Reliability Gang Podcast
Welcome the #Reliabilitygang Podcast! I would like to welcome you all to my reliability journey. I am passionate about reliability and I want to share as much as I can with everyone with my experiences. Stories are powerful and my aim of this outlet is to gather as many insights and experiences and share them with the world. Thanks for joining the #reliabilitygang.
Reliability Gang Podcast
Electric Motor Management Revolution
When most people think reliability, vibration analysis is the first thing that comes to mind. But what about the electrical side of your motors. That’s where a lot of hidden damage is happening and it’s often ignored.
In this episode we get into shaft current. It’s been around since variable speed drives first hit the scene but it’s still not well understood and definitely not well managed. These stray electrical currents eat away at bearings and windings, quietly cutting motor life down by decades. The scary part is that from what we are seeing around 90 percent of drive installs haven’t been designed properly to deal with it.
I reference the teachings of electrical condition monitoring expert Mark Gurney to break this down. We look at how to actually detect the issue with oscilloscopes, Rogowski coils and the SKF TechEd tool. We explain the difference between common mode and circulating currents and why quick fixes like grounding rings aren’t always the silver bullet people think they are. Done wrong they can even make the problem worse.
We also talk about Motor Current Signature Analysis MCSA. For me this is the future of motor testing. It gives you the ability to pick up not just electrical faults but mechanical ones too. Bearings rotor bars even belt tension can all be identified just from reading the motor’s electrical signal at the MCC. If you’re serious about IoT and advanced monitoring MCSA is a game changer compared to just scattering sensors everywhere.
The bottom line is this. Why are we okay with motors failing after 10 years when they are designed to run 30 or 40. It’s time to move beyond only looking at vibration and start treating electrical testing as a key part of reliability. Our most critical assets deserve that level of care.
Hello, welcome back to another episode of the Reliability Gang podcast. I'm here with my right-hand man, will Crane. How we keeping buddy Doing very well, thank you. How's your week been, mate?
Speaker 2:Busy.
Speaker 1:We've both been on our side quests. Again, I feel like I don't really speak. We do speak in a week, but I feel like we're so full on what we're doing and we come Friday and it's just like we're just very focused on making sure we're delivering that that top our service to their customers like even our little catch-up is not a little catch-up, is it?
Speaker 1:it's like literally 15 minute about the whole week and I'm like, oh, and then I do mine and all the rest of it. But I mean, yeah it's. It's been great because I feel like you're more on the reliability quest, really helping people with the reliability stuff and really formalizing the roadmap for certain customers, and I've kind of been on more of an electrical kind of quest we have with with shaft current and current signature analysis, teaming up with mark gurney and I thought you know what great podcast to talk about that journey today.
Speaker 2:Yeah, because for me just thinking about it has been a bit of a whirlwind and it's coming up more and more in conversation in the reliability chats I'm having with customers, which is well, what about? Are we thinking about the electrical testing side of things, like we've got all this mechanical stuff, we've got the va, we've got the oil analysis, but what else can we be doing on some of the motors? We actually had one at one of the customers I was at this week which they had a stirrer that stopped working. They went to change the motor, got Got a bit of an issue there with problem solving. They changed the motor and didn't realise that the gearbox was also broken. But they changed the motor and it still didn't work. And then they've changed the gearbox and we're doing the RCA on that.
Speaker 2:But interestingly, this motor, when we got to it, although the gearbox had failed the motor, we couldn't turn the shaft at all. Really, absolute bound up. So you couldn't turn the shaft at all, really Absolute bound up. So me and the reliability engineer on site, we actually split the motor right and the motor had gotten so hot that the VPI resin had melted, oh wow, right To the point where it left like white residue all around between the stator and the rotor.
Speaker 1:Do you know what that white residue is? So it's the insulation papers that you get, usually like an insulation class, it's like a plastic right, and when that gets, I mean it's rated to quite high temperature stuff okay, but when it does get to that particular temperature, when it dries again it's it was like white ash, it's like and it was like white glue, but it was so bad we couldn't get the rotor out.
Speaker 1:Wow, literally I'm sledgehammering the rotor out. We've had that before as well when we've been in workshops, that things have obviously motors burnt out the windings, but now and it's got that bad, yeah. But again, this is also another kind of discussion as as well, because when we do look at most failure modes in any plan and whenever we're doing anything, we're always looking at vibration as the number one. I get why because it is a great indicator of many failure modes in one, isn't it? Yeah, but one thing that we do really don't, especially with the bigger motors and different sides, is look at kind of the failure modes of the actual electrical side of it, because it is a very inherent failure mode.
Speaker 1:Like well, when we do for me, because on these particular kind of assets which we have done this year, we've been working quite close with quite a big manufacturer and they're really looking to be able to really get the top end of reliability. They've been on a journey for a while different. They've got a few sites in the uk, but they're all at different stages as well, which is quite interesting. But one of the particular sites has very robust CMS systems, a lot of information going in there. The maintenance system is being driven to drive out the PMs, and the guys are really kind of on top of a lot of things. They've got good condition-based maintenance in there, but they're now looking to improve. Because I feel like when you've got all of that, you get to a bit of a ceiling, yeah, and you're like what next? How do we improve? How do we now get above this ceiling?
Speaker 2:And if you don't really do some real sound for me because I understand the next level of where the failure modes are almost hidden when you're overlooking them, you're not going to be able to really improve from that point, and I feel like this topic of how do we measure the electrical condition of our motors and manage them better has become more and more of a hot topic as people are becoming more and more aware of the potential damages from inverters yeah, my drives as well, like the revolution of inverters, in terms of where we are in the industry, was crazy.
Speaker 1:Because obviously technology is it moves fast and obviously back in the day when you get some of these old plans, dc mowers were the thing that that was the way that you control speed and that was the way you did belts and pulleys or pulleys and in different ratios. And you know what we had to do back in the past is actually over manufacture to be able to vary speed. You know dcs are very difficult to overhaul. I've done a few dcs in my time not much, but even when I was winding in the workshop to do it interpole winding or field winding, that's kind of different in the poles were bar wound. But be able to do that properly took a lot of time.
