Sign in to follow this  
gyenesvi

[WIP] Mid-scale PU off-road chassis variants

Recommended Posts

Hi Folks,

As you may have guessed from my previous posts of my models here, I like off-roaders, including classic cars using live axles, so naturally I tinker a lot with various ideas for building them. This time I'd like to work on some mid-scale 4x4 cars, by which I mean wheel sizes in the range of 60-80mm, with 11-13 wide wheel bases, because that's a reasonable scale for lego electronics, things don't get too heavy and bulky yet, but still challenging features can be implemented. Many people have built great models before me on this scale, I especially like the chassis designs of Madoca, RM8 and Sheepo because they are compact, yet packed with interesting suspension and drivetrain features, besides nice bodyworks. However, their models are from the PF era, and now we have PU electronics and new parts related to that (such as CV joints, heavy-duty differentials and planetary wheel hubs), and I'd like to design my versions using these parts as well.

I'll probably end up with a few different variants; for example axles with planetary wheel hubs / regular wheel hubs / built wheel hubs, 11 or 13 wide axles, various motor and drivetrain setups, with or without gearbox, manul vs RC gearbox, various suspension geometries, BuWizz battery vs Technic Battery, etc. Just like some of the above mentioned builders' models, I'd like to keep things realistic where possible. By that I mean things like (somewhat) realistic drivetrain, (engine and gearbox placement), realistic steering setup (servo off the axle, linkage based steering), and a possibly clean interior (no motors or battery box in the middle of the cab). However, I'll probably also experiment with some less realistic arrangements as well (like drive motors at the bottom). I'd also like to have some other functional elements related to the RC functions, like the steering wheel or the gear stick even when it's RC controlled, but this is more like a bonus when possible.

So without further ado, let's get started. The first step for me is to provide an acceptable solution to two issues: 1) building a compact front live axle using planetary wheel hubs and 2) a motor setup that could result in a realistic drivetrain and steering arrangement.

Given the scale, I need at most 13 stud wide axles. Since the planetary hub can only be mounted with towballs when steered, and towball arms are very limited in lego, this gets tough, especially if I want to retain acceptable ground clearance. From my Zetros alternate models I have learned that tilted lower towball arms can work well and retain acceptable ground clearance. With 5L towball arms and a 3 wide differential between them, 13 wide axle is the shortest theoretically possible. So I had to find out if there is a stable way to actually build it (without becoming too bulky, which I don't like). Here is what I came up with:

800x600.png

800x600.png

800x600.png

The key was to find a stable way to mount the lower towball arms under the 5x7 frame. The wheel hubs become a little tilted vertically, but it is unnoticeable in reality and does not effect the working of the CV joints. As you can see, there are connection points for 3 links, a panhard rod and a steering link. The axle is quite solid against bending, better than I expected. The spring is mounted with only one stud offset forward from the axle center (not possible right at the center), and about mid-way sideways from the center to the wheels. This results in quite okay articulation relative to the scale even with these short shock absorbers.

Next up is the motor setup, which should ideally be above and behind the front axle to keep it realistic. As noted above, others have already made interesting setups before me that I really like and would take that spirit on. RM8's chassis uses a PF servo motor for steering in the front and a PF XL motor behind it for drive. Unfortunately, in case of PU the L motor that is used instead of a servo is 1 stud longer, and the XL motor is 2 studs longer, making this setup impractically long (the drive motor would end up inside the cabin space). Sheepo's chassis on the other hand uses a servo motor sandwiched between two L motors, which is compact length-wise but much wider. This is an interesting setup that I might try to replicate with PU L motors later, but for now I settled for one that's kind of a mix of the two: a PU L motor for steering and a PU XL motor for drive, side-by side, both of them offset from the center. Interestingly, the shape and mounting points of the motors allow to do this while retaining some symmetry although the two motors are different, while also leaving just enough space for mounting the shock absorbers of the suspension (and even the third top link and the panhard rod onto the bottom of the XL motor).

800x600.png

A further advantage of this setup is that the steering motor is not centered but offset to one side, resulting in a more realistic steering linkage crossing from one side to the other. I did have to try quite a few variants for the linkage to arrive to this one that has the right geometry (minimizes bump steer) is compact enough not to collide with other parts, such as the panhard rod (which is preferred to be close to the front of the axle to avoid the steering linkage pulling the whole axle sideways), and the springs (which also need to be mounted in front of the axle center to allow moving the motor assembly as much to the front as possible). Furthermore, this collision-free movement has to be true in all steering positions and suspension compression states. The variant here has absolutely no collisions in any combination of steering / suspension positions and the suspension moves smoothly (built and tested in real life).

