a cylinder as a single body

Hi,

I need to create(or import) a cylindrical-shape body and as I see in any case this cylinder is presented by an assembly of facets which means several bodies.

However, what I want to do is measuring the viscosity of a material. It means we apply a moment on a cylinder surrounded by spheres and we measure the shear rate (angular velocity).
So, the problem is how can we treat the cylinder as a single rigid body?

Thanks .

Question information

Language:
English Edit question
Status:
Solved
For:
Assignee:
No assignee Edit question
Solved by:
Solved:
2015-02-01
Last query:
2015-02-01
2015-02-01
 Jan Stránský (honzik) said on 2015-02-01: #1

you have more options:

1) use O.bodies.appedClumped(facets) to create a clump, which you can aplly
force to. Currently it does not work with facets, but it should not be
difficult to make it work with non-spherical shapes. I don't have time now
to work on it..

2) use normal bodies and use RotationEngine on them, each (or each a few)
time step checking forces on individual facets and apdate angular velocity
according to the value (i.e. creating PyRunner doing the same as
NewtonIntegrator :-)

I like no 1 most (if it works), you can ask some other developers for the
help :-)

cheers
Jan

> New question #261529 on Yade:
>
>
> Hi,
>
> I need to create(or import) a cylindrical-shape body and as I see in any
> case this cylinder is presented by an assembly of facets which means
> several bodies.
>
> However, what I want to do is measuring the viscosity of a material. It
> means we apply a moment on a cylinder surrounded by spheres and we measure
> the shear rate (angular velocity).
> So, the problem is how can we treat the cylinder as a single rigid body?
>
> Thanks .
>
> --
> You received this question notification because you are a member of
>
> _______________________________________________
> Post to : <email address hidden>
>

Hey Jan,

Yeah, no.1 seems a good way to work on. The facet bodies, I think must be set as dynamic bodies, though.

Anyway, thanks for the answer. I'll work on it.

cheers,

 Jan Stránský (honzik) said on 2015-02-01: #3

I am not sure where the problem is, after appendClumped, the values of
clump.members (i.e. relative position and orientation of clump members are
NaNs..). Setting mass and inertia to facets does not help

Probably nobody needed such feature before, but definitely it may be
considered as a bug, feel free to report it, at least we would not forget
to fix it :-)

cheers
Jan

> Question #261529 on Yade changed:
>
>
> behzad confirmed that the question is solved:
> Hey Jan,
>
> Yeah, no.1 seems a good way to work on. The facet bodies, I think must
> be set as dynamic bodies, though.
>
> Anyway, thanks for the answer. I'll work on it.
>
> cheers,
>
> --
> You received this question notification because you are a member of
>
> _______________________________________________
> Post to : <email address hidden>
>

How can I find appendClumped function to modify it?

 Jan Stránský (honzik) said on 2015-02-01: #5

pyBodyContainer::appendClump
calling pyBodyContainer::clump
calling Clump::add and Clump::updateProperties (see core/Clump.*pp). I
guess the problem is in the Clump methods..

Jan

> Question #261529 on Yade changed:
>
> behzad posted a new comment:
> How can I find appendClumped function to modify it?
>
> --
> You received this question notification because you are a member of
>
> _______________________________________________
> Post to : <email address hidden>
>

 Christian Jakob (jakob-ifgt) said on 2015-02-02: #6

Hi,

updateProperties function assumes, that clump members are spheres. it will not work for facets atm ...

regards,

christian

 Jan Stránský (honzik) said on 2015-02-02: #7

@Christian: Ok, I see.. the correction for non-spherical particles should
not be too difficult

@Behzad: how much is it urgent for you?

Jan

2015-02-02 7:31 GMT+01:00 Christian Jakob <

> Question #261529 on Yade changed:
>
> Christian Jakob posted a new comment:
> Hi,
>
> updateProperties function assumes, that clump members are spheres. it
> will not work for facets atm ...
>
> regards,
>
> christian
>
> --
> You received this question notification because you are a member of
>
> _______________________________________________
> Post to : <email address hidden>
>

I need to have the modifications(or another way to deal with this problem) in two weeks.

 Jan Stránský (honzik) said on 2015-02-03: #9

Ok, I will have a look during the weekend, is it ok?
Jan

> Question #261529 on Yade changed:
>
> behzad posted a new comment:
>
> I need to have the modifications(or another way to deal with this problem)
> in two weeks.
>
> --
> You received this question notification because you are a member of
>
> _______________________________________________
> Post to : <email address hidden>
>

Sure! Thanks.

 Jan Stránský (honzik) said on 2015-02-07: #11

the non-spherical clumps should be possible to use in latest trunk version
(r3576). I added an example, in principle doing exactly what you want,
cheers
Jan

> Question #261529 on Yade changed:
>
> behzad posted a new comment:
> Sure! Thanks.
>
> --
> You received this question notification because you are a member of
>
> _______________________________________________
> Post to : <email address hidden>
>

Hi Jan,

yes, I tried it and it does what I wanted. Thank you!

cheers,

Hey Jan,

Can I share some files with you?
I'm importing two bodies as stl files in YADE and I define them as clumps. They're having unrealistic interactions.

 Jan Stránský (honzik) said on 2015-02-11: #14

sure, you can also (preferably) place the files somewhere on the web such
cheers
Jan

> Question #261529 on Yade changed:
>
> behzad posted a new comment:
> Hey Jan,
>
> Can I share some files with you?
> I'm importing two bodies as stl files in YADE and I define them as clumps.
> They're having unrealistic interactions.
>
> --
> You received this question notification because you are a member of
>
> _______________________________________________
> Post to : <email address hidden>