Hod Carrier

I hope the board will forgive me for duplicating a lot of information already posted, but as it's mostly been scattered across the board in various other topics I thought it would be a good idea to gather it all together in one place for ease of reference and discussion. Other threads were this has been discussed are: LMS Articulated Railcar 1938 - Development of the idea and application for an articulated train My LEGO Renfe S-130 - Use for a Talgo design I had been thinking for a while whether it might be possible to come up with a way to close couple LEGO trains in a way that looks more like real life. After spending a while fiddling around with some Technic parts I came up with the following arrangement that offered a possibility. This design gives a floating pivot point that holds vehicles close in to each other on the straights while still permitting enough movement relative to each other to allow for bend and points/switches. The idea initially was to use this design for conventional bogie vehicles that would need this amount of movement and still centre correctly on the straights, although I am still to test this idea to see how it works in practice. The first application of this idea was in an articulated railcar, but this presented the first problem. I wanted to isolate the rotation of the bogie from the movement of the cars, as would be the case for an articulated train in which neighbouring vehicles share a pivot point, but such a design would not centre itself automatically. The initial solution was to use an elastic band to pull the vehicles back into line, but this was not a very successful idea and actually caused a few more problems than it solved. What the idea needed was another adaptation. The solution was to add an additional diagonal linkage across the car mounting points. This prevents the sideways motion of the linkages that is desirable for bogie vehicles but not on an articulated train. It's an awkward length and requires careful set-up and securing as it's important to get the length absolutely right in order that the cars centre correctly. Get the length wrong and you get an offset between the cars. It turns out that this design also works well for Talgo trains as, in addition to providing close coupling between the cars, it also gives the axle steering effect needed to ensure reduced friction and prevent derailments you might otherwise get with a fixed axle.

I should flag one issue with the design, which I suspect will be a problem whichever solution you choose. The test train struggled to take the points/switches on the test track at low speed and I didn't want to try it any faster because of the risk of derailing. The issue is that the axle is being steered by the movement of the cars relative to each other rather than by the geometry of the track as would be the case with a convention bogie. This means that sometimes it's not being steered in the correct direction to follow the bends. Therefore I do not recommend having lots of corners, bends and points/switches close to each other. It may be worse with my railcar because the cars are longer than would be the case with a Talgo, but testing your finished train on different track configurations will prove the case.

I'm ahead of you. I've done a very quick mod to the LMS Railcar to turn temporarily into a Talgo. This is how I (illegally) attached the wheelset to the close-coupling system. There are a couple of 1x2 jumper plates (in red) hiding deep inside the render which are joined to the pin holes on a pair of 3L Technic pins. I'm not suggesting that this is necessarily the best solution or the one you should follow, but I wanted to make the modification easily with the minimum of new parts to fit the existing layout of the Railcar. The top half of the mechanism isn't shown because I've not changed anything there. So does it work? You know when you've been Talgo'd. (Click on the image for the video) Look between the cars and you can see the self-steering effect. It's not perfect but it's working OK. I suspect that there is an issue with the unequal length of the cars or maybe the weight distribution that could be affecting things. Well, for now I shall wish you luck in your plans and wait to hear what happens.

No, it's complete. The only thing that is missing is the railcar bogie which, because it's a two axle bogie and needs to rotate freely around the central member, is attached using a tile with pin as described last night. The design uses non-friction pins where necessary (all shown in light bley) to ensure free movement. Unlike the double hinge design you've used so far, this design does not give a precise turning point. Instead, the geometry of the parts and the way they turn creates a floating pivot point, which is why it can allow for close coupling compared to a conventional fixed pivot (see first proof of concept photo below). For a Talgo you don't want a free pivoting bogie with just a single axle, as you will have no control over steering. The same as @Ferro-Friki's double hinge design, mine gives you the steering you need to make sure that your train goes around corners smoothly and without derailment. All you need to do is to attach the wheels to the central member (white) in a similar way so that it gives the axle the steering inputs that you need. Looks like a few of us here might be getting a cut, then.

