modelling a drinks pouch

I need to model a drinks pouch

- see http://www.gualapack.it/segmenti_e.htm for an example

- my pouch will be like the "Cheerpack"

Does anyone have experience of modelling similar things - especially
with respect to getting the volume accurate.

Thanks,


Kev

 

How you approach this is primarily a function of what you will use the
model for. If it is just a "looks like" model for documentation, it may
be more difficult, depending on how much realism or accuracy you need.
I'm going to guess and say that the thickness of the material is
relatively unimportant compared to the volume, so it might be ok to just
model the outside dimensions. At the worst, a usable approximation might
be to simply multiply the surface area of the solid by the thickness of
the material to subtract the volume of the material from the overall
volume. You may also find that starting out with surface tools is more
appropriate than solid features. If you go the surfaces route, you
cannot get a volume determination until you make a solid from it.

On a highly flexible part like this, it seems like the bulge is going to
account for a fair amount of the volume. I don't know if you have a way
to prototype a few to get an average projected volume.

In terms of modeling to a specific volume, any time I have done that
with bottles or buckets, it has always been an iterative process. Unless
you have a very specific shape with an accurate volume formula, you are
looking at an iterative process. The easiest is if you have the
flexibility to use a scale function rather than needing to tweak dimensions.

Then again, if a company like the one on the link you posted is helping
you, they have probably been through the process many times.

 

As Matt mentioned the manufacturer probably has a pretty good idea of
the volume of any particular design. On something proprietary like this
that is usually the case.

If a really accurate volume is needed one could go the FEA approach and
model the actual deflections. Cosmos/M would be pretty good at this as
would ABAQUS, ANSYS or Nastran.

 

As Matt mentioned the manufacturer probably has a pretty good idea of
the volume of any particular design. On something proprietary like this
that is usually the case.

If a really accurate volume is needed one could go the FEA approach and
model the actual deflections. Cosmos/M would be pretty good at this as
would ABAQUS, ANSYS or Nastran.

 

"If" you can model the pouch in solids to its "full" condition you can run a
cavity routine(see help). Then use mass properties to get the volume of the
interior.

Mike

 

Hi Guys,
Thanks for the thoughts...

Well It's pretty much as I thought. The models (4 sizes) will be used
to check the volumes and to send to the graphics guys for the arty
stuff to be added. I'm not too concerned about the absolute volume as
this has a +10% tolerance on it - so I think I'll just scale them up.

The bigger concern for me was the "bulge" And I guess at this stage
I'll take a trip to the local supermarket, buy some and measure them
to get an approximation as a potential supplier has not yet been
contacted.

The way I normally model rigid bottles - if the shape allows, is to
model the volume then shell outwards - this means the material
thickness can be increased/decreased without re-sizing the volume.

These pouches are new to me and I was wondering if there was any
cunning trickery .

Cheers,

Kev