Speaker 1:So the time it took for me to to rewind ac motor the whole thing say, for example, a four pole 15 kilowatt motor we could probably do that in four hours. Do you know what I mean? That's from start to finish as a rewind. Okay, quite quick to be able to do that. You could probably do that within the day. Okay, that's still going to cost the customer quite a lot for that rewind because you've got the parts, you've got the materials and the copper and all them other things that are associated with it. That associated with it an interpole or a field winding. Just you've got four in each winding for a dc, a field or interpole. And if you want to do a bar wound interpole, you're talking two or three days to get that really good and that's just one interpole.
Speaker 1:So when you had a full rewind for a dc you could be taking up to a week to do that properly and and that's a long time to be able to do so over the equipment of ac AC motor. So obviously the introduction to these drives was revolutionary because what we could do now is start to actually say well, we can start to replace some of these DC motors out for motors. The spares is a lot cheaper, it's a lot less risky to the business if something should fail and it was a lot easier as well to test as well. An AC motor is a lot easier to test in the dc motor in terms of the winders, because you've got one wind in the ac mower and you've got different types. You've got a commutator bar as well with, with a dc loads, different little fading modes in terms of the electrical side of the dc. So I can see why the revolution was so and the energy efficiency was huge, massively.
Speaker 2:That was a massive job, even just for drives that were on standard motors, that were just needed a slower speed, or even some motors that they were like, oh, we're having to run that at full speed, but we don't actually need full speed, we'll put it on inverse save massively on energy as well and all the rest of it.
Speaker 1:So the need for them inherent massive. You know, technology did take that leap forward. But one thing that we miss within this whole scenario was understanding how inverters affect, you know, shaft current, how they affect the winding, not just the bearings, and how also real, simple ways of grounding can mitigate these from the get-go, from a design point of view. But again, this thing is what we've studied, has been so misinterpreted by the whole industry.
Speaker 1:We were coming across this quite a while back and of my first shaft kind of current fluid journey started even my old company, all right, and this we found some electrical fluid on the compressor. We've got that customer now, but we've had the issues with that since. But when I first saw that was the first time I ever see it and that was a good you know good eight, nine years ago. So this, this shaft, has been around forever, ever since drives have existed. But the problem is what we've not done is understand what is really causing it, what damage it does, not only just to the bearing, what other implications does it have further on down as well and that's something that we've learned this year and also what are the mitigations around it? What are the other things that design stage need to be done correctly to be able to actually, you know, eliminate the defect altogether or the root cause of it. And unfortunately, I'd say 90% of the drives that we've got within the whole industry, that design bit hasn't been done.
Speaker 2:No, because even on our journey at the beginning, we started off with the idea of the fact that, okay, we've got this shard current. I don't think we really understood how it was necessarily being generated, but we knew it was in the rotor and we knew it was discharging through the bearing. And the main kind of products on the market then to solve this problem was grounding rings, grounding brushes, different solutions that were being applied to the shaft, and that's what we were looking at doing. Yeah, I mean I think because when, but first, so you know to add on to that, but now, thinking about it, even when we were doing that, we still were saying at that point in time was we still wanted to be able to measure this somehow. And we, I remember we're having so many conversations and buying lots of kits and scope meters around it?
Speaker 1:we were, because I remember the first journey we had. We was we was measuring the voltage off the shaft. That was the way to measure, because what we didn't like to do is just put something on randomly and and assume that the problem's being solved. We're vibration analysis engineers. We we live our life on data. We wanted to see the good or the bad, and then it good, that's it. That's that's what we wanted to do. But the thing is, because of our lack of knowledge within in that particular subject and also the industry's lack of knowledge, it wasn't just ours, because what was we doing when we're finding the problems? Right, we're going to go online, yeah, reset.
Speaker 1:So what we found was a lot of the shaft ground and current companies were selling devices that only warranted their product to prove that that product was doing something, yeah. So when we look at business drivers for companies now, we've always got to look at a business driver now, we always got to prove that that product was doing something, yeah. So when we look at business drivers for companies now, we've always got to look at a business driver now, we always got to do that. What is that business there to do? What is their outcome and what they, what do they want to push and what do they want to sell and what does that actually correlate to their business driver and a lot of these things. That again that we come across, especially nowadays, is to sell a product. That's what they want to do, that's how they make their money.
Speaker 2:Which is a solution and we have learned since then, which you'll come on to a bit later with mark is that they have somewhat of a place, but there's a lot of extra things that can be done.
Speaker 1:First, exactly and I think as well as it's always looking at root cause. Or we do root cause analysis right when, wherever we find a problem, what do we want to do? We want to go far as back as we possibly can to say what is the actual root cause of this issue. We do it in reliability. We do a lot of in vibration analysis and again, vibration detect has detected these problems and we've found flu in a week through our journey of try to go back. But how do we fix the? How do we fix it? How do we go back to the actual root cause of the issue? We know it isn't generated from the inverter. But also, what mitigations do we do from that point that also allow us to be able to be far back as possible to that and try to mitigate that so it doesn't cause any further damage?
Speaker 1:And again, you know mark gurney is a huge. Well, he's basically mentoring me now. Basically that is my mentor when it comes down to electrical side of condition monitoring. Because again, you need these people in the industry that have done their own extensive research but also really understand the principles beyond the level of just selling a product, and that's why I've got a big mark up here really massively, because we just even had a conversation yesterday and he's been on the podcast. We've been on the podcast. When he came in february, that's when we did the training. Okay, we understood that. We didn't understand.
Speaker 2:It's the level that we needed and even like, even on, like my journey with the reliability piece. When I first met mark at the mobius conference in manchester and he delivered his presentation, I can remember everyone in the room was like okay, whoa? Yeah, because for the first time ever I felt that everyone in that room really started to understand and everything Mark was saying made sense. Yeah, he conveyed it to make sense, didn't he? And he took a very complex topic and he made it simple to understand. And that's also when we then, as we were doing more of the ARP and we had more awareness with Mark, was just how big of an influence Mark's had in the reliability space as well, with the work he's done with Jason at Mobius, is that he is a massive mentor to us now at Maintain and what we're trying to do, but what he is doing is he's leading the way in electrical motor management.