You mean the red one? I got those set to 58 degrees, but I had to work it out using trial and error. What is more important when you build the train for real is to get the length of the red bar correct otherwise the cars won't centre correctly. The axle is about 3.5 studs long, but no such part exists so I have used axles 3 studs long and glued them into the pin connectors to fix the linkage to the correct length. If you mean the black arms, these are set at 30 degrees. The mounting to the cars is 5 wide where they attach and 3 wide where they link to the central member (white). The half liftarms are just there to join the various components in the central member together. There's a Technic axle and pin connector (32184) in the middle of it and the bogies for my railcar connect and pivot through that part using a 2x2 tile with pin (2460). For a Talgo you would want to mount your axle to this white section so that it is fixed to it. You can use different parts to make this happen if it's easier, as this specific design is really for articulated trains with bogies like my railcar. Focus on those first and come back to this later. Employers will be more interested in your exam results than in your ability to build Talgos out of LEGO. It's up to you. If you're just testing out the articulation between the cars, you can just push the train by hand.

@bogieman Thanks Dave.

This is part of the reason for building and testing, so that you can see how things work and whether or not they will work in your own build. It's part #60483. The axle hole is threaded onto the tan axle while the diagonal linkage fits into the pin hole. It will articulate very nicely and give you the steering effect that you need for the rodales. It's basically just a double-hinge mechanism with the diagonal linkage ensuring that movement on one side of the hinge is replicated on the other. On the subject of the articulation and coupling between the cars, I know that both you and @Ferro-Friki were looking at ways of trying to reduce the gap between the cars. As mentioned earlier, I have been working on an adaptation of the close-coupling system that I used on my LMS Articulated Railcar that might also work for Talgos and would be an alternative to the design that @Nikonissen came up with. Scroll to the end of that thread to see the design. I haven't tested it with a Talgo wheel arrangement because the length of the cars is much longer than for your train, but if you swapped out the regular bogie for a fixed axle it should work just the same. The only thing I would say is that I would try and avoid any tight S-bends, such as my test track has, as I fear the system would cause a lot of friction and possibly a derailment too. Coupling to the cars is using the upward-pointing pins and axles, with the pins joining the mechanism firmly to one car while the neighbouring car rests on the axles. All you need is some pin connectors securely fastened into the structure of each car to carry the weight of the car and any forces through the coupling.

Sorry for bringing this back up from the depths, but I wanted to post an update. As I mentioned during the building phase of the project, I was not entirely satisfied with the close coupling system I had employed for this build as it wasn't very well optimised for an articulated train. Part of the problem was that the geometry did not automatically centre the cars which meant that found myself having to resort to using strong elastic bands to achieve this, plus it did not fully separate the articulation of the cars from the rotation of the bogies. It was an OK solution but not ideal as it caused a few unwanted side-effects, such as causing the fully-articulated middle car to lean alarmingly outwards on any curves due to the change in the weight distribution as a consequence of the pivot point shifting with the articulation. It also needed a very strong pull to achieve any centreing. After a good long break from the build I decided that I needed to revisit the design and address these problems and try to come up with something a bit better. And here is the fruit of my labours. The innovation is the diagonal linkage (coloured red). This provides the force to centre the cars and also acts to stabilise the middle car to prevent it from leaning. It has necessitated a bit of a redesign inside the railcar to accommodate all the changes, but these alterations now allow for the coupling gap to be shortened if necessary for larger radius curves. I have also altered the design of the central member (coloured white) to make it more compact. The results are just as I had hoped. The cars now articulate independently of the bogies and the middle car is now stabilised. It can still negotiate R40 radius curves and handles points/switches with ease. Here's a couple of very short videos of it in action.

