Inside a Railway Research Lab! We go behind the scenes at the University of Southampton

Added: 2026-05-06

[00:00:00]Britain has a proud history of advanced railway research.

[00:00:04]British rail engineers came up with the high-speed train, arguably the most successful high-speed diesel train the world has ever seen.

[00:00:13]British rail research came up with the tilting train design of the advanced passenger train until BR management scrapped the program only to see it reappear years later when it was sold back to us by a French-Italian consortium in the form of the Class 390 Pendolino.

[00:00:29]The railway technical center in Derby was a global center of excellence. And then, at privatization, we rather lost the plot.

[00:00:36]The research arm of BR was sold and research and development became fractured.

[00:00:42]Fortunately, though, the story did not end there and I have come to the University of Southampton to find out a bit more.

[00:00:48]We're going to see firsthand some of the groundbreaking research undertaken here including the soil mechanics laboratory where, amongst other things, they research that fundamental of the permanent way, ballast.

[00:01:01]The testing of wrought iron samples from old bridges by putting them under stress to breaking point and beyond. The centrifuge which increases gravity to recreate full-scale stresses within scale models in massively shortened time frames.

[00:01:15]The towing tank where we make some waves.

[00:01:18]The high-voltage lab where they can simulate lightning bolts.

[00:01:21]The anechoic chamber and its alter ego, the reverberation chamber.

[00:01:27]And finally, a six-axis motion platform that, amongst other things, can simulate the ride qualities of the not-so-terribly-missed BR Pacer.

[00:01:36]Professor William Powrie CBE is Professor of Geotechnical Engineering at the University of Southampton and a fellow of the Royal Academy of Engineering. And he explains a little of the history of rail research post-BR privatization as a general overview of the current focus of the group's work here at Southampton University.

[00:01:55]I'm, by background, a geotechnical engineer. So, that's a branch of civil engineering which deals with structures made in, on, or off the ground. So, in the railway contact context, that's very much about embankments, cuttings, retaining walls, tunnels, and latterly, the track itself because, of course, conventional track sits on a bed of ballast. The ballast is essentially stones and stones are a soil-like material even though the particles are very much bigger than in most natural soils.

[00:02:24]So, that's really where the focus of the work of the group here has been. And it all kicked off round about sort of 2003 when the research council, the Engineering and Physical Sciences Research Council, recognized that with the privatization of the railway system, what used to be British Rail Research at Derby wasn't really producing new stuff.

[00:02:45]So, they had a deliberate um initiative to try to kick-start railway-relevant research in British universities. So, a number of groups emerged out of that and in the sort of geotechnical side, the civil engineering side, the track side, uh that turned out to be Southampton. Of course, research is not cheap. And I asked William how, in general terms, rail research in 2026 is paid for and by whom.

[00:03:11]Obviously, this lab is quite expensive to run. So, one of the things that the rig behind me does, it does cyclic loading tests so we can look at how much uh a typical sleeper ballast arrangement will settle over perhaps 3 million loading cycles. So, even if we only charge a penny a loading cycle, that ends up to be quite a significant amount of money. But it actually costs quite a lot of money to run. So, it's important for us that the research is funded.

[00:03:36]And if you look at the funding journey, then initially, it was funded by the research council and that was about sort of getting the basic science going. And then, latterly, so we've had a strategic partnership with Network Rail for the last, probably, 8-10 years. So, where we have actually started to put this fundamental scientific understanding into practice looking at particular problems. But it's still necessary, of course, to make sure that it's applicable in a range of places, not just a solution for a specific place.

[00:04:05]So, uh a lot of our funding latterly has come from Rail Safety and Standards Board, Network Rail. And then, we do do work, we've had commissions from SNCF, we've done some monitoring work in Sweden, for example. And we've also done some collaborative work with railways in Japan, Spain, and so on. But uh our big funder at the moment is essentially UK railways, yes. Time now to have a look around. It's going to be quite the tour.

[00:04:29]We start with the underpinning of the very fabric of the railway. Why is the study of soil mechanics and ballast so key? One of the very practical applications is understanding how issues such as soil moisture deficit are affecting the railway.