Speaker 1:But we need a leader in that sense, because the thing is, this problem is becoming bigger and bigger because the more drives are introduced without sound testing and the acceptance testing as well. That's another thing as well. New drivers yeah, new drivers, but new drives as well. We've got to be testing how much shaft current is actually coming off the drive to begin with, before we even connect it up, because when you connect that drive up and it's and it's producing too much shaft current, you're causing damage straight away. Yeah, so this is the thing about reliability.
Speaker 2:it's about trying to, at that design stage, be able to mitigate as many failure modes as possible and I feel like from the conversations and the training that we have had with mark, obviously you'll go into all the testing, how we that, but a lot of the common mode and circulating currents are literally avoided from good design and installation Massively. Yeah, and I was even thinking now while we're talking about it you know, if you get that bit right, a lot of the potential damages can be avoided. You don't need to put a grounding device on necessarily, and in some instances, instances that can make it worse and I even thought my reliability brain's going at the same time which is, you know, when we do install a grounding brush and this goes back to the customer. That where we first experienced this if we do install a grounding brush or a ring, what we actually are now doing is we're introducing another failure mode that needs to be captured within the formica ring because that ring might break. And what happened to that customer that we went to?
Speaker 1:We did and that happened. What we did initially was put an Aegis ring on, working perfectly fine to begin with, so it was diverting the current away from the bearings. Unfortunately, because the shaft current was high, there would have been an attack in the winding as well, but we didn't really know about that. We had no awareness of that. So that has obviously taken a bit of a hit from that. Has it failed because of that? No, but has it reduced the life expectancy of that winding? Possibly, but the problem was with that is that when they were lubricating that motor on the drive end, they were over-lubricating it. And the Aegus ring, how it works, is loads of tiny little carbon brushes around the shaft as a circumference. But when, as soon as you get any grease, dirt, dust, dust and, let's be honest, most of the environments that we operate in, yeah, are at least dusty, at least you know?
Speaker 2:and this particular motor was a little bit interesting in that it was completely the winding is completely open, isn't it? It might be pulled. Pulled through the winding rather than just, uh, the cooling is pulled through the winding rather than go over the top of the external fan.
Speaker 1:So what it does it brings, it draws air through and it's also got open end shield to be able to to cool the mower down. Obviously great for cooling, but very poor for dust and grit. I don't know who designed that. Because, shirley, having all of that dust, because when we opened that motor up as well, we had to spend a lot of time just trying to get a lot of compressed air because we changed. Bearing on that, and I remember that job it was one of our first jobs that me and you did together and it was so fun, what a job. When we lined it it purred in. At the moment we were just like what a win. I went home that day very, very proud of what we did. But to this day, that's running, still running, it's still running nice. But again though this is another discussion we know they're shaft current. Now we're now revisiting, going back to actually test the inverter source to see if we can now, yeah, put some emf calls on as well.
Speaker 2:Well, so obviously, from what we've learned from Mark is, we know that there's a lot of mitigation that can be done at design and installation stage and we'll probably go into that. But what are some of the things that we now can do or should be doing to manage our motors better? What are some of the tests that we now can?
Speaker 1:do so in terms of that design stage. One of the first things that we do see a huge issue with and I think the way that we can mitigate the quickest action to be able to start to reduce this problem massively is just how a motor is earthed correctly. You know how many motors have you seen with the earthen point on the frame completely unattached? Yeah, not even earthed Loads. And also the mitigation of understanding is the the cable actually screen cable? Is it actually inverter rated cable?
Speaker 2:yes, being utilizing you on the earth in one. What tests could people do to try and find out if it's working properly?
Speaker 1:so simple tests. So so what mike advised us to buy was a current clammy. So what that clammy does? It can test the impedance resistance and also have a look how much current is flowing through the PE, the conductor. So all we're doing is basically putting a clamp over the actual main earth into the terminal block, main earth into the terminal block and as well, what we should have on that particular mower is earth from the frame to a sufficient earth as well. But also again, though, we have to then consider if that's earth to the frame, is that frame earth correctly as well?
Speaker 2:Because you've got a fan, you've got to make sure the chain is perfect.
Speaker 1:The chain is perfect to earth.
Speaker 1:But remember as well, a lot of things are earthed as a final point of call to the actual frame. But if the frame is then on anti-vibes which are rubber to the floor, you haven't even got a good earth. So the idea is how can we lower that path of resistance from the actual source of the inverter to earth as a lower path of resistance as quick as possible? Because we can put a shaft, ground the ring on if we want to and we might be able to actually ground the rotor. But if we can't dissipate the actual, it can't be anywhere to earth quickly. It's going to recirculate again some way or going to cause damage or travel to another part of the machine, and this is where actually a grounding device can cause more damage right, yeah, 100%.
Speaker 1:So if you've not really earthed your actual motor correctly, right, and you've got a shaft device on, that particular it might travel to the driven machine.
Speaker 1:It will actually start to try to find a lower path, but that generally can be through a compressor or pump or gearbox, and people don't understand that this shaft current can travel. I think it's hard for people to get their head around, and it was for me at first, because the problem is we're mechanical engineers. I mean, I was actually electrically trained, but then I went into vibration analysis so it was like I understood the principles, but I was quite lucky that I also was a winder. I understood free phase theory. I understood how, how things were wound. I understood this. So also when mark was telling me that the shaft current that appears can really attack the first turn within your winding, yep, like really bad on startup and that's because we talked about this in the track.
Speaker 2:This is because of reflective voltages. Yeah, in that first court turn.
Speaker 1:Yeah, so what? What a lot of winding manufacturers do now is on averted driven motors, is another layer of insulation around that first turn. But if that winding has been rewound and it hasn't followed that practice, which is quite common, so if you don't tell your rewind company if or they're not experienced or whatever exactly?