Ah, I get it now. You've got two problems with attaching the skirts to the bogie. The first one is the half stud offset I discussed yesterday, which should be easy enough to fix. The second is that big issue that affects all 7-wide builds, which is that you have a quarter plate left over on each side and there isn't a part with a quarter plate thickness to deal with it. This is a much tougher puzzle. I always like to try and keep things simple if it avoids having to deal with a really tricky problem head-on. The first thing I would explore is whether the side-skirts can be attached to the body of the car and still leave enough space inside for the bogie to move as much as you need it to. If you can make this happen it will completely avoid all the other problems as well as look better when the train is running. Only when I have exhausted all the possibilities over this would I start to consider how I might try to attach the skirts to the bogie.

Dark blue is VERY dark. I have some parts in my bin in that colour, and it's hard to tell them apart from black. Looking at photos of Sir Nigel together with Mallard, I would say that they are both almost the same shade of blue. There isn't a huge difference between BR Blue (which really doesn't look purple in real life) and LNER Garter Blue. They look different because Sir Nigel no longer has the streamlined wheel and motion covers which accentuates the amount of black visible. It also has banding over the casing, whereas Mallard is simply plain blue. Also, bear in mind that looking at photos online can also be misleading where colour is concerned, as there are any number of factors that can affect how it can be presented and reproduced. In some pictures, Mallard can appear almost azure. I vote standard LEGO blue for both. I think you've done a grand job at capturing the shape of an A4. The only nit I would pick over the styling would be that the lower front casing should taper inwards on both sides where the buffer are fitted. I'm also a little confused about the configuration of the drivers. Do you not need more than one set of flanged drivers?

That might not be a massive problem. If the weight is being supported by the bottom link the attachment at the top is there really just for dynamic stability (in other words, so that the cars don't wobble too much), so you may not need to build it as strong. It's possible that I might just have something in the pipeline that could help you gentlemen. I have been working on an adaptation of the close coupling technique I used on the LMS Articulated Railcar, specifically for articulated trains using Jacobs bogies. Depending on how this works out, it might just work for a Talgo also. The problem at the moment is that I don't have a lot of spare time to get the thing built and tested, but when I do I will report back and let folk know how it went together with some ideas about how it might apply to Talgos. I mean, it's a high-speed train so it needs bigger curves so that it can stretch it's legs a bit. It's nice if a train can be made to work even on R40s because then it can run anywhere, but you are right about them being limiting. If you want to get close to anything like scale modelling with LEGO they really are useless for anything other than trams. I'm unsure if I've understood what you're trying to make happen. Are you trying to get the light bley brackets holding the side skirts to connect with the bogie arm? Is there not enough space inside the skirts to have them fixed to the body and still leave enough room inside for the bogie arm to swing? You might have to approach the problem from the opposite end. Sometimes the solution to a design issue is not obvious and there could be more than one solution to the problem. Brackets, as with any part that has an open stud (as opposed to a solid stud), can provide you with the means to create a half stud offset. Any plate that is 1 x N has little studs on the back which will fit inside an open stud and secure parts together, which is a feature that Stud.io supports. So rather than starting off by fitting the brackets to the inside of the skirts and then stressing about how you can attach them to the bogie, maybe fix the brackets to the bogie first and then see what options you have for fitting the skirts to the brackets.

That looks good. I think that it should work fine like that, subject to real-world build and testing. I know you said you were thinking of removing the top link, but I see you have found another way instead of attaching the top of the mechanism in a way that suits your design better. Well done. I did a quick and dirty mock-up in Stud.io to see what the problem might be. I don't know if I got all your parts choices, measurements and spacings correct so I might be way off, but I managed to get to the magic 16 degrees without affecting the look of the model. The problem is not the slopes themselves but the supporting parts that the slopes are attached to; specifically the dark bley inverted slopes underneath. You just need to create some space for the corner of the bogie to move into, as the bogie part will move sufficiently far over the top of the inverted slopes to give you those extra few degrees of movement. I have found them to be OK and not too weak, and I think others would agree. Your solution does look much stronger for sure and would take away any worries about performance of the part, especially on a heavy train. Ultimately it's up to you to choose the design that works best for you.