[00:04:42]Kevin Briggs is an associate professor in geomechanics. So, as a university, we've been instrumenting a number of earthwork sites over many years. So, what I mean by that is cuttings and embankments and we have different sensors in those earthworks measuring soil moisture changes and pore pressures, which is another type of sensor.

[00:05:01]And the benefit of that is we gather data of how the earthworks are actually behaving to inform our models. And when I say many years, we have nearly two decades of information from an earthwork just north of here in Southampton near Newbury.

[00:05:15]And that's invaluable because you could go out and monitor an earthwork for 1 year, but it's not that useful because how wet was it that year? How dry was it that year? We can put those years in context.

[00:05:25]And then, look at the return period of how wet. So, 2006, for example, was a very, very dry summer and we have data from that. And that can inform numerical models of earthwork behavior and within a big project called Achilles, which was led by Newcastle University, but Southampton was part of that, we could then forecast the future. So, looking at climate change scenarios with these models that have been um calibrated against the 20 years of past data, we can then forecast 120 years into the future um looking at soil moisture deficit, which, as you say, meant measures um change in moisture in the ground, but that affects the volume change. So, shrink-swell. The study of ballast is particularly important and Dr. Madhu Murthy explains things about ballast I had literally no knowledge of.

[00:06:09]Right.

[00:06:11]Well, that's smoother.

[00:06:13]That's that's Well, I'm guessing that that one's used.

[00:06:17]Yeah, yeah. So, you don't see much difference unless you actually feel it.

[00:06:21]>> You really can feel that, though. Yeah.

[00:06:23]I mean, I it's obvious really. I suppose it's why pebbles on the beach get the shape they do, don't they, over over millions of years. But Yeah, makes makes sense. But, you know, when you put it in touch with each other, they initially rub rub each other, but then, once they find a comfort zone, They stop moving.

[00:06:39]They stop moving.

[00:06:40]So, this is the extraction of the ballast that has been CT scanned. So, obviously, it's an assembly and we can individually take it apart. Yeah. And we can uh like back in 10 years ago, we used to create potential particles, but now we can create the real ballast uh into what we call as discrete element modeling.

[00:07:00]So, obviously, it's a granular assembly.

[00:07:02]Granular assembly, you cannot Okay, now it's going to be technical.

[00:07:07]You can't model this as a finite element, that means continuous. We have to model it as a discrete. So, individual ballast will have its own characteristics like shape, etc., becomes more important.

[00:07:19]So, you can see the here I'm particularly interested to hear that the team's research has been able to show that reused ballast is mechanically just as good as brand new ballast, potentially offering major savings.

[00:07:30]So, the problem with ballast is that the more I know, the more I know I don't know. Brilliant.

[00:07:36]So, whether we, you know, we have found out a lot about ballast in the last sort of 10-12 years, but we've also found out quite a lot that we don't know.

[00:07:44]So, um one of the things that the the rig behind me there, that was originally set up in our old lab um which we moved from there to here in about just before the just before the pandemic. So, it's about 2019.

[00:07:57]Um so, the original rig was set up to look at the effect tilting trains. So, a Pendolino curves at a higher speed, so it exerts a higher lateral force. And so, we were looking at what is the effect of the weight of the train plus the curving force for a Pendolino and does that have any adverse impact on the um sleeper ballast system?

[00:08:21]And it was partly prompted by the observation that at the time that certain locations on the West Coast Main Line, the ballast would remove itself from the high end of the sleeper and end up in a little heap on the low rail because, you know, the curve is canted, so there's a low rail and a high rail.

[00:08:37]So, you ended up with this little heap of ballast. And we did an investigation into that with our field measurements.

[00:08:42]So, you know, we're here in this lovely lab, but the field measurements and the analysis is an important part of what we do.

[00:08:48]And it turned out that this really only occurred where you had got some feature like a a sleeper which wasn't properly supported, so it moved too much or possibly a a knocking welds. So, as the rail went over, you'd get an extra thump and then that would actually potentially cause the ballast to shift. Um so, in the rig, though, we have looked at ways that we can improve the lateral resistance of the ballast. We've looked at ways in which we can reduce the gradual settlement which occurs over over time.

[00:09:17]There is also a view that after a certain number of loading events and a certain number of tamping cycles, the ballast is spent and has to be replaced.

