skipping the Thanksgiving,
week, and then we'll be back in the

(47:30):
in the, in December, and then we have
shows lined up for the new year already
as well. So thank you for being a
listener, a viewer, and, please, wherever you're consuming
the show, whether it's on YouTube or on
the automation blog or at iTunes or Spotify
or Google Podcasts
or anywhere,
please give us a thumbs up and a

(47:51):
like or a five star review because that
really helps us expand our audience and find
new vendors to come on the show. And
with that, I'm gonna end by wishing you
good health and happiness. And until next time,
my friends,
peace.

All Episodes

Episode Transcript

Popular Podcasts

Stuff You Should Know

Las Culturistas with Matt Rogers and Bowen Yang

Crime Junkie

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}OTee Virtual PLCs (P252)

Episode Transcript

Popular Podcasts

.css-r6mb8g{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:1;overflow:hidden;}Stuff You Should Know

Las Culturistas with Matt Rogers and Bowen Yang

Crime Junkie

All Episodes

OTee Virtual PLCs (P252)

Stuff You Should Know