You're very welcome. This is the beauty of the community. Everyone is willing to pitch-in and help share their ideas, experience and expertise. Never be afraid to ask whenever you feel you need some help or feedback. It looks promising. It's a little hard to see what you've done inside from that angle, though. That depends on a number of things, including the wheelbase of the vehicle (the distance between the axles). The longer the wheelbase or the tighter the curve, the more movement you will need. I don't have a mathematic formula that allows you to calculate how much movement is needed and arrived at the required amount through trial and error, which is the reason why sometimes you have to test build things and try them out. Happily the wheelbase of your end car looks similar to that of my freight wagon. For that model, I found that I needed around 16 degrees of movement for a wagon with a 25 stud wheelbase to match an R40 curve. You could get away with less if you're prepared for the axle not to turn all the way into the curve, but the trade-off with that approach is that the wheels will "scrub" across the inside of the rails and cause friction. It looks very chunky. I'd keep things simple and just use a 2x2 turntable plate attached to the four holes on the bottom of a 3x3 plate built into the floor of the end car. It's surprisingly strong and gives you the half-stud offset you need for a 7-wide design. You're welcome. There may not be a huge amount in the discussion that applies to Talgo designs, but you may still find it interesting. Sounds like a great plan. We can have a pan-European Talgo-Fest. I might have to have a go at building one myself.

@Imanol I've been reading the thread and looking at the various renders to try and understand what the issues are in order to make some sort of assessment. I have no direct experience of replicating the Talgo wheel arrangements, but I have spent a fair amount of time looking at other articulation and axle arrangements. The rodal design shown by @Ferro-Friki looks sound and should give you the axle steering that your train will need. However, as he says, you will need to make sure that you allow enough space between the cars for them to articulate without the car bodies striking the axles (which I suspect is what is stopping his Alvo from using R40 radius curves). The rodal design also provides for car-to-car coupling, so there isn't any need to go looking for another solution. If you look at his cut-away render of his Alvo, you can see that one side of the articulation is built into the car on the left side while the car on the right side sits on the red axles and is joined by lowering the car onto these points. This connection will be more than strong enough to support the weight and keep the train together provided that the attachment points inside the car are built into the structure. I would personally use something like a Technic Axle and Pin Connector Hub with 2 Axles (PN:27940) with a Technic Brick 1 x 2 with Axle Hole (PN:32064) on each side built strongly into the structure of the car, but you may need some other solution that fits better into your design. In your own render, you have made an error by making the design of the articulation the same on both sides and have also made both sides 1 stud longer than necessary. As for the weight of the cars, I still don't think you should worry so much about it. Excessive weight really only has two downsides; it requires more power to move it and it can make any problems you have with friction worse. Weight doesn't really have a negative impact on the coupling, especially for a vertical coupling system like this one. Carrying the weight simply means ensuring that it is transmitted down through whatever structures you have into the wheels that are required to carry it. In this case the design works fine because there is a weight-bearing attachment on the top of the axles to help carry the weight. In addition, each side of the rodal is going to be balanced when the cars are coupled because all the cars are likely to be very similar in weight. Your implementation of the rodal design currently intrudes 5 studs into your cars, but this is with a 1 stud gap between the cars. As mentioned already, you can lose 1 stud off each end straight away just by copying the design more closely, which will bring the intrusion down to 4 studs. If you extend the gap between the cars to 3 studs as recommended the intrusion is brought down further to just 3 studs, which could possibly be reduced even more by changing some of the parts and the way that it is secured inside the cars. Looking at your earlier cut-away renders of your cars, I think you could possibly live with a 3 stud intrusion without having to sacrifice too much of the interior, which you can achieve by following the proposed rodal design more closely. If you can successfully work these ideas into your design you may not need to discard the top link at all and retain the extra dynamic stability that this will give your train. I like your idea for the end car coupling and am glad that you have avoided the mistake of using a pivoting magnet on this side of the coupling. Having pivoting magnets on both the locomotive and the end car would be fine under tension (pulling) but not under compression (pushing), as the coupling bar to the axle under the end car almost certainly would be deflected to one side, pushing the axle out of alignment. My only concern with your proposal is that it doesn't appear to have enough articulation for small radius curves like the standard R40. You may have to consider cutting back the sides to give the axle more space to turn. I think you're getting close to a buildable design now and, by implementing @Ferro-Friki's ideas, you should have a train that looks cool and runs well. Talgos are extremely tricky trains to engineer, but it looks like you're on your way to a winner. If any other thoughts come to me I will add them to this thread. Any questions, just yell. I don't recommend this. I tried using free swiveling axles on a long-wheelbase freight car and they don't work. There needs to be some mechanism to allow the axle to steer and turn when it needs to (and to make sure that it turns the right way) and to bring it back to the central position when it goes onto straight track. Free swiveling axles just flop about in any direction they feel like going and then get stuck in the wrong orientation. If you feel like wasting a couple of hours you can read more about some of my adventures in axle articulation here.