[00:09:27]But recently, Network Rail have developed a pretty good circular economy in reusing ballast. So, they take it out of the track, they clean it. Uh the stuff that's broken, they'll get rid of, but the good stuff, they will mix back in with fresh ballast and then use that in new jobs. So, we've done quite a lot of work proving that the settlement characteristics and the lateral resistance of this reused ballast is generally at least as good as the fresh ballast. And part of the reason for that is that the weaker grains tend to break early early doors and get weeded out. So what you're left with with the fresh ballast is um essentially grains that are quite strong.

[00:10:06]There is a little bit of loss of roughness. They lose a little bit of angularity, but they actually keep their fundamental shape. So if you look at a bit of ballast which has been through the mill as it were, you can tell that it is worn. That actually mechanically it's okay.

[00:10:23]Network Rail has some 30,000 bridges, tunnels, and viaducts on the railway network. Many of course are built to modern standards and are relatively new.

[00:10:30]However, many date back to the Victorian era when the materials used included stone, brick, and wrought iron. And the axle loads of trains, notably freight trains, were not the same as today's services.

[00:10:43]In the testing and structures research lab, PhD student Saba Ghasemi is testing samples from a wrought iron bridge quite literally to breaking point to understand the material properties after decades of real-world use.

[00:10:57]So these samples are taken from two old railway bridge and we cut them into that one shape coupons and we want to test them here. Tensile test and fatigue test to see their material properties. These samples have faced decades of real-world condition like rain, cold, stresses from train passes.

[00:11:22]Now we want to see how much the corrosion, how much the these fatigue change their material properties or >> Is that is that a piece sample has been exposed to the elements there? Right, so that's right. Okay, yeah. And after testing, we just realized the material is wrought iron. Yeah.

[00:11:40]You can see the laminar nature of the wrought iron by your just naked eye.

[00:11:44]Yeah. And wrought iron is really interesting material, very interesting.

[00:11:49]Because we have it has a really low carbon.

[00:11:53]So because of that, we see a different range of ductility.

[00:11:58]And also it's less stronger than the steel. A lot of aging bridges has this material.

[00:12:05]>> Right. So when we want to assess them, we need a very huge data set to validate these results. Actually, we use these results to validate our numerical modelings.

[00:12:16]I do some simulations of real aging railway bridges. And we scan all of these with a 3D laser scanner. You see these marks? It's because of the to see the pitted corrosion. Right.

[00:12:31]>> Exactly see where we have a corrosion and mass loss and everything.

[00:12:35]Professor Piry explains why this research has such a critical real-world application. So that's in the context of some work that we're doing look at the looking at the impact of heavy axle weight traffic on historic bridges. So we looked at masonry arch bridges and also metallic bridges and some of the old metallic bridges are wrought iron possibly even some in cast iron. There aren't so many of those now. And early steel or wrought iron and trying to look at, you know, what will the effect be of perhaps having heavier axle weight traffic going across that. And that's some work that we hope to continue into looking at the impact of increasing the axle weight on the embankments and actually also on the track itself. And that's in conjunction with a tool that's being developed with colleagues who specialize in this sort of thing, which is actually helping to plan a route and looking at for example for how long could you operate, let us say, heavy freight trains over this route over these assets. And we actually see that tool as being a way of potentially sort of capturing a whole lot of information about these assets and thinking about, you know, how do we maintain them? How do we repair bits of them in a timely way so that you can start to get things in time before something happens which is catastrophic or destroys your ability to run trains.

[00:13:54]One of the challenges of any research seeking to understand how assets or conditions can change, perhaps deteriorate over many years, is that fieldwork and taking constant measurements takes, well, many years.

[00:14:07]So the centrifuge at Southampton University is a vital piece of equipment. Dr. Katherine Squire explains how through the wonders of scaling factors, it is possible to model 25 years of weather impacts, for example, in just 24 hours. Yeah, so this centrifuge can go up to 130 times earth's gravity.

[00:14:27]130 times >> And it can spin one ton worth material up to that sort of increased gravity fields. So the speed at which it's going at is around for 100, it's probably just over 200 km/h. Wow, that's going real fast. Wow. Yeah. And the way it works is that you have a small model of could be a railway embankment, for example.

[00:14:50]Let's say that it's 1/100 scale.

