Jump to content
Hash, Inc. Forums
itsjustme

Car suspension rigging

Recommended Posts

Very nice, Mark! I'll have to check out the project. I got a very big grin when I saw that you went to the trouble of setting up the drive shaft...that is very cool.

Share this post


Link to post
Share on other sites

If that's not what you want, than I could probably setup just a wheel and body installation rig.

 

Mark,

Wheel and body would be great, thanks.

 

I would still like to consider the suspension aspect as it will allow for greater detail.

 

David

Share this post


Link to post
Share on other sites

Here's an installation rig. This is just wheels, axle (if needed) and body. This is an easy install, it took me 10 min to install it in the porche model from the cd.

 

Import the rig into your model. From the front view, scale the model bone in the X axis, using the scale manipulator, so the tire/axle bones are centered on your tires. Now from a side view, you may need to translate the model bone (using the translate manipulator), in the Z axis only, so the tire bones are inbetween your tires, now scale the model bone in the Y and Z axis (using the scale manipulator) so the tire bones are centered on your wheels. Translate the body base bone centered between the front and rear wheels (Z axis) and to the bottom of the body mesh (Y axis), also translate the body base target null to the same height as the body base bone (Y axis). Assign the cps to the right bones (use body bone, not body base bone).

 

Now that that is done, there are 2 relationships to edit. First is the body constraints, the 2 translate to constraints need to be compensated (compensate each one when the other is at 0% enforcement, set set them both to 100%). Next is the auto wheels expression, but first you need to find out the diameter of your tires (zoom in on a tire and use markers on the ruler, subtract one number from the other to get your diameter), now edit the relationship, click on the expression and change the last number to the diameter of your tire. The car is all set, now for the choreography.

 

Create a new choreography, create a new path (make sure it is just below the lowest point of your ground plane), drag and drop your car model and your ground plane (if you have one) into the choreography. Now you have 7 constraints to setup. The bones you need should be showing. Start by constraining the model bone of your car and the auto steer target to the path, set the ease on the auto steer target so it is right between the 2 front tires, subtract that ease from 100% and set the model bones ease at the end of the path to that number and the ease of the auto steer target to 100%. Next is the surface constraints, there are 4 surface target, constraint each one the the ground plane (surface target) and the surface aim bones (surface aim target) and turn ON 'translate only' in each surface constraint. The last is to translate the ground target translate bone to the ground plane. Your finally done, play your animation.

Share this post


Link to post
Share on other sites

Hey Mark,

I rendered out your Choreography yesterday from a different angle and stretched it to

90frames.

Pretty cool!!!!

Thanks for putting it together

 

Toon Render....

about 1.9megs

http://www.vrcops.com/Vids/Trucker1.mov

 

 

Mike Fitz

www.3dartz.com

Share this post


Link to post
Share on other sites

I forgot to mention that, the diameter you need to plug into the expression needs to be in (cm).

 

That looks cool Mike, I like the toon render, but the tires seem to be rotating in the wrong direction, I'll have to take a look at that.

Share this post


Link to post
Share on other sites

I don't think that the wheels are spinning the wrong way...

this came up a few weeks ago.

I think it is something to do with the spin being faster than the frames per second.

Ofcourse, that is the caveman explanation, so anyone with a more "advanced" explanation should jump

right in!!!

I notice it on cars in real video as well....

 

 

Mike Fitz

www.3dartz.com

Share this post


Link to post
Share on other sites

I found a problem with the way I set it up and how the expression works. The target bone is on the right, I move it in bones mode so it was at the same xyz as the model bone. I translated the bone to the ground plane, now rotating the model bone the wheels spin, even though the distance has not changed. But actually it does, because according to the model bone it should be in front of the truck, but when you rotate the model bone, it actually is translating to the rear of the truck. Any thoughts?

Share this post


Link to post
Share on other sites

I don't think that the wheels are spinning the wrong way...

this came up a few weeks ago.

I think it is something to do with the spin being faster than the frames per second.

Ofcourse, that is the caveman explanation, so anyone with a more "advanced" explanation should jump

right in!!!

