#6: Smarter bridge maintenance with EIS: ProRail and Bjond Innovation on condition-based maintenance

How do you move from scheduled maintenance to maintenance based on the actual condition of bridges and other structures? Jo van Montfort of Bjond Innovation and Jos Kronemeijer of ProRail discuss EIS measurements, smart sensors, digital twins, predictive maintenance and extending the service life of infrastructure.
In conversation with:
• Jo van Montfort, founder of Bjond Innovation
• Jos Kronemeijer, materials technologist and systems specialist at ProRail
Host: Roel van Gils


Een zakelijke vraag aan BJOND INNOVATION? Neem contact op
Volledig transcript
Alle hoofdstukken (41)
- Maintenance based on actual condition
- Jo van Montfort and Bjond Innovation
- Jos Kronemeijer and asset management at ProRail
- The scale of the rail network
- Preventive maintenance and RAMSHEEP
- Why visual inspection is not enough
- Electrochemical Impedance Spectroscopy
- Natural ageing of materials
- Making EIS accessible to inspectors
- Working with C-Cube International
- A comparison with an ECG
- Baseline measurements for new coatings
- Establishing the baseline condition
- Non-destructive measurement and scarce expertise
- From measurement data to predictive models
- Measurement intervals and requirements for a paint system
- From practical application to standardisation
- Premature maintenance is wasteful too
- Finding the optimum time for maintenance
- Retaining old protective coating layers
- Abrasive blasting, primers and new cleaning technologies
- Laser cleaning as an alternative to abrasive blasting
- Remaining service life and replacement requirements
- Technology Readiness Levels
- Continuous monitoring with EIS
- Monitoring without a possession
- From demonstration projects to wider application
- Digital twins and just-in-time maintenance
- Real-time measurement and action thresholds
- Six stages of coating degradation
- Concrete as protection for steel
- The similarities between coatings and concrete
- The Zeelandbrug as a real-world example
- A coating system that has been working for decades
- Smart infrastructure at ProRail
- New materials call for new monitoring
- Low-carbon cements and steel grades
- Geopolymers and unknown long-term behaviour
- What will maintenance look like in ten years?
- Extending service life, reusing and recycling
- The technology still depends on expertise
Maintenance based on actual condition
Roel van Gils Welcome to another episode of Beton en Staalbouw, the podcast. The Netherlands has thousands of bridges, viaducts and other structures that are used intensively every day. Many date from the 1960s and 1970s. They are gradually reaching an age at which maintenance is becoming increasingly important. At the same time, asset managers face a huge challenge. There is more infrastructure than ever, maintenance is becoming more expensive, possessions are scarce and the impact of disruptions on society is only growing. When can we move away from maintenance based on a fixed schedule and switch to maintenance carried out according to an asset’s actual condition? That is what we will be discussing today with Jo van Montfort of Bjond Innovation and Jos Kronemeijer of ProRail. Welcome to you both.
Jo van Montfort Thank you.
Roel van Gils Perhaps you could briefly introduce yourself first, so we know who’s here.
Jo van Montfort and Bjond Innovation
Jo van Montfort I’m Jo van Montfort. I’m the owner and founder of Bjond Innovation, an engineering consultancy specialising in steel and concrete. We mainly deal with difficult issues involving extending the service life of these materials, their deterioration and arbitration cases. That’s a brief overview of what we focus on day to day.
Roel van Gils I’m sure we’ll come back to that later. Some great examples. Thank you. Jos?
Jos Kronemeijer and asset management at ProRail
Jos Kronemeijer I work at ProRail in the Asset Management domain as a systems specialist. I help draw up design requirements, specifications and guidelines for materials used in structures such as bridges, viaducts and tunnels. Maintenance and management issues are part of my remit. I’m a materials technologist. I studied Materials Engineering and work in the field of cementitious composites, including cement, grouts, mortars, concrete, coatings and so on.
Roel van Gils I think so.
Jos Kronemeijer Yes.
Roel van Gils Right. Welcome again. I’d like to start with ProRail. Of course, many people see the trains running, but don’t realise how many structures they travel over or under. Can you give us a rough idea of the scale of the maintenance challenge?
Lees het volledige transcriptTranscript inklappennog 38 hoofdstukken
The scale of the rail network
Jos Kronemeijer There are around 7,000 kilometres of railway line in the Netherlands. And every few kilometres, there’s a culvert.
Roel van Gils That’s a huge amount, yes. And in steel, concrete and all sorts of other materials, of course.
Jos Kronemeijer Yes.
Roel van Gils And what makes a railway bridge different from, say, a provincial road bridge?
Jos Kronemeijer The fact that trains run over it, at a very high frequency in the Netherlands, so the usage intensity is enormous. One important difference in the loading is that salt isn’t spread on railway viaducts and bridges, for example. The structures we build are different, and fatigue in particular, long-term behaviour in terms of fatigue, is very important for us. There is also a strong link here with the maintenance systems we use, as well as the products we apply, such as coatings, which we’ll discuss later. These have a strong correlation with fatigue, among other things.
Preventive maintenance and RAMSHEEP
Roel van Gils Yes. So, if we look at how maintenance is traditionally carried out today, how does that work?