[00:14:53]If you spin it up to 100 times earth's gravity, then it's as if it's the real thing.

[00:14:58]Yeah. And some processes are sped up sort of just like William was talking about. So for example, your model is smaller and so the water in the model has less distance to travel.

[00:15:08]And so things like settlement, movement of foundations, that can be sped up in terms of time. So one hour of spinning at 100 times earth's gravity can be in real time like as if it's one year.

[00:15:23]We then move on to one of my favorite test facilities we see on our visit, the towing tank.

[00:15:28]Dr. Rodolfo Oliveira is a senior enterprise fellow and he's going to give us a first-hand demonstration.

[00:15:35]But first, William explains just how important perfect calibration of equipment such as this is to producing reliable test data. So it's really to study the hydrodynamics primarily of ships, but there is a potential railway application which I can come on to.

[00:15:51]But what's interesting is that at 138 m long what happens is that the model ship is towed by a carriage which runs on these rails. So these rails have to be set absolutely precisely to follow the curvature of the earth because at 138 m you get a small number of millimeters of curvature of the earth on the water surface.

[00:16:12]>> It's it's it's just millimeters. It's just millimeters.

[00:16:14]>> is that important? Because because the water presumably >> Because that's the the water follows the the curvature of the earth and then the carriage needs to be a fixed distance above the above the surface of the water.

[00:16:26]But the sort of thing that you can do, so ordinarily you might test, for example, a high-speed train in a wind tunnel.

[00:16:33]But essentially it is a body with a fluid flowing past it.

[00:16:36]>> Yeah. So in the towing tank, and what happens in the wind tunnel, is that the body is stationary and the wind is blown past the body. What happens in the towing tank is that the body moves, which is as it would in reality, and the fluid is stationary, which is the water.

[00:16:52]But the water has different properties.

[00:16:53]It has a different density from air, it has a different viscosity, but all of that can be dealt with by the scaling rules. So by having a combination of the density, the viscosity, and the speed, you could actually model, for example, the aerodynamics of a high-speed train by looking at the hydrodynamics of a high-speed train shape in the towing tank. Well, um good news and bad news. So bad news that the model is out of the water because they are changing the setup.

[00:17:23]Good news, we can send bigger waves and go faster.

[00:17:27]Yes.

[00:17:30]The model is a free-running model that will try try to catch up with the carriage.

[00:17:35]That's the setup that we have. Usually there will be two beams here, dynamometer, tow post, and a vessel just as you saw on the video at the entrance.

[00:17:44]And then we tow it against the water at this still.

[00:17:48]So practically zero turbulence.

[00:17:51]And you can measure drag, side force heave, like basically it's performance against whether it's a regular wave or a sea state. So there will be an alarm and after that will be 5 seconds of a delay and then we will like all of this thing will move.

[00:18:11]Perfect.

[00:18:46]So that was 3 m/s. We can go up to 10.

[00:18:50]You can go 10? Yes. We we wouldn't be able to go with this configuration because there's like a number of people that can be maximum on the carriage and seated and well, so but we can go that.

[00:19:05]So we'll send some waves and when they are roughly midway down the tank, we will follow them.

[00:19:11]And you will notice that if you see a wave, you will be traveling at their speed individually, but as a group we will pass it by the time we go to the end. It's something counterintuitive, but you will you will see it, okay?

[00:19:41]It's something like 40 or 50 second delay.

[00:19:51]So we are at speed.

[00:19:53]If you see, we seem to be traveling with the waves at the same speed.

[00:19:58]But, the group is traveling half of the speed of the wave. It's almost like a because they are in circular motion, it's almost like a conveyor belt that they are going faster in the group, but the group is going half of that speed. We are coming to the end of the group because we are going faster than the group itself.

[00:20:20]But, we are still on the same wave.

[00:20:24]So, we It's almost like a tie.

[00:20:26]Here, we arrived at the same time, and then they go on and they hit the end.

[00:20:33]In the Tony Davies High Voltage Laboratory, Charlie Reed, High Voltage Test Engineer, talks to me about pantograph tests designed to improve the operational efficiency of this vital piece of equipment for any train looking to make use of overhead line equipment.