800x600.png

800x600.png

I have tested the steering assembly and how it works with the PU app, it seems fine so far.

800x600.png

The drivetrain will start from the XL motor on the back of this assembly, and probably go to the center and 1 stud lower with a 20T - 16T gear mesh. From there two more vertical gear meshes are required to reach the floor where the main driveshaft will be, which leaves room for some sort of gearbox as well.

That's it for now as a start, I'll move on with the rear axle (and the driveshaft) next, which should be much simpler than the front.

Update 2022.02.27

On the rear axle things are simpler, however, there is an important decision to be made: whether to connect the driveshaft coming into the rear axle with double joints, which would allow the axle to be level, or just a single joint, which would make the axle have a positive caster angle (in case of the front axle, double joints are necessary, the single joint would result in a negative caster angle, which is a no-go for me). The decision also has an effect on the way the shock absorbers need to be mounted to allow them proper freedom of motion. For now, I go with the single-joint option for two reasons: it results in a 3 studs shorter driveshaft (shorter axle base), and the shock absorbers can be placed lower, this way they won't protrude into the trunk. However, even having decided this, I still have two options: use the ball-joint part designed for this purpose, or not. Although the ball-joint simplifies things by eliminating the need for additional sideways stabilization, it has some drawbacks as well; it needs to be mounted in the middle of the chassis, hence it limits the options to build the middle of the chassis where the gearbox goes, it reduces ground clearance in the middle, and it uses a U-joint, which transmits power with a non-constant velocity (though not sure that would be noticeable at a small angle, and when only one joint is used). For these reasons, I decided to leave the ball-joint out, and test how the rear part of the drivetrain would work without it. I have not seen this technique used with the new CV-joints before, only with U-joints, but I have tested it in my Jeep Wrangler alternate of the Zetros and seems promising.

On the actual axle, mounting the planetary wheel hub is not as trivial as mounting an older non-steered wheel hub because its connection points are not designed for slim axles. However, I wanted to retain both ground clearance, and space above the axle to be able to lower the rear part of the chassis. Here is what I ended up with.

800x600.png

800x600.png

800x600.png

Because I left out the ball-joint, it will need a panhard rod to be mounted somewhere to stabilize it sideways. The hub mounting is a bit asymmetric because I shaved off some material from the rear end to give way to the shock absorber as the axle tilts (lesson learned from previous builds).

As for the driveshaft, I'd like to start with something around the minimal length that's possible to achieve (aiming for short wheelbase 2-door models), that's why I went with the single joint option on the rear. Because I want to use the stronger new CV-joints instead of U-joints, that does pose some limitations on it, because there aren't many ways they can be mounted and connected in the middle. Here's what I'll go with for now, this has 26 studs from axle to axle (so the axle center distance is 25 studs). That results in exactly the wheel base that my Toyota FJ40 alternate of the Ford Raptor has. In the future, I also want to try the double-joint option on the rear, resulting in a more realistic live axle, but that may require a longer model (for example a 4-door body).

800x600.png

800x600.png

800x600.png

Although this is 1 stud longer than the theoretical minimum because of the 3L axle joiner in the middle (could be a 2L as well), it leaves the option for a straightforward gearbox in the middle (two gears next to the joiner, and one stud on each end to mount the driveshaft).

As a side note, there would be some ways to reduce this wheel base with alternative parts. For example, if the male part of the new CV joint was available with a 2L axle as well like the old CV joint, that could be used to reduce the shaft both at the front axle side, and at the chassis side in the front for example (I assume using the old CV joint is a no-go as it would easily pop apart in an RC build). That would still leave room for a gear on the rear part of the middle section to transmit drive from the engine (but without allowing the simplest form of a 2-speed gearbox). With the 3L middle joiner swapped for a 2L joiner, that would be a total of 3 studs reduction in the wheel base, which would even work for models on a smaller scale, like Arocs wheels. At some point, I might consider making those parts for myself by cutting some axles (if TLG does not release that part, which would also be useful for building a narrower driven independent suspension).

Update 2022.02.28

At first I thought the chassis would only have room for a trivial gearbox with 1:1 and 5:3 gearing ratios, with the two shafts situated under each other in the middle, arriving to the driveshaft. However, I realized that with the current driveshaft length, there is just enough space for a 1:1 and 3:1 ratio gearbox using a 4L differential piece, where the gearbox is laid out horizontally on one side. This setup could work well with the planetary hub, as the 1:1 gearing results in very slow movement (with a lot of torque), thus a geared-up second speed is useful, so I went for this one as a start.

800x600.png

This would result in a manual gearbox version, with the gearstick positioned quite reasonably in the middle of the cabin space. Furthermore, I thought the space on other side could be used for controlling the gear switching with the orange changeover catch piece and making it RC. I started wondering where I could put a PU L motor for controlling the gearbox. There seemed to be only one reasonable place, somewhere in the back and raised with one gearing to give space for the rear axle (the view is from the other side here).

800x600.png

And then a nice idea struck! I could place a BuWizz right next to it, and connect the two to make a 7 stud wide assembly, just like the two motors in the front.

800x600.png

So the next logical questions were: 1) can I frame this in a solid way, and 2) will this leave enough room for the rear floating axle to move properly? The BuWizz sits lower, so that's the critical one here. Moving it up one stud would not be a good idea because it would loose the connection between the two, and on the outer side as well. That connection on the outer side proved critical, as that's a great point for the rear shock absorbers to mount. Framing wasn't too difficult, and I was just lucky with the rear axle's movement as well. It's very tight, but it just flies.

800x600.png

The back of the gearbox motor can also be used for mounting point for the rear panhard rod.

800x600.png

On the rear axle I had to make a half stud vertical offset for the panhard rod to get to the right position.

800x600.png

Because of the low clearance above the rear axle, the panhard rod could be placed only at the very end of the rear axle (actually, that's where it's most useful). Because of that, and because of the mounting point on the motor/BuWizz required for the springs and attachment to the chassis, there is not much option to move the battery and the gearbox motor further back (maybe 1 stud, but that would result in not so ideal mounting position for the springs and the panhard rod). I was a bit worried that this would not allow the seats to be moved back enough, but this setup is just one stud more to the front than my Toyota FJ40 alternate, in which it's pretty much back, so I think this will be okay.

Here are more renders of the whole chassis with the axles.

800x600.png

800x600.png

I have built this and gave it a test drive. I mean I tried to, but the BuWizz App doesn't seem to be prepared for this yet; I did not find a way to control the gearbox. The steering kind of works, but it is quite jerky (calibrated fine, latest app version, firmware updated). I tried to play with the settings a bit, but there was not much difference. At least I know it runs with a decent speed in high gear, and seems to have good torque in low gear.

Because of this, I think for now I'll try to build a version with the Technic Hub in place of the BuWizz and the gearbox motor, removing the RC control of the gearbox (leaving just manual), and wait for the BuWizz app to be updated with the gearbox module.

Let me know your opinions so far!

Cheers,

Viktor

Edited by gyenesvi

Share this post


Link to post
Share on other sites

Looks really good! Did you try it with real parts?

One thing: The lower liftarms with ball joints are angled, so the wheel hubs are angled, too. As the connecting liftarm between them is connected with pins, is there too much friction/bending of the pins?

Share this post


Link to post
Share on other sites

That's a great start to a chassis. I like it. I need to remember this next time I build one of these vehicles to copy it :grin:

Share this post


Link to post
Share on other sites

Thank you!

8 hours ago, Jundis said:

Did you try it with real parts?

Yes I built this, works smooth, it really feels like the axle is floating :)

8 hours ago, Jundis said:

As the connecting liftarm between them is connected with pins, is there too much friction/bending of the pins?

Do you mean the tie rod of the steering? Since the tilting of the hubs is very minimal, and those pins are frictionless that have some play anyway, there is absolutely no stress or bending there. I think it would even work with friction pins to further reduce the play in the steering system.

Share this post


Link to post
Share on other sites

I have updated the initial post (marked by the date) with some progress on the rear axle and the driveshaft. I will keep updating the initial post like this to have a continuously readable flow, while I'll always post the latest status as a new comment.

The current status is this (only showing the driveshaft and the two axles with the motor assembly hidden):

800x600.png

Share this post


Link to post
Share on other sites

I managed to add an RC gearbox with 3:1 gearing ratio, and a BuWizz to the back in a fairly slim overall packaging. This is where it stands now, I have updated the initial post with the details of the gearbox.

800x600.png

Share this post


Link to post
Share on other sites

Another small update. I have swapped in a Technic Hub in place of the BuWizz and the gearbox motor in the back (the framing is not so elegant, but managed to make it solid thanks to some flip-flop beams), since the BuWizz is not able to control the gearbox, and the steering is also problematic for now. The Technic Hub also fits nicely, although the performance suffers significantly. At 7.2V (with rechargeable batteries) it is quite sluggish in high gear. A 9V it is okay-ish, but struggles a little in high gear when turning. Low gear works well though.

Because this 3x up-gearing ratio may be too much for the single XL motor, I also tried a simpler gearbox, with a 1:1.66 gear ratio. Obviously, that way it works much better in high gear as well.

800x600.png

Share this post


Link to post
Share on other sites

Really nice progress. Looks great.

For your Buwizz setup: perhaps try the Brickcontroller 2 app? It also works with Buwizz afaik and it has good calibration options for steering and gearbox switches.

Share this post


Link to post
Share on other sites

Looka very fresh! How much time did you spend with making all this renders? And which program did you use? I am making renders directly from Studio.io and they are not as good as yours.

Share this post


Link to post
Share on other sites
20 hours ago, Appie said:

For your Buwizz setup: perhaps try the Brickcontroller 2 app? It also works with Buwizz afaik and it has good calibration options for steering and gearbox switches.

Thanks, I have been thinking of trying out BrickController 2, but I don't have a physical controller. I want to buy one though, just not sure what brands work well with the app. Can you suggest some?

Edited by gyenesvi

Share this post


Link to post
Share on other sites
13 hours ago, Daniel-99 said:

Looka very fresh! How much time did you spend with making all this renders? And which program did you use? I am making renders directly from Studio.io and they are not as good as yours.

The renders did not take much time, I used Studio. This was designed first virtually, and I built it physically only afterwards, so the virtual model was already done, so I just had to make some renders from a few angles. For the inner parts, I just hide parts that I don't want visible and that's it. I think the grey background helps make the images look better :)

1 hour ago, Mechbuilds said:

I'd love to see this with power functions instead. Is it possible?

Well, given that the whole essence of this is to use Powered Up, I don't have plans to try that. I am using some key properties of PU here, so it would be harder to do with PF, if possible at all. For example, the shape of the motors and the hub make them better suited as structural elements. Even the BuWizz 3 can be used here better as a structural element because of the position of its pinholes, which I make crucial use of. And you could not use a PF L motor as a gearbox control motor.

As I noted in the intro, many people before me have made similar chassis builds using PF (like Madoca, RM8, Sheepo), so I don't want to replicate those. However, when it comes to an RC gearbox for example, things get cumbersome with PF. I am trying to make use of the advantages of PU here, and also modern parts, like the orange shifter part, planetary hubs, etc. A planetary hub with a PF XL motor would be super slow for example (PU XL motor is 50% faster, and it's still slow).

Share this post


Link to post
Share on other sites
2 hours ago, gyenesvi said:

Thanks, I have been thinking of trying out BrickController 2, but I don't have a physical controller. I want to buy one though, just not sure what brands work well with the app. Can you suggest some?

Most if not all bluetooth controllers work afaik. Playstation controllers work and the latest (from the Series S/X) Xbox work, but also the cheap ones like from Ipega (I got one of those for like 20 bucks, works great).

Share this post


Link to post
Share on other sites
1 hour ago, vascolp said:

Or... you can try Remote BlaBla... 

To control a BuWizz? That's my problematic point here.. You probably meant for the Technic Hub version, but for that I have the PU App.

While I like the concept, unfortunately I don't have that controller either, and the fact that it cannot do proportional control is a big minus for me (as far as I understand it cannot). I understand it can increase/decrease the speed/angle in steps by pushing the button multiple times, but that's another not so controllable option. For me PU is about precise control. Thanks anyway!

Share this post


Link to post
Share on other sites
8 hours ago, gyenesvi said:

While I like the concept, unfortunately I don't have that controller either, and the fact that it cannot do proportional control is a big minus for me (as far as I understand it cannot). I understand it can increase/decrease the speed/angle in steps by pushing the button multiple times, but that's another not so controllable option. For me PU is about precise control. Thanks anyway!

Well, the main purpose of Remote Bla Bla is to be a general purpose tool that avoids the need for programming. So, I guess it will never be perfectly precise.

And the lack of remote is of course a no go...:classic:

I am already working on a newer version which includes modes for motor speed control (instead of controlling motors by power), so maybe later you can give it a try...

Thank you for your answer!

