Spring constraint bug

Perhaps it would be better to post this in #bug-reports instead, if you think it’s a bug.
Proper format won’t hurt either :slight_smile:

1 Like

@Veldaren He is just a basic user

@Player1unknown If you find a bug please relay it to a full member. Bug reports do not go in this category

3 Likes

I saw this bug reported internally. This was my answer:

We definitely have a bug with the interaction between springs and external forces. I think I know what it is but it will take time to fix. Here are a few workarounds to ensure that the vehicle stays upright and doesn’t flip around the corners:

  1. A first good workaround for your problem is to increase the masses of the Front_Left, Front_Right, Rear_Left and Rear_Right parts instead of using a downward force. This will stabilize your vehicle and prevent it from flipping to some extend.

  2. A second workaround is to use AlignOrientation constraint with PrimaryAxisOnly option selected on the body of the vehicle aligning it with an axis that’s vertical (the second attachment on something anchored like the a baseplate). This will force the body to be upright but allow it to rotate.

  3. A third workaround is to use the Forces on the wheels themselves.

3 Likes

Yeah I honestly do not know how this thing works because I joined had basic user settings. then I applied got accepted yet still a basic user and kinda gave up on the forums till I ran into this bug which i need fixing so yeah… Didn’t know what to do.

chefdeletat the cars drive upside down and do loops so we can not do those methods otherwise we would have done that. What I want to know is if this bug is being worked on "obviously I don’t expect it to be fixed by tomorrow but I would like to be informed about the progress of this bug. Which so far you told me that it was a known bug and from what I just read you are working on it atm. I am stressing over this as our entire game is dependent on this one thing and its delaying us significantly in our progress hopefully we can still launch before December 27th if the issue is fixed before then.

Alright you can still use option 2 and 3. Make sure you use VectorForce and not BodyForce, and for option 3 select the ApplyAtCenterOfMass.

Also please stop using Legacy Body Movers. We are not actively supporting them anymore and they can entirely be replaced by the new Attachment based forces and Align Constraints (see under the Constraint menu).

1 Like

VectorForce causes the same issue as body thrust we just tested it.

What happens if you apply VectorForces on the wheels themselves? At the Center of Mass. (As opposed to applying them to the Knuckle)

Not sure we will test that next.

1 Like

Just realized you meant the property going to try that.

Ok we just tested it with vector-forces on the wheels themselves with center of mass selected and it works for straightaways but it does not work with the loops and that is only if we toggle the option for relative to World otherwise the car fly’s around. With relative to world as soon as the car starts going up the loop it gets pushed back down. The other two settings just made the car fly around when driving. Your second work around only works on straight aways as well doesn’t work on loops or slanted parts.

Yes I can see that your initial solution is best - if only the spring didn’t react in this way. You’ll have to experiment with other options. Here is what you can do: you wanna apply a force on the wheels in a direction given by an attachment on the knuckle.

  1. Use VectorForce on the wheels, set the Attachment0 on the wheel and select ApplyAtCenterofMass
  2. Under “RelativeTo” select Attachment1
  3. Place an attachment on your Knuckle (the Front_right, Front_Left, etc) with main axis pointing down
  4. Set this attachment as Attachment1 on your VectorForce

I had to make some other fixes to your model:

  1. remove the old surface motors and replace it with a HingeConstraints
  2. Swap the direction of the prismatics in the suspensions (it’s more stable when the Attachment0 is on the heavier part)
  3. Made the knuckles a bit heavier for stability (otherwise you’ll get into high mass ratio issues and jittery suspensions)
  4. Increased the friction on the tires so it holds better upside down.
    Sending your model in a pm.

As for the Spring vs Force bug, the fix might not come soon unfortunately. From past experience the ETA is about 3 months. This is because we need to do extensive testing on impact of such a change.

1 Like

? That is not our chassis that is a free-model chassis used to show that it wasn’t a design flaw by us. ours was the porschee

Taking a look at the model right now I will post if anything changes or if this works as I can not post anything else today.

The same ideas should apply, right?

Ok your method worked, thank you for answering and proving a way to work around the issue.

Cool. Please mark this thread as answered.

Uh did I do that? Not sure.

A fix to this bug was enabled today.

You can click this button under a post to make it a solution. It will indicate the problem described in the thread has been solved.
chrome