Akshat Sharma with ITT, Inc

SUMMARY KEYWORDS

data, vibration, sensor, hexagon, motor, fft, device, dazzled, talk, machine, faults, world, industrial, information, asset, put, collecting, sharma, operator, type

00:03

Welcome to the industrial talk podcast with Scott Mackenzie. Scott is a passionate industry professional dedicated to transferring cutting edge industry focused innovations and trends while highlighting the men and women who keep the world moving. So put on your hard hat, grab your work boots, and let's get all right,

00:22

thank you very much for joining an industrial talk a platform that is dedicated to industrial professionals all around the world, because you know what I'm gonna say you're bold, you're brave, you dare greatly, you innovate. And you are definitely making the world a better place. We are on side Las Vegas, broadcasting from hexagon live. And if you want to know anything about technology, anything about innovation, anything about all of that, you need to be at hexagon live, it has been a absolute wonderful venue, wonderful conference full of incredible people, and all focused on making the world a better place as well as solving problems. They are collaborating, it is a great, great opportunity to meet some great incredible professionals like the one that's sitting right next to me. I shot is his name Charmin his last name. And his company is ITT and he's in the hot seat. Are Chad, how you doing? Good. Scott,

01:22

how about you?

01:23

No complaints? Yeah, got a vibration device there. But we're not going to talk about that. Would you think about that last night, the keynote keynote was

01:34

pretty good. It was really great presentation very inspiring about how you know, hexagons, helping make a lot of the world a better place, and especially how things like we can help on the industrial machinery and reliability side.

01:50

It's all sort of mesh together. Right? It seems to me that you just can't be a cowboy. And just go off on your own. I think that it's becoming more and more collaborative you need. You need to have companies like yours and others to be able to come together to come and solve problems, because not everybody has the complete solution. Right? But you put gotta pull it together he got and I think they were hitting some great points like community. We're in a part of an industrial community, this neighborhood these this ecosystem, right, anyway, and I thought it was, what were those little light things?

02:28

Yeah, that was interesting. I mean, I guess they've got like a little wireless receiver on there. And they're just coded for different lights at different

02:36

times. Yeah, it was, wasn't it? Yeah. Seems pretty neat. Even that was digitized. Get some magic out there. But once you walk away, it's now it's a paperweight. Oh, yeah,

02:47

I know, I left with one by mistake. And I was like, Oh, this doesn't do anything anymore.

02:52

So true. All right. For the listeners out there. Give us a little background on who Oxshott is. Yeah,

02:57

so my name is Akshay Sharma, I've been working with a variety of different companies in industrial control wireless communications. Actually, I've been working in the IoT space before it was called IoT.

03:10

What was it called? That's a good

03:13

question. I think it was more just instrumentation and sensors, right? data collecting device. Yeah. And then the IoT name came about? And it's like, oh, this is the same thing. It's just a different

03:23

name. It is. And whoever came up with that. Could have been earning a lot of money. Right? Yeah.

03:30

So I've been doing a lot of that, you know, different types of communication systems, wireless, radio, satellite monitoring, for different applications, for industrial machinery, for environmental monitoring, right, all that kind of stuff, basically, throughout my whole career. And that's my background is more on the electrical engineering side.

03:51

And in fundamentally speaking, it's important for these assets that would be able to communicate the data and the data in a way that is health related. And that gives the business the operation, the ability to be able to proactively do what they need to do and or whatever, make the make the changes. And, you know, of course, there's no way to shoot holes in that. That sounds That's right. Strategy, right thinking. Yeah, exactly.

04:15

And, you know, now, with sensors being a lot lower cost a lot more accurate, right, so one sensor technology advancing, now you can really switch to rather than just having this general maintenance schedule that says, okay, every six months, or every so many operating hours, I'm going to do these tasks, you can look at the actual condition of your machine, and vary the maintenance based on the actual condition, right, which is huge. Yeah, bottom

04:40

line value. Yeah, if you can extend the or prolong the operational life of that asset and also sort of saying, Hey, we don't have to do all that maintenance right now. We can just sort of sweat the asset a little bit more and a little longer to be able to do that that exactly, and

04:58

then the flip side as well right, which Is that when it really does require it and maybe sooner than normally? What comes up because of how often has been running it or how you've been running it, you can do that rather than missing it and having some sort of catastrophic failure. Yeah.

05:14

And that you don't want that either. Right. So it makes sense. It's the right thing to do. How do you How does ITT How do you and we're gonna be talking about this device? I got this, it looks like a VW bus or an old VW bus device. That's the only way I can describe it right now. But it's, it's pretty cool, quite frankly. How do you work with hexagon with all of this,

05:38

so we're partnering with hexagons, J five, software. So J five part of their J five software is operator rounds, right, which enables the ability to digitize that operator round process, where operators can go around, you know, collect the necessary amount of information, make sure they're actually going to machines and collecting that information. And we work with them through the Eilert, where instead of having to manually take a vibration reading, or take no vibration readings on the machines at all, they can use their mobile device simply to connect to the sensor, pull that live vibration data and put it into there operate around record.

