Hi Brent
I do not think you should apply the AnyForce3D in the two nodes, the
AnyForce should be used instead.
The AnyForce3D is used for adding a force in a node as a vector; they can
not be used for the ligaments.
The best approach will be to use the AnyForce and apply it on the
AnyKinPLine, the AnyForce3D can not be used for this. This will
automatically take care of the directions of the forces, and make sure that
the forces are applied in both end nodes. You can have as many via nodes as
needed and some force will also be exerted on these nodes if the tendon path
has a “kink” at the via point node, if the line is straight no force is
exerted, so the via points works as “needle eyes”.
You can visualize the tendon by using the AnyDrawPLine object.
Best regards
Søren AnyBody Support
From: anyscript@yahoogroups.com [mailto:anyscript@yahoogroups.com] On Behalf
Of blulrey
Sent: 22. august 2006 00:33
To: anyscript@yahoogroups.com
Subject: [anyscript] Re: Standing Model: Rib Cage and Ligaments
Hello Søren,
I finally got around to trying this (it was my turn for a vacation).
If I used AnyForce3D, I would have to apply equal and opposite forces
at the origin and insertion nodes in order to model a ligament force.
Will this method apply a force to both nodes automatically? What if
I use a via node? Will the loads be applied to only the origin and
insertion nodes in the correct directions? And how do I determine the
direction the force that the force is being applied? I find this hard
to visualize because AnyDrawVector will not draw the generalized force.
Thank you,
Brent
— In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“AnyBody Support” <support@…> wrote:
>
> Hi Brent
>
> I just relized I made a small error in the suggested script in the
> last message. I forgot to apply the force to the kinematic measure
> This could look like:
>
> //kinematic model of the ligament
> AnyKinPLine Lig={
>
> AnyRefNode &ref1=origin node
> AnyRefNode &ref2=insertation node
>
> };
> //force model of the ligament
> AnyForce ligforce={
> AnyKinPline=.Lig; //apply the force function to the geometrical lig.
> model
> F=t*.Lig.Pos; //very simple time and kinematic dependency
> };
>
> Søren
>
> — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“AnyBody Support” <support@>
> wrote:
> >
> > Hi Brent
> >
> >
> > No what i mean is that you have access to the time in the
> > AnyLigamentModelPol, but it is not allowed to apply it on the
> > constanst like f1 etc.
> >
> > You can can create an AnyForce which is time dependent and
> kinematic
> > dependent as you have done, and then apply this force on a
> > AnyKinPLine which is then the kinematic model of the ligament. In
> > this way you have created your own ligament using only AnyForce
> and
> > AnyKinPLine. I Think this will solve your problem.
> >
> >
> > It could look like this:
> >
> > //kinematic model of the ligament
> > AnyKinPLine Lig={
> >
> > AnyRefNode &ref1=origin node
> > AnyRefNode &ref2=insertation node
> >
> > };
> > //force model of the ligament
> > AnyForce ligforce={
> > F=t*.Lig.Pos; //very simple time and kinematic dependency
> > };
> >
> >
> >
> > Best regards
> > Søren, AnyBody Support
> >
> >
> > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com> ps.com,
“blulrey” <blulrey@> wrote:
> > >
> > > Søren,
> > > I was attempting to create an AnyLigamentModelPol with a
> > > time-dependent nominal strain by multipling by ‘t’. This was
> > > resulting in errors. I tried time-dependent F1 and L0 values as
> > well,
> > > and these also gave errors.
> > >
> > > I have created a seperate AnyForce3D that is time-dependent.
> You
> > are
> > > saying
> > > that I can still use AnyLigamentModelPol with a time-dependent
> > force?
> > > This would be good because then I could use via points to
> > construct a
> > > ligament. It seems that whenever I include ‘t’ in an
> > > AnyLigamentModelPol it causes errors.
> > >
> > > Brent
> > >
> > >
> > > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com, “AnyBody Support” <support@>
> > wrote:
> > > >
> > > > Hi Brent
> > > >
> > > > The time should be available both in the geometry model
> > > > AnyViaPointLigament and in the force model
> AnyLigamentModelPol,
> > i am
> > > > a bit unsure which one of those you refer to.
> > > >
> > > > You should be able to write expressions in both objects making
> > use
> > > > of the time, but it is not possible to initial for example F1
> in
> > the
> > > > AnyLigamentModelPol object using an expression involving the
> > time,
> > > > maybee this is your problem?. The variables in the
> LigamentModel
> > are
> > > > static values.
> > > >
> > > > If you would like the force in the ligament to be dependented
> on
> > the
> > > > time and some kinematics you can do this by creating an
> > expression
> > > > for F in a AnyForce object which is timedependent and
> kinematic
> > > > dependent.
> > > >
> > > > I hope this made it more clear otherwise please write again.