Speaker 1:they're not going to do that because when they, when they obviously spec a winding to rewind it, what they do essentially is count the pitch and count the coils from the top. When they spec it out, they'll cut the head off, burn the motor around, pull the cores out. But once you burn it off, you're not going to be able to see that insulation on that first hand anyway. So again, even winding shops has a responsibility to understand. Are we copy winding this to the actual specifications? Or you can have a motor that's been installed that's not inverter driven before, and then they'll re-round it to be yeah, or or they've changed, like, for example, that one we was at the other week yeah, this mower's not designed to be run on an inverter, but they've then changed it to an inverter.
Speaker 1:Obviously, that nameplate is not going to say it's inverted, it's inverted duty. So if it goes away, exactly but this is the thing as well is getting the context around every element. Like, how many times have we seen a spare motor with a roller bearing be put in as a you know what I mean? Like a spare in there, same motor, but it's designed for a bell driven application thrown into a straight coupled application? It's not got the right bearing for the application. We see it all the time in industry, so that earth clamp that we spoke about earlier.
Speaker 2:That's obviously a really good measure to see whether or not you've got any current going back down to the Earth. It's just measuring the Earth leakage. So, on that main Earth coming out of the terminal block, you are expecting to see that going back 100% and if it's not, that's a concern.
Speaker 1:That's a concern. And then what we can do is, as well, what we've seen sometimes is just from where the frame is to another suitable Earth point. We measure it and if it's very low the current, we it's very low the current, we know that there's some form of issue, and sometimes it's literally the case of taking that earth off, cleaning the surface, and then you can see it, and then putting it back on to see a difference, and that's what Mark did.
Speaker 2:his last conference in Barcelona was that shaft current can be as simple as scraping a bit of paint away.
Speaker 1:Do you know Mark's huge advocate for it? It is because he generally says do you know how many problems and issues just come from that first simple way of earthing a motor? And obviously the thing is, though you also, you can't just go do it without testing. Does that make sense? So you can't just go, I'm going to chuck an earth on here, chuck an earth on here, no-transcript cable, okay, okay. So when we've got that in the drive, what we do is get. We get emf kind of dissipate through the cable as well, and that, if you've not got screen cable, can actually jump to other lines and cause shaft current in different areas.
Speaker 1:So Mike was telling us that one time, with one of the scenarios he had, they had basically changed a similar scenario to the one that we've just dealt with. They both were normal direct online motors and they changed one of them to an aver for efficiency reasons. They didn't change the cable though, right, but what was happening was, because the cable wasn't screened, the shaft current was traveling down both lines, affecting the bearing on one of them, but it was also affecting the bearing on the sister motor because both cable runs were coming from the same panel and they were run side by side charging. The other would charge it in the other line. It was so what the screen cable does. It protects it and it doesn't allow it to also jump, if that makes sense, and it keeps it contained and I remember Mark saying as well unusually, unlike other maybe electrical grounding with inverters you have to screen both ends.
Speaker 1:Both ends exactly. So what you get on a lot of applications is when that screened kind of cable runs from one end of the inverter to the other. What we want to be able to see is that screen part, all of that mesh, earthed correctly at the inverter end Good earth but also on the other end, where it goes to the terminal block, really well earthed. But the problem with that is as well. What we've seen is that some terminal blocks have gaskets.
Speaker 2:So if you've earthed it even sufficiently at the gasket, well, if it isn't earthed correctly from that terminal block to earth, then it's going to try to travel back up, and I suppose that's why on some really big motors they have the earth cables between the actual terminal block and the frame as the motor. They have that. You've seen that as well on most applications I remember Mark mentioned this a little bit as well and you start to build the picture. So much more about this. Well, what about the gland that goes into the terminal block? Isn't there special glands as well?
Speaker 1:Yeah, so them glands obviously clamp all the way around that screen. If that's a plastic gland, well that ain't getting any more earth in, is it? And we've seen, we've seen inverted duty cave, one with a plastic. Glad it's like. This is the whole point of this is completely pointless now. Yeah, but also, what you can get is these lugs that come off that ground. You've seen them little. Yeah, how many times have you seen them just drill into the terminal blocks? Yeah, the whole reason for that is to be able to take a lower path again to earth. So we should be seeing another cable from that to the earth point. Okay, and that way we also can measure the leakage there as well.
Speaker 2:And we've got a new tool from SKF as well that does a lot. What is that?
Speaker 1:Right. So the TechEd, okay, so the SKF TechEd tool tool. This is a really great non-intrusive tool to be able to measure them, emf generated from the inverter on the cables, whatever to. It's a non-contact probe and what it does? It measures very high frequency, counts basically where whether that shaft current is coming, because that shaft current is extremely high frequency.
Speaker 1:Yeah, this is why it's very difficult to see it within our spectra as well, because we only measure in a certain frequency range. We can see some inverted frequencies, but the damage is coming from the harmonics of these frequencies, not the actual frequency themselves. So when we're looking at harmonics of that high frequency noise I mean, we've often seen sometimes our loop factor reading that we've got within our spectra is roughly about 600,000 to 900,000 cycles per minute. We've seen initial inverted frequencies within that range sometimes, right, but what we're looking at is the harmonics of that particular frequency which we can't even see. Ultrasound, high-frequency ultrasound, can detect this as well, which is quite interesting. But again, when we're looking at an oscilloscope which we'll get onto in a minute, we are really looking at the source signal, but that sample is extremely high frequency. I think they're megahertz, some of these oscilloscopes, the one that we've got, that Mark got us to get, is megahertz. The frequency is a bit. It's happening so fast.
Speaker 2:It's really difficult to wrap your head around and in vibration we're just dealing most of the time in kilohertz.
Speaker 1:Yeah, exactly. So when you're talking megahertz it's a whole and it's really hard for me to even think how quick that's happening. Does that make sense? Because when you look at electronics and how quick things happen, it's a different world to what we're dealing with with mechanical vibration analysis and the frequencies that we're looking with. So, yeah, the tech heads are really great tool to be able to just see non-intrusively when you can use that along the cable and and near that gland that we were then talking about.