06:20

You're gonna have to expand upon that J five and set up a mobile.

06:25

Yeah, so J five is an overall software piece. Part of it is mobile, right, so the operator in the facility can take his phone or tablet or iPad out in the field. So before when they're doing their rock rounds, right, they've probably got like a little notebook, and they're writing everything down on a piece of paper,

06:44

right, and the pencil Lipinets standard stuff.

06:48

And that's probably the best case, right, they're actually doing the rounds, they're not just skipping a couple of things. Now, they've got this tablet that kind of instructs them, hey, on this piece of machinery, this is the information they need to collect. And they can do that, you know, right there while they're at the machine. And so once they're done, you know, everything's kind of collected, it can go out in a nice little report format, it can go back to the J five database, right? Everything's available there, they don't have to input it anywhere else. So with that digitisation concept, comes the ability to now just automatically connect to our islands sensor, and bring in that vibration data as well.

07:29

So when we start talking about vibration, let's shift to the device. When we start talking about vibration. Of course, there's a community of vibration technicians, right? category one through four, or five, or whatever it is, and, and, and they live in the world of analyzing the vibration, similar to like data analytics, where they're living in this world where there's a tsunami of information coming out. And then they're, they're trying to figure out what's going on. Right, right. There's, there's a level of just detail with that whole manual approach. Does that device sort of take into consideration that level of detail or how, you know what I'm getting? Right? Yeah, so

08:13

usually, let me step back a little bit with vibration, there's kind of almost two levels, right? There's the overall vibration, which just says, Hey, this machine is shaking at a quarter inch per second, right? Normally, it shakes at maybe point one five inches per second, right? So now you can see, hey, it's been trending higher, or all of a sudden, it's spiked high, something's wrong. I don't really know what's wrong, but something's going on. So that's, you know, obviously, our high alert sensor does that very well, right. But the alert goes beyond that. And it captures what's called an FFT. So it takes that point to five inches per second, and breaks it into the different frequency components. And now when you've got that the FFT is a fast Fourier transform. And you can see that entire spectrum of data. Yeah, that's what those vibration analysts look at to see. Okay, well, you know, this point two, five inches per second, most of this energy is coming from the running speed, right of that sheet, which means okay, maybe based on the type of machine it is, and what I know about it, it's a misalignment issue, right? Versus if all of that energy is maybe coming from seven times running speed. And I happen to know this is a pump that has seven veins, there's probably something going on with the impeller, right? So that's what the eye alert does. It does that FFT data as well. And it does it very accurately, and it does it all the way out to a wide frequency range all the way up to 5000 hertz. So the more information you can get a better you can diagnose you know what's actually causing the issue.

09:48

Is there a proactive or a notification that says, from your Eilert to a person say, Hey, you There is an issue. It appears to be this investigate.

10:08

Yes. So that's actually one of the really cool things about our

10:11

system. So I didn't even know I said that to you. Yeah,

10:14

I know, quite insightful. This, what we're looking at here is just our basic sensor, right? It's the Eilert. Three. But once you get the data out of this, we've got our own web platform or cloud platform. It's called our AI platform. And part of what it does is exactly that it does the analysis of that FFT, and it runs it through a rules engine. So it knows like, Okay, on this type of equipment, these are the types of faults that can occur. And this is what it looks like. So it's constantly using all that processing power in the cloud to compare those FFTs to what a fault looks like. And it can detect them and say, Oh, look, I think there's a misalignment issue, you need to go check the alignment of that pump motor. Yeah. So I liken it to probably about like a category two vibration analyst

11:03

know what, yeah. So that's, that's pretty robust. Yeah,

11:08

that's pretty powerful. Right?

11:09

It is.

11:10

And we can do that, actually, we can link it. So for example, when you've got a sensor on a pump at a motor, it will take into consideration data from both sensors. And not only will it be able to detect faults on the pump and the motor, but also faults that are related to both of them. So if there's misalignment between the two, right, it will take in data from both sensors to be able to detect them.

11:34

Because now Now, this is what I see. Costs are going down. That means I can get I got a question, we're gonna go, but I'm gonna paint this picture costs are going down. So these devices are more available, right. And it's going to do a great job at identifying the vibration of a particular or the condition of that asset from a vibration perspective, proactively notify the any technicians or whatever the whatever the process or the workflow needs to go. And, and then be able to get out there, do whatever is necessary, or they analyze it. And maybe it's not really at a point where we need to go send a technician or whatever, whatever the the action is, but it's very tactical in its nature. There are conversations that are wrapped around cyber. From, from this perspective, you've got a Bluetooth or even a cellular type of connection. Tell us is that protected? Tell us a little bit about the cyber capabilities. Yeah,

12:39

so we've designed this really, with security in mind, right. So when the data goes up to the cloud, by default, it actually stays out of the customer's network. So that means one, it's really just separated, there's no way for us to access an end user network. But also that data is encoded and encrypted. So this data coming in wirelessly out of the sensor over Bluetooth, that's all encoded. As soon as it gets to, we have this gateway that transfers it up to the cloud, it then gets encrypted. So as it's being transferred across the internet, yeah, it's all encrypted, as well as encrypted when it's stored on a database as well.