Jos Kronemeijer Like all major public asset managers, we use the RAMSHEEP method, among other things. It’s a kind of risk and performance analysis tool. The acronym RAMS stands for Reliability, meaning the reliability of the system; Availability, because the assets need to be available; Maintainability, meaning how easy they are to maintain. That’s the central theme we’re focusing on now. And Safety, for the surrounding area and users. Then, of course, there’s the interaction with the surrounding environment.
Roel van Gils Yes, but do you largely work to a fixed schedule?
Jos Kronemeijer Yes, we have to. We can’t make it, so to speak, dependent on failures. By then it would be too late. It would be chaos if we didn’t do that.
Roel van Gils But perhaps preventive maintenance?
Jos Kronemeijer Yes, on a large scale.
Roel van Gils Yes, that does happen. So it’s not just based on a fixed calendar?
Jos Kronemeijer A combination.
Roel van Gils A combination, yes. Right. Because visual inspection alone is no longer enough, of course. Jo, your approach ties in perfectly with that.
Jo van Montfort Absolutely.
Why visual inspection is not enough
Roel van Gils Why is visual inspection alone no longer enough?
Jo van Montfort Well, as Jos said, when it comes to coatings, you can obviously see when a coating fails. Anyone can see that. But by then, the coating has deteriorated to the point where you’re into what’s known as corrective maintenance. For most clients, particularly ProRail, that’s a huge problem because track possessions are very short. That means repairs can often only be carried out at night. Or an entire stretch of track has to be closed, with all the consequences that entails, including rerouting all the services that use it and all the disruption that comes with that. So that’s not desirable. We need solutions that extend the service life of the existing protective coating or system on an asset as much as possible. And that means you need to know when it’s going to fail. So you want to assess the coating using data, not just by looking at it.
Roel van Gils So visual inspection has its limits, because a coating can look perfectly fine on the outside even though deterioration has already begun?
Jo van Montfort Absolutely. Someone can look perfectly healthy and be admitted to A&E with a heart attack that same afternoon. Or think of tooth decay. It can go unnoticed for a long time, even though the cavity is already there. Then, at some point, it gets through and the decay reaches the nerve. You feel it then, but by then it’s too late. That’s why you go to the dentist for preventive check-ups. They might look and say: ‘Hey, that enamel is starting to wear away and erode a little.’ It’s no different with coatings. You can measure that too. But in that respect, the coatings sector is a bit further ahead. We’ve developed techniques to measure deterioration, the degree of deterioration, or the health of the coating, in more detail. And non-destructively.
Electrochemical Impedance Spectroscopy
Roel van Gils What is that technique called?
Jo van Montfort It’s called Electrochemical Impedance Spectroscopy, a rather impressive name.
Roel van Gils It’s a term that’s widely used in the field. It sounds pretty complicated. But what do you actually do? Explain it to us.
Jo van Montfort Well, it’s a non-destructive method. As I said, you place two electrodes on the surface of the steel. You pass an alternating current through them and vary its frequency. To keep it simple, you’re measuring the resistance and barrier properties. Ultimately, the barrier properties are what you get out of it. You have to bear in mind that a coating like this is 300 micrometres thick, which many people don’t realise. Not even half a millimetre. The specified thickness can even be 350 micrometres. Those are the requirements. That’s about the thickness of a hair on your head. And it often consists of three or four layers. That’s what has to protect the steel for 20 or 30 years. That’s what we’re measuring. We’re essentially measuring that thin layer. That’s more or less how you should picture it. The result is a percentage reduction in condition. Is it green, orange or red? That’s essentially what you get.
Natural ageing of materials
Jos Kronemeijer Perhaps I can add something to what you said. Everything we build ages over time. Natural deterioration, exposure to UV, weather, wind and so on. Among other things, specialists like Jo look at, as do we as asset managers, are the chemical influences from outside. They cause these materials to age, quite literally. The sooner you catch that… that’s part of preventive maintenance. Periodic inspections are on the preventive side. If something happens between those preventive checks, you carry out what’s known as corrective maintenance. You then take action.
Roel van Gils And how do you decide where to take the measurements?
Making EIS accessible to inspectors
Jo van Montfort That also depends on your knowledge of the material you’re looking at. So there’s certainly that to consider. We’re now working on standardising and formalising the process, because at the moment it still requires real experts and specialists. We want to bring it to the same level as coating thickness measurement. In the 1980s, coating thickness was measured by scratching the paint and using a magnifying glass to see how thick the layer was. Then a magnet came along and the man who invented it was thought to be crazy, but in the end he was proved right. Today, all measurements on paint systems are made non-destructively using a coil and a magnet. They’re accurate to the micrometre. That’s where we are now with EIS, too. We want to formalise an existing technique so that, in 80 to 90 per cent of cases, anyone who has completed a standard coating inspection course can carry out and even interpret the measurement. Of course, there will still be 10 to 15 per cent of cases that are really difficult. But that means Bjond still has a reason to exist.
Roel van Gils Yes, but was the EIS method developed by you, by Bjond?
