multiplying a rotation between two shafts

[jannikfea]

Hello,

I am confronting a problem where I'd like to multiply a rotation of a shaft so that one half of the shaft is rotating faster than the other end.

I tried this with the flow converter connector elements as well as with linear equations using the mpc constraints. With both I manage to convert/ multiply the rotation, but the moments are mulitplied as well. (Since I think its handled like a regular transmission). Is there a way to eliminate the magnification of the moment, so that just the rotation is magnified? Maybe other than flow converters or mpc linear equations. Thank you very much!

 

[FEA way]

Could you share a picture of your model ? Which type of elements do you use (solids/shells/beams) ? How is it done in real life ?

*Equation seems like the best option here. Apart from using it or certain type of connector it will be difficult to achieve similar effect. You can manipulate nodal rotational DOFs but this influences moments as well.

 

[jannikfea]

thanks for your answer!

i'm using beam elements. In real life, there is no such effect. Its just one shaft that is rotating (in a gearbox). SInce the transmission is very high, the second shaft is turning very slow, in the first step its is lowered by the factor of 10. To accelerate the simulation, i just want to accelerate the rotation of the second shaft by cutting it in two parts, where the first one is doing one rotation and the second one is doing e.g. 4. So far i am achieving this goal, but the moments are multiplied as well, which generates different loads in the bearings. I know that in reality,there is no such thing, i just want to accelerate the simulation so achieve a certain situation in the gearbox after a shorter amount of time while keeping the forces and moments constant. I hope you can understand my problem.

 

[FEA way]

I would rather use different methods to reduce the time required to complete the simulation in this case. The choice of these methods depends on the type of analysis (which type of step is it ?) but also other factors such as elements and features used in simulation (do you have any other elements than beams ?). For example you can increase the time increment or reduce the number of requested output variables as well as decrease the frequency of output.

 

[jannikfea]

since it's a static simulation step where i'm simulating a fully functional transmission in a 2 step gearbox, its rather difficult to increase the increment time. Because of that, increasing the speed by accelerating to a situation, that would happen anyway, seems like the way to go for me. The Output requests are already limited to the absolute minimum.

Its hard for me to accept, that there really is no way of excluding the magnification of moments while magnifying the rotation. Nearly anything is possible in abaqus, why would that not?

 

[jannikfea]

like you can imagine for a fully fuctional gearbox, i'm using multiple element types (Shells, Element Surfaces, Solids, Connectors, Constraint etc.)!