Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

Deactivate Motion - Motionsolve

Hi all,

i have a motion which needs to be deactivate at time=16.1221. it is a very simple model but as soon as i deactivate the motion i can see a velocity drop which should not be there. i can also see the velocity jump graphically in the animation

all other elements are deactivated for debugging (force, sensor, etc). Joints have no friction, stiffness or anything.

i would have expected to see a constant velocity due to inertia, of course gravity is acting against the rotation…

any ideas what the reason could be??

thanks for your help - Michael

Find more posts tagged with

MotionView/MotionSolve

Templates

Accepted answers

GTT Adam

@Orestis_YB, @Christopher_Fadanelli, @Michael_V8,

For additional context, here is the animation in question:

Table rotates starting around 13 seconds via a Motion defined by a rotational acceleration curve:

The Motion that drives this rotation is deactivated via a template at 23.7867 seconds.

If we refer to @Michael_V8's curve values, we can see that at that exact time step the table is to be accelerating:

As noted, the rotational velocity shows a reaction at the time of deactivation:

It is important to look at the torque the Motion is applying over this period, as when it is deactivated it is not applying 0 Nmm, but nearly -1E+7 Nmm:

If we look at the exact instant the force is removed, we can see that (as expected) these instances align.

What is interesting to note is where this deactivation occurs:

The red turn table is travelling counter-clockwise at this instant. As you can see, the inclination of the table will cause the table to accelerate due to the position of its COG, thus increasing the velocity it reaches its lowest point. So this 'blip' in rotational velocity when the Template turns off the Motion is likely because the table is just accelerating under gravity. The Motion was constraining it to rotate at a set speed, but with it off it is allowed to accelerate.

Hope this helps!

Adam Reid

All comments

Orestis_YB

Hello @Michael_V8,

unfortunately there is no much information to understand the reason behind this. If no forces (or constrains) are applied to the body then after deactivating the motion it should continue with constant speed. I would suggest to first check the log file of your simulation.
1) The model should not be over constrained. If it is, then MotionSolve will try to remove some constrains and might affect the behavior of your model.
2) Double check that the motion is indeed deactivated.
Are there maybe other entities in the model that could affect the dynamics of the part that you are monitoring?

Best regards,
Orestes

Michael_V8

hi, thanks for your quick answer.

as said, the model is very simpel. i have discovered, that, when switching the rotational joint to "virtual" than it works without problems. No idea whats wrong with the "normal" joint…

Christopher_Fadanelli

Hello @Michael_V8,

What is the goal you are trying to achieve that the motion needs to be deactivated? Would it be possible to have it as a force expression. For example until 16.1221 seconds you have a non zero value for your force. Once it hits 16.1221, the force becomes 0?

Best,

Christopher Fadanelli

Michael_V8

i have a motion (rotation) as acceleration provided by the customer. after the time 16... i need to switch it of and have to apply a counter force (moment, with a step function starting at 16….) to decelerate the system. i need to know when the system stops

Christopher_Fadanelli

Hi @Michael_V8,

Would it be possible to share more information about this model (potentially share the model). It is unclear to me what the entire end goal and model configuration is? If this is confidential information. I would suggest to create a support case and one of us can help you out!

Best,

Christopher Fadanelli

GTT Adam

Hi @Michael_V8,

There is a unique condition where if the simulation is run Online, and the Template runtime is shorter than the Analysis runtime, the simulation technically restarts. Please see the animations below for results.

In short, this is only an issue if the model is simulated Online. Please run the simulation Offline and either method will be fine.

In this example, the Template turns off the Motion at 1 second. The second Transient simulation is only set to 2 seconds. The Analysis end time is set to 5 seconds. When run Online, the simulation jolts after 2 seconds (Red Plate).

In the second example, the second Transient simulation is set to 5 seconds. The analysis end time is also set to 5 seconds. When run Online, no jolt. (Green Plate)

Alternatively, if both are run Offline, there is no issue.

Hope this helps!

