huge pressure difference

[prabu_Karunakaran]

Hi everyone,

                          i did one small comparison study with simple rectangle  geometry 150*60mm with 1.5mm thickness

i have choosen Lexan 121 material .

i found huge pressure difference between both studies

please find the picture.

only gate position is different for both studies, i wnt to know the reason?

[madhukeshwart]

It is difficult comment, without seeing the study files and please do share study files 

[prabu_Karunakaran]

Hi,

                   I dont know how to attach the study. This is simple rectangular geometry the dimensions of the part is attched here.

Mesh type used : 3D with 6 layers

Material used Lexan 121 (MF tested)

Please try it by your self .

Thank you,

Prabu.k

[mppkumar]

Hi

I have tried with the sizes as mentioned by you and noticed similar pressures.

Request to cross-check the process parameters set by you once again.

Thanks

M  P Pradeep Kumar

[prabu_Karunakaran]

Hi ,

         Here i have attached the study files .....

[mtrujil1]

hi M  P Pradeep Kumar!

i tried the same scenarios and my results were same that the results obtained by Prabu. but my part design was with 150mm x 60mm x 1.5mm

Please see the images attached.

i tried with different Scenarios

1 Mesh type 3D, Mesh generated in MF2012 (DD and after 3D), Aspect Ratio <6, Match Perc. 100%, Edge Length: 2, E.L Through Thick: 2.5. Gate: Left Side, Part Dimension 150 x 60mm x 1.5mm

2 Mesh type 3D, Mesh generated in MF2012 (DD and after 3D), Aspect Ratio <6, Match Perc. 100%, Edge Length: 2, E.L Through Thick: 2.5. Gate: Right Side, Part Dimension 150 x 60mm x 1.5mm

3 Mesh type 3D, Mesh generated in Ansa & Mf2012 (ANSA DD and after MF 3D), Aspect Ratio <6, Match Perc. 100%, Edge Length: 2, E.L Through Thick: 2.5. Gate: Left Side, Part Dimension 150 x 60mm x 1.5mm

4 Mesh type 3D, Mesh generated in Ansa & Mf2012 (ANSA DD and after MF 3D), Aspect Ratio <6, Match Perc. 100%, Edge Length: 2, E.L Through Thick: 2.5. Gate: Right Side, Part Dimension 150 x 60mm x 1.5mm.

5 Mesh type 3D, Mesh generated in Ansa & Mf2014 (ANSA DD and after MF 3D), Aspect Ratio <6, Match Perc. 100%, Edge Length: 2, E.L Through Thick: 2.5. Gate: Left Side, Part Dimension 150 x 60mm x 1.5mm

6 Mesh type 3D, Mesh generated in Ansa & Mf2014 (ANSA DD and after MF 3D), Aspect Ratio <6, Match Perc. 100%, Edge Length: 2, E.L ThroughThick: 2.5. Gate: Right Side, Part Dimension 150 x 60mm x 1.5mm

[mppkumar]

Hi mtrujil1

Thanks

M P Pradeep Kumar

[MartinCavalry]

Hi All,

I can confirm I have similar findings to what were reported by Mppkumar.

Here the computation conditions are based on 3D solver, MF2014, same Lexan 121 mf tested material with 150x60x1.5 mm geometry.

I have checked the files that was attached in the beginning of the thread and noticed something mush have been dodgy with the meshing (check the fill pattern the flow front looked very unusal).

My results are as follows.

The VP pressure using an STL geometry, mesh is 6mm, unstructured

gate in +X dir.: 90.9 MPa,

gate in -X dir. 89.3 MPa. - identical values

The VP pressure using an IGES geometry, mesh is 6mm kept only as a reference value but controlled by divisions, structured.

gate in +X dir.: 93.3 MPa

gate in -X dir.: 87.3 MPa - almost identical.

Regarding accuracy the best is to perform a mesh sensitivity analysis to contrast accuracy and computational time.

An unstructured mesh, provided it`s fine enough does the job perfectly and should provide sufficient accuracy.

Best regards,

Martin

[mtrujil1]

Hi Pradeep Kumar

i already ran your sdy file, and the pressures results were very similar with both scenarios.

your Global Edge Length selected was 3mm.

i tried to simulate with your globalk edge length, but instead to use the 3D mesher Advancing Frront with 4.5 value (Rule from Moldflow fundamentals Training "Use a value between 0.5 and 1.5 times the global Edge length for  Edge Length through thickness parameter"), was selected 3D Mesecher Legacy.

And the Results for Bot scenarios were similar too. see the images

Martin, which 3D mesher option was selected for your analises.? Why the difference between these option?

Regards

[harald_goetz]

I would like to reply with a question here:

The advanced front mesher is the default option, right?

So it should bring better results.

Could it be, that with gate refinements your results might be enhanced?

Because what I saw here so far seems odd.

Best regards

Harald

[mtrujil1]

hI hARALD

The advanced front mesher is the default option, right?

Yes, the Advanced Front is a Default option, but the results are not according with the logic.

Could it be, that with gate refinements your results might be enhanced?

i tried too, i drew the gate with beams elemtns (6 beams) and the results were the same :(.

Could you simulate the exercise? Part dimensions (150mm x 60mm x 15mm)

regards

[PascalGosset]

Hi.

I'm also really interested by this problem.

I made some simulations to try to understand.

Here are the results I got :

                             Gate position  Side Y-    Side Y+

initial simulation                             27,4         42,6

Gate size imposed : 1 mm             32,3        48,5

mesh 12 layers instead of 6          40,4         45,3

PC : Makrolon 2405                      29,6         42,9

Gate not in the edge but near       27,7  

Part turned in the plane XY           27,4  

Mesh Midplane                             45,6          45,6

Mesh Dual Domain                       47,3          46,5

So, I also found big differences of pressure with the 3D meshing.

Regarding the results got with the Midplane and DD , it seems the problem comes from an underestimation

when the gate is in the Y- position.

I hope someone will give a good answer to your question.

[madhukeshwart]

 1. If Study with feed system ...... I doubt there will not be any differences.
 
2. If study without feed system ....It can happen due automatic gate concidered and thickness at gate region.
 
3. It should not be difference in both the case.
 
4. Gate contact diamter may be a culprit for 3D model witout feed system