hexagon logo
Register Sign in
  • State Not Answered
  • Replies 8 replies
  • Subscribers 11 subscribers
  • Views 171 views
  • Users 0 members are here
Options
Related

GCON unable to solve function that include model parameters as a variable. Suggestions to run two coupled shafts with variable speed ratio?

ashok
Hi, In my model(attached) I am trying to drive a pump shaft(slave) with respect to the motor shaft(master) based on a variable efficiencies. I have coupled the two shafts using GCON.
 
The efficiency is calculated using the output torque and the input speed. The coupler(GCON) seems to be not working when the model parameters are used to calculate the ratio at which the pump shaft should spin. It works fine with some arbitrary generated variable that dose not involve model parameters.
 
Any suggestions to run two coupled shafts with variable speed ratio?
Attached Files (1)
Parents
  • 0
    I assume that what you are trying to model is some sort of belt drive? Or CVT?
    Anyway, there could be a possibility to use a good control element to do this. Is reaction forces on the driving axle important to you? Or will that be motion driven?
    The problem is that forces tends to be a bit noisy, so good solver tuning will be important. I will see if I can get something together.
     
    But I think there are a couple of logical mistakes in your setup.
    First, you have the GCON defined as
    AZ(master.MAR_3, ground.MAR_1)+AZ(slave.MAR_3, MAR_2_reaction)*VARVAL(Calc_Vol_eff)
    which means that the output shaft will rotate in opposite direction and faster than the input shaft as the Calc_Vol_eff < 1.
    You see this clearly when rewriting the rewriting the equation as
    WZ(slave) = -WZ(master)/Calc_Vol_eff
     
    The other mistake is to drive the input with a torque and have a torsion spring on the output. The output spring SILL wind-up until the shaft stops, no matter what, and the reaction torque on the driving shaft WILL stop that as well. It becomes a very hard system to control.
     
Reply
  • 0
    I assume that what you are trying to model is some sort of belt drive? Or CVT?
    Anyway, there could be a possibility to use a good control element to do this. Is reaction forces on the driving axle important to you? Or will that be motion driven?
    The problem is that forces tends to be a bit noisy, so good solver tuning will be important. I will see if I can get something together.
     
    But I think there are a couple of logical mistakes in your setup.
    First, you have the GCON defined as
    AZ(master.MAR_3, ground.MAR_1)+AZ(slave.MAR_3, MAR_2_reaction)*VARVAL(Calc_Vol_eff)
    which means that the output shaft will rotate in opposite direction and faster than the input shaft as the Calc_Vol_eff < 1.
    You see this clearly when rewriting the rewriting the equation as
    WZ(slave) = -WZ(master)/Calc_Vol_eff
     
    The other mistake is to drive the input with a torque and have a torsion spring on the output. The output spring SILL wind-up until the shaft stops, no matter what, and the reaction torque on the driving shaft WILL stop that as well. It becomes a very hard system to control.
     
Children
No Data
© 2025 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement