Does Moldflow can not predict Orientation effects in 3D element?

[takafumiaoyama]

We analysis shirinkage of test peices which has 6mm thickness , like letter of P, and was made by PA66GF50.

I checked 2 kinds of packing pressure as 10MPa and 80MPa on same inject conditions and calculate Differential shrinkage and Orientation effects.

I think that if test pieces has same inject conditions, it shows difference only in differential shirinkage but same fiber orientation and orientation effect.

But in analysis result MFI2019.0.5 which use 3D element, residual stress model, we get different orientation effects in spite of 2condition has almost same orientation tensor.

How to understand it?

[takafumiaoyama]

My analysis result are here, all resurt has dependence for packing pressure.

[raalteh]

Dear  Takafumi-san,

Moldflow does predict orientation effects in 3D and your plots actually show it the Orientation effects both show more or less the same value at the high end, but the transition form 'Red' to 'yellow' is happening in different places.

You indicated that the fiber orientation is the same between the two scenarios but when you look at the fiber orientation plot for 3D, it show (by default) the fiber orientation at the skin layer. With the injection phase being the same, the skin orientation is expected to be nearly identical because the skin orientation is 100% dominated by the part filling (and the skin freezing immediately). To see the orientation at the core, you will need to create a cutting plane. With a different packing pressure, I expect your core orientation (the orientation at the center of the part) to be different --> different orientation effects in warpage. (alternatively you can look at the 'Average Fiber orientation' to see some differences.

Something unusual about your analysis is that your orientation effects are of about the same magnitude of the differential shrinkage effects. For a normally molded Nylon with 50% fiber ... I would have expected the fiber orientation to be an incredibly dominant factor compared to packing ... so that suggests one of two two things to look at:

1. If you look at the material file, does it have 50% glass fibers in it in the filler tab? Is this material in the Moldflow material database?
2. Is this part very badly under packet and do you have a LOT of differential shrinkage? if you have a significant amount of frozen. 

[takafumiaoyama]

Thank you for Answers.

I checked my analysis setting with your advise and I found that Fiber orientation has small changes and large size of differential shrinkage occurs because its nozle seal.
But I still can not understand the reason why fiber orientation change is small but orientation effect change are large.

1. Material constant are default DB in MFI2019.0.5, Zytel70G50HSLRBK509 which has 50%glass fiber.
So, I agree that fiber orientation is more dominant than differential shrinkage and I checked fiber orientation in probe XYplot.
It shows that 1st component of fiber tensor show change about 0.05 at the position near from gate, and shows about 0.01 far from gate.

2. In pressure at freeze, it shows nozle seal, its under go bad filling.But I cannot found backflow.

In condition of small packing pressure, resin shows 0MPa in center of test pieces at freeze, and in large packing pressure, it shows 30MPa. 

With compare of all effects, test pieces has large shrinkage.
small packing pressure : all ->0.6mm orientation ->1.8mm shrinkage -> 2.1mm
large packing pressure : all ->0.3mm orientation ->1.2mm shrinkage -> 1.4mm

  

[raalteh · Accepted solution]

Dear Takafumi-san

Sorry for the delay in response.

1) its' good that your fiber orientation in not drastically different with a different packing profile, that's consistent with that I would expect (unless you had some very unusual case). So ... that is in line with expectations.

2) The 0 pressure at freeze is a bit of a red flag. That means that the polymer's volumetric shrinkage there will be quite significant. Its not completely straight to give you a very precise answer with the information I have here, