Front suspension with geometry (MOD)

[amorti]

First things first: This post isn't about them (and I'd prefer if it doesn't go that way) but it will show constructions which are obviously built using them. So, if seeing Chinabricks upsets you, best look away now.

Following from @astyanax's Koncept Mantis MOC thread, I have been playing with some of @nicjasno's ideas for incorporating real suspension geometry into a model.

Nic (and others!) - I'd really appreciate any feedback to the design, construction, etc.

I've taken a bunch of pictures, here for full resolution gallery.

Let's start by looking at the original suspension. 

It's actually not bad at all. It uses two 9L wishbone parts as the lower control arm, which gives it great stability and looks perfect for the car. The 9L steering links are used and it doesn't have any bump steer.

Clearance is around 1.75 studs at the front, and it uses about half the spring travel before the car grounds out.

Non-friction pins have been used instead of ball joints, they have a little play but are actually really pretty good.

The top wishbones are shorter than the bottom ones like a real car. Although the lack of ball joints makes me wonder if it's "proper", it all works within the limits of the actual suspension travel without any binding.

The springs have been attached at the bottom wishbones in really the only available place, although I did put those two red 3L thin beams in place of a 32184 connector, and mounted the springs there, which I preferred that as it softened the spring rate and gave a bit more clearance.

Here's a comparison side by side with what I've come up with:

It has (possibly too much?) caster angle. It's 1.5 studs further back at the top than at the front. I think ideally I'd like to move the bottom ball joint half a stud back, need to look again at whether I can get the wishbone mounted solidly in that spot.

 

Which should also let the steering rod run completely parallel to the wishbone. 

That misalignment doesn't seem to be enough to cause a noticeable bump steer, though.

The top wishbone is 6 studs long, the bottom wishbone is 9 studs long. 

The steering knuckle has some King Pin Inclination (KPI) which is shown by the grey 5L thin liftarm. That lets the tyre's scrub radius be just about where it wants to be just inside the middle of the tyre. Which is kind of a shame because this guy was a better part for mounting the top ball joint than the similar 3L part which only has pin holes rather than an axle hole in the end.

The 5L thin beam being used as a brace isn't quite in system, I think. It gets hard to tell when you're dealing with angles all over the place. Either way, it's staying, because there's not much other chance to brace the two wishbones to each other and trap the ball joints.

The spring is now a single. The car has around 3.5 studs of ground clearance, which is nearly 2 studs more than the original design by a trick of the knuckle (see below). Once it uses all the travel it still has about 1 stud clearance. It sags about one stud into the travel on the car's own weight, which seems good to me, however I can't get an accurate read on that until I do both ends and both sides... which will mean ordering up another pair of "rear" wheel hubs! I only got out of my Dark Age this Christmas and don't have that big of a collection, hence buying some Chinabricks to get it moving.

The lower wishbones need to run pretty much flat to the floor of the car, but I wanted more clearance. You can see here that the wheel is not in the middle of the ball joints, rather it's been offset lower to give more clearance.

The steering knuckle is sturdy enough, but still it could be stronger, I think.

I have paid attention to Nic's pointer that it should have pins going through the hub not axles, but the top and bottom aren't the most solidly connected. Unfortunately I ran out of patience trying to make it hold together better whilst fitting inside the rim.

There are more pictures from more angle in the BrickSafe Album.

So gents, what do we think?

As above, I'm really looking forward to feedback on this, and maybe suggestions for improvements. 

Edited September 8, 2020 by amorti

[Jurss]

Wheel turning point is on the weel side, not in the center anymore.

Connection of that 5L halfbeam also is not nice. 

In needs to be hardly developed.

[amorti]

3 minutes ago, Jurss said:

Wheel turning point is on the weel side, not in the center anymore.

 Connection of that 5L halfbeam also is not nice. 

In needs to be hardly developed.

I was just looking at that. That wheel has a reduced steering lock compared to the other which isn't such a bad thing, but IDK what other effects that would have. I could put the ball joints in the centre line of the knuckle without a problem, but I'll have to look again at whether the wishbones can slide along to make that work inside the wheel arches.

Like I said - I know it's not nice, but you need something to hold the ball joints together and it's the only spot to do it. I can probably make it a little nicer by using an axle with stop, but it's never going to be a beauty feature.

I know it's not perfect - help me make it better :)

Edited September 8, 2020 by amorti

[amorti]

5 hours ago, Jurss said:

Wheel turning point is on the weel side, not in the center anymore.

Connection of that 5L halfbeam also is not nice. 

In needs to be hardly developed.

Well I think I answered all your points here.

The wishbones are thicker and now supported on pins instead of axles. Even so it's not nearly as sturdy as the original version.

The steering rod is straight and there's no bump steer, albeit the ball joint is not properly mounted in the picture (hanging down by half a stud).

The ball joints align with the wheel hub's centre. There's 1 stud of caster, taken care of by the wishbones being offset only.

Probably still pretending to be about a 5.1L thin liftarm, but much closer to legitimate.

Now the ball joints are in line with the hub's centre, the thing has gained more steering lock. That's good, but also a problem. Since the top ball joint is 2 studs inboard from where it would be on a normal steered hub, the wheel "wags" quite a bit, and I'd have to redesign the whole wheel arch constructions for it to work.

I'm coming to the conclusion that while this type of 3D puzzle is definitely possible and really interesting to do, it's not easy, and it's not necessarily suitable for RC models.

Edited September 8, 2020 by amorti

[nicjasno]

The gray 5l liftarm in the last pic is unnecessary and just hinders everything.

[amorti]

Just now, nicjasno said:

The gray 5l liftarm in the last pic is unnecessary and just hinders everything.

Maybe!

The two bigger problems are mounting the steering ball joint properly, and the wheel "wagging" causing a clash with the bodywork. The knuckles could be stronger too, although in use they don't at all look ready to fall apart.

Other than that, it's basically complete at this point.

[nicjasno]

Just now, amorti said:

Maybe!

 

Definetly.

[amorti]

1 minute ago, nicjasno said:

Definetly.

Alright I'll take it off if you can solve the other problems 

[nicjasno]

Give me more detailed pics of what you have and i make you a good suspension tomorrow. Strike that... after looking closer i can see what you have. Will do something tomorrow.

 

Edited September 8, 2020 by nicjasno

[amorti]

48 minutes ago, nicjasno said:

Give me more detailed pics of what you have and i make you a good suspension tomorrow. Strike that... after looking closer i can see what you have. Will do something tomorrow.

 

https://bricksafe.com/pages/A_morti/konceptmantis/geometry-front-suspension/for-nic

Few more pics anyway including an angle of the bottom, you can see where the wheel geometry "wagging" causes it to hit the bodywork but the other one doesn't. Some of the bodywork can be moved or taken away, however it's already not got a lot of mounting points holding it together.

Is it worthy of a live stream? I can be online tomorrow.

I'm interested to see what you get, and how close I got.

[nicjasno]

By "wagging" you mean the whole thing moving too much in the lateral direction? Try triangulating the lower wishbone. I'd even make it a multilink at this point.

[amorti]

Some of it is flex in the lower wishbone (it's not really any kind of final version) but really I mean because of the length of the top ball joint mount, the wheel doesn't pivot within its own volume, it flops back and forward while moving. Somewhat like the old wheels do.

Sorry, not a great description I know.

 

Multi link would be cool too. I do have plenty of 6L steering links and 6L wishbones spare (I think 8x of each), but only 2x 9L links, 2x 5L wishbones. Of course, bricklink has more.

Edited September 8, 2020 by amorti

[nicjasno]

Would doing a pushrod setup be acceptable?

[amorti]

Basic brief: the car is really fast, so the suspension should be sturdy and bring stability. Ideally it would fit without changing the bodywork too much. I think that might also be interesting to you to prove a point that any technic car can have proper suspension and space constraints isn't a good argument.

Everything else is open. Think of me like the dog: I will eat what I'm given, and be very happy.

[nicjasno]

Would it be ok if i develop this on todays live stream?

[amorti]

1 minute ago, nicjasno said:

Would it be ok if i develop this on todays live stream?

Absolutely! I'll get in the chat.

[nicjasno]

Deal!

[amorti]

On 9/9/2020 at 8:10 AM, nicjasno said:

Would it be ok if i develop this on today's live stream?

Huge thanks to Alex at LPE Power! @nicjasno

https://www.dropbox.com/sh/lapgx5lqw1urm0f/AAB0nR6K2cOXD1XOnt4QwHzUa?dl=0

@astyanax - how cool is that??

I don't have enough 9L steering links to build both sides, but I'll have a go at building one side myself tonight.

[amorti]

@nicjasno

It was a lot of fun thinking along with the live stream, and I've now built one side which is also fun to see how it works and what happens when you move one reference point a stud in or out or up, or down...

Unfortunately I won't order the parts to make the other side, as there are too many compromises.

The ball joints have too much chance to pop off. This in contrast to the very sturdy "ordinary" suspension the model was designed with, which is pretty decently form locked.

I can't find a way in the available space to form lock the pushrods , which keep coming apart under the weight of the car. I could glue them but as mentioned on the stream, it's cheating and doesn't feel good.

The edge of the wheel rim rubs the steering rod in one direction, and the front wishbone in the other. It would be fine with land rover wheels but I don't have any.

The two lower wishbones clash at full right steering causing the knuckle to rise.

The one problem that doesn't seem like it would be easy to design a way out of, is that the wheel has moved forward in the arch almost 2 studs. The front bodywork can be moved forward to accommodate but the car starts looking too long, and even then it won't really be possible to fit a front hood. It would be possible to move the whole front wheel assembly backwards 2 studs, but then the top wishbones are fighting with panels.

Still, I thank you for the master class, Alex. It was very enjoyable. If one day the feeling takes me to build a non motorised car, this will be the way to go.

[nicjasno]

Too much would need to be redesigned. Tried to fit it into the existing structure. I think it was an interesting thought experiment. I always enjoy challenges like this and we have all learned a bit from this.

 

[astyanax]

Holy cow you guys! Very cool stuff, and lots to learn for me. @amorti or @nicjasno if you settle one day on a strong alternative suspension, let me know and I might update the PDF building instructions.

[amorti]

I think during the ~3h live stream Alex (@nicjasno) got as close as possible.

It worked pretty well, and was actually pretty steady without much slack at all.

As I looked at it on the bench one last time, I realized the ball joints popping off isn't really a big deal as they only want to pop off when the suspension is above where the car decks out. Even so, Lego small ball joints are never going to be as sturdy as pins in a crash scenario, and when you add the geometry you get all sorts of angles and fractions of stud lengths, and you can't brace stuff easily.

If you've seen any of Alex's dodge charger streams (or tons of his historical YouTube videos) you'll know it definitely can be done, and made sturdy even for a pretty big car. It just didn't work out in this case.

Maybe the very long wishbones were a problem? Or just trying to fit it in the existing mounting points? Not sure, maybe Alex can say.

[nicjasno]

Trying to fit into the existing structure was the main hurdle. If designed from scratch, this would be much better.