gyenesvi

I don't think either that you need absolute positioning for that, why would you? And as I mention above, no need for hydraulics either, pressurized air may work well enough, and that's much more plausible within the Lego system. See the conclusion in this thread: @allanp, 3x3x3 servo is not really possible I think, definitely not with meaningful power, smallest we could hope for would be 3x3x5, like a GeekServo, but the lego version of that is actually 3x4x5 (small angular motor), because it needs a position encoder, unlike a real RC servo (though I think real RC servo control would be possible with the Technic hub, just would need proper firmware). Anyway, the beauty of the peristaltic pump is that it does not need a switch. Instead of a servo for every switch, there could be an M motor for each pump. True that it's not realistic, but at least would be plausible and playable. Definitely not 10 functions, I could imagine max 4. Maybe a 4-way gearbox could drive 4 pumps. But then there would be no simultaneous movements possible. It's a trade-off, like everything.

It's true that breaking the gearbox into multiple sub-parts may help, especially separating front/rear functions, but even then the sheer amount of output axles that need to be routed would make it pretty complex. And this too.. What I'd really like to see in this direction is a peristaltic pump based RC pneumatic solution that we concluded would be possible to run with pressurized air based on the experiments of @2GodBDGlory. Anyway, two of these huge hubs would not fit into a backhoe I think (next to all the motors that drive the actual functions).

Wonder if that is really true. It might enable 8 speeds, on two parallel shafts, but for 8 functions you need to route the 8 functions in 8 separate directions, which would be super hard to do I guess..

Nice model, it looks good with those tires. As far as I understand, the width difference between 81mm Lego tractor tires and FisherTechnic tires is very important here not just aesthetically but also clearance-wise; the FisherTechnic ones being a bit slimmer allow the longer steering arm to clear. This is a pain point with 81mm Lego tractor tires, some larger steering arms/linkages don't work with it, which is a pity as they are less useful with actual off-roaders. Luckily they work better with Audi tires, though their thread pattern is less aggressive. @Tatrovak, do you know a way to purchase those FisherTechnic tires separately? Those should fit in terms of clearance, but agreed that design-wise not so much.

Why don't you try it in Studio and experiment with whatever you think might fit before ordering parts :)

Creative solution, but another problem this has besides the quite bad ground clearance is the anti-ackermann steering geometry.

Thanks! And that's exactly what the diff casing does if you drive its two sides (normally the outputs).

I don't think its old, it's an Xbox Series controller. It's definitely not latency, it does not stop indefinitely. And I don't remember having this issue before, it just does it recently..

Yeah, maybe sometimes, but I often find that part needed in very tight places where this new one would be too big. Still have to see where it would be useful for me.

Oh thanks, I meant in general. Silly me, I knew that it averages, I just didn't think it is used that way for gearing :) But what I was actually wondering the other day, is that if it has two inputs, and only one is driven temporarily, it essentially halves the speed, but does that double the torque at the same time? I guess it should, as in total, power should not be lost (not counting loss in friction), right? So if I have two motors driving the two sides of a diff, and I stop one (brake it) then the speed will become half, but the torque will be the same (only one motor, but torque doubled), is that right?

What ratio is achievable with a differential? I was wondering about that recently, but could not really find out how to calculate that..

Thanks guys, I appreciate it!

I wonder if anybody had issues with BrickController 2 (on Android) where it does not stop the motor after releasing the joystick (with PU hub). It does that quite frequently for me nowadays, which today resulted in my model slowly moving while I did not pay full attention, flipping down a slope, and then tripping over a ledge and dropping from 5 meters high, to finally break into pieces in the middle of a bush.. It was a painful experience to collect the parts and put it all together (though only lost/broke a few parts). @imurvai, anything to do about that?

I thought the same, hope it is not because they don't ever want to make a 3L flip-flop, that would be very useful too. That was my second thought indeed :) Wonder if this is the same 'small beam' mentioned to be in the space sets in the interview video. Other than that, I don't yet know how useful this part would be, never found myself needing something like exactly this.

One thing about MOC reviews or feedback, I do think they tend to be somewhat biased. Most positive comments are about the looks, or some well implemented functions. Negative feedback comes in only when something is really not working as people don't want to discourage others. Sometimes, there is negative feedback about looks, when something that models a real-life counterpart differs a lot in shaping for example. Only some people who know the source material really well call these negatives out publicly. However, there is very little feedback about functions that are not well implemented or weak (but work somewhat or just don't matter too much). Very few people understand specific technicalities that are included in niche MOCs, and they don't necessarily want to call those out. For example I do see a lot of crazy suspension lifts on FB that I think are wrong on many levels, yet I don't start debating them.

Agree that a U-joint in the ball joint is probably less prone to break than when it is not protected that way. No, I was asking about experience with the new CV joints, but you answered it already. So I do think it is worth trying because I have really good experience with those, pretty sturdy ones (can withstand power of two coupled Buwizz motors) and not much slack. But you are right about the major down-side; it is bigger and that's a problem in many places. Also, they are limited with the axle lengths, so harder to incorporate, even though the recent addition of a shorter version in 42160 is really good, opens up new possibilities.

Thanks for the model file! If the steering system was not such a let-down, I would have bought the model. Even if the slack is fixed, for me the main problem here is that the servo is still not proportional. With a proportional servo and a proportional remote (that Cada does have but as far as I know is not included here), this would be a great little set. Or does this set have the joystick remote included?

