Discussion Groups
New Idea
5 Kudos
xkoja.tomoyasu

Sink marks estimate results

Status: New Idea
by xkoja.tomoyasu on ‎06-19-2014 12:33 AM

Our customer request you to display the Sink marks estimate results in the thick portion other than the boss geometry and a rib geometry.
Because he want to see the shrinkage of the thick portion.

2 Kudos

Often, after starting a typical Cool(FEM)+Fill+Pack+Warp sequence, and while still at an early stage (i.e. Cool(FEM) running), one may change his/her mind on some options re further-on behavior; for example demand Warp to isolate the components. But when you change ANYTHING in Process Settings - regarless on whether it would really affect the current analysis state - we're presented with this familiar "Create copy", "Delete Results", or "Cancel" warning window. Less expereinced useers get frightened off by the necessity of running the sequence from scratch, so they cancel and do not introduse the desired changes...

 

It would be nice if the program "considered" the situation first, and only display the warning window if the user-introduced changes would invalidate the already produced/being produced results, or require the running analysis to alter the way it's executing. We the more experienced users know that if none of the above would happen, pressing "Delete" will actuallyNOT stop the sequence while introducing our changes to the way the consecutive analyses should execute.

 

Do you consider to introduce a more intelligent, context-sensitive use of the "Create copy", "Delete Results", or "Cancel" dialog window?

 

Thanks

 

Piotr

5 Kudos
aholman

Simplified Modeling for Baffle

Status: Under Review
by aholman on ‎03-13-2014 02:22 PM

Summary

 

A proposal to simplify modeling Baffle features in a cooling system.

 

Problem

 

Currently, every baffle requires two separate lines to create a baffle. Users can either make a small V configuration or the two lines can be coincident. However, if at any time the nodes get merged on the baffle line then the solver interprets the baffle as a dead-end. This issue is especially problematic if using HyperMesh to generate a Moldflow mesh file. 

 

Solution Proposed

 

Eliminate the second line needed to generate the baffle. The necessary pressure drop should be automatically calculated by the solver. The solver should also be able to tell which direction the water is moving based on the coolant inlet boundary condition. This is similar to other softwares on the market.

 

 

Status: Under Review
We had a similar discussion at the closing session of the moldflow power track at AU 2013, which got some good support. We have this under consideration. In principle you could create a simple bubbler and baffle modeling API script to simplify the modeling.
2 Kudos

There must be a provision to simulation cooling for ' thermoplastic overmolding' process

 

i.e  'Cool + fill+ pack + over molding fill + overmolding pack + warp'  option

 

Probably Autodesk team should develope the same.

3 Kudos

 A  SimCFD license is required for using Channel (3D) elements in a Cool (FEM) analysis.

 

It should be available without need of seeparate licence. It allowed user to import cooling channals 3D model

directly from tool design and do analysis.

3 Kudos
madhukeshwart

Shrinkage analysis for 3D

Status: Comments Requested
by Mentor on ‎04-01-2014 03:45 AM

Shrinkage analysis should be available for 3D mesh, as most of the things are available in 3d mesh now, which were not possible in previous versions

7 Kudos
Jacky1187

Hybrid Meshes

Status: Comments Requested
by Jacky1187 on ‎12-13-2013 01:25 AM

Hello all together,

 

to speed up the simulation, hybrid meshes would be advantageous.

 

For example Sigmasoft takes voxal meshes and this mesh is more tolerat of CAD problems. The calculation time for a compled Mould is very fast.

 

To decrease the preparation time and  computation time hybrid meshes can help.

 

It would be possible to simulat very large models in 3D.