[00:20:48]Paul Lewin is Professor of Electrical Power Engineering and Director of the Lab. He explains how groundbreaking research that this lab undertook showed how electrification could be done without the need to always raise bridges. This significantly reduced the cost of electrification in Cardiff, on the Midland Main Line, and on the recent electrification between Wigan and Bolton.

[00:21:09]We've been working with Network Rail for over 15 years, primarily with that as our end objective.

[00:21:16]Fundamentally, what we're trying to do is to find safe operating distances between overhead line equipment and other infrastructure.

[00:21:24]Sometimes that means we have to introduce additional insulation.

[00:21:27]Sometimes it means we need to use equipment such as surge arresters, which will always act to limit the maximum voltage any overhead line equipment will see.

[00:21:37]Um ultimately, it means that when it comes to rail electrification, the costs of actual electrifying existing rail will be substantially reduced because we have We won't need to remove bridges or widen increase diameters of tunnels or move stations because we will have equipment will be will be at much safer, shorter working distances, and therefore will limit the cost of civil engineering.

[00:22:05]I um actually started my career as a civil engineering trainee with British Rail, and I was based in the Eastern region, and I did do some of the work preparing for the East Coast Main Line electrification. So, bridge raising and that kind of thing.

[00:22:18]And people sort of talk about Well, you know, the East Coast electrification falls over, but it's fault falls over all the time, but it's not really the structures. I mean, the head spans are difficult because they're wonderfully elegant because you just have a cable stretched across the line, and that supports the overhead wires. But, the problem is in the event of a dewirement, just kilometer suddenly hundreds of meters of it just just comes down.

[00:22:39]So, I think that there was a desire to kind of make the structures themselves very robust. I think they're probably designed for 140 mph, and there's also a desire to sort of make them almost like a Meccano kit, I think. But, where we came in was in the the pile foundations.

[00:22:57]So, the We actually had a call from somebody in Network Rail.

[00:23:02]actually came from somebody I used to teach when I was at Queen Mary, I think, in London.

[00:23:06]And uh he said, "Um my boss says that we've never installed any pile more than 5 m, but our designers are telling us they've got to be twice that. Why is this?" And sure enough, at that point, you would go up through Reading, and you would see all of these pile foundations sticking up 3 or 4 m out of the ground because they tried to drive them to these big depths, and they'd refused. So, you know, not only not only was this twice the depth of foundation which anybody had ever had in the past, but also they couldn't physically get them in.

[00:23:38]So, we did a a series of tests. It turned out that actually designed these foundations historically using a bit of an empirical method, but it was based on lots and lots of tests that had been carried out in Europe. Um you know, sort of in the 1950s and 1960s. And I think it was actually used to design all of the piles for the East Coast Main Line and also for HS1. There isn't a single instance of one of those failing due to inadequate pile length. There might be a couple that have got swept up in an embankment failure, but that's a different thing.

[00:24:09]>> going to say that. Consequential failure.

[00:24:10]>> thing. That's a consequential failure.

[00:24:12]So, uh the one slight problem was that that this was an empirical method, and it was based on all of these tests. The loads associated cuz there is a the bigger you make the structure, a lot of the load that comes onto this structure is the wind. And the bigger you make the structure, the bigger the wind load. And actually, it's it's the wind load that gives you the overturning, which is a lot of what the pile foundation has to resist.

[00:24:35]So, um it did turn out that the loads on these big structures were actually outside the evidence base that they had for the empirical method. So, we did two things. We did um a number of comparative analyses where we showed that if you used the sort of a more modern design method, which still has a degree of empiricism in there, in in particular in converting from 2D to 3D 3D, but um if we used that more modern design method, uh we actually got comparable results with the traditional method. And then we went out on site to a disused embankment at High Marnham, and we actually designed some foundations for the Great Western type loads using the traditional method, put them in an embankment, and tested them until they moved too much. And what we found was that the traditional method was perfectly adequate. So, at that point, the order went out, you go back to using the traditional method, they cut off the excess length of the piles, and cracked on with the job.

[00:25:33]>> And saved a load of money.

[00:25:34]>> And saved a load of money, and it also eventually enabled them to restart the Midland Main Line. So, it was part of a It was one of a number of industry-wide savings to try to get the cost of electrification back under control. The study of noise is a crucial area of research for railways, whether that's noise within vehicles such as railway carriages or noise generated by railway systems such as from wheel-rail contact or aerodynamic effects.