Unfortunately I tend to agree with @Lipko that the situation with MOCs is not so rosy either. There are only a handful of MOCs out there that I find really well designed, and that applies to looks / functions / sturdiness. Maybe my standards are too high, or we have differing preferences, but I do see many solutions that are too hackish for me, only work more or less, or the looks could be greatly improved by adding less clutter. I think official sets are not really limited by inventory either, maybe in color choices, especially when it comes to recoloring smaller connector parts, TLG tends to save const on those.. And of course they won't use useful but retired parts, while some MOCs can. I don't get this, since official sets also have many design iterations before they are finalized. I guess more so than MOCs. Of course one advantage of MOCs is that people may throw a lot of electronics in there because they are not trying to save costs, but official sets can't really do that otherwise they come out too expensive. Also, too much power can damage parts, which MOC builders don't care about or are even okay with greasing, but that does not fly in an official set. In terms of better capabilities, it boils down to realistic building techniques I think. In this regard, I think MOCs can have an advantage as builders experiment more and strive more for realism than TLG does. Suspensions are a good example of that, TLG really does not push the limits there, which is sad as it could be mechanically more interesting / educative and could result in cool new parts.

Of course the ball joint gives extra stability, requires less space and is easy to work with, I totally see that. However, I was never convinced that the single U-joint in it, albeit being protected by the ball joint itself, is a sturdier solution than the double U-joints. I am always using new heavy duty CV joints, and they have less slack and are stronger than a U-joint. Also, the long male part is telescopic and still holds the axle very well due to its length. So I never had a problem with those, though I never built really heavy model, just medium, but I bashed it hard with Buwizz motors. Have you had any negative experience with those?

Something like that, it would be useful if they were interchangeable for example. On the other hand, there's a crucial difference; the steering arms are centered vertically on the planetary, which I do find useful and easier to work with in many cases (for example, not being in the way for things like springs).

Those interviews are cool, good to have some insights into the design process! I'm curious about the new short beam mentioned in the long interview. Any clues? I don't think there are that many possibilities for a new short beam. I'm holding my fingers crossed for a 4L beam or a short flip-flop one :D

No worries, there's nothing offensive here, good to have a discussion with someone who understands the trade-offs in such designs. I think you see things quite similarly how I see them, in terms of possibilities and trade-offs. I agree that a steering system is most responsive if the servo is on the axle, and that the biggest challenge about the linkage based steering when the servo is off the axle is the clearance it requires (furthermore it requires a Panhard rod as well to be stable). However, I don't quite understand why a 4-link would be less stable and less usable for a heavier truck. In my experience it can be built quite stable, especially with stronger links (built from towball arms) and using a Panhard rod. Sure it is quite a bit more complicated than using the ball joint, but it's exactly that challenge that I am looking for when building them :) Also, what I found is that when the servo is on the axle, it requires a hole in the front of the chassis, and that makes it less rigid. When the servo is in there, it can be used as a structural element as well. But I understand we have different building styles / goals. I did not know that the real one also has some negative angle, good to note. I'm actually okay with using the ball-joint in the front when it is almost horizontal or only has a minimal amount of negative angle (what I don't like is steep negative angles). And trucks are better candidates for that.

So what's your definition of a good MOC? What makes them better than some official sets? A few things I could think of is maximizing the functionality put into the model at the given scale, and the realism of that functionality being a fairly good looking (maybe scale) representation of something realistic (does not necessarily have to be a licensed model) For me, these two points are some common shortcomings of official sets mainly due to cost saving I think. It is understandable that they need to produce smaller / cheaper / simpler models as well to cater for younger kinds and beginners. But I don't really fancy when they make something empty / missing functionality (42126 rwd) or unrealistic in functionality (like 9398 and 42099 suspension, or the huge unrealistic supercar gearbox layouts), or when they pick an iconic real world model and build it on a too small scale to be able represent it properly (like 42122, both looks and functions). So for me a really interesting model is something that has realistic and enough functions put into the available space for the scale, but not too much crammed, while also being a good looking one given the possibilities of technic panels, something like 42128.

I just realized that it was the planetary hub that you had to cut to make this work. I wish Lego would have made that hub that way, it becomes so hard to use as a steered hub because of that extra bit being in the way all the time. That's what happens when they want to make a part usable for multiple purposes; it becomes much less useful for its original purpose. At least with the Audi hub, they made it better. The truck looks nice, especially the cab, I'm not a fan of big boxy stuff, so I prefer it without the rear end :) On the other hand, I don't like the front axle construction, because of the use of the ball joint causing negative caster. Did you consider building a proper 4-link suspension for the front? And moving the servo off the axle through a linkage? At this large scale, it should be possible. Do you want 11 stud wide axles (between steering pivots)? I don't think that's possible with current parts, but the 13 wide you have here can be made fairly stable (while having okay ground clearance like the one in this post), you just need to get rid of that servo on there, move the cross bracing on the top closer to the center, and maybe turn it into a proper 4-link suspension :) Something like I used in this build.

I totally agree with this. Though quite a few of those questionable techniques can be spotted by looking only, very few people do. Most often people just comment "it looks good" without thinking much about the functionality. I sometimes kind of await such questions when I present a MOC and I know that there are some weak points in it, but rarely get those questions. A bit relieving, a bit sad at the same time :) Exactly my thoughts.