Steering wheel force/torque values will be zero in function-expression references

[BlazPlestenjak]

Hi,

I am currently working on a full vehicle analysis. Kinematics and compliance simulation works fine (steering turns as it should). But when i run any of the dynamic events i ge t the same error:

fdm::ActJoiMot
Setting function on: jms_steering_wheel_angle(steering::steering_wheel_angle)
command: MOTION/4, FUNCTION=0.0
command: SFORCE/1, FUNCTION=0.0


fdm::ActVar
Connecting: jms_steering_rack_travel(steering::rack_travel)
To : vas_steering_demand(driver_demand::steering)


fdm::ActVar
Setting function on: vas_steering_demand(driver_demand::steering)
command: VARIABLE/92, FUNCTION=DZ(428,440,440)
command: DEACTIVATE/SFORCE, ID=2
Deactivated model.testrig.jfs_steering_wheel_torque.force, force/torque values will be zero in function-expression references.
command: DEACTIVATE/MOTION, ID=3
Deactivated model.testrig.jms_steering_rack_travel.motion, force/torque values will be zero in function-expression references.
command: DEACTIVATE/MOTION, ID=4
Deactivated model.testrig.jms_steering_wheel_angle.motion, force/torque values will be zero in function-expression references.


fdm::ActJoiFor
Setting function on: jfs_steering_rack_force(steering::rack_force)
command: SFORCE/4, FUNCTION=DIF(4)
command: SFORCE/5, FUNCTION=-(DIF(4))


fdm::ActVar
Setting function on: vas_throttle_demand(driver_demand::throttle)
command: VARIABLE/94, FUNCTION=USER(976,3)\ ROUTINE=abgVDM::var976


fdm::ActVar
Setting function on: vas_brake_demand(driver_demand::brake)
command: VARIABLE/96, FUNCTION=USER(975,3)\ ROUTINE=abgVDM::var975


fdm::ActVar
Setting function on: vas_gear_demand(driver_demand::gear)
command: VARIABLE/98, FUNCTION=(4.0)


fdm::ActVar
Setting function on: vas_clutch_demand(driver_demand::clutch)
command: VARIABLE/100, FUNCTION=0.0


vdm::SteadyState::MiniInfo

Static Task : 'Straight'
Lon. Acceleration : 0.000 [m/s^2]
Initial Velocity : 11.111 [m/s]
Perform Linear : No
Include Damping : Yes
Halt on Failure : Yes



fdm::ActVar
Setting function on: vas_clutch_demand(driver_demand::clutch)
command: VARIABLE/100, FUNCTION=1.0
command: ACTIVATE/MOTION,ID=4
command: DEACTIVATE/SFORCE,ID=1
Deactivated model.testrig.sse_steering_force, force/torque values will be zero in function-expression references.
command: SIM/STAT


Begin Static Solution

TIRE ID: 2
TYR900 -> Pacejka 89 Tire Model
Using 2D Road
One point Contact

TIRE ID: 4
TYR900 -> Pacejka 89 Tire Model
Using 2D Road
One point Contact

TIRE ID: 6
TYR900 -> Pacejka 89 Tire Model
Using 2D Road
One point Contact

TIRE ID: 8
TYR900 -> Pacejka 89 Tire Model
Using 2D Road
One point Contact


STATICS: Original static solver will be used with the Sparse linear solver.

Static Solution converged in 4 iterations
Residual imbalance is less than 1.45761E-06

End Static Solution

Simulate status=0

command: DEACTIVATE/JPRIM,ID=2
Deactivated model.testrig.josinl_body_stake, force/torque values will be zero in function-expression references.
command: DEACTIVATE/JPRIM,ID=3
Deactivated model.testrig.josper_yaw_stake, force/torque values will be zero in function-expression references.
command: DEACTIVATE/JPRIM,ID=6
Deactivated model.Front_Tires_rev1.jolper_sse_jprim, force/torque values will be zero in function-expression references.
command: DEACTIVATE/JPRIM,ID=7
Deactivated model.Front_Tires_rev1.jorper_sse_jprim, force/torque values will be zero in function-expression references.
command: DEACTIVATE/JPRIM,ID=8
Deactivated model.Rear_Tires_rev1.jolper_sse_jprim, force/torque values will be zero in function-expression references.
command: DEACTIVATE/JPRIM,ID=9
Deactivated model.Rear_Tires_rev1.jorper_sse_jprim, force/torque values will be zero in function-expression references.


fdm::ActVar
Setting function on: vas_clutch_demand(driver_demand::clutch)
command: VARIABLE/100, FUNCTION=0.0
command: ACTIVATE/JPRIM,ID=2
command: ACTIVATE/SFORCE,ID=1
command: ACTIVATE/JPRIM,ID=6
command: ACTIVATE/JPRIM,ID=7
command: ACTIVATE/JPRIM,ID=8
command: ACTIVATE/JPRIM,ID=9
command: DEACTIVATE/MOTION,ID=4
Deactivated model.testrig.jms_steering_wheel_angle.motion, force/torque values will be zero in function-expression references.

acar/vdm_gra_mod

Setting gravity vector in g's to: 0.00000E+00 0.00000E+00 -1.00000E+00
command: accgrav/igrav= 0.00000E+00,jgrav= 0.00000E+00,kgrav=-9.80665E+03

Apply a two steps statics:


Begin Static Solution


STATICS: Original static solver will be used with the Sparse linear solver.



---- START: ERROR ----

Static equilibrium analysis has not been successful.
The maximum number of iterations 25, has been reached.
For the last attempted iteration:
The equation with the largest error was model.Powertrain_rev1.powertrain_gse Comp = 0 Error = 354750.951720
The variable receiving the largest increment was model.Powertrain_rev1.gel_diff_output Comp = 10 Delta = 12990386845.840965.
---- END: ERROR ----

Static Solution failed to converge

End Static Solution

Simulation failed



---- START: ERROR ----
Sim/stat cannot be achieved ...Stopping in <apre_ssssta>
ID = 1020

---- END: ERROR ----
----------------------------------------------------------------------------------------------------------------------

I have tried to use diffrent solvers, but with no success. With GSTIFF Solver I got the vehicle to move a little bit in acceleration event, but then it suddenly turned and event stooped. I do not know where the source of the problem is coming from.

Please help!

 

[GregLocock]

So the last bit is the important bit.

---- START: ERROR ----

Static equilibrium analysis has not been successful.
The maximum number of iterations 25, has been reached.

The run fails to start because the new config with the real rack is not in equilibrium. I gather you are using /Car. So, your approach is either start from a model that will complete the event in question, and substitute subs until it breaks, or try and brute force the static equilibrium. To do this you need to change Solver settings. Typically this involves changing MAXIT, error and various other settings, rather at random I am afraid. Unfortunately /Car seems to have been written to appeal to managers and salespeople and not engineers and debugging the models is not easy. You may find that switching debug/eprint on is sufficient to get the model to run, showing how poorly stable the whole pile of code is. debug/eprint may also give you better clues as to what is going wrong, but usually doesn't.



Cheers

Greg Locock


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

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