Issue with IMPVEL applied to RBODY (no movement)
Hello,
I am having some issues with (I assume) IMPVEL applied to an RBODY in that there appears to be no motion after running the simulation. Everything in the model has previously worked fine when I had INIVEL applied to the master node of the RBODY, but when I switched over to IMPVEL (also applied to the master node), there is no movement.
I have tried increasing the velocity of the IMPVEL to see if maybe what I had originally defined it as was too small (and maybe the runtime was too short) but the results have still been unexpected... in one iteration of the model it was almost as if the RBODY passed through the parts in its path but the part that the RBODY was assigned to did not move (that is, only the RBODY element itself moved).
I have attached the 0000.rad and 0001.rad files for reference, as well as a screenshot of the definition of IMPVEL for quick reference (do I need to be defining a frame/skew (unlike with INIVEL which accepts not defining this)?)
At this point, I feel as though I have made enough small changes that I am not quite sure if one of those changes along the way is causing additional issues, so any help or suggestions would be much appreciated.
Thank you in advance for your time.
Best Answer
-
Your 'platen' RBODY isn't connected to your platen (it's secondary nodes are coincident with the platen nodes, but not the platen nodes), so the platen doesn't move when you move the RBODY
There are a lot of other issues in the model though I think
Your units also seem to be very odd? Everything seems to be 10s of orders of magnitude wrong to me?
e.g. at the velocity you have requested, it would pass through the whole part by a factor of 150 in the run time, within one animation cycle it would already have gone 2x the height of the lattice
Your density gives a Lattice Mass of 300Million Tonnes and Your material modulus is basically rigid?
What units are you trying to use?
0
Altair EmployeeAnswers
-
Your 'platen' RBODY isn't connected to your platen (it's secondary nodes are coincident with the platen nodes, but not the platen nodes), so the platen doesn't move when you move the RBODY
There are a lot of other issues in the model though I think
Your units also seem to be very odd? Everything seems to be 10s of orders of magnitude wrong to me?
e.g. at the velocity you have requested, it would pass through the whole part by a factor of 150 in the run time, within one animation cycle it would already have gone 2x the height of the lattice
Your density gives a Lattice Mass of 300Million Tonnes and Your material modulus is basically rigid?
What units are you trying to use?
0 -
Paul Sharp_21301 said:
Your 'platen' RBODY isn't connected to your platen (it's secondary nodes are coincident with the platen nodes, but not the platen nodes), so the platen doesn't move when you move the RBODY
There are a lot of other issues in the model though I think
Your units also seem to be very odd? Everything seems to be 10s of orders of magnitude wrong to me?
e.g. at the velocity you have requested, it would pass through the whole part by a factor of 150 in the run time, within one animation cycle it would already have gone 2x the height of the lattice
Your density gives a Lattice Mass of 300Million Tonnes and Your material modulus is basically rigid?
What units are you trying to use?
Actually 3000 Million Tonnes...
0 -
Paul Sharp_21301 said:
Actually 3000 Million Tonnes...
Hmmm I wonder why using INIVEL worked but switching to IMPVEL did not, even though this is all that changed - I'll take a look at the secondary nodes/platen nodes and make modifications there. Thank you for pointing this out.
Units are as they appear in the BEGIN_CARD - m, kg, s - though I am really working with everything on a mm scale.
The velocity is excessive in relation to everything as I was trying to figure out if it was just too small for me to see in the simulation (I really want it much smaller). Runtime is appropriate for the smaller velocity, I just never got around to changing it when I was playing around with the velocity magnitude.
I'm not sure where 3000 million tonnes comes from. If my material density is 1150 g/cm3, this gives 1150000 kg/m3, which is the value listed for the lattice density. I agree that the mass calc for the lattice gives a rather large number - 3.017 kg - for its size but I do not think it's millions of tonnes, unless I have really misunderstood the units.