Working with C-Cube International
Jo van Montfort It wasn’t developed by Bjond. It’s a method that was developed in laboratories in the 1970s and 1980s, mainly by Americans, and was adopted by European laboratories and further developed there. A company like C-Cube in Delft, for example, is a university spin-off. They’ve turned it into a practical system, and we work very closely with them. Their coating… Sorry, the measurement system they’ve developed is very robust, portable and reliable. We’ve tested it ourselves, of course, because we want to know what we’re working with. In the outdoor environment, in the field.
Roel van Gils But do you attach it to the structure?
Jo van Montfort Yes. So how does it work? First, you place a number of small sponges under a magnet on the structure, on the coating, to let the coating acclimatise to its surroundings. It needs a conductive medium, so you have to dampen it slightly. You leave it there for a week or so. Then you come back after a week, place your measuring electrodes on it and take a reading within two minutes at the point where you want to know the properties. Each measurement takes about one to two minutes per point. So within a few hours, you know exactly what condition the coating is in, on the underside, the top, the side, the inside, wherever.
A comparison with an ECG
Jos Kronemeijer Yes. There may be a useful analogy with the example Jo just gave. When you have an ECG, you stick electrodes to your chest and they need to make good contact. These sensors need to make good contact too.
Jo van Montfort Analogies, yes. That’s right.
Roel van Gils You’ve already touched on it. How often do you take measurements?
Baseline measurements for new coatings
Jo van Montfort The measurement frequency… So you take a baseline measurement at a point in time of your choosing. Ideally, you do that when the structure is newly built.
Roel van Gils So, on the newly applied coating?
Jo van Montfort Yes. The system is also powerful enough to determine the quality of a newly applied coating. I often say that a coating like this is very thin, but also very difficult to apply. Painting a steel structure involves quite a lot. Conditions have to be right. The temperature has to be right. The humidity has to be right. The substrate has to be clean. There are a lot of requirements.
Roel van Gils You mean things can already go wrong at that stage?
Jo van Montfort A lot goes wrong at that stage, very often in practice. Especially when there is also a possession-free period on ProRail projects.
Roel van Gils It has to be done quickly, of course.
Jo van Montfort It has to be done quickly. So you have to apply specialist products in difficult conditions, and then they still have to cure properly. You’re dealing with a curing process and materials that react with one another. If the temperature isn’t right, which is almost always the case in the Netherlands… Last week it was far too hot, and before that it was far too cold, or it’s too wet. It’s never right, never ideal. The products aren’t ideal either, so you’re always compromising on a number of factors. But what you really want to know is how good the finished coating is.
Establishing the baseline condition
Jos Kronemeijer Yes, maintenance begins in much the same way as with a car fresh from the factory. It starts the moment it rolls off the production line. You measure its condition so that you can then track its degradation over time.
Jo van Montfort Yes, exactly. You can also use it to establish the baseline quality. Then you can make clear agreements with your contractors, saying: look, this is what we’ve agreed: it looks pretty good. The colour should be right. The thickness is sufficient and it still adheres well. Testing adhesion is destructive, though, because it damages the coating, which is something you don’t want. So you have to repair it afterwards. This technique is non-destructive. You can establish the condition at the outset and say: is it at 80%, or have you actually achieved the 90% you need to reach?
Roel van Gils That seems rather low to me. You’d expect to achieve 100%. But it’s never 100%, then?
Jo van Montfort As I just said, applying it is so complex that we could only achieve 100% theoretically, in the laboratory. That’s where a paint system like this is tested, through laboratory tests. I always say, rather unkindly, that these tests are carried out on flat panels. But a real structure… Just look at the corners, joints and welds.
Roel van Gils Where coverage is actually limited.
Jo van Montfort Yes. And corners and joints that are hard to reach, where conditions are unfavourable for the coating’s eventual performance. It’s the coating’s performance that matters. And you want to know that at the start of the whole cycle: what is my baseline?
Non-destructive measurement and scarce expertise
Jos Kronemeijer Perhaps I can add something. In the past, determining the condition was almost always destructive. Sometimes you cause a little damage while assessing the condition, and then you have to repair it. That means more work. Monitoring systems that you simply stick on and then remove without causing any damage, so that the process is non-destructive, offer a major advantage. You also need the expertise to carry this out in the field. Smart monitoring systems can provide that, and demand for them is growing. The number of qualified technicians with the specialist knowledge needed to do this is falling dramatically. We need to address that too.
Roel van Gils Several things are happening at once, and they all come together in these kinds of new techniques.
From measurement data to predictive models
Jo van Montfort Yes. You get a huge amount of data from the structure. That data tells you something, but you still need to be able to interpret it. An expert has to develop the prognostic, predictive model. As I said earlier, we’re working to standardise this as much as possible. It turns out that standardisation is feasible. Many of the paint systems used on the structures around us, bridges, lock gates, flood defences, you name it, as well as chemical plants, tanks and so on, are similar. That’s because paint suppliers keep an eye on one another and borrow ideas from each other. The basic binders are often the same, so their basic behaviour is often similar too. That should make it easier, in inverted commas, to develop a good predictive model.
Measurement intervals and requirements for a paint system
Roel van Gils Jo, you say that in an ideal world, you should take a baseline measurement. What about the intervals between measurements? Are they longer at the beginning or during the initial period, and shorter later on?