Adam Reid

Michael_V8

So… i have tried the online/offline with different end times. the behavior is still the same and still not as expected. Let me try to be a little more specific.. (different load case, but similar behavior)

Description:

The car moves tot the (red) turning table, which rotates to the exit track

Analysis end time in this load case is 28s

There is one rotational motion (acceleration) on the turning table

Curve: at the green time i stop the motion with the template

When the motion is inactive, i would expect the table continues to rotate with a constant velocity

but when checking the velocity (blue) and the acceleration (red) i can see a jump which should not be there. This is exactly the time when i switch of the motion with the template

Michael_V8

… the force entity was temporarily deactivated for this run

Orestis_YB

Hello @Michael_V8,

You said that the cart moves towards the red part of the track. I believe that the change you see in the velocity and acceleration may be due to the change in the moment of inertia of your system towards the center of rotation. You should see indeed constant velocity only if the inertia remains constant as well.

I've attached a simple model of a rotation cylinder at the end of which there is a sphere with a translational joint. A motion is defined at the revolute joint of type Acceleration to increase the acceleration from 0-10 rad/s^2 in 0.5 sec. In t=2s motion is deactivated (there might be some noise in the result depending always on your model set up and solver settings). There we see the velocity remaining constant. At t=4 sec we start a linear motion of the ball towards the center. We see the impact of the in the acceleration and velocity due to the change in the moment of inertia f the system. At t=7s ball stabilizes again and the rest of the system as well.

Hope this helps!

Best regards,
Orestes

Michael_V8

thanks for your feedback. i will check is. regarding your "guess". the cart is already sitting on the table, there is no change in the inertia. the motion looks like this:

0-5sec: cart moves to the table
5-20sec: table with cart rotates down
20-28sec: table with cart rotates up → motion deactivates at 23sec

Orestis_YB

Hello @Michael_V8,

I suggest to share the model then so we could check the model set up. To properly share a MotionView model go to File→Export→Model and then check the archive and zip options. In case you don't want to share the model publicly in the community page I suggest to create a support case!

Best reagrds,
Orestes.

Michael_V8

uploaded, maybe you can have a look at Case CS0705429

GTT Adam

@Orestis_YB, @Christopher_Fadanelli, @Michael_V8,

For additional context, here is the animation in question:

Table rotates starting around 13 seconds via a Motion defined by a rotational acceleration curve:

The Motion that drives this rotation is deactivated via a template at 23.7867 seconds.

If we refer to @Michael_V8's curve values, we can see that at that exact time step the table is to be accelerating:

As noted, the rotational velocity shows a reaction at the time of deactivation:

It is important to look at the torque the Motion is applying over this period, as when it is deactivated it is not applying 0 Nmm, but nearly -1E+7 Nmm:

If we look at the exact instant the force is removed, we can see that (as expected) these instances align.

What is interesting to note is where this deactivation occurs:

Hope this helps!

Adam Reid

Michael_V8

Hi Adam,

Thanks for your explanation. Its true that there is a additional torque due to gravity load. But unfortunately o cannot folly agree with your explanation. If i apply a force or take it away, i should not have a change in the acceleration anymore (newton...) but as you can see on my output (velocity and accel. above) you can see zhat there is infinate steep jump. If gravity accelerates the table, than there should be a rather smooth transition phase until the counter moment applys..

Michael

Michael_V8

https://community.altair.com/discussion/comment/189556#Comment_189556

🤣🙈i have to correct myself. Of course, when I apply a force, I accelerate my system. When i take the force away, then i don't accelerate it anymore. What I have ment is that I don't have such a step either way. .. its too late

Michael_V8

Hi Adam,

we have discussed it internally and did a very deep review. Questioning everything... It seems that you are right. The the acceleration is given by curve. When i switch it off, gravity overtakes and creates the jump.

GTT Adam

Hi @Michael_V8,

That's great to hear.

I hope the remaindering development of your model goes well.

Thanks,

Adam Reid