[00:26:03]Martin Towler is a senior engineer with ISVR Consulting at the University of Southampton, specializing in noise and vibration.

[00:26:11]We begin with noise and look at two very different facilities for measuring it.

[00:26:16]Yeah, it goes down to very low frequencies where it's anechoic. Right.

[00:26:19]So, at the moment we've got this hard floor on here. Yeah. Underneath that is this kind of creaky floor here. Yeah.

[00:26:24]But, we can take that up as well. So, all you're left with is the wires and like a trampoline floor.

[00:26:28]And then you get a very anechoic environment. So, with the door shut, you can hear your heartbeat. It's quite loud.

[00:26:33]>> Is that right? Yeah, yeah.

[00:26:35]Cuz the advantage of an anechoic chamber is you make a sound, it goes off in one direction, it doesn't come back again.

[00:26:40]Yeah. So, that way, if it's making different noise around it, by measuring it, you can see in in the directivity. So, where you know, where the sound's going.

[00:26:49]So, um if you don't do an anechoic chamber, you need a big field where it doesn't return either. Right. Okay. Yeah, so you don't get any echoes. So, anechoic, no echoes.

[00:26:57]>> Yes. Kind of for railway applications, again, you know, there might be bits within a train that you might test in terms of looking at how much noise it produces or the directivity.

[00:27:07]Um another thing we can do with this is that wall there can come out, and we can have a very um a very quiet um airflow across, and we can look at aerodynamic noise sources as well. So, we've tested um scale pantographs and bogies.

[00:27:22]So, we'd normally do a combination of doing um some kind of modeling like um computationally uh computational fluid dynamics CFD modeling, and then um then do some kind of mock-ups in here, and then we've also done some testing on HS1 and things like that of different pantograph um designs uh for noise. So, this is a a microphone array.

[00:27:42]So, it's 4 m diameter.

[00:27:45]So, this kind of uh this was set up on HS1, and that gives effectively, if you imagine like a thermal image of of the train going by, this is like an acoustic image. So, it shows the the noise coming from the pantograph or the bogies.

[00:28:09]And then you can use that to separate the various sources, but we can also do various modeling approaches as well. Um We can do numerical models of the track and um uh which include the track and train system to try and then identify those different sources.

[00:28:26]If an anechoic chamber is a space that is free from echo, reverberation, or sound reflection, then a reverberation chamber is, well, pretty much the opposite.

[00:28:37]Very hard floors, hard walls. If you look, none of the walls are quite um a particular angle so you don't get standing waves set up, and that's what these things here are suspended from as well to break up the sound field to make it as diffuse as possible.

[00:28:51]>> Yeah. So, there's kind of three applications main applications use this.

[00:28:54]First is where you can attach an aerodynamic make really loud noises and fatigue test things like satellites and stuff like that, acoustically fatigue test things.

[00:29:03]>> Right.

[00:29:04]Um the other thing is we can make a a noise in here, loud noise, and we can build say a wall, for instance, if you had the wall of a carriage of a train, build that in that where that blue doors are there, and we've got another chamber on the other side of that.

[00:29:18]And then we can measure the transmission loss from this room to that room in terms of frequency.

[00:29:24]And then the third thing is looking at absorption. Um so, one of the things we've done is fill this up with ballast up to 30 cm, measure the reverberation time, take the ballast out, measure it, and see how much absorption you get from ballast.

[00:29:36]And another thing we've done, as you can see here, is that we can track absorption.

[00:29:42]So, these are uh layers that go on top of the track to try and reduce as we said but yeah how much reflection you get off the slab. We end our visit at a facility that is unlike anything I have ever seen.

[00:29:56]This is a six-axis motion simulator, the purpose of which Martin explains. It looks a slightly scary sort of torture style chair but I'm sure that's not What's the purpose of this piece of equipment then?

[00:30:09]Yeah, I mean you're you're not far wrong about the experiments [laughter] we do here pretty close to torture. So yeah, this is a six-axis simulator so it can move in the free translational XYZ and then roll pitch and yaw so it can go up and down a meter side to side half a meter and then 10 to plus minus 10 degrees in roll pitch and yaw.