Jo van Montfort Yes, it depends on a few things. It depends on the ultimate requirements the client sets for the paint system. There’s OGOS, an organisation of major clients in the Netherlands that is also involved in this technology. It may be worth mentioning them. OGOS has a number of members: Tata Steel, Provincie Zuid-Holland, ProRail, Defensie, de Vlaamse Overheid, Gemeente Rotterdam, Gasunie and, of course, Rijkswaterstaat.
Roel van Gils And what do they do?
Jo van Montfort They consult with one another, sharing this knowledge and initiating projects to test the reliability and reproducibility of these techniques in practice. They can see that this offers major advantages over simply inspecting the paint.
From practical application to standardisation
Roel van Gils So is it still only being used in pilot projects?
Jo van Montfort No, it is being used. But to make it truly standard practice, to standardise it properly, so that the 80/20 or 90/10 rule becomes a reality. We are fully behind making that happen, of course.
Jos Kronemeijer Perhaps this partly answers your question. Bringing that knowledge together builds experience with different coating systems across a very broad range of applications. That helps with calibrating your system at the outset, whether it’s an application for ProRail, Rijkswaterstaat or somewhere else. Based on the product system, people can estimate what the inspection intervals should be. You set those intervals by calibrating the systems at the outset, at T0 when they are commissioned. Then the expert system and the experience tell you: okay, we’ll choose those intervals. We know that if we schedule them then, we’ll be in time.
Jo van Montfort Whereas now, it’s often too late.
Roel van Gils Exactly.
Jo van Montfort Yes. But on the other hand, sometimes it’s too early.
Premature maintenance is wasteful too
Roel van Gils Does that happen too? Surely that’s better, though?
Jo van Montfort Yes, but it wastes a huge amount of material, of course. It drives up costs.
Roel van Gils What do you mean? Measuring too early, or…?
Jo van Montfort Carrying out maintenance too early. A bit of rust is enough for someone to decide that the coating is no good, simply because they don’t know. If you don’t measure, you don’t know. As the EIS measurement clearly shows, parts of a structure, such as the underside of a bridge, may have coating that degrades faster than the coating on top. You might not expect that, but it’s often the case because the underside stays damp for longer and more dirt collects there. That’s less of an issue on ProRail bridges, but it is on ordinary bridges exposed to salt. These are also findings from the research into paint systems. We have clear examples showing that a different system should have been chosen for the underside than for the top.
Roel van Gils And at the moment, the same system is used everywhere?
Jo van Montfort Or even the wrong way round, because the assumptions are wrong.
Finding the optimum time for maintenance
Roel van Gils Yes. So it comes back to the knowledge you mentioned earlier.
Jos Kronemeijer Yes. You can also compare it with other fields. When your car is due for a service, it needs fresh oil. Doing it on time is good for the vehicle, but it also means taking it out of service, reducing its availability. Leave it too late and you’ll cause damage. So finding the optimum time also depends on how it is used and the operating conditions. A car will use more oil in the desert, say, than in a cold region, for example.
Roel van Gils So you look at each structure differently, individually?
Jo van Montfort Absolutely. As I said earlier, the requirements for the structure in terms of service life and use are very important. Its environment matters a great deal too: is it by the coast, or in a chemical environment, exposed to chemicals? The complexity of the structure also matters. How is it put together? Is it a clean, simple structure, or does it have a lot of latticework with riveted connections? You’ll be conserving it. These structures often still have very old protective coating layers.
Retaining old protective coating layers
Jo van Montfort Well, Jos knows… People often say those layers have to come off. If we measure them, that isn’t always such a good decision.
Roel van Gils They have to come off?
Jo van Montfort So we also looked at a number of projects together with ProRail where the conclusion was: okay, if you repair it, it could easily last another 50 years. Whereas older protective coating systems that are initially thought to need stripping off… The new coating systems applied in their place fail within 15 years. And the old ones have simply been there for 50 or 60 years. The drawback is that they pose occupational health and safety risks. So if you have to strip them off, you naturally come across substances such as lead and chromium-6.
Roel van Gils Oh, then you have different problems, different challenges.
Jo van Montfort But as long as you leave it in place and it does not leach into the environment, it is not in itself a problem for people or the environment. But it does provide extra protection for the old structure. And this technology is also very effective at monitoring that. Of course, it is up to the client whether they want that or not. Some clients say: I don’t want it, I just want it stripped off and to start with a completely clean slate.
Abrasive blasting, primers and new cleaning technologies
Jo van Montfort And that brings us to the point where, if you’re starting with a clean slate, to pick up on that for a moment, if that’s okay.
Roel van Gils Yes, go ahead.
Jo van Montfort Then you say: right, we’ll abrasive blast it. We’ll blast it all off, with lots of dust and a lot of hassle. Then we’ll apply a primer and build up the paint system again, layer by layer. But there are now moves towards saying: well, we’re not going to remove it using abrasive blasting. Because when you use an abrasive, it’s an active method in which you propel particles at the steel. You throw something at it, and that makes something come off. You can imagine that dust and particles are left behind on the surface. The primer is meant to ensure that the paint system adheres properly afterwards. If I could get the surface perfectly clean, I wouldn’t need it. I could leave out the primer. That would save material, and give better adhesion too. New laser-based cleaning technologies are being developed for this. They are under active development. Until recently, they were considered too limited because they were too slow. Abrasive media have to be brought in. They have to be blasted onto the surface. Then the abrasive media, along with the contaminants, have to be removed. And there is the load on the scaffolding, which can be carrying hundreds of thousands of kilos.