> > > >
> > > > Best regards
> > > > Søren, AnyBody Support
> > > >
> > > > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com, “blulrey” <blulrey@> wrote:
> > > > >
> > > > > Hello Michael,
> > > > > Thank you for the explanation of the time variable.
> > > > >
> > > > > Am I correct to state that time is not available in a
> ligament
> > > > object?
> > > > >
> > > > > Brent
> > > > >
> > > > > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com, “AnyBody Support”
> <support@>
> > > > wrote:
> > > > > >
> > > > > > Hi Brent,
> > > > > >
> > > > > > It’s Michael. I will give you an answer on the ‘t’ issue.
> > > > > >
> > > > > > Time ‘t’ is not global variable you can access everywhere.
> > In
> > > > > > fact ‘t’ in a local member of objects, just like most
> other
> > > > > > variables in the model. The reason why it may look like a
> > global
> > > > > > variable (I understand your confusion) is that a member
> > > > called ‘t’
> > > > > > exist in a lot of objects, in fact in all mechanical
> objects
> > > > > > (AnyMechObject-derived). Therefore it seems like you can
> get
> > > > access
> > > > > > to ‘t’ in most places (but not all. Main for instance does
> > not
> > > > > > have ‘t’ predefined).
> > > > > > Actually you do not access the same ‘t’ from evertywhere
> if
> > you
> > > > just
> > > > > > write ‘t’ n an expression; you access the local ‘t’ member
> > in
> > > > the
> > > > > > given object you are in.
> > > > > > During the an analysis, the current study (e.g.
> > AnyBodyStudy)
> > > > > > updates the local times ‘t’ for all the objects, that are
> > > > included
> > > > > > in the study, thus all ‘t’ have the same value.
> > > > > >
> > > > > > The reason for this design is exactly that time is
> > controlled by
> > > > the
> > > > > > analysis you make, i.e., something inside a study, and a
> > model
> > > > may
> > > > > > contain many studies. Therefore we cannot have one time
> for
> > the
> > > > > > whole model; we need one for each (mechanical) study.
> Since
> > a
> > > > > > mechnical object can be part of several studies, they also
> > need
> > > > > > there own time, but as I said before all these times are
> > > > > > synchronized during each step of the analysis before ‘t’
> is
> > used
> > > > for
> > > > > > anything (including expressions you have defined).
> > > > > >
> > > > > > The ‘denied access’ property of ‘t’ simply means that you
> > cannot
> > > > > > access ‘t’ for initialization in Script. Even if you
> could,
> > you
> > > > > > value would just be overridden by the study during the
> > analysis.
> > > > > > In general, AnyScript does not deny access to the values
> of
> > > > > > variables, only access to initialize them is denied when
> the
> > > > > > reference manual states so.
> > > > > >
> > > > > > Best regards,
> > > > > >
> > > > > > Michael
> > > > > > AnyBody Support
> > > > > >
> > > > > >
> > > > > > — In anyscript@yahoogrou <mailto:anyscript%40yahoogroups.com>
ps.com, “blulrey” <blulrey@>
> wrote:
> > > > > > >
> > > > > > > Hello Søren,
> > > > > > >
> > > > > > > Thank you for the response.
> > > > > > >
> > > > > > > I managed to figure out that the thorax and rib cage
> were
> > one
> > > > > > segment
> > > > > > > prior to your reply by looking in the BRep files and
> > playing
> > > > with
> > > > > > the
> > > > > > > RGB values of different segments.
> > > > > > >
> > > > > > > As for the ligaments, I will most likely supply my own
> > material
> > > > > > > properties, but it would helpful if attachment nodes
> were
> > > > > > available.
> > > > > > >
> > > > > > > I noticed that the time variable “t” is not available in
> > many
> > > > > > places
> > > > > > > in a script; e.g. “t” cannot be accessed in Main or a
> > ligament
> > > > > > object,
> > > > > > > but it can be accessed in a force folder such as
> > AnyForce3D.
> > > > In
> > > > > > the
> > > > > > > reference manual, “t” is listed as a denied-access
> member
> > for
> > > > > > > AnyForce3D. So why is “t” accessible in AnyForce3D if
> it
> > is
> > > > > > listed as
> > > > > > > a denied-access member?
> > > > > > >
> > > > > > > Regards,
> > > > > > > Brent
> > > > > > >
> > > > > > >
> > > > > > > — In anyscript@yahoogrou
<mailto:anyscript%40yahoogroups.com> ps.com, “AnyBody Support”
> > <support@>
> > > > > > wrote:
> > > > > > > >
> > > > > > > > Hi Brent
> > > > > > > >
> > > > > > > > Sorry about the late reply.
> > > > > > > >
> > > > > > > > > I have some general questions about the standing
> model.