 

Greeting Michael

Status: Comments Requested
Hi Michael and Harald, My personal goal is to remove 'meshing' as a skill we have to teach the user in order to get fast and reliable answers. And we do spend time every release to improve the meshing and the solvers, but this is certainly not a 'done' problem. I do want to provide a little context though. 1) Voxelation provides a very fast and robust mesh generation path. We actually developed what is probably one of the fastest voxel meshers available for other reasons, so we are very familiar with the technology. We did not develop it to replace our normal meshes as as there it comes with very significant sacrefices in appearance and run times (you generate very very large meshes). 2) Within Moldflow we can generate several different mesh types, and each works a bit differently, and depending on the starting geometry type, the meshing path may be different (this expalins why meshing on an STL or a Solid CAD model will give you different meshes on the same gemoetric part). Within the meshers there are typically a few steps that predeed or follow the 'mesh generation' like mesh imprinting and matching. These post meshing steps can take significant time compared to the brute mesh generation, but in the end pay back in better analysis results. For some pieces of some meshing paths we license technology (not Distine :-) ), but very significant pieces were developed internally and are continued to be developed. 3) In the last few releases (since 2013 I believe) we have improved the tolerance of the solver for dealing with meshes that are not pristine and perfect. Appart from this being a good idea, there is simply no way that we can guarantee that the initial starting geometry is good enough to create a pristine mesh. More work still needs to be done, but I have seen many customers spending too much time fixing the meshes up perfection, whereas the 'default' mesh would have given nearly the same analysis result. 4) Lastly, I agree that the meshing controls in Moldflow could use an overhaul. It's not as interactive as you would hope it would be. Thank you for your valuable input on this topic.
12 Kudos

Tiger strips/a surface defect, which often occurs in the injection molding, is characterized by a cyclic sequence of shiny and matt stripes at the flow front.

 

 The physical cause of the occurrence of tiger stripes, is the re-melting of already frozen outer layers the thermally damaged material reaches the surface and leads to spots with a matter shininess in appearance. 

 

It is possible with calculated plots with some relations, which is being used by some raw material suppliers.

 

Mold flow should introduce Tiger strips occurrence plot.

Status: Future Consideration
We are aware of this request and have been looking for methods to predict TigerStriping. There are several competing theories out there. This will probably requier some more time.
1 Kudo
madhukeshwart

Flow length in flow path plot

Status: New Idea
by Mentor on ‎05-05-2014 11:37 PM

Flow length plot

 

There must a be plot, which shows flow length from the injection point or gate in the flow path or material travel distance  

13 Kudos

 

An Option to Dump results in between the analysis

 

In Sigma soft (competitor software similar to mold flow),

there is a option to dump the analysis in between, then stop the analysis. The, next time you open the study and re- launch the analysis, it will start analysis from where you stopped it.

 

In mold flow, we need similar option

i.e In Between; we can stop analysis instead of aborting and close the study. Then,we can re-run from, where we had stopped it. So that we can save time, incase we need to abort the analysis & re-run after some time,

Status: Accepted
Valid request. We have some experience with some of this (through restart files). This is a significant development and hard to get to work well. For the end user, this is clearly a valid request.
6 Kudos
madhukeshwart

Simulation of air cooling

Status: Comments Requested
by Mentor on ‎09-18-2013 11:00 PM

Simulation of air cooling

 

1. Air as a coolant

2.Simulation of air sprayed cooling systems

Status: Comments Requested
I've heard the use of air as a coolant medium a few times recently. The main reason that for its use seems to be that air as a cooling medium has low pressure requirements and would allow very thin coolant cross sections. I'd be interested in hearing more on this. Could you provide more insight into 'air sprayed cooling systems'.
3 Kudos

Everytime that I start SIM360 Moldflow the mesher and the solver are set by default as "in  cloud". It would be very useful to have the possibility of setting them in as "locally" by default.

Status: Accepted
I think you have a good point. Thank you for bringing this up.
4 Kudos
madhukeshwart

3 shot molding- moldflow simulation

Status: New Idea
by Mentor on ‎09-11-2013 06:00 AM

Moldflow should introduce -3 shot molding-  simulation

5 Kudos

we are not gettting 'Sink marks estimate' results in "3D - overmolding / 2 shot molding analysis"

 

I am not sure, whether it is available in 2014.