Roel van Gils Of course, you have to enclose everything.
Jo van Montfort Enclose everything, manage the dust. If you factor all that in… With a laser cleaner, you just shine light onto it.
Roel van Gils Literally.
Jo van Montfort Literally. The coating evaporates. Just that thin layer, a coating as thin as a hair, evaporates. I’d vacuum it up with a little vacuum cleaner so it doesn’t end up in the environment or in my lungs. That means you can work much more cleanly.
Laser cleaning as an alternative to abrasive blasting
Roel van Gils But how far along is that technology?
Jo van Montfort It is very advanced at this point. It is widely used in the United States, partly because dust generation is prohibited on many projects there. Unfortunately, Europe, and Western Europe in particular, is lagging well behind. I would strongly urge people to take a good look at it. This technology has a very promising future, especially where abrasive blasting is simply not possible. But it could be used on a large scale, including on important structures. Particularly for preventive maintenance, where it gives you a perfect substrate. And if you have a perfect substrate, you can imagine that you’ll also have a perfect protective coating system.
Roel van Gils A protective coating system in the end.
Jo van Montfort And I can monitor it with EIS to see whether that is indeed the case. So I can use a system like that to monitor very effectively whether what I’m saying now proves true.
Remaining service life and replacement requirements
Roel van Gils Right, okay. How does ProRail view these kinds of solutions?
Jos Kronemeijer To put it very briefly: in the same way. Perhaps I can offer a useful perspective, because Jo has said a lot of things that ring true. When you build new assets, bridges, viaducts, tunnels and so on, the design incorporates an intended service life. Put simply, the design life. In practice, the service life is sometimes shorter. But we live in a time when large-scale replacement often has no budget available. If you have to postpone replacement, monitoring the condition becomes even more important, of course. Monitoring lets you establish how degradation is progressing, so that experience can tell you: okay, a critical point will arise in two or three years. If you carry out monitoring like that continuously, you get a clearer picture. You no longer have snapshots, but rather a continuous film, making it easier to extrapolate into the future: okay, the downward trend has begun. The better we can predict where that point lies, the better we can determine the remaining service life. Knowing the remaining service life tells you when major repairs or, if necessary, replacement will be needed. And if you know that, you can budget for it and make the issue manageable. That is essentially why asset managers want monitoring systems to bring greater intelligence into the system, enabling them to manage costs: not too early or unnecessarily, but certainly not too late. Because large-scale replacement simply cannot be funded in the coming years.
Technology Readiness Levels
Roel van Gils But does that extend beyond the protective coating system, in terms of monitoring too?
Jos Kronemeijer Yes, definitely. You also asked just now whether this technology is new. Perhaps I can offer a useful perspective. New technologies entering the market, innovative technologies, have the drawback that they often lack a historical track record of reliability. This means the people behind them have to prove that they can be used in the field. After all, they are the ones who developed the technology. There is a system they use alongside this to assess it. It is called Technology Readiness Levels. A new technology or innovation can be ranked on Technology Readiness Levels from very low, level one: an idea in your head, tested at the kitchen table. Technology Readiness Level two: tested in the garage, on your own workbench. Right through to level eight, when it is established in the field. And level nine: available everywhere, with regulations in place. So you also need to factor in these levels. The technologies you are talking about are at levels eight to nine.
Roel van Gils Oh, they’re that far along already?
Jos Kronemeijer Yes. And that makes them interesting to asset managers, because the teething problems, limitations and strengths of these systems are much clearer. New technology is always useful and important, and you should always keep an eye on it, even if only because it can help solve future or existing problems more effectively. If you want to apply it, asset managers will be interested to know what level the new technology has reached.
Continuous monitoring with EIS
Jo van Montfort We were just talking about measuring condition using a handheld method, which I’ve just explained. And that is really what this discussion is about. Perhaps we can add that we have also developed a method for continuously monitoring coating condition. It was piloted at Philips Stadion. One of the pylons there is 40 to 45 metres high, and it has an aluminium casing around it. So it is a steel pylon, and the entire roof is suspended from it. There are eight of these pylons. We chose one that was relatively easy to access. You have to go up 14 metres in a cherry picker to install the sensors. They are then connected to a control unit via a wireless connection. This means you can monitor a structure that is difficult to access and cannot be inspected… There are countless examples. Think of high-rise buildings, bridges that are difficult to access, lock gates, you name it. There are also quite a few projects offshore, including offshore wind. The idea is to create an alert system. It is not meant to cover the entire structure. It is like in your car: if the oil level is about to get too low, a warning light comes on. It is much the same here. If the condition level is about to fall below a certain percentage, we get an alert. Then you know: okay, something is going on. You can also monitor the microclimate inside the pylon very effectively, at the same time.