I don’t think that supporting the weight will be an issue with the current design, as long as the ball and socket parts are built strongly into the structure of the car bodies. But yes, I think your biggest problem with the current design will be derailing and friction on curves. I understand that you don’t have track or bricks to try things out, but sometimes it’s cheaper to do this first and build any changes into your design before building it than having to alter everything afterwards. It’s a big train which is going to be expensive to build anyway, so trying out some of the design features first might save you money overall.

Yeah, I agree. Thank you, sir. That's very kind.

I fear you're going to have some derailing and/or friction issues if you use fixed axles as shown in your design, as the length of the cars is quite big. You may also need to consider using some kind of articulation between the cars to steer the wheels around curves. There are some ideas from other builders on Flickr if you search for "Lego Talgo". Have a go at building some basic chassis to test and see what happens. You may find that the articulation provides a way of coupling the cars more securely than by using towballs, which would solve two problems in one. As for strength and structural integrity, I wouldn't trust the stability function in Stud.io as it doesn't always correctly identify areas of weakness. Try and use lots of overlapping parts to give strength and check your build as you go.

It's a fair representation of a Hall, perhaps with better proportions than the Emerald Night. It is maybe the best we can expect as an official offering at this size, but I'm not blown away by it. The loco may be nice, but the coaches are horrible and the station is a take-it-or-leave-it for me. I expect it to be a roaring success. I appreciate that this is damning with faint praise, but that's my feeling.

Thanks for the feedback folks. I'm also liking how it looks, especially when all lit up. It's certainly a distinctive vehicle and would have made for a brilliant addition to the streets of Oslo. Do you mean the sleek modern styling and the way that it combines the aesthetic with the practical, or that it divides opinion along intensely fought lines and is incompatible with systems that went before it?

This is another dip into the vaults for a design that been sitting on the books for quite a long time. It was almost exactly 5 years ago that I had the idea to design and build a tram. It was something that I hadn't tried before and I wasn't sure how I would do. Part of the problem for me is that they don't look particularly distinctive, unlike trains that come in a bewildering variety of sizes, shapes and colours. I spent a good amount of time trawling through the internet looking for inspiration, which is when I came upon the 2020 Oslo Tram concept. The story of this design concept goes back to 2015 when Sporveien, the public transport operator, started the tendering process for a new tram to replace it's ageing fleet. Two young Norwegian designers, Taral Jansen and Mikkel Brandt Bugge of design studio Norwegian Attitude, submitted a design for a brand new high-capacity tram customised to suit the needs of Oslo while reflecting Scandinavian design principles. More information about the project can be found on the project's Facebook page. My first attempt at reproducing the design was not especially good. One of the main problems was that there was a mismatch between the width of the cabs and the width of the car bodies, which just didn't look right to me. I tried making the cab front wider but it just made the problem worse. The other issue was that, when I rendered the design, a lot of the features simply disappeared. This is something I have experienced a few times before when trying to render black and trans-black parts. I was very underwhelmed by the whole thing and so I just put it away and pretty much forgot all about it. That was until just a few weeks ago when I visited Oslo for a few days. I already knew that the 2020 concept would not have been realised as it would have required bespoke vehicles that would have been incredibly expensive. However, I was excited to see what types of vehicles would be running on Oslo's tram network given the amount of time that had passed, and whether anything of the concept design had been incorporated. So imagine my disappointment when I came out of the Oslo Sentralstasjon to see that the backbone of the Oslo tram service is still being operated by Duewag SL79s, some of which are already over 40 years old. Don't get me wrong, I enjoyed riding these older trams, but I had expected something a bit more modern. With my interest rekindled I came home and decided to revisit my model and see what improvements I could make. The first item on the work-list was to reduce the width of the cars so that they matched the cabs. Using a combination of brackets inside the roof and floor of the model, I was able to reduce the width by 1 plate on each side. Now the width is consistent all the way down the tram cars, going from 8 studs wide to half a plate over 7 studs. I also substituted some parts that were not available for others that are in order that the design can be buildable and replaced the windscreen with a much deeper alternative. I have also learned a thing or two about the rendering process in Stud.io in order to try and get the design features to be more visible. All that said, the most striking aspect of the design is still not visible when rendered. The "illuminating lines" around the cab windscreens and along the sides of the cars were just not showing up. So, by swapping out the trans-clear for luminous parts and adjusting the background colour and light settings in the render menu, I was able to replicate the effect of the lights. Applying the same trick to the inside of the cars meant that I could also simulate the effect of internal lighting. I'm unsure if I shall ever build this model, but I am at least happy now to present it for feedback.

Thank you for the very generous name-check. I like how you’ve developed the coupling concept and made it work for you. But I especially like the beautiful railcar you have built. It’s a real visual delight and a joy to watch in motion. Congratulations on yet another spectacular build.

I suspect that at least some of this is as a consequence of the requirements set down in the Bricklink Designer Program Guidelines, which stipulates that certain criteria must be met (click for details). Having browsed the Series 1 entries, I thought I'd have a look at the submission requirements to see if it's something I'd like to have a go at for Series 2. As such, I've been reading through and trying to absorb the guidelines and I'm not sure that I'm sufficiently tempted. While I can understand that some of the requirements exist for a reason, I can also see that it can result in a somewhat dissatisfying experience, as has been reported here. It seems that the requirement for clear instructions has resulted in a situation where rigid adherence to the guidelines produces unnecessarily bloated instructions where a slightly more flexible approach could have dealt with this problem. That said, I think the designer may have slightly misunderstood part of them. The requirement for the first step is to have no more than 5 elements, but from there on out it can be up to 5 different elements and a maximum of 10 elements per step. If there are 17 glass panes to fit, this is going to take at least two steps depending on whether or not other elements are required to be used and if building them up extends the construction by more than 12 studs in any direction. Exceeding any of these limits would necessitate an additional step. I also wonder whether or not the designer tested the instructions using a third party (and probably less experienced) builder to ensure that they were adequate and to gauge the quality of the building experience. It's a shame to hear the negative comments about this design, as it promised so much. Looking at it, I had pigeon-holed it with as a Modular building rather than as a Train-theme design. I'm sure that some of the negative aspects could be addressed by modding the kit, though.

@LEGO Train 12 Volts Thank you. You know me. I just keep going until I get it as close to reality as I can. @Ashi Valkoinen Thanks for the welcome. I was a little worried that you'd also been working on the UK version of the FLIRT and I didn't want to tread on your toes.

@Darkkostas25 Thank you. That’s very kind of you to say.

Thank you, gentlemen. @Asper @Feuer Zug It’s a model using parts ancient and modern. These were the best techniques I could find to recreate the various vents on the roof. @Ts__ I agree that the side window is the wrong size and shape, but that was the compromise that I had to make to ensure that the slope and taper of the cab could be made. There’s quite a lot of snotting and supporting structures going on inside that had to be accommodated. I spent a lot of time wondering if I could have done that detail better, but concluded that the only way I make it better was to make the rest of it worse, so I decided that I could live with it There have been a lot of compromises made, such as the inset headlights. The original design was closer to the real train, but it would not have been possible to include LEDs. What you see here is probably the most buildable version in the closest design I could make. No news on the prize as yet.