How to simulate/create a spring in an already generated Block Diagram of a CAD Model in Simscape Multibody?

Hello there,

i am trying to simulate a double wishbone suspension. For that, i created a CAD Model in Solidworks and exported it to Simulink.

The wishbone connected with a spring damper, is the part i am having trouble with. In fact how can I simulate the spring itself? Since the spring and damper component can not be visualized, how do I add the needed joints or blocks to the already generated block diagram of the imported CAD Model?

The requirements are that, the spring and damper component is not represented by a cylinder/piston (like shown in various videos) so only by a spring.

So again, how do i connect a spring with the wishbones so that the wishbone can "oscillate". Which joint(s) is/are required?

Thanks

Answer 1

[Yifeng Tang]

You'll need to first define how the two ends of the spring is connected to the wishbone part and the vehicle body, including the locations, orientations, and joint types. The locations and orientations is defined by Rigid Tranform blocks, and the joint types are probably two ball joints or one ball joint and one revolute joint.

Then the longitudinal motion of the spring can be constrained using prismatic or cylindrial joints. From there, you have two options to model the spring & damper force: (1) inside the joint itself, OR (2) couple with Simscape translational mechanical domain in 1D. For (1), you need to specify the "internal mechanics" in the joint, while for (2) you need to enable the force actuation and velocity sensing in the joint and a translational multibody interface block.

[Nayla]

Hello,

Firstly thanks very much for the quick answer.

I tried to implement the first option.

For the beginning here a simplified visualization of a wishbone and block to which i added a prismatic joint (see attachement Simplified visualization).

The wishbone is connected to world frame with a revolute joint and the block should be fixed. But when i run the simulation, it shows the error *** Model not assembled: position violation *** Resolve this issue in order to simulate the model. (See attachment First_trial to see its block diagram)

But when i try something else (see attachment Second_trial) there is no error message but the assembly does not work as desired.Meaning, the block is swinging along with the wishbone. Rather the block should be fixed and the wishbone should move with the spring (prismatic joint) while being fixed with a revolute joint.

What am i doing wrong?

Thanks in advance.

[Yifeng Tang]

In the First_trial, you need a spherical joint where the primatic joint is connected to the wishbone AND where it's connected to the ground. Otherwise the system is over-constrained, which is what the error message is saying. Depending on the exact locations of the two joints, one or two of them may be replaced with revolute joints, but starting with spherical joints makes things easier.

In some pictures of wishbone suspension, you can see the joints connecting the spring/damper to the ground and wishbone, but you may also see some picture where the connection looks rigid. In the latter case, the spring may bend sideways to account for the DOF needed, which we can't do with with prismatic or cylindrical joint. Adding spherical joints to the two ends is easier and better.

[Nayla]

Thanks very much for the suggestion. I was able to implement it and now there is no error message.

Right now i have 2 questions:

• is it possible to add a spherical joint to the other end of the wishbone (see attachment Trial_new) becaus for the moment the end without any joint (F2 of wishbone) is loose and doesnt swing according to the movement. Ive been trying to connect F2 to an other world frame but it shows the same error message i had at the beginning. (-->The corresponding block diagram see attachment Block_diagram)
• is it possible to restrain the rotation of the spherical joint to only one axis (for example make the wishbone rotate just around the z-axis of the following frame)

Thanks again for the quick responses.

[Yifeng Tang]

Yes and yes, I think. Just need to figure out the location of the joint. In case of a revolute joint (i.e. spherical joint with only one axis), you also need to figure out the orientation using rigid transform. The joint defined in your model should match what's in an actual design, so it helps to look at some pictures or under your car :)

[Nayla]

Hi,

i have been trying to implement the knowledge on a more advanced model and was able to make the wishbones swing. (See Blockdiagram_normalstate and its visualization in mechanics explorer Normal_state)

But when i try to connect the ground and wishbone with the spring (prismatic joint), then it shows the error message *** Model not assembled: position violation *** Resolve this issue in order to simulate the model. " (See Blockdiagram_leading_to_errormessage and Result_errormessage)

A wishbone is conntected to the chassis with a spherical on one side and telescoping joint on the other side.

What do i have to change? What is the cause to the problem?

[Yifeng Tang]

You probably need a rigid transform to align the axes in the correct direction. Prismatic joint moves in the z directions of the two frames, and the revolute joints rotates along the z-axis of the two frames as well. there is a 90 deg rotation needed somewhere I believe.

[Nayla]

Hi,

i have been trying really hard to make this simulation run, but it just dont work even after adding the rigid transform. Error messages are still shown.

So, to ease the process could you please have a look into the file attached.

Thanks !!

[Yifeng Tang]

missing a "Baugr_test_DataFile.m" file. I believe it's generated when importing the CAD.

[Nayla]

Here's the missing file

[Yifeng Tang]

You were pretty close actually. So here are the list of minor fixes I put in:

1. replace telescope joints with spherical. longitudinal DOF not necessary.
2. replace revolute joint with spherical. direction and location were too hard to determine by hand for the revolute joint. If you have CAD data, it will be easier.
3. add mass to both ends of the prismatic joint, to avoid zero mass degenerate system. Also made it ugly so it's very noticeable.
4. Turn on simscape data logging, run the model, to find out the initial position of the prismatic joint.
5. Use initial position of prismatic joint to set up internal mechanism with a spring. stiffness and damping randomly picked.

Now the model runs. The lower wishbone oscillates to a stop but the upper one is swinging. I suppose you need some kind of mechanical connection to bind the two wishbones together.

[Nayla]

Thank you very very much :) :) .

Very helpful!!!!!!!