Roel van Gils So you’re taking it a step further, really?
Jo van Montfort Yes. You get that extra data as well. What we don’t know at present is that dirt accumulates there and temperatures change. All of this affects the coating’s service life. We add all that data to the model, which we may talk about later, so that you can make the predictions increasingly accurate and easier to produce.
Monitoring without a possession
Jos Kronemeijer Perhaps I can add something quickly. If you want to assess condition for ProRail, you can do so physically within arm’s reach, or through monitoring using sensors installed on site. Taking the railway out of service is very difficult. It is not safe for the people who have to do the work. The impact of a possession is greater than that of a road closure. You cannot just put up a barrier. You cannot simply close a lane so that traffic can keep flowing, or send it a few kilometres around. That is not an option on the railways. Freight and passenger services all run on the tracks. This means that planning a possession has a major economic and social impact. The associated costs are enormous. So anything you can do through continuous monitoring, with more detailed information about degradation, its rate and its nature, is a clear benefit.
Roel van Gils Yes. But what needs to happen for this to be applied on a large scale at ProRail?
From demonstration projects to wider application
Jos Kronemeijer The concept has been under consideration for some time. It is already being used. This particular technology, however, is not yet being used on a large scale, because it is still relatively unfamiliar. There have been demonstration projects, and there is interest in it. Every asset has its own unique challenges. Take an offshore wind turbine, where access is an obvious challenge and distance makes things very difficult. Drones may allow us to do some things. But taking physical measurements on the coating systems themselves is what gives you the most reliable overall information. You have the overall condition and the unique local condition. Many people focus on a patch of paint starting to peel in a corner where water has been collecting for a long time. That can also be tackled in other ways. But the overall ageing of the entire coating system is another aspect that makes this approach so valuable.
Roel van Gils Yes. And it goes a step further still, at least in what Jo says about Philips Stadion: you’re also measuring the health of the asset.
Jo van Montfort Yes, exactly. Standard inspections are not going away. So let’s make that clear. It’s not that you no longer need to do them. You absolutely still have to, but you get an extra tool. And an early warning. If you use the permanent system, you can also take measurements remotely. So that adds another dimension to managing your assets.
Digital twins and just-in-time maintenance
Roel van Gils Yes. And you’ve already made the connection to a digital model yourself.
Jo van Montfort Exactly.
Roel van Gils What does that mean for a bridge, for example?
Jo van Montfort You literally create a digital twin.
Roel van Gils You’ve taken the words right out of my mouth.
Jo van Montfort Yes. The pylon is physically located in Eindhoven. And there’s a digital version of it on my computer. I can climb inside it and all around it. I can see what signal each sensor is producing and how it’s doing, so to speak. And I think that, looking to the future, you’ll be able to do this for a great many assets. Look at the data. But, of course, the future is an automated system running on that data. Think artificial intelligence, systems you can apply to it that recognise patterns very quickly: is something going on here or not? And probably issue an alert straight away. You’ll still need an expert, of course, or at least a person to do the final validation. And we’ll certainly advocate keeping that in place. But creating digital twins makes it very easy to monitor those assets and carry out just-in-time maintenance.
Real-time measurement and action thresholds
Jos Kronemeijer Yes, perhaps I can add something else. Those sensors and systems allow you to take measurements in real time. You can determine the condition at any given moment and make decisions based on that later. If the operating conditions are severe, you can shorten the intervals and increase the frequency. But it’s real-time, with remote access. If a certain threshold is exceeded and action is required, in other words, if an action threshold is reached, the asset manager might see a red light flashing on a dashboard. Or someone in the asset management organisation might receive a text message: at such-and-such a location, on this particular asset, we’ve established that the action threshold has been exceeded.
Jo van Montfort And then you respond, reactively, but in good time.
Six stages of coating degradation
Jo van Montfort It actually consists of six stages. First, there’s a stable stage. That lasts 15 to 20 years. Everything is still fine. You can’t see any difference either. The coating is still in good condition. In the second stage, it really starts to degrade. It starts to deteriorate a little. You can’t see that, because the coating looks perfect. There are no cracks in it. It still looks fine, but the coating is already damaged. So moisture can get in. I could detect that immediately with EIS, but I can’t see it. Then I reach stage three. That’s what you call propagation. That’s when things start to happen. It really starts to break down slowly. You’ll also see a little bit of delamination and, in brackets, a little rust. But then comes stage four. That’s where most inspectors say: now there’s a problem, now I can see something. But by then we’re already at stage four.
Roel van Gils The tipping point has already passed by then.
Jo van Montfort You’re already too late. It’s already past that point. The coating then deteriorates very rapidly, and its quality declines. It effectively pulls itself away from the steel. You often see that too, a coating curling up at the corners. They call that underfilm corrosion. But the coating itself is also coming away, because it literally shrinks away from the surface. And then, of course, it’s lying there. We never get to stage six. Sometimes we do, but then it has come away completely. This can be prevented by dealing with it properly.
Concrete as protection for steel
Roel van Gils Yes. People often associate coatings with steel, at least I think most people do. But Bjond does a lot more than that. Can you tell us a bit about it?
