Assembly sudden steering issue

[BlazPlestenjak]

Hi,

For the past few months I've been trying to implement our Formula Student suspension into Adams Car FSAE template. Our formula car has so called decoupled suspension, so i had to make my own templates for front and rear suspension and ARB. My assemby behaves very weirdly and I cannot pin-point the problem.

For example:

I ran an accleration test. At the second increment simulation stops. To exaggerate the problem, I have increased the error number and number of steps. Result of the acceleration is shown on the pictures below. Car accelerates and the suddenly turns. Car has the tendency to steer (don't know if steering is the problem) into one direction.

When I tried to simulate hand free test, car drives foward with slowly turning to the right. I am attaching link to adm and msg files of the acceleration test I ran:

https://we.tl/t-UBG8wFFBiG

Has anyone seen anything like that before? I am really desperate to finish this project.

Please help!

 

[GregLocock]

Sorry, I can't get onto that download site, use the the engineering.com upload instead. However the disgraceful format of the /Car adms means I won't learn much.

So, in a straight line acceleration the steering controller will probably be set to 0 SWA, so your acf/xml files might be more useful.

SLA (if that is what you've run) is a weird event, it is a static equilibrium at each acceleration increment. I'd start with straight line drive. Anyway, I think you'll find the steering controller isn't correctly linked to the steering wheel.



Cheers

Greg Locock


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

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

 

[BlazPlestenjak]

Thank you for the answer.

So, I am using a steering template that is psrt of the FSAE templates that you can find on the internet. I have only changed hardpoints, so I don't think there should be any issues regarding the steering? Or?

I have ran a Straight-Line Drive instead. The result was basically the same, only that the wheels were turned in different direction. I have noticed that the wheels behave very weirdly and tend to loose contact with the ground. Could the problem be here? (picture below)

 

[BlazPlestenjak]

I am linking the acf file, as well as the video of the simulation

 

[GregLocock]

from the msg file

Degree-of-freedom analysis identified a NEW set of redundant constraints:
-------------------------------------------------------------------------
- deactivating constraint equation Zi.Yj in model.testrig.gyro_body_parallel_axis
- deactivating constraint equation Zi.Xj in model.testrig.gyro_body_parallel_axis

So, it is removing some of your model because it is overconstrained. I'd have thought there were warnings about this in whatever course you did to learn /Car.


Cheers

Greg Locock


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

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

 

[BlazPlestenjak]

Hi Greg,

I've had seen that the Adams is unable to satisfy the constraint for: SphJnt/30. However I was hoping that there is some additional info in the log files that implies error in the steering / suspension. SphJnt 30 is actually a balljoint between steeringrod and upright (picture below), which is realistic.

Can you tell me, which constraint should be between steeringrod and steering rack? Currently i have hooke. Could this be a problem?

Reagrds

 

[GregLocock]

Hookes is fine. it is very hard to tell why it chooses to suppress any particular DoF.

So if you run your model with the standard steering subsystem is it OK?

Cheers

Greg Locock


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

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

 

[BlazPlestenjak]

So for the last few days I have been trying some things.

1. If I use standard steering system from the ADAMS template I still run into the same issue
2. For powertrain, tires and chassis I am using the default FSAE templates as well
3. Inputs and outputs are all connected as they should be
4. I have run Suspension Parameter Measurement Machine and test on Vertical motion, Roll motion and steering motion. It all works and looks fine.

So basically there should be an issue when the car drives as opposed to Static tests of the suspension. Is it possible that constrains in suspensions are the problem even though that the static tests run? In kinematic test (when I encounter error, such as discussed above), both, Front and Rear axle have tires looking out of place.

I am really in a dead end here. Is there some other help possible from ADAMS?

 

[GregLocock]

Sure, there is a large and active community at the Octagon/MSC site

So from 1 and 2 if you use the standard FSAE model you get the same problem?

The usual approach in ADAMS which would be mentioned by any serious introduction is to start from something that works, then substitute new subassemblies in until something breaks.

Incidentally, personally I think using ADAMS in FSAE is going to use up too much of the student's time for very little gain compared with simpler tools, as the real learning curve for ADAMS is 6 months full time.

Cheers

Greg Locock


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

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

 

[Vojtech Brychta]

If the model runs without control then I assume it runs with "locked" steering as well. It might indicate just an issue with the steering control. Have you checked the gain on the .testrig.pvs_steering_ratio or the steering rack's reduction gear? Maybe just an added negative multiplier will do the trick.