 

Share this post


Link to post
Share on other sites

It's been a while I have last posted in this thread, but now I have something worth putting here. After the chassis posted above, I have recently been experimenting in various builds with a different motor setup using two L motors on the floor for drive. I have tested both PF and PU versions of this setup. The reason I really like this setup is its cleanliness and simple but solid chassis and suspension geometry.

800x600.png

800x600.png

This is a popular setup in many MOCs, and as always it has its advantages and disadvantages. As an advantage, it takes little space away from other components, such as the steering servo, has low center of gravity, and the motors are conveniently placed so that the drivetrain is quite simple. Also, a gearbox can easily be sandwiched in between the two motors. As a disadvantage, the drive motor placement is not realistic, and it takes the space away from an RC gearbox as placed in the previous chassis, though it would be still possible to add an orange selector in place of the gear stick and control that with a small motor.

About the suspension. The axles are 13 wide, with somewhat similar build as above (though they got a bit more slim). The front axle has a complete 4-link setup with an additional Panhard rod that is placed quite low and results in minimal sideways sway upon flex. The steering is via a linkage directly from the servo, this is my favorite setup due to its simplicity and realism. The springs are placed fairly close to the middle, so it has a good amount of flex, about 2 studs at the wheel. The rear axle is a 2-link setup with positive caster, and an additional Panhard rod, also placed very low to minimize sideways movement, which would be bad for the drivetrain as it only has a single CV joint (like in the previous chassis above).

An interesting property of the chassis is that it can easily be built both with PF and PU components, and this way I had the chance of making a comparison which had some unexpected results for me. At first I tried to build with PU components, but the length of the L motors (8L) was just not perfect for the gearbox / suspension combo. Since I had many builds with PU components and I wanted to compare them with PF ones, I checked how things would turn out with PF components. Well, I have to say many things are much better than with PU..

First, the length of the L motors is only 7L, which is just perfect for sandwiching a gearbox between them as the length of the gearbox is also 7L, which is important if one wants to build only using new CV joints. The 8L PU L motors can be hacked in, but they will be in the way for the mounting points for the rear suspension links, so those have to be moved back 1 stud, which will result in slightly misaligned geometry. No big deal, but I like when things line up well :) Also, a big plus for the PF L motors is that they are about 30% faster than the PU L motors, hence when used with planetary hubs, they result in a more playable speed while still having plenty enough torque (for a model this size).

Second, the steering. The shape of the PF servo is just great for this kind of steering, it is possible to make its output low, but the whole motor is not too low to take space from other stuff. Furthermore it has a rear output, which can be used to route it to a working steering wheel. Last, but not least, it really seems to me that the PF servo is quite a bit stronger than the PU L motor for steering. (At least, based on some testing, my guess is that the control software do not make full use of the total power of the PU L motor, but it can make use of the power of the PF servo; I guess the difference is in the complexity, as the PU L motor is not designed for servo function only. Anyway, as I have tested both options, the PF servo was able to steer under a complete build with bodywork while stopped on rough terrain and the front axle being flexed out, while the PU motor was lacking here..

Here are some more images. This time I have a fitting bodywork as well, coming soon in its own thread :)

800x600.png

800x600.png

Finally, any kind of battery can be simply placed in the trunk behind the seats, which again is an advantage for building / recharging, but raises the center of gravity somewhat.

Edited by gyenesvi

Share this post


Link to post
Share on other sites

You made a great investigation here! That was a pleasure to read through. I have 2 C+ hubs, Mindstorms 51515 set and a bunch of C+ motors, but have not interest in using them, cause I have a plenty of PF stuff: from almighty XL to the fast BM ;-)

Share this post


Link to post
Share on other sites
3 hours ago, damian_kane_iv said:

Have you tried this setup with harder springs?

Yes, in this Bronco build I was using hard springs and that worked well also. It really depends on the weight of the whole model; the Bronco's body was fairly big and heavy, but the body I am putting on this one is quite light-weight, so it needs softer springs to allow it to flex properly.

 

3 hours ago, Daniel-99 said:

You made a great investigation here! That was a pleasure to read through. I have 2 C+ hubs, Mindstorms 51515 set and a bunch of C+ motors, but have not interest in using them, cause I have a plenty of PF stuff: from almighty XL to the fast BM ;-)

Thanks, glad you like it! Yeah, I also have plenty of C+ stuff, but now I am tempted to build more with PF stuff, luckily I have some components to use :)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.