Jump to content

Recommended Posts

Hi everyone,

Im building a suspension system and trying to apply some testing datas to model. The datas include airspring property mesured by static test and damper property mesured by dedicated tester. I run the simulation and using verical acceleration of chassis as output. The matter is when I using the properties available in directory, the output is smooth. But when changing to testing properties, there were a lot of rough in my result. I wonder do I need to do fitting step before adding these kind of testing data to model? The identification tool just uses for bushing, isn'nt it?

Thanks in advance!

Share this post


Link to post
Share on other sites

Did you build half-car models first, so you can evaluate the spring data?  When using curve data to model Airsprings, if you do not set things up correctly, with the correct sign conventions, etc., this can lead to problems.  Also, not having enough data to define the curve properly can cause problem.  Even having too much curve data can create problems.  But, the springs are easy to evaluate with a half-car model, since the body is fixed to ground, and we can plot wheel vertical force vs. displacement and verify that are wheel rates are reasonable.

The same tips in general also apply for shocks...if you don't have the shock force vs. velocity defined correctly, the shock (damper) can actually add energy to the system rather than dissipate energy.

 

For simple airsprings and dampers, there is no need for the MIT fit tool.  And you are correct, this is really only for frequency dependent bushings.

Share this post


Link to post
Share on other sites
On 6/19/2020 at 8:16 PM, Chris Coker said:

Did you build half-car models first, so you can evaluate the spring data?

I've tested components individually like air spring, damper, bushing. It's behavior looks well. But when combining them together the issue come up. Actually, I know there are problems in my input data (airspring/damper property) but I don't know how to fix them. Because it's the data I collected from testing. I want to know to get the reasonable results, do I need to to do something with the testing data or just put it to the model with the correct format?

It will focus more on the way to solve these kind of problems.

Thanks!

Share this post


Link to post
Share on other sites

The general answer to your question is "you don't need to do anything to the testing data".  But without seeing the actual data, it's impossible to really pass any good or bad judgement on it.

 

Since you are not a university user, I would recommend you reach out to your local Altair office, and request support.  Someone should be able to review your data in detail, and offer you suggestions.  Your questions are not very specific, so it's not really possible to give you a specific answer

 

FYI, we have a class "Vehicle Modeling with MotionView/MotionSolve".  You may benefit from attending that class.

Share this post


Link to post
Share on other sites
On 6/22/2020 at 9:06 PM, Chris Coker said:

The general answer to your question is "you don't need to do anything to the testing data".  But without seeing the actual data, it's impossible to really pass any good or bad judgement on it.

Well, I see. I get the point. Thanks.

Share this post


Link to post
Share on other sites
Sign in to follow this  

×
×
  • Create New...