Issue with mass "breaking" ADAMS Car model 2

[OTRace]

Hello all

I am working on an ADAMS Car template and am trying to run a constant radius cornering simulation with vehicle accelerating. The template comes with suspension hardpoints already defined that I have modified to meet my needs. There also is a weight and moment of inertia tensor that I need to change to match my expected vehicle properties. The issue I'm having is that the unchanged template works fine and the results I get from the post processor view make sense. Keeping mass the same, I changed the suspension and steering hardpoints, as well as my spring rates and tire dimensions, then ran the same simulation again and still no problems everything makes sense.

The problem is when I change the mass and principle moments of inertia on ADAMS to what the CAD file I am referencing my design off of has. To change mass and principle MOI tensor, I went to Adjust > Adjust Mass, image attached below shows mass and MOI tensor inputs compared to what I want from CAD.

To make sure changes are reflected correctly on ADAMS, I went to Tools > Model Diagnostics > Aggregate Mass and the MOI tensor and mass looked alright. When I run the simulation, graphs I plot on post processor make no sense, and then looking at the animation, it looks like the vehicle model bottoms out fully (frame hits the ground plane). (Forgot to grab pic with my model before closing it without saving but here is the default model with the same effect)

This is the default model

Default model breaking

Kind of looks like instead of the mass of 265kg the car is feeling a mass 100 times that amount. This only happens as soon as I change the mass in either the default template model or in the modified template model with the different hardpoints, springs, tires, etc. To fix it, I close the program without saving and reopen so everything goes back to the default settings but the template model does the same thing even with the default mass. Only way I can revert all the changes to make it work again is by deleting the folder containing the template, removing the session from ADAMS, and redownloading the folder onto my computer to start from scratch

As I was writing this I was playing around with some more sims to try finding a root cause. I changed mass of the default model (changed only mass) and it doesn't replicate the issue mentioned above, but it gives me the following error message and gives me a small animation of the car facing almost perpendicular to the track instantly. I thought because I was changing springs it was messing with my model but even without changing the springs I still get this error so first and foremost it has to be an issue with the mass

Next I kept the same failed condition from above but this time I put the softer springs I wanted to test (changed nothing else) and I was able to replicate the issue I mentioned in the beginning. However, this time the car animation actually completed the entire simulation. I got the same error as above and the graphs still look very wrong.
Refer to this image again for how default model looks "crushed"

Its worth noting that the original mass of default model is about 249kg and the mass I'm testing is 265kg so not a huge difference, original springs are also around 285 lb/in and the springs I'm changing to is 225 lb/in so also not a crazy difference that should make the model look like its on Jupiter. What this tells me is something has to be wrong with the way I put in my mass as my suspension hardpoints and tires have no effect. I'm not sure about the springs however, the 285lb/in vs 225lb/in shouldn't have such a drastic difference in the model

Let me know if you have any ideas on what could be wrong! Thanks

 

[GregLocock]

OK, take a deep breath and just tell us explicitly what you changed between the two. I am fairly confident you've ended up with a weird inertia matrix, with negative off-diagonal elements. If you want to do it my way export the adm from each of the two new configurations and post the two adms as files.


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[OTRace]

Hey Greg

So the issue occurs when I change the mass. I'm assuming that it should fix the issue I'm also facing with the spring once mass is sorted out. I tried changing the diagonal and off diagonal elements of the matrix to according to what my CAD assembly has them as but still not able to get it to work on ADAMS

I attached the file with the default configuration that is available on MSC website, is this what you mean? I'm trying to change the mass in the fsae_full_vehicle.asy

Thanks

 

[GregLocock]

No that's not what I mean. If you look in your working directory each test builds an adm file, which is written in Solver language. I need the adm from a run that worked with the original mass, and one that didn't, where you have modified the inertia.


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[OTRace]

Ah I see. Below I attached zip folder containing the two .adm files. New_Model_w_Template_Mass is the file with the mass and inertia tensor that the template comes with by default, it works fine. The other file retry_new_mass is the same model but with the updated mass. Let me know if this is what you wanted. Thank you!

 

[GregLocock]

Yes, that's the right file. You've got stupid values for the inertia matrix in the part 'fsae_chassis.ges_chassis' with moments of inertia about 10^4 times larger than the original. Your Ixy and Izy (I think) are negative which is physically impossible. You've also got the CG of that part in rather a strange place, 300 mm forward of the original one.

In future make your changes one at a time until something breaks.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[OTRace]

I've been trying to change the mass and COG location through the Automatic Mass Adjustment option. Does that "suppress" all the other masses in the assembly? When I go to check the overall mass of the entire assembly it shows whatever values I put for mass, COG, and inertias in the Automatic Mass Adjustment. Basically I want to input my mass and inertia data from a CAD assembly that contains all the necessary vehicle components, into the ADAMS model. Hence why I thought Automatic Mass Adjustment would let me do that provided I have the COG and inertia tensor.

Regarding Ixy and Izy, doesn't it being negative refer to their directions? My principle inertias are still positive.

Yeah the CG location for the working file is using the template CG location but with my suspension hardpoints. The suspension hardpoints I want in my model are further away from the ADAMS GCS origin than the template model hence why the huge discrepancy.

 

[OTRace]

Ah so I figured out why my moment of inertia is 10^4 larger than original, I was putting in the wrong units! Fixed the units now and the sim is running fine but I'm still unsure about the data I'm getting from post processor view. Will play around with model a little more to make sure mass and the inertia tensor are no longer the issue.

 

[GregLocock]

When you increase the mass AND move the CG in X you'll need to retune your springs to get a reasonable ride height. Otherwise the vehicle will end up sitting on its jounce bumpers or rebound springs, assuming it has them. It should still assemble, and may even run, but it will be meaningless.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[OTRace]

Ah I completely forgot to even consider that thanks for bringing it up! I think that fixed all the issues I was having with the model. Thanks for your help Greg!