Full Body MBD of a FSAE car with Electric powertrain.
I am trying to do a full-body MBD analysis of an FSAE car with an Electric powertrain FMU system imported from the commercial car model. I was successfully able to run the car but not how it's supposed to move. watch this video to get what I'm saying.
Also, I would love to hear your suggestions on all the things I could extract from analyzing the full vehicle to improve the car.
Answers
-
Hi Shrikar,
When watching the video, the first thing I notice is that a steering component (likely the steering rack) falls below the vehicle. I would expect that there is a missing cylindrical joint between the rack and the vehicle body. I would start there and see if the steering performance improves.
In terms of event performance, you can create event-specific reports upon completion of any simulated event to review the model's performance.
Hope this helps!
Adam Reid
1 -
Adam Reid_21142 said:
Hi Shrikar,
When watching the video, the first thing I notice is that a steering component (likely the steering rack) falls below the vehicle. I would expect that there is a missing cylindrical joint between the rack and the vehicle body. I would start there and see if the steering performance improves.
In terms of event performance, you can create event-specific reports upon completion of any simulated event to review the model's performance.
Hope this helps!
Adam Reid
Well, I checked with the commercial car model and there are all the necessary joints defined in my model.
0 -
SHRIKAR WADADARE said:
Well, I checked with the commercial car model and there are all the necessary joints defined in my model.
Hi Shrikar,
As Adam mentioned the steering rack is not constrained properly and rack falls down and letting the wheels to steer freely. You can use topology viewer to understand the connections. Check the MotionSolve log for any redundant constraint related to steering rack being removed.
Regards,
S. Ganesh1 -
Ganesh Shanmugam said:
Hi Shrikar,
As Adam mentioned the steering rack is not constrained properly and rack falls down and letting the wheels to steer freely. You can use topology viewer to understand the connections. Check the MotionSolve log for any redundant constraint related to steering rack being removed.
Regards,
S. GaneshThis is the topology view. Where do you think the issue is?
0 -
SHRIKAR WADADARE said:
This is the topology view. Where do you think the issue is?
Hi Shrikar,
Just like what Ganesh mentioned, you will have to check the LOG file after the simulation completes to see if it is deactivating any constraints that it thinks are redundant. Also, the model you downloaded from an FSAE library might not be 100% correct.
My guess from the pictures shared is that the primitive joint 'Rack Dummy Orient Joint' between the Rack and SteeringDummyBody is not functioning as intended. I would expect the Rack and this dummy body to be fixed together, and then they attach to the 'Rack housing' via the translational joint shown.
Hope this helps!
Adam Reid
0 -
Adam Reid_21142 said:
Hi Shrikar,
Just like what Ganesh mentioned, you will have to check the LOG file after the simulation completes to see if it is deactivating any constraints that it thinks are redundant. Also, the model you downloaded from an FSAE library might not be 100% correct.
My guess from the pictures shared is that the primitive joint 'Rack Dummy Orient Joint' between the Rack and SteeringDummyBody is not functioning as intended. I would expect the Rack and this dummy body to be fixed together, and then they attach to the 'Rack housing' via the translational joint shown.
Hope this helps!
Adam Reid
Hi Shrikar,
The topology view is not complete looking at the available data, You may have to add a spherical joint between rack and SteeringDummyBody at the orientation primitive location, as a designer you should be aware of your vehicle configuration.
Regards,
S. Ganesh0 -
Ganesh Shanmugam said:
Hi Shrikar,
The topology view is not complete looking at the available data, You may have to add a spherical joint between rack and SteeringDummyBody at the orientation primitive location, as a designer you should be aware of your vehicle configuration.
Regards,
S. GaneshThanks. I think I have fixed the steering issue now by verifying all the joints but the car is still not completing the event.
On checking the output file it shows an error during integration.
ERROR: At time=9.214E+00 the integrator failed to proceed.
ERROR: Simulation failed due to error encountered in analysis!
Forced termination from API_SendOffCommandSimulate!
ERRMES:USER [336001]
<< ALTAIR DRIVER :: Error running TRANSIENT/TRANSIENT Simulation for Maneuver 1
** EXECUTION STOPS DUE TO ENDFLG IN USER SUB **
ERROR: Error encountered in processing xml input file!
0 -
SHRIKAR WADADARE said:
Thanks. I think I have fixed the steering issue now by verifying all the joints but the car is still not completing the event.
On checking the output file it shows an error during integration.
ERROR: At time=9.214E+00 the integrator failed to proceed.
ERROR: Simulation failed due to error encountered in analysis!
Forced termination from API_SendOffCommandSimulate!
ERRMES:USER [336001]
<< ALTAIR DRIVER :: Error running TRANSIENT/TRANSIENT Simulation for Maneuver 1
** EXECUTION STOPS DUE TO ENDFLG IN USER SUB **
ERROR: Error encountered in processing xml input file!