Jo van Montfort Yes. We sometimes say that steel is protected with a coating, of course. But in the concrete world, we protect steel with concrete. I sometimes call concrete a pretty good coating, rather irreverently. I don’t mean that in a bad way. It’s because I look at things through the eyes of a materials scientist. And from a materials science perspective, concrete is actually nothing more than… A coating is really just micro-concrete, as I call it. Why is that? Well, concrete is made mainly of sand, gravel, cement and water. The result is a stone-like material, which is ultimately called cement stone. This forms a thin layer of concrete around the steel. Steel is a crucial element in a concrete structure. And because the pH is very high, it puts the steel into a passive state. As a result, the steel no longer rusts. So that cement skin effectively provides protection.
The similarities between coatings and concrete
Jo van Montfort Just like a coating. A coating also consists of three main components. An A component and a B component, which you mix together. They contain some fillers that provide colour, as well as other fillers. And if you look at them at a microscopic scale through a microscope, they look very similar. So if you showed me a photograph under a microscope of a coating and of concrete, in black and white, you’d have trouble telling the difference: are you looking at concrete or at a coating?
Roel van Gils Oh, interesting.
Jo van Montfort That’s why. But of course, the material properties are very different. And we don’t think of a coating that way, but actually it is like that too. It’s also very hard and stiff. And brittle epoxy becomes increasingly stiff over time. That happens less quickly with concrete. Concrete has a slightly different deterioration mechanism. You get chloride ingress and carbonation. But ultimately, the properties of the outer layer of the concrete determine whether the steel inside it will corrode. So it’s the same. Physically, from a materials science perspective, it’s the same.
Roel van Gils So could you use EIS to measure the condition of the steel, as well as that skin on the concrete?
Jo van Montfort Yes, and that’s already possible. You can measure that too. A lot of research has been done on it. And as we speak, serious projects are being set up to take measurements in the field, just as we do for coatings. Yes, it’s at a lower Technology Readiness Level than it is for coatings, but I think it’s past level three. Four, perhaps even five.
The Zeelandbrug as a real-world example
Roel van Gils But you told me that you’ve done something similar, or something comparable has been done, at the Zeelandbrug too?
Jo van Montfort Yes, I’ve been involved in the problems at the Zeelandbrug for more than 25 years. The Zeelandbrug was built in the 1960s, in the early 1960s. The problem there is that chloride ingress occurred quite quickly, making the chloride content in the skin, the outer layer, which is called the cover, too high. If you compared it with today’s standards and requirements, you’d say: well, just demolish that bridge, because there’s far too much chloride in it. It’s rusting away beneath your feet, as they say in construction. Of course, that’s not the case at the moment, because in the late 1980s my former colleague at Intron and I came up with a clever solution to stop the corrosion. The problem is that there’s a lot of chloride in it. So, to compare it with a patient, it has terminal cancer. But we’re going to stabilise and contain the cancer. That’s essentially what we’re going to do. And that’s what we did.
A coating system that has been working for decades
Jo van Montfort It was an experiment at the time, a large-scale experiment. It’s been there for 35 years now and is still working very well. Yes. We were able to restrict the ingress of moisture and oxygen to the point where the corrosion process proceeds slowly. Yes. And, well, under current regulations, such as CUR-Aanbeveling 118 and similar documents… You couldn’t do that. But this is a good example of how solutions already exist for maintaining a concrete bridge of this size with a coating system. We’re talking about a five-kilometre bridge with a few hundred thousand square metres of concrete surface area. What’s crucial now is that the coating keeps doing its job. That’s the main thing I focus on. I’m not so much concerned with the concrete. That was the initial focus. We looked at the concrete. Now the question is: does the jacket covering the bridge, to put it simply, keep doing its job? We want to know whether the coating system remains in good condition. At the moment, we still use traditional methods. That means taking samples and testing them in the laboratory, looking at all kinds of properties. Chloride profiles and things like that, but also more complex tests to examine the properties of the polymers in the coating, and so on. Yes, you could also do that using EIS. We’ve carried out tests for that. So some exploratory tests have been done. They were relatively successful. But the project has currently been put on hold due to various circumstances, including budget constraints.
Smart infrastructure at ProRail
Roel van Gils And yes, do you see opportunities to apply these kinds of principles at ProRail too?
Jos Kronemeijer Yes. I’d like to pick up on a few of the keywords I’ve heard Jo use. He’s mentioned quite a string of them. If you can monitor the maintenance condition with fewer people and get a clearer picture of degradation rates than you can through occasional inspections, you’re essentially creating smart structures. They tell you themselves when they need maintenance. Just as a modern engine management system in your car tells you when the tyre pressure is low and you need to do something about it, and so on. The need for this will explode in the coming years. There are a great many possibilities when it comes to smart sensors. There are sensors we can install in road surfaces under normal traffic conditions that can tell you that the rear-left twin-wheel assembly has low tyre pressure. So some spectacular things are happening in the field of sensors. They’re getting cheaper all the time.