 

Moldflow should introduce 'Sink marks estimate' results in "3D - overmolding / 2 shot molding analysis"

6 Kudos

moldflow should be able to estimate the time for analysis.

 

when ever, we are launch an analysis, moldflow should be able to pop up tentative time to complete the analysis.

 

it should be based RAM/ memory / processor availble/solver settings and number of nodes/elements

 

 

 

 

Status: Comments Requested
I would like to have other people's perspective on this; if this is a significant amount of work involved (that we would for instance not dedicate to for instance accuracy) and the time estimate would be a very very rough ball park, it this worth doing?
4 Kudos

hi

when defining a sequential gating and the valve gates are controlled by flow front and trigger nodes and the user is changing f.e. a location of a injection location AND is using "global merge" the nodes are renumbered.

afterwards all the triggernodes have  to be defined completely new.

 

when simulating big parts with lots of injection locations this can be annoying.
or the filling behaves in a unexpected way because the triggernodes have changed. when simulation time is very long this is recognised to late.

 

solution 1.: lock the trigger node numbering. so the trigger node keep their number, even when all nodes are renumbered

 

solution 2: highlite the triggernode location by f.e. an clearly visible arrow

 

solution 3: change the way to define triggernodes or triggerlocations.
at the moment the user has to "remember" the trigger node. then swich to the property of the valve gate or change to the  valve gate timing tab and enter the trigger node.

this could be more comfortable. like f.e. after highliting a node and rightclick, a menue could appear and the user could define "define this node for valve gate XY"

 

 

 

 

 

 

Status: New Idea
Fair request. Not trivial. But in general I agree that this is currently not ideal. This would probably require some significant re-architecting. But the request is fair
3 Kudos
M.Brassel

Throughput result for 3D

Status: Under Review
by M.Brassel on ‎06-28-2013 03:38 AM

I'd like to see a throughput result for 3D flow analysis, as it sums up the Flow rate, beams result for the whole analysis.

Otherwise I would have to do Dual Domain flow analysis in parallel just to get this result. Or did I miss anything?

Status: Under Review
1 Kudo
p.hammer

New Analysis for mold heating up

Status: New Idea
by Contributor p.hammer on ‎10-09-2013 02:51 AM

HI everybody, 

 

in reality before you can start with injection moulding the mould has to be heatet up to working temperaturen.

And then starts the injection molding process.

With bad cooling layout you already get here very different mold surface temperatures.

 

Today cool fem starts with constant temperature in the mold and it calculates with cooling of the melt even when you start it from 25°C.

And this is not real.

 

 

So it would be very good if there is a new analysis sequence in moldflow, perhaps called mold heating,  

that calculates the temperature just in the mould prior starting a cool fem analysis from production start up and uses the temperature

distribution in the mold as starting boundary conditions for cool fem from production start up.

 

- more precise results for cool fem

- more precise warpage results

- getting real information how long it takes to get the mold to working temperature

- and how long it takes in njection molding to get a stable process 

 

Philipp

1 Kudo

Currently, there are hard coded defaults for the conformal cooling setup.

 

A user would like to be able to define the number of layers of mesh enhancement and switch the turbulence model.

1 Kudo

I would like an 'All' option for Intermediate Results in the Fill+Pack solver parameters.  I find it frustrating that solver timesteps are solved but not outputted for results.  As it stands, you either need to:

  • Use the 'Write at constant intervals' option, which will miss timesteps for fine timestepping.
  • Use the 'Write at specific times' option, but this requires us to know when all timesteps will occur before running the analysis, or to assume a very small time interval.

I understand that computing resources will be tied up for results output and file storage, but this option would still be of use to me.

Status: Under Review
Announcements
IdeaStation Guidelines
Review guidelines and best practices
before posting a new idea.