Speaker 1:We can use it along the cable, we can have a look how you know vicious it is and how many counts we get in, generally, what the settings on it does, and it's not a kind of trendable measurement, if I'm honest.
Speaker 1:It's more of a just an indication indication. But what I do is set it to maybe 10 or 30 seconds and then I can almost not trend but maybe have a look at the severity across a certain amount of moments. So what we'll do is set the tecad to 30 second, count, see how many counts it rises to over 30 seconds and we can kind of gauge a kind of a severity through that as well. And we're already starting to gauge what's good and what's bad ourselves from our own data that we're bringing in. So what's really good is to see that from the cable, have a look at what's coming off the cable and you will get some coming off the cable, depending if it's screen cable or not. But what we are looking for is really is it traveling through the motor? So the non-drive end, going across the whole circumference of the motor, seeing if the counts are high and going across the drive end?
Speaker 2:And if you're seeing counts at your motor, you're very, very likely to be having very poor shaft, and that really then instigates really going and delving in deeper now, which is why we have the scope meter and the Rogowski core.
Speaker 1:So important because even what we've found we might have high shaft current come from the inverter, but the good earths might be taking the majority of that away. We want to try to reduce that anyway, because it's best practice to. But the problem is as well just because you've got no counts being measured doesn't mean that your bearings are not getting damaged.
Speaker 2:And from all these tests so far, we've obviously got some indication that shaft current is happening. But there is two kind of types of shaft current right that we can detect, and it's the scope meter that helps us see what type they are and what what we need to do to fix them, because it's not just grounding, is it? There's another type in which people may be more familiar with, with the insulated bearing right.
Speaker 1:Yes, circulation, so circulation current. So it's really difficult for us to know what type is there if we don't actually have a look. There's a source with it, with the scope, right. So the scope will do two things. It'll measure basically a voltage, peak to peak voltage, because it's looking at a sine wave. It's looking at the, obviously the, the line frequency coming in. Okay. But what we'll be able to do, a bit like vibration analysis, is about to look at the patterns that you've got from that sine wave and what you've got. Okay. So good example of it if we're seeing a really clean sine wave with no spikes coming off it, we know that that actual signal is pretty, pretty clean in terms of the way it is, in terms of the common mode, but the circulating is quite high, isn't it?
Speaker 2:if it's got the sine wave? Yes, if you're around all three phases. That is because it should be. The idea, I think, when mark said, is that it should be flat line, because they should all just balance out to zero right.
Speaker 1:So circulated currents will show. If you're seeing a more of a sine wave, yeah, shaft current, common mode current, company of that. That essentially that's the spiky bit. The spiky bit, the very high frequency harmonics of it. If you're watching this on a video, hopefully we might be able to, I might be able to pull a few. Go on, you can pull a few, we've got a few, we've got a lot of data.
Speaker 1:Now, actually, where you get these little spikes coming out of the signal, they're common mode currents and they're the harmonic frequencies of them, inverter frequencies, and they're the ones as well that will cause your peak to peak to be higher. Because, just like you've got in a time waveform, if you've got low RMS values but high peak to peak values, your crest factor will be high. So it's almost like crest factor to a degree, and the higher these peaks are coming out, the more significant. Exactly, but that is going to give you a higher peak to peak count. But that also is going to correlate to a lot more current, because what we're looking at shaft current, we're not measuring voltage, because the ragowski coil is looking at current. Yeah, but they're actually plugged into the scope, which is plugged into the scope, yeah. So we get a voltage read off of the scope me and we convert it to current for an equation of what the regal c core set to, yeah, and we set it to one amp to every every. I think what every one volt yeah is is 0.33 of an amp, gotcha, okay. So then we can do that calculation and then generally, what we're looking for as as a measure, we really want under 10 amps in terms of shaft current coming off.
Speaker 1:That. Yeah, that's a safe zone. Anything over that is considered too much and potentially can cause damage. Yeah, with a a safe zone. Anything over that is considered too much and potentially can cause damage. A lot of drives will be over that, because all the drives that we've tested, generally most of them, are above, which is crazy. It's mad to think of. But what we're now doing is, regardless of whether that's causing issues down the road, we're still trying to reduce it because an EMF core will reduce that.
Speaker 2:So I was going to say so in terms of, obviously, that that sine wave is an indication that we've got circulating currents and and that is for an insulated or an insulcote bearing, can help mitigate that, to break that path, because the problem is what we've seen everyone's like it's insulated bearing.
Speaker 1:It's still got fluting. It's because, remember, insulated bearing is only going to really cover you for your circulation currents.
Speaker 2:And what I've understood from Marcus as well is that obviously that insulated bearing is for that sine wave circulating current. That spikiness is down from the inverter and the shaft current but that will peck away at that insulation on that bearing and once it goes first time on the winding and once that goes actually the circulating current is more damaging and it takes it out so much quicker Because you've then you've broken the insulation from that one particular earth path, from the non-driving of the bearing.
Speaker 2:Yeah, so you talked about those EMF cores. So what are they doing?
Speaker 1:So the EMF cores is all. These also have been actually in some designs of some inverters. One of the customers that we were talking to would get sent these EMF calls with the inverter and they would just say what's this? Really, they didn't know what it did. They didn't know what it did. It was one of these things and they just chucked them. Yeah, so some drives have been designed to be able to have these mitigated Not all drives, but some drives have these in this particular occasion. But what they're there to do is actually reduce the amount of this sharpness, spikiness and smooth out the signal with a coil. So basically, what it does is like an inductance. It takes that high frequency, it stores it in the actual, actual coil itself. The core heats up a little bit and it stops. It prevents it from moving any further from that point, if that makes a little bit like.
Speaker 2:I don't think it's an exact thing, but when I was looking into it, it's like you know, when you sometimes get uh data transfer cows and they have that love on magnetic, magnetic. It's a little bit like that. That's the same, yeah, I think it's the same concept.
Speaker 1:Oh, you can get down. You can take that off as well, don't?
Speaker 2:get away that shaft current and spikiness and those emf cores. That's going to kind of try and smooth that out a little bit.