New materials call for new monitoring
Jos Kronemeijer And the properties you measure… The properties you measure in coatings and the properties of concrete that are relevant are sometimes just a little different. And yes, there is a lot of analogy between the composition and performance of coatings and concrete. Concrete as a construction material is many times cheaper than building in steel. And untreated steel needs protection from the weather, otherwise it will corrode. Concrete acts as a preservative for steel. The difference is in the price and the service life. Steel is of course more maintenance-sensitive than concrete. There are examples of concrete structures, albeit exotic ones, that are sometimes 2,000 years old. The Romans were the first to build with cast concrete. However, some concrete structures were put together rather crudely during production, with a few mistakes creeping in, meaning the concrete does not adequately protect the steel. The thickness of that covering plays a role, as it does with coating systems, but porosity is particularly important. And that can vary significantly from one location to another. When you take measurements of it, this technology also seems to show that it is highly suitable for doing so.
Low-carbon cements and steel grades
Jos Kronemeijer That brings us to the new types of concrete that are on the way. We can monitor not only existing structures, but a wave of new materials is coming to the concrete and steel industries. Steel grades are going to change. They will also require different coating systems. From now on, we will produce steel with a much lower environmental impact than before. Spectacular things are happening there. It is just taking a little longer. In the world of binder systems, meaning cements for concrete, one of my specialisms, the new binder systems with much lower carbon emissions, those low-carbon cements and low-carbon types of concrete, will behave differently. The thing is, we do not know how they will behave. We do know that we do not have time to keep working with the old conventional materials. Conventional cement production is incredibly damaging to the environment. The same applies to steel. And if you want to monitor both of them at the same time, while expertise in the field and maintenance budgets are declining, you need to be able to set up a smart monitoring system. Yes. And when you use different materials and the whole ageing process changes, you will also need to monitor those coatings more intensively. Sometimes, you may not want to use a coating at all, but instead apply the same system directly to the bare concrete surface, to tell you what is happening in the substrate as well. So the need is only going to grow. If nothing else, because the world is changing. Access to certain raw materials in Europe is changing. We will work with different materials more often, and everything will have to be built in less time anyway. At the same time, we want to extend the service life. And so on.
Geopolymers and unknown long-term behaviour
Roel van Gils Well, great. Anything to add?
Jo van Montfort No. Well, in the sense that Jos is right. These new types of concrete, the geopolymers, are becoming increasingly common. And indeed, their long-term behaviour is unknown. And we do not have time to wait and see. What we do have is this technology, which allows you to monitor them. And therefore to intervene at the right time. To build up data in the field so that, ultimately, you can make predictions. And ultimately improve and further develop the products. Think also of reusing old materials and old structural components. This is an increasingly important topic too. You see it coming up more and more in the literature and trade press. But that poses a challenge as well: how good is it? And how sound is it all? And how do we establish that? Do we load it beyond its limits, or what do we do? This is where these advanced techniques, as I’ll call them, including EIS, and especially their non-destructive nature, play an important role. They can be an important part of gaining a better understanding. These are also issues that are certainly attracting attention across the entire construction and civil engineering sectors.
What will maintenance look like in ten years?
Roel van Gils After that masterclass in materials science, if we sit down together again in ten years, what do you think maintenance will look like? Can you sum it up in a few lines?
Jo van Montfort I think a great deal will be monitored using remote sensing. Smart sensors will be built into most structures, and increasingly advanced and new materials will be used, as Jos just said. Application methods will change too, as I just mentioned. We will definitely no longer be grit blasting in ten years. I actually think we should have stopped doing it already, but anyway, that is just my opinion. There are more advanced techniques available now, and they will continue to be developed and improved, in combination with smart monitoring. And all of that linked to smart systems and artificial intelligence. I think that is the future. We will still need to apply coatings. Unfortunately, we cannot prevent corrosion. But we can manage it much better than we do now. That is my view.
Roel van Gils There we are, a great way to finish. Jos?
Extending service life, reusing and recycling
Jos Kronemeijer Yes, I’d like to build on the same picture. We already need to extend the service life of the existing stock and ensure it continues to function reliably. We will increasingly remove complete structural components and reuse them in a circular way, making their condition even more important. The expectation is that they will then be able to function without failure in their new setting for, let’s say, 25, 50 or 75 years. And that is only possible if you know the condition of the components you remove. We will recycle more often, reducing materials to their raw-material level and putting them back into use. Those materials may also find their way into coating systems. So if we can intelligently strip coating systems from structures and reuse them, that will also reduce the demand for raw materials. In short, we need to do more with less material, fewer people and lower costs, and make it last longer.
Roel van Gils There you go.
The technology still depends on expertise
Jo van Montfort Yes, so perhaps it is worth making clear to listeners that, fundamentally, this is not about EIS and EIS sensors. Ultimately, it is about the people who work with them. And it is about those people knowing where the first risks arise. That is what you want to know. If you have that firmly under control, the future is bright.
Roel van Gils Thank you both for all those great insights. And thank you for listening. Beton en Staalbouw, the podcast, is produced by Louwers Mediagroep in Weert.
Bedrijven in dit verhaal
Blijf dit volgen
Bij een nieuw verhaal krijg je een e-mail. Geen account nodig.
- ThemaBruggen en kunstwerken langer mee
- OnderwerpGeneral
- BedrijfBJOND INNOVATION



