13:20

So there it is, well, so I can install it, I can put it out on that device, I can begin to pull data, how quickly can I pull data.

13:29

So you can go down to every 30 seconds on this. And that's you know, that's a good point, by default, were taking data every hour, but you can change that interval down to 30 seconds. So I think like you mentioned, you know, having the vibration analysts already taken the data versus something, now, you're getting that data at least once an hour, right. Whereas before, if you're manually walking around to collect it, maybe you're getting it like once a month. So if you think about that, that's about 750 times more often that you're collecting the data, it's gonna be really hard to have a human go analyze all that data by themselves, right? You've got to use technology and the computing power that we have available today to do that sort of first pass of analysis before passing it on to a vibration analysis expert if you need it. But

14:18

that's a heck of a value proposition. You're absolutely correct. But if I wanted to, and I say, hey, I want that device out on my this, this motor. And you come out and you say yeah, okay, cool. Plop, how quickly can I begin receiving? I mean, just like, it's all connected now? That's

14:38

a great question. So the the setup of the sensor is literally about five minutes, right? It's most of it is physically bolting the sensor to the motor. And then you use a mobile app to activate the sensor. And you can download the data right through the mobile app. You can install the gateway, it starts collecting data immediately. And the other beautiful thing about our analytics on the web platform is because it's rules based, it doesn't have to learn a fault, right? As soon as it starts getting data, it's going to be able to start recognizing issues and faults immediately.

15:10

Do you do any? Let's say I have Acme Acme motor here. It's in the south, and then Acme motor. Here in the northern part. Do you guys sort of see how the performance of similar devices and and then say, It's okay, because we got this we got, right. Do you do that?

15:34

Sort of. So it happens on our analytics side, right. So the analytics by default, when you tell it, hey, I'm installed on this particular type of motor, it creates what we call a synthetic baseline. So it just says, Hey, on this type of equipment, this is based on the historical information we have, this is what normal looks like. So it then starts looking for stuff outside of that normal to recognize assaults. And you can actually enhance that by then saying, Okay, on this particular motor, this is what normal data looks like. So then it will kind of we perform or enhance those recognition rules to say, well, now I'm gonna go off of this baseline on this particular piece of equipment. AI

16:18

has been a major conversation lately. Is is? Are you looking at having a sort of an AI? solution with this? Yeah, so

16:30

we're looking at AI where it can be beneficial. In terms of especially I think one of the areas that's useful is being able to recognize some of those baselines, recognize some of those feature characteristics for the machine that we need for that fault detection system to be accurate?

16:49

Because you want it to learn, right? Not every motor is the same right now. And they might have some unique quirks about it, but it doesn't mean that the asset is failing, it means it's just like, it's got a quirk. Right, right. Yeah, exactly. For lack of a better term. So

17:06

I think that's the important part for us is using the AI in the machine learning to be able to detect that baseline.

17:12

Okay. You're here all week? Ish.

17:16

Ish until Thursday.

17:18

Oh, yeah. That's one of the thing wraps up and yeah, where you stand you stand?

17:23

I'm staying over at the link. There it is. Okay,

17:26

we're neighbors. Oh, okay. Nice. All right. How do people get a hold of you there? They

17:31

can visit our website at Tai dash alert.com. Or they can email us at sales at iDASH alert.com.

17:39

There it is. Oxshott. You were great. It's been conversation. And by the way, you could use this if you want to try to make it look like a bus. I'm telling you. That's that's. That's the next level.

17:53

That's that's a good idea.

17:54

I'll just put a little wheels on

17:56

it here. Yeah, VW retro edition on there.

18:00

All right, you were absolutely wonderful. All right, listeners, we are going to wrap it up on the other side. Remember, we are broadcasting from hexagon, live. Put this on your calendar for next year. I'm telling you, you will not be disappointed as our child was saying. Yeah, last night's keynote was sort of it was dazzling. I had a lot of information. Oh, yeah. Anyway, it was it was good. Anyway, thank you very much for joining. We're gonna talk and wrap it up on the other side. So stay tuned, we will be right back.

18:28

You're listening to the industrial talk Podcast Network

18:37

All right. Once again, thank you very much for joining industrial talk and thank you once again for your continued support of a platform that celebrates industry professionals all around the world that was x shot. Ox shot Sharma i t t is the company and the reality is they got a great device. They got they got incredible solutions. You need to reach out to that that's your that is your to do list thing is to reach out to arc shot. We're going to have all the contact information as you know out on industrial talk. But what a great event, hexagon live. If you want to be dazzled with innovation, hexagon live, if you want to be dazzled with solutions, and individuals that are dedicated to solving problems. Yes, hexagon live, put that on your calendar. Put that in your budget for next year to attend. And yeah, you will be dazzled by put by all standards. All right, be bold, be brave. You're greatly hanging out with oxide change the world. We're gonna have another great conversation shortly.