Speaker 1:That's the idea with those. That's the idea. What we're trying to do is take that damaging kind of current that really will attack your bearing and insulation away and mitigate that source. And the good thing is, what we can do is we can test with a Rogowski coil after that, before and after oh, okay, so we can then see what's different, see it, smooth it out, yeah. And also what we'll do is take a when that coil gets a little bit hot. We know it's working. So we can use thermal imaging as well to be able to make sure.
Speaker 2:Should that then be? Do these need to be inspected as a check, just to make sure they're not getting too?
Speaker 1:hot. What I'd say is we're in line with your thermal imaging surveys to make sure that at the same time, and maybe even trend it, you know, has it got too hot? If it's getting too hot, is that getting affected? But I do also believe you should still do shaft current tests on these as a condition model, Just to make sure it's all there. Because what we can do is and we'll get on to the motor current signature analysis as well, which is a really good way to measure certain running attributes of your motor, is that you can still take an MCSA test with a Rogalski coil and you can do them tests together. Do you know what I mean? One after the other.
Speaker 2:This spikiness that we're seeing in the inverter? Are some inverters worse than others?
Speaker 1:Yeah, all different, so brand that you buy are, but you could have two identical inverters, the same brand and they could be, admittedly, completely different shaft. Really, I've tested it, I've had it and I'm now I'm still on my journey, right, I'm. I'm in test phase, I'm in experience phase, I'm right, I'm having a look, going across these different sides, and I've tested a lot of inverters now already, even within a short space of three months, and I'm starting to see patterns of what we've got and I'm also starting to see the different waveform patterns as well, but also what we can also see from that which I spoke to mark about. You know that waveform was talking about. We had one where it was more of a triangle wave. Okay, so he mark was like well, we've definitely got an issue here with potential, the issue with the rotor, because it should be sinusoidal and it's more of a triangle.
Speaker 2:Yeah, so then now the different patterns can now point to different failures of the actual motor or the rotor itself and I suppose this factors into now one of the things that mark's had a lot of support with and and it is definitely becoming more and more it was always a topic that we spoke about in vibration analysis. It was always okay how do we do electrical motor testing? But it's definitely become a lot more pioneered, especially with the motor doc, with howard penrose and what he's doing on linkedin and he'll be sharing a lot of things. He's him and mark have been in communication quite a lot, yeah, but this is motor current signature analysis now and it's becoming a lot more as a tool isn't it.
Speaker 1:I think this is the future, personally, and I think it's really difficult for people to get their head around, if I'm honest, because I think vibration analysis at first, even when you explain that to people, is black magic. But when you start to go into electrical signature and you start to say we can start detecting different faults and different things and bearing defects and bearing defects. The problem is it's really hard for people to get their head around mechanical faults to be detected electrically, yeah, and it is difficult because you know your subconscious belief system doesn't cover. No, they don't relate, they don't mix. But again, you've got to be open-minded in this game and I do think that there's a huge place, place for mcsa in the future. We've now teamed up with mark as well. We've got um, a motor dock on the way. Yep, we've got 80.
Speaker 2:Uh and I think for the for people trying to understand the difference. So vibration analysis is using a vibration transducer to measure that acceleration value through a piezoelectric crystal motor. Current signature analysis is using the variability of the air gap and that current variable to generate that same sine wave of a frequency and a spectrum is generated, isn't it? Within the motor dock.
Speaker 1:Yeah, so it's FFT that is generated from the electrical signal, opposed to FFTs can generate from any signal.
Speaker 1:Right, so we can get a spectrum out of that particular signal. But the beautiful thing of getting a spectrum out of that particular signal is that you can measure now the the actual damage of your rotor bars, because the higher the sidebands you get through, that electrical signal will give you a really clear indication whether you've got issues with your rotor bars. It's really difficult to do that from a oscilloscope perspective. It's a bit like it's a bit like trying to detect faults in your time waveform with vibration analysis. Really difficult to do because you just got lots of different signals playing together.
Speaker 1:But what mcsa is now doing is being used as a spectral analysis but also because it can measure that frequency in that air gap. It can also measure certain elements like vibration analysis can go as far and even like a bearing defect, because it can still identify them fault frequencies as a frequency as as an event that's happening through that variable air gap, because remember as well how high it can actually measure and how quick it can sample is it a lot, a lot more high frequency in sampling in mcsa compared to yeah massively, yeah, so that's why I can also can take a real high number of different failure modes and it's incredible what we can get out of it and what.
Speaker 1:I don't understand it fully yet, but some of the things I'm researching into and obviously speaking to mark and following howard's kind of you know research there's still a lot more to learn with it but now, non-intrusively, you can start to detect a lot of issues. So say, for example, you've got submersible pumps, doesn't matter. Say now you've got a borehole pump. Say you've got assets that are so hard to reach, and if we can start to mitigate these failures just as effectively as we can, vibration analysis and detect the electrical frequency faults, potentially the winding the rotor. Who tests their rotors?
Speaker 2:nowadays and what interests me as well I was talking about this with the customer the other day is that we're talking about with this particular customer about IoT and how things are moving forward and AI and the limitations. And technology is moving so quickly right now, but one of the limitations that we have with vibration analysis and IoT is you've got sensors all dotted around the factory gateways all over the place communicating. Well, if MCSA becomes a place where it is as good as or gives enough of an indication of a problem, of a bearing on a motor, or even and in some cases I've seen in the motor dock, we're measuring vane pass rates and blade pass and gear mesh frequencies as well, not even that Belt Belt.
Speaker 1:It can let you know. Michael was on the phone to me the other day. Yeah, they did a test and it could say that the belts were too tight and you tell me how va is going to do that, there's certain failure modes that we struggle with.
Speaker 1:Yeah, yeah, and I, and I'm trying to get my head around that. How is it done that? And and again, it's just another level of understanding, because there is science behind it and when you go deep enough into the science, it so you think, though, if this does become a very important technology in the future and the world is moving towards AI and acquiring data.