> > > > > > > > >
> > > > > > > > > 1. Is the rib cage and sternum just for
> visualization,
> > and
> > > > not
> > > > > > a
> > > > > > > > real
> > > > > > > > > bone segment? I don’t see it anywhere in the
> segment
> > > > > > definitions.
> > > > > > > > >
> > > > > > > >
> > > > > > > > The ribcage and sternum is counted as a part of the
> > thorax
> > > > > > segment
> > > > > > > > in this model, this is why you do not find them as
> > separate
> > > > > > segments.
> > > > > > > >
> > > > > > > > > 2. If the ribs are not a bone segment, how would I
> > model a
> > > > > > load on
> > > > > > > > the
> > > > > > > > > chest? Would the load have to be applied to a
> spinal
> > > > segment
> > > > > > > > instead?
> > > > > > > > >
> > > > > > > >
> > > > > > > > The load should be applied to the thorax segment, if
> > there
> > > > is no
> > > > > > > > node at the location where you would like to apply the
> > load
> > > > you
> > > > > > can
> > > > > > > > add an extra node at this location.
> > > > > > > >
> > > > > > > >
> > > > > > > > > 3. I see that there are some ligament nodes defined
> in
> > the
> > > > > > trunk,
> > > > > > > > and
> > > > > > > > > that ligaments were originally planned to be part of
> > the
> > > > > > standing
> > > > > > > > > model. Will the framework for all the trunk
> ligaments
> > be
> > > > > > available
> > > > > > > > > soon?
> > > > > > > > I can not say when the ligaments will be available, i
> > think
> > > > the
> > > > > > main
> > > > > > > > problem with the ligaments is to obtain the mechanical
> > > > > > properties.
> > > > > > > >
> > > > > > > > I am attempting to build a model that includes these
> > > > ligaments,
> > > > > > > > > and could supply my own material properties. Of
> > course I
> > > > > > could
> > > > > > > > supply
> > > > > > > > > my own nodes as well, but was wondering if this will
> > be
> > > > > > available
> > > > > > > > in
> > > > > > > > > the near future.
> > > > > > > >
> > > > > > > > No i do not think it will be available soon, if you
> have
> > the
> > > > > > > > material properties you might be able to make use of
> the
> > > > > > ligament
> > > > > > > > nodes already defined in the model. I will try to ask
> > the
> > > > Mark
> > > > > > de
> > > > > > > > Zee which developed the Spine model, what is the
> status
> > of
> > > > the
> > > > > > > > ligaments.
> > > > > > > > >
> > > > > > > > > 4. Is it possible to simulate long loading periods?
> > For
> > > > > > example,
> > > > > > > > if
> > > > > > > > > the end time is set to 3600, would this be one hour?
> > > > > > > > >
> > > > > > > > In principle this should not cause any problems i
> think,
> > but
> > > > if
> > > > > > you
> > > > > > > > have a fast motion you will need very many timesteps
> to
> > > > capture
> > > > > > the
> > > > > > > > nature of the problem. If you have to few timesteps it
> > may
> > > > cause
> > > > > > > > problems for the wrapping muscles which uses the
> > position of
> > > > the
> > > > > > > > previoustep as the starting guess, and if this is very
> > > > different
> > > > > > > > position it might cause a problem. I think it would be
> a
> > > > good
> > > > > > idea
> > > > > > > > to devide the period into smaller sections.
> > > > > > > >
> > > > > > > >
> > > > > > > > > 5. Is it possible to control loading variables with
> > > > respect to
> > > > > > > > time?
> > > > > > > >
> > > > > > > > Yes you can create a interpolation function which is
> > time
> > > > > > dependent
> > > > > > > > and apply this a load, please use the AnyFunInterpol
> > keyword
> > > > for
> > > > > > > > this.
> > > > > > > >
> > > > > > > >
> > > > > > > > > If I want to model an 8 hour work day, including
> > breaks
> > > > and
> > > > > > lunch
> > > > > > > > > hour, could this be simulated in one model? This
> > would
> > > > include
> > > > > > > > > periods of loading and unloading.
> > > > > > > > >
> > > > > > > >
> > > > > > > > In principle yes but i think it will be best to devide
> > it
> > > > into
> > > > > > > > several analysis. Without knowing your problem in
> detail
> > i
> > > > think
> > > > > > you
> > > > > > > > shoudl consider finding some descriptive tasks and
> start
> > by
> > > > > > doing
> > > > > > > > the analysis on these instead of an entire working
> day.
> > Once
> > > > you
> > > > > > are
> > > > > > > > satisfied with this you try increasing the time period.
> > > > > > > >
> > > > > > > > Best regards
> > > > > > > > Soeren, AnyBody Support
> > > > > > > >
> > > > > > > > > Thank you for any assistance,
> > > > > > > > > Brent
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
[Non-text portions of this message have been removed]