I believe the material modulus for the lattice - 1400000000 - is in Pa unless I have misunderstood the units for this as well. 1.4 GPa for a material modulus is fine I think.
0 -
Autumn said:
Hmmm I wonder why using INIVEL worked but switching to IMPVEL did not, even though this is all that changed - I'll take a look at the secondary nodes/platen nodes and make modifications there. Thank you for pointing this out.
Units are as they appear in the BEGIN_CARD - m, kg, s - though I am really working with everything on a mm scale.
The velocity is excessive in relation to everything as I was trying to figure out if it was just too small for me to see in the simulation (I really want it much smaller). Runtime is appropriate for the smaller velocity, I just never got around to changing it when I was playing around with the velocity magnitude.
I'm not sure where 3000 million tonnes comes from. If my material density is 1150 g/cm3, this gives 1150000 kg/m3, which is the value listed for the lattice density. I agree that the mass calc for the lattice gives a rather large number - 3.017 kg - for its size but I do not think it's millions of tonnes, unless I have really misunderstood the units.
I believe the material modulus for the lattice - 1400000000 - is in Pa unless I have misunderstood the units for this as well. 1.4 GPa for a material modulus is fine I think.
Ah, ok, sorry, I think it is because your model has parameters, they aren't converted for me with units on import so everything is off, so yes ok, if you are in m/kg/s Your modulus would be ok because, you are using Pa,
What material are you using with a density of 1150 g/cm^3? Do you mean 1.15 g/cm^3
Steel is 7.85 g/cm^3 so that is nearly 150x the density of steel
INIVEL wouldn't have worked set on that node group either, have you got the inivel model for reference?
0 -
Paul Sharp_21301 said:
Ah, ok, sorry, I think it is because your model has parameters, they aren't converted for me with units on import so everything is off, so yes ok, if you are in m/kg/s Your modulus would be ok because, you are using Pa,
What material are you using with a density of 1150 g/cm^3? Do you mean 1.15 g/cm^3
Steel is 7.85 g/cm^3 so that is nearly 150x the density of steel
INIVEL wouldn't have worked set on that node group either, have you got the inivel model for reference?
Attached has RBODY corrected and runtime/outputs set to values more suitable for velocity
0 -
Paul Sharp_21301 said:
Attached has RBODY corrected and runtime/outputs set to values more suitable for velocity
I'm using a nylon-carbon fiber material - I went back and checked my notes for the density and it looks like somewhere along the way 1.15 g/cm3 turned into 1150 g/cm3 which then turned into 1150000 kg/m3 in the actual model... it's now been corrected - thanks for pointing this out!
I do have a "base" model that I copied to create this one, though I don't have/can't find the in-between model that had INIVEL but not IMPVEL, unfortunately.
Thank you for the .rad file updates - I did go back to modify my .hm model and checking the preliminary ANIM files from Radioss makes it seem like it's now working properly, even at the slower IMPVEL velocity that I was originally hoping for.
0 -
Autumn said:
I'm using a nylon-carbon fiber material - I went back and checked my notes for the density and it looks like somewhere along the way 1.15 g/cm3 turned into 1150 g/cm3 which then turned into 1150000 kg/m3 in the actual model... it's now been corrected - thanks for pointing this out!
I do have a "base" model that I copied to create this one, though I don't have/can't find the in-between model that had INIVEL but not IMPVEL, unfortunately.
Thank you for the .rad file updates - I did go back to modify my .hm model and checking the preliminary ANIM files from Radioss makes it seem like it's now working properly, even at the slower IMPVEL velocity that I was originally hoping for.
One other modification I would make, since you have element failure defined, you should set the flag 'idel' = 2 on each of your 3 contacts.
Eroded elements are not removed from contact without this and it will cause stability problems after element erosion if not set
You could also constrain the main node of the platen in the other 5 dof, currently is is free to move/rotate where it wishes outside of the loading direction
Good luck!
0