[00:30:27]So what we can do is record motions in an environment and then replay them very precisely on here.

[00:30:34]And depending on what you're looking at you can look at different factors. So at very low frequencies typically below 1 hertz you can have motion sickness.

[00:30:43]Um so Zach over there is doing a um his kind of part of his PhD is trying to look at understand motion sickness in military environments. And at higher frequencies um you begin to get comfort effects so the just the seating comfort and also particularly something like on a train where now it's becoming a work environment how well you might write or use a laptop and things like that may also depend on the on the frequency. And often then there's a um there's a kind of balance of static and dynamic comfort. So one of the things is that um things like the old slam door trains you had the sprung cushion seats and typically it was about twice as uncomfortable sitting on the seats as it would have been sitting on the floor.

[00:31:25]Um you know dynamically um and so there's been a trend towards thinner cushion seats so that makes it more comfortable dynamically. Um maybe gives you a bit more legroom as well and over a journey you'll be in theory more comfortable so that's why maybe in something like the Class 800s and things like that I think often people think Yes the seats are really hard and firm.

[00:31:46]Isn't it fas- fascinating that because that is exactly what people think? Yeah.

[00:31:51]And yet you're saying the science behind it the physics behind it really is actually the the the dynamic comfort will be better on the thinner harder um seat.

[00:32:01]>> Yeah certainly over a longer journey. So maybe the initial comfort getting in but also there's this balance because you've got the dynamic comfort and the static comfort. It is the first thing you often feel when you get into something like that is you go "That seat's really hard." but actually actually whether you feel the same at the end of the journey.

[00:32:16]And then again if you're trying to do a job on there or like you know write on the laptop you know you could literally see on those you know the you know the slam door trains people bouncing up and down on the seats going up and down.

[00:32:26]You know so actually trying to read a newspaper or something like that you know not that people do that anymore you know would have been quite challenging.

[00:32:31]So in a minute you can go on there and what we can do is uh put different frequencies of motion on and see how easy it is to write if you're happy to.

[00:32:39]And there's no better way of seeing how it works than becoming a lab guinea pig.

[00:32:44]Oh brilliant. Okay.

[00:32:49]So either do it in your lap or on the table you could try both even. Yeah. Um but um should get the other bit down there. Never used to get these on Pacers that's for sure. No.

[00:33:00]Emergency stop button here. If you press that it will just come down in a kind of controlled way shouldn't do anything unusual but yeah so um Zach's going to power it up so it's going to go up to roughly the level of the floor.

[00:33:12]Yeah just building magnitude so wait a minute or two and or a few seconds at least.

[00:33:17]Is that It's not much of a straight line it has to be said so.

[00:33:25]My writing's bad before it started so So this is okay?

[00:33:29]>> This is 3 hertz. So the magnitude is the same in each case in terms of acceleration magnitude this is just the difference is the response of the body and the well the body on the seat so the phone also.

[00:33:41]So yeah that's 3 hertz.

[00:33:48]Okay that's slightly >> is yeah So this is kind of pretty close I mean this is round about where the the natural frequency of the body.

[00:33:56]Where you've got a resonance it always increases the vibration there's not much you can do with a train to make it better because of the nature of the frequency of vibration coming in.

[00:34:03]>> But what's interesting about that is the worst one was 5 hertz.

[00:34:06]>> Yeah which is the seat person resonance 4 to 5 hertz so yeah.

[00:34:15]So you can escape now.

[00:34:22]Well you can almost see it here you can see the sinusoid in there. If you write if you if you run your hand at a constant thing so you can see there's a slightly higher Absolutely and I and I was trying hard to keep that straight but I don't I don't have the best handwriting in the world anyway but that Yeah and the 5 hertz and yeah you and you I mean obviously when you see the play it back you can visibly see yourself bouncing up and down on it.

[00:34:44]Fantastic.

[00:34:45]Superb.

[00:34:47]We've come to the end of the visit. I am mightily impressed not just by the sheer brilliance of the minds at work here but arguably even more so at the genuine real world applications that this work can inform.

[00:34:58]I ask William what comes next and how important it is that the work keeps going.