Speaker 2:well, it's a lot easier to acquire motor current signature analysis data from the MCC, from the electrical cabinet. You don't have to have nodes across the whole factory, it's all done in the motor control center. It doesn't have to be done out of the boa exactly.
Speaker 1:I mean you. You consider we've got a customer now looking to be able to to go to that 4.0 and and what we? What we're doing in the moment? We have to consider all options of vibration and motor current signature analysis together. But imagine now you don't have to cable run Everywhere.
Speaker 2:A million senses Gateways all over the factory. It's all done in the MCC, of course.
Speaker 1:Like for me, obviously, that motor air gap measure is in the motor. Like my testing that we're going to do is see how effective it can go further Further, yeah, how far is it? That for me, okay, I admit the motor, I'm quite confident it can do it. But I think one thing I'm trying to get my head around whether this is right or wrong. I don't want to make any statements because I've not done any tests. But what I think MCSA can do very well is identified as a problem. But what I think vibration analysis is really good at is uncertainty, and that is really key, especially with really so the way I see the future happening, like vibration analysis is not too expensive to do now. It's getting cheaper and cheaper. I do see these things probably going in line and each other and each other as well depending on if anything that we do in condition based maintenance don't rely on one technology.
Speaker 1:No, no, you know, and again we've done for me, because on certain assets and this has completely changed the way that we're actually doing condition monitoring because condition monitoring shouldn't just be vibration analysis. Condition one is monitoring the condition of the winding, the rotor, the other elements that potentially could fail to cause that asset to fail. Doesn't mean we do every month, but like we should be doing motor current signature analysis to check if the road is okay every six months as a minimum, we should be doing shaft current testing every six months because, say, for example, even you've put an emf core or maybe it's not working anymore and if you can mitigate failure, it's not difficult to do them checks on your very critical assets twice a year.
Speaker 2:And with a good maintenance strategy, with the right PMs in place, so that you're not wasting your time and you're being effective within your team. You can upskill, you can get the training. Mark can come with us. We can bring the tools and equipment to be able to do this. Or, if you haven't got the resources or experience, we can do it.
Speaker 1:We can help with it and that's what we're doing now. So when we're doing Femekas, what we're doing is identifying well, the rotor is a part of the asset. How are you demonstrating that the rotor's in good condition? Because if you're just doing an insulation resistance test, you're only testing the winding've got a whole moving part there that's really intrinsic to that whole function. We're not even testing it. We don't even know the condition. So this is where motor current signature analysis is so important. It is a driver to be able to test that rotor, but also it now has a full host of other things and all of this ultimately boils down to the criticality of the machine.
Speaker 2:You know, if this yes, you got any little conveyor, we're not going to worry about all that.
Speaker 1:But what we're doing, though, is the a8 assets, the really important ones, ones. You know ones even, yeah, and we want to be able to say have we extended the life of that motor for maybe five years, 10 years, to 30, 40? And if you mitigate that shaft current, if you ensure that you're doing the right test electrically, you reduce it a massive amount. You ensure the bearings are lubricated sufficiently and on condition and done with the best yeah, you know ones that don't bleed, and you can do that, and you can make sure that mowers is good, and we can extend the life of these assets for a very, very long time. Problem is, we've got so used to dealing with sailing, accepting failure, accepting the fact that these things, oh, that's lasted 10 years, that's done, great. It's like, well, what was it supposed to do? 30. Well, is that great? No, it's not, is it? If you look at that, it does not even reach it's expanding the Dell 10 life, and the problem is we've we've, just we're we're accepting failure way too too much. We've got to change our expectation of what failure means, and that means we do need to do these tests, we do need to understand what's important, we need to go to the end of the earth to understand the failure modes. And then also when we've done the for me, because we've got to put a maintenance plan in place that covers them. How many times have companies done Femicas? But also when we do them we've got an extra layer of knowledge. So when we've understood that, okay, you've got a motor and the rotor could fail or the winding can, we can't just say I'll just do a mega test, just do an insulation resistance test, because it doesn't actually cover the winding, because even that winding alone has different ways it can fail. Face-to-face winding, because even that winding alone has different ways it can fail. Face a phase, turn to turn. You've got all these other failure modes even in the winding game. Now, when I was winding, we'd get winding in and there'll be a different fault for each winding in the pattern. So you see, for example, one winding with just the phase burnt out well, that's the overload of the phase. You'll get one winding where you'll get turn to turn faults, right. You'll get one. You'll get one winding where the whole thing's burned out as an overload. So there's different fault conditions for the winding, but the main common one that we found was turn-to-turn failures, right.
Speaker 1:And a mega insulation resistance test, if you're testing between phases or to earth, doesn't even cover that one failure mode. So a lot of people that are doing this test, they think they're just mega in it. They, lot of people that are doing this test, they think they're just mega and they think, oh, that the motor is absolutely fine. Your insulation resistance test is nowhere near good enough to actually ascertain whether that wind is in good condition and you've got a static test. So we've now we invested in the adx tester, yeah, first, which we use in the workshop, using the workshop.
Speaker 1:But again, surge testing is a good way to understand whether there's any turn-to-turn faults, but you are inducing a voltage into the winding. Personally, if you're doing that in a workshop environment, I think that's fine, because if it fails, it fails, you deal with it. It's not good enough. Anyway, you don't want that going back out. Yeah, and you need to rewind that motor, simple as that, okay. But if you're doing that in the field and it fails, and it fails, great, you found the failure, but oh, great, now we've got an issue right.
Speaker 1:We've got a problem because you potentially, if you've you've stressed the- wine stress it too much, are the company in a good enough position to have a spare or whatever? Yeah, on a very large amount. I think it's too risky. So we've now bought all test pro tester which allows us to test a turn to turn fault through frequency rather than energising the winding Gotcha. So now that one failing mode of turn-to-turn shorts, which is a very common failing mode with windings, we can test and we can trend, yeah, and now we'll have a database with all of these assets the electrical test, the resistances, the impedances, the resistances to earth. We can still do insulation resistance, but we can also do that frequency test that allows us to take a check for them. Turning to their failures, so now, when we do them, comprehensive tests, we can say confidently that Moa winding is okay, right, as soon as we see any deviation from that, we see a trend drop or whatever. So we can trend with the all test. We can trend it. We trend them just like we would vibration analysis. We trend it.