I notice it on cars in real video as well....

 

 

Mike Fitz

www.3dartz.com

 

Yeah...that's a common phenomenon called 'wagon-wheeling'...it does have to do with the fps and the fast rotation of the wheel creating an illusion that it is spinning the wrong direction, and it happens all the time. Many times, an animator will counter the problem by actually spinning the wheels the wrong way, and the illusion corrects the action...

 

I see this illusion all the time in real-life, as I watch cars go down my road thru the vertical blinds of my house.

 

BTW- I sure am enjoying this thread and played with the rig a little bit... it's GREAT but I tried delving into it to see how it works, a bit over my head... made THIS tho---

truktest3.mov

Share this post


Link to post
Share on other sites

John, I noticed the reversed wheel rotation by stepping through his animation, I could be wrong though. Also, did you disable the dust from the front tires or is broken?

Share this post


Link to post
Share on other sites

Well, guys, as the guy who started this whole subject in the original thread, I've been conspicuously absent. My hands have been full with tending to such minor things as a sister's attempted suicide and then a friend's. Just another month in the life of Nosferatu...

 

I just wanted to thank you guys for all the great work you've done. Now I can finally work on the sample projects you've provided and I'm really looking forward to it.

 

I love to see the spirit of invention that this and other threads have brought out in many of us!

 

Nos

Share this post


Link to post
Share on other sites

Regarding the "Wagon Wheel Effect" (temporal aliasing), there is another consideration that I rarely see mentioned.

 

When the wheel is rotating under a stroboscopic lighting condition (such as photographing at 24 fps), the mind assumes that it is watching the same individual spoke rotating backwards.

 

For example, let's assume we are watching spoke #1. If the fps matches the wheel rotation exactly, we will see spoke #1 appear to stand still. Of course, this applies to all the spokes, but we are only interested in spoke #1.

 

When we change either the rate of rotation of the wheel or the fps, we will see spoke #1 apparently start to rotate backward or forward. As a matter of fact, the brain is assuming that we are watching spoke #1, when in fact we are seeing other identical spokes each time a frame is shot. It looks like spoke #1 is slowly moving backward, but in actuality we may be seeing other spokes "masquerading" as spoke #1 just because they happen to be positioned in the right place when the next frame is taken.

 

I used to work on mail processing machines that moved envelopes around at extremely high rates of speed. It was very hard to diagnose certain problems because the high rate of speed made everything a blur.

 

One day I made a strobe light with a very wide variable flash speed and with this I was able to "freeze" the motion of gears, pulleys, and envelopes. I could make the gears seem to stand still, or rotate backwards or forwards. I could make it look like the envelopes were standing still inside the machine, because the strobe light would flash every time a new envelope was at point "X". In reality, I wasn't looking at the same envelope standing still, but rather, I was looking at a different envelope at point "X" which was taking the place of the one ahead of it.

 

Yours in ever-increasing confusion,

Nos :blink:

Share this post


Link to post
Share on other sites

Mark,

 

I am having a bit of a hard time getting the rig to work.

 

I followed the instructions to the best of my ability (which would be a 1 on a scale of 1 to 10) and can't get it to work.

 

Any revisions to your rig?

 

Thanks,

David

Share this post


Link to post
Share on other sites

Sorry David, no upates to the rig. I haven't looked or worked on it since I posted the rig, maybe there is a problem with it in v13 (I assume you are using it, I think I built that in v12) I'll take a look.

 

You can e-mail me a prj file to look at, if you want, to see if you set it up correctly, if it's not a rig problem. I'll PM you my e-mail address if you are interested in doing so.

Share this post


Link to post
Share on other sites

Thanks David

 

That's strange though, the poses work fine in v12 and v13, I downloaded it from this thread on a different computer and they works fine.

Share this post


Link to post
Share on other sites

I'm not sure either...I didn't re-download the file, I just used the one I had on my drive that I downloaded when it was first uploaded. Maybe you reposted it over the original post and I downloaded it in between?

Share this post


Link to post
Share on other sites

Hmmm, I never reposted it, oh well, no big deal.