[00:35:03]So one of the things which is quite gratifying so I'm probably going to come up with three things but this is it might be like the Spanish Inquisition it might be before you know. So um so one of the things is we've got a number of projects working with particular routes on Network Rail which is really trying to sort of apply a lot of the fundamental scientific knowledge that we've that we've gained over the last decade plus into practice. So under sleeper pads for example are generally pretty good. So they reduce the rate at which the track settles.

[00:35:36]And one of the things that we found with the work that we did on rail stress management is that they also increase the lateral resistance of the sleeper.

[00:35:44]And we have a a project with one of the routes which is looking at whether if you put on an under sleeper pad can you then get rid of the little heap of ballast at the shoulder because those costs would balance out and if you do that you get at least the benefit in terms of increased lateral resistance from the under sleeper pad and you get reduced settlement because it's kinder on the ballast. So we've got a number of projects we're also looking at the particular case of the SMD and its relationship with the track geometry in the context of the West of England main line. So what's quite nice is that we've got a number of projects where we're sort of pulling through even further the results of the recent research into practice which will hopefully see some results.

[00:36:27]We're also quite keen on extending the heavy axle weight work so really understanding what's the effect of running faster trains heavier trains more frequent trains and extending that from the metallic bridges and the masonry arches through to earthworks and indeed the track itself so I think that is another piece of work we'd like to do.

[00:36:48]And the third thing is very much about how can we use new technology so robotics automatic inspection of assets perhaps digital image analysis maybe analysis of image using AI uh how can we sort of integrate that into the sort of physically based understanding that we have from our experiments and our observations in order to start having intelligent sort of maintenance across the piece. We've got some work we've been doing with colleagues in our optoelectronics research center which is about using optical fibers to pick up um rail strains so that could be both the longitudinal strain due to temperature effects but also potentially strains due to settlements and issues associated with the track bed. And uh if we could sort of extend that to use kind of fibers that are already there that would be a big win. That's a challenge and there's a number of other people working on that. So that would be the sort of third area is really sort of integrating new technologies and the sensible use of new technologies numerical analysis and digital twins because they're not going to solve the problems for us human ingenuity is still going to solve the problems for us but what we do need to do is make use of this amazing technology at disposal to enhance our ingenuity not replace it.

[00:38:07]Professor Powrie was recently awarded the CBE for services to engineering.

[00:38:11]And in a double celebration the University of Southampton also won the Queen Elizabeth Prize for education a highly prestigious award. Both were richly deserved.

[00:38:23]This has been a thoroughly enlightening day. I confess I had been concerned that ever since the demise of BR Research at Derby the vitally important work they carried out had withered away somewhat but it appears that fear was premature.

[00:38:35]Together with work underway at Birmingham and Huddersfield Universities what the University of Southampton are doing under Professor Powrie's leadership is vital.

[00:38:44]And it's vital because not only is it expanding our knowledge enabling us to be more proactive it's also saving money real money.

[00:38:52]The work in the anechoic and reverberant chambers is directly linked to demonstrating that the noise difference between slab and ballasted track is less than first thought reducing the cost of noise mitigation measures on phase one of HS2 by some 65 million pounds.

[00:39:09]In October 2020 and referring to overhead electrification David Clarke technical director of the Railway Industry Association said "I estimate that the research at Southampton on foundations will have saved the industry in the region of 600 million pounds by the 31st of December 2020."

[00:39:27]And in February 2025 Martin Frobisher group safety and engineering director at Network Rail said this.

[00:39:34]"The University has delivered breakthrough research which is significantly reducing the costs of our capital investment program. The high voltage lab at Southampton University has identified high-tech solutions which enable us to reduce the number of bridges and structures which we need to replace when electrifying railway lines.

[00:39:50]The geotechnical team have enabled us to safely reduce the size and cost of foundations.

[00:39:55]And the university team have found practical and durable solutions to many of the track engineering challenges we face.

[00:40:03]The need to improve the efficiency of spend on the railways, whether it is related to maintenance, renewals, or enhancements, and to understand and prepare for the challenges we will face in the future, such as through climate change, mean facilities and programs like those at the University of Southampton are more vital than ever. They need to be funded and supported because if they are, we all benefit.

[00:40:28]>> [music] [music] [music] [music]