Speaker 1:You need to trend, you need to have a baseline, we need to know when it started to come out of tolerance, because if it is, we can then plan. Yeah, yeah, we need to get a new motor. And let's be honest, who wants to be keeping a 300 kilowatt, 500 kilowatt spare? You don't want to sit in there. It's expensive, it deteriorates. But I understand why people still carry it, because it's a fear. But the thing is, if we can be so confident in what we're doing, our detectability right, we can mitigate the risk. We don't need to have all them spares sitting on the shelf, yeah, but then we can act really early. But it does need consistent testing and good condition-based maintenance, good discipline, discipline as well, and just saying oh well, that's been and this is another argument oh, it's been fine for five years, we don't need to test it anymore. It's like this is the most insane thing I've ever heard, because defects can occur randomly and that's what we test for.
Speaker 1:All it needs is to have a bit of load vibration, and then the winding gets loose in there and you get something happen and if you just say, okay, we're going to do it once every two years, then again it's a bit like vibration oh, we'll do it once every year or once every three months, then you're risking the risk. But I'll be honest, these tests don't cost a lot of money to do from the the assets, do you know? I mean company like us can set up a plan for it. We can say, look, this is how often it needs to be done. We can write it into your pms and we can make sure and close it off, close it off, and that's what we're going to do for a major company.
Speaker 1:We're looking at all of their electrical tests on their critical motors. But when you look at everything across the whole business, each particular site, there's not a huge amount because we're only doing it on the big players, the big critical assets and the ones that have a criticality RPM score and where we can reduce it with this particular technology. So this is where we need to go. It's not just vibration. Analysis is going to mitigate this. Now, if we really want to be reliable, we need to look at our electrical systems, managing our electrical motors effectively, exactly Because I think a lot of people have kind of probably not mastered vibration but a lot of people have kind of probably not mastered vibration, but a lot of people are doing it now.
Speaker 2:they understand the mechanical side of things and and there are a lot of people that have that good mechanical knowledge, but there's less people with the, the, the detailed electrical knowledge that mark's been able to bring to the table with this me, yeah, and it is becoming more and more important and I hope that for anyone listening to this has been able to come away and may rethink their maintenance strategy or look at their. For me, cause go, I didn't consider that and that is why we do this podcast.
Speaker 1:That is why we're listening to it, cause, I'll be honest, we didn't have this year. My knowledge, like you get areas within this. You know you're following the same strategy and all the rest of it, and you learn a lot. But the last year, how much I've learned with this electrical side. It's all coming full circle, though, because I used to be a winder and all of these things make sense now. I'm like ah, that's why we see that often, and now I'm coming back with this new knowledge. It's almost getting reinforced what I knew before, but also now, as you can see, I'm quite amped up and excited about this. I am, I'm getting quite passionate about this because I hadn't noticed. Oh, sorry, sorry, I cut myself down, but like now's the time to act. We are too slow off the mark. We're too slow Like everyone's talking AI. We want to do this, we want to do that. If we're not gathering data right now good data, sound data, doesn't matter about AI, you're never going to make it work.
Speaker 2:But you think, for the big companies, you know, with the companies that have got these really important assets, where's the value? Ai or doing good motor testing?
Speaker 1:Exactly. Where's the value? What are we doing now to make sure we're doing what we should be? What's the outcome of what our actions are doing? It's got to be always driven. How do we make that asset live longer? That's got to be always driven. How do we make that asset live longer? That's got to be the number one talking point, because the minute all we're doing is just saying when's it going to fail, that's not good enough. That is not what we need in this industry. If we want uk manufacturing to be here and be healthy, be strong and be sustainable, we've got to be thinking how do we make these assets live longer? Because if we can make them assets more reliable, it's more sustainable. So sustainable. We are so wasteful by just even if we detect things we're replacing, replacing, replacing and the line is still running. But also not even that. You know motor current signature analysis can see how efficient we're running our motors.
Speaker 2:Yeah, that one really interests me. So interesting With the motor dock what we're able to do is it can tell how much that motor is costing to run. We can laser align it and we can see the cost difference afterwards, which inherently has been quite challenging to measure.
Speaker 1:That it has been and I think there's some equations that we've been from aqua or whatever. But you want the proof in the pudding. You want that as a return on investment. Because how, how are you ever going to gain trust to say, we need to align all of your motors across the side? They're just not going to be bothered by it if they don't know what the impact is going to be and if it's going to be big enough for them to do. But I guarantee you're right. If we did that on two, we can model that and then you can say well, if we did this across your whole plant, this is how much energy you're going to go up. Yeah, we're living in hard times right now. Like, I think a lot of people forget the cost of living and everything we're going through. Like inflation's massive. It's huge. The cost of parts is huge. Lead times lead times is huge.
Speaker 2:Now, companies now are having to they're forced to look into reliability, and I think that's why the reliability has taken off. So you know they're having to do something because I'm not being funny.
Speaker 1:Back in the day when energy prices were next to nothing, parts were easy to access, do you know? I mean, plants could afford to run unreliably and still make a lot of profit. Do you know I?
Speaker 2:mean they didn't have to worry about the lead times, did they? Whereas it's like what do you mean that one tiny part is going to be 12 weeks?
Speaker 1:I need to make sure this asset is reliable and again, the end goal should always try to be making the asset reliable, because there's so many hidden benefits like sustainability and the cost of things, and if you do that across the world, then it does make a huge impact at the way that we live. But again, the problem is what is a lot of business drivers driven on? It is cost and it is money, it is profit. I get there's all these other sustainability things and goals and stuff around the company, but I'll be honest, the number one driver generally in the business drivers to be able to how do we make as much product as possible and how do we sell it for as much as we possibly can? But remember that is directly balanced out by good reliability, because if you don't do that, then you're not going to achieve your goals anyway and I think and on that note, it's a good way to end it. So, guys, thanks for tuning in. I'm going to just now sit down for it.