Hi Shrikar,
The error that appears in the command window and LOG file is due to the suspension explosion that happens at the 25 second mark.
The initial position of the vehicle shows a dramatic amount of vehicle toe:
Is this intentional? When entering the second gate, the left wheel is pointed directly ahead while the right wheel looks to be at a maximum steering angle. That type of suspension geometry will make steering events difficult. I would revisit that initial position to ensure the front suspension is constrained properly.
Hope this helps!
Adam Reid
0 -
Adam Reid_21142 said:
Hi Shrikar,
The error that appears in the command window and LOG file is due to the suspension explosion that happens at the 25 second mark.
The initial position of the vehicle shows a dramatic amount of vehicle toe:
Is this intentional? When entering the second gate, the left wheel is pointed directly ahead while the right wheel looks to be at a maximum steering angle. That type of suspension geometry will make steering events difficult. I would revisit that initial position to ensure the front suspension is constrained properly.
Hope this helps!
Adam Reid
No, the toe was not intentional. I rechecked the front suspension and steering but couldn't find anything so reimported the systems and gave coordinates again. looks like the issue got fixed but still the vehicle isn't able to complete the event.
Also, The main event that I want to analyze is the constant radius event with my custom radius and velocity inputs but I am not able to build the event itself.
Then when I tried to build this event I got the following error:
I thought this was due to the initial velocity which I had given as zero. So I adjusted that and now I am not able to either build the event or run the event. It does not even throw any error nothing happens when I run/build the event. What should I do?
0 -
SHRIKAR WADADARE said:
No, the toe was not intentional. I rechecked the front suspension and steering but couldn't find anything so reimported the systems and gave coordinates again. looks like the issue got fixed but still the vehicle isn't able to complete the event.
Also, The main event that I want to analyze is the constant radius event with my custom radius and velocity inputs but I am not able to build the event itself.
Then when I tried to build this event I got the following error:
I thought this was due to the initial velocity which I had given as zero. So I adjusted that and now I am not able to either build the event or run the event. It does not even throw any error nothing happens when I run/build the event. What should I do?
Hi Shrikar,
DLC
When I watch the video you posted, the vehicle appears to have completed the event, albeit slowly. What are you hoping to occur differently?
Constant Radius
It should have rejected the 0 velocity input and caused an error, but I'm not sure why you are no longer able to generate an event. I would recommend trying to build any other event and see if the issue persists. If it is just the constant radius event that is causing issues, your 0 velocity attempt may have effected something. Delete all events and then export the model as an archive (.zip) using file >> export >> model >> archive model + create zip file. This will hopefully create a 'fresh' model to work with, where you can try to make the event again.
Hope this helps!
Adam Reid
1 -
Adam Reid_21142 said:
Hi Shrikar,
DLC
When I watch the video you posted, the vehicle appears to have completed the event, albeit slowly. What are you hoping to occur differently?
Constant Radius
It should have rejected the 0 velocity input and caused an error, but I'm not sure why you are no longer able to generate an event. I would recommend trying to build any other event and see if the issue persists. If it is just the constant radius event that is causing issues, your 0 velocity attempt may have effected something. Delete all events and then export the model as an archive (.zip) using file >> export >> model >> archive model + create zip file. This will hopefully create a 'fresh' model to work with, where you can try to make the event again.
Hope this helps!
Adam Reid
Actually, in the video, it seems as though it has been completed but after the second lane change, the solver is failing showing the same integrator error. Now, when I tried deleting and reimporting the event again similar thing is happening that is the car is turning but not able to complete the event. In the DLC at least, it seemed that it was going off track so it was giving errors but in a constant radius, it is not able to complete despite being on track. Are these errors due to simulation settings?
0 -
SHRIKAR WADADARE said:
Actually, in the video, it seems as though it has been completed but after the second lane change, the solver is failing showing the same integrator error. Now, when I tried deleting and reimporting the event again similar thing is happening that is the car is turning but not able to complete the event. In the DLC at least, it seemed that it was going off track so it was giving errors but in a constant radius, it is not able to complete despite being on track. Are these errors due to simulation settings?
Hi Shrikar,
For a model of this complexity, I would suggest using the default settings for the solver. The DLC and Constant Radius events, unless attempted at extreme speeds, should not be computationally intense for the solver.
To find out more about what is happening before simulation failure, you can enable 'Write Debug Info' in the Output Settings in the Run panel. There may be a joint failing to converge due to kinematic issues in your model, and this debug info will help to highlight a location in your model that is carrying the highest amount of residual error.
I would always recommend trying these events with a default model first to confirm that your issue is localized to this custom model you are attempting to use and not to you entire installation.
Hope this helps!
Adam Reid
0
Altair Employee