 

I just checked the original file, it worked fine. Must have been some forum strangeness or something at the time I originally downloaded.

Share this post


Link to post
Share on other sites

Hi, guys,

 

After realizing that the inner wheel of a turn has a smaller turning radius, that meant that the inner wheel should turn more than the outer wheel in a turn. To see what the calculations would be like (after a friend of mine and I spent a couple of hours playing with the math), I did some googling.

 

I found some information on Ackermann Steering Linkages, and after doing a somewhat bad rigging, I went back to find this post. I wonder if one of you guys might want to add this behavior to the rig?

 

I'll upload my not-completely-working rig that showed how it might be done in principle. This is just for the mechanism, I did not figure out a way to make it properly scale based on the length to the rear axle using Ackermann angles and how to have the tires be straight-ahead when adjusting. Anyways, if you create an action off of it and move rotate the bone called "Cross", it gets almost reasonable steering behavior. (You can ignore all the hidden bones, where were a failed attempt at making the angles adjust based on where the rear axle was placed. Sorry, I should have cleaned it up a bit more. The only relationship that is needed is Right Ack End's Kinematic Chain to Right Base.)

Ackerman_Steering_Model.mdl

Share this post


Link to post
Share on other sites

It doesn't seem to do anything, can you post an action with the mechanism working?

Share this post


Link to post
Share on other sites

Hmm. Hopefully this will show the behavior.

 

There's probably a lot that I've taken for granted and not anticipated a description being necessary. One thing I just thought of is that "Left Ackerman2" and "Right Ackerman2" are the bones that should be pointing at the rear axle. They are supposed to be at a constant angle from the tire they are to be connected to; hopefully that will be clear from the diagram in the Wikipedia article I posted.

 

Do let me know if I am crazy. :)

Ackerman_Steering_Action_Test.act

Share this post


Link to post
Share on other sites

I was trying to use the rotate manipulator, that's why it wasn't doing anything.

 

As for the mechanism, would that even be noticable in an animation? Would it be worth the extra rigging involved? It's certainly doable.

Share this post


Link to post
Share on other sites
As for the mechanism, would that even be noticable in an animation? Would it be worth the extra rigging involved? It's certainly doable.

 

Good question, possibly not. Even without 100% accuracy it probably still achieves "believable" (aka still looks good) side of things. Probably best to just leave it out. Thanks for the feedback, though!

Share this post


Link to post
Share on other sites

I tried it out on my school bus. I've attached a video, as well as the model and choreography. One difference in the model: I put all of the wheel bones under a top-level bone of their own. One difference in the chor: I did not constrain the 4 wheels to the ground surface, because the terrain in my example is flat.

 

Two things I noticed:

1. The wheels don't rotate/spin as fast as the forward movement in the very beginning. Does this mean I need the primary motion to go in the Z axis within the Choreography?

2. Near the end of this video it is evident that the wheels turn before the front of the vehicle does. I think it is because the path is translated to in the middle, and the wheel target is in front of the vehicle. I wish there were a way of getting the front of the vehicle to go along the path, and have the rear wheels drag behind; after playing with it a while, I could not find a way.

 

Next is the auto wheels expression, but first you need to find out the diameter of your tires (zoom in on a tire and use markers on the ruler, subtract one number from the other to get your diameter), now edit the relationship, click on the expression and change the last number to the diameter of your tire.

Could the expression be changed to use 2 * (front_axle.Translate.Y - steering_base.Translate.Y) or something along those lines? Or do those change while the vehicle actually move?

 

Hope this is helpful.

School_Bus_Wheel_Body.zip

WheelTest1Avi.zip

Share this post


Link to post
Share on other sites
I wish there were a way of getting the front of the vehicle to go along the path, and have the rear wheels drag behind; after playing with it a while, I could not find a way.

 

Perhaps an expression could be used to make one null trail another at a fixed percentage of a path's length. Then the front of a vehicle could costrained to the leading null and the body of the car aimed back to the trailing null.

Share this post


Link to post
Share on other sites
Perhaps an expression could be used to make one null trail another at a fixed percentage of a path's length. Then the front of a vehicle could constrained to the leading null and the body of the car aimed back to the trailing null.
As far as I can tell, that is more-or-less the mechanism in the current version of this rig. I think part of the problem is that (ETA: In a real vehicle) the rear and the front of the vehicle travel along different paths; putting them both on the same path would be/is rather noticeable.

Share this post


Link to post
Share on other sites

Perhaps a more accurate solution would involve putting the main bone of rotation at the back axle and the front of the car swinging at the end of that, which more like what happens with a front steering car. Or is it?

Share this post


Link to post
Share on other sites
Perhaps a more accurate solution would involve putting the main bone of rotation at the back axle and the front of the car swinging at the end of that, which more like what happens with a front steering car. Or is it?

I think there are probably two approaches. I don't know if either will work with the available set of constraints.

 

1. What you are suggesting sounds like what it would be like for rear wheel drive. I think one important aspect is knowing what the curve is like at the rear wheels. The bone itself should keep to the tangent of the path. The front tires must then be turned enough to compensate for the turning that is about to happen, so it looks like they are what is causing the front of the vehicle to swing over. The only danger I can foresee here is that a sharp curve in the path would make the front of the vehicle swing quite rapidly.

 

2. If you try to mimic front-wheel drive, the big question is how do you get the rear wheels/axle to drag behind the rest of the vehicle.

 

I think approach #1 might be the closest to the existing rig.

Share this post


Link to post
Share on other sites

Unless the wheels are skidding i don't thing front or rear wheel drive makes a difference. There's still a fixed geometry between the front and rear axles whether the front or rear wheels are providing the power.

Share this post


Link to post
Share on other sites

Yep. From a rigging point of view, the difference is where you attach something to path. If you're pushing from the back, you need to get the front to behave correctly (which I think might be doable). If you're pulling from the front, you have to get the rear to behave correctly (i.e. drag behind; I think this might not be doable with the existing set of constraints A:M offers, although the rigging gods on the forum are probably much more knowledgeable than me about that).

Share this post


Link to post
Share on other sites

Now, I'm thinking that making the front the center is better instead.

 

But say you have a car pointed north. Then you turn the front wheels to the north east and keep them pointed northeast as you drive.

 

How long before the body of the car has swung behind the front so that it is also pointed northeast? I suspect it's related to how far the car has traveled and that the distance required is the same for any reasonable speed of the car. I also suspect that most of the swing happens sooner rather than later.

Share this post


Link to post
Share on other sites
Now, I'm thinking that making the front the center is better instead.

 

But say you have a car pointed north. Then you turn the front wheels to the north east and keep them pointed northeast as you drive.

 

How long before the body of the car has swung behind the front so that it is also pointed northeast? I suspect it's related to how far the car has traveled and that the distance required is the same for any reasonable speed of the car. I also suspect that most of the swing happens sooner rather than later.

I think this would be very nice. However, from a bone-and-constraint perspective, how do you get the rear axle to follow the front wheels? Getting it to aim at the front wheels is easy, but the following part is hard. (At least given what I know about rigging.)

Share this post


Link to post
Share on other sites

The "body" would be a bone that originates from the center of the front axle and points backwards.

 

 

Basic motion, animated by hand.

 

Wheels.mov

Share this post


Link to post
Share on other sites

The biggest issue is the length of the wheel base, the longer it is the more slipping you will have with this rig.

 

Try this:

 

Move the model bone to the rear axle. Add a child bone to the WheelBodyRig null, that is located at the rear axle as well. Make this bone a parent of all the bones except the "auto_steer_target" and the "ground_target_translate". In the chor have this new bone aim at the "auto_steer_target" (no compensation). You will need to readjust the ease on the path constraints.

 

This will help, but not solve all the issues.

Share this post


Link to post
Share on other sites

Also note, this version of the rig does not rotate the wheel properly when moving in the X and Z axis at the same time and/or the opposite direction. I have an updated rig that fixes these issues, but no installer. I never posted it, I'm too lazy to write up instructions.

Share this post


Link to post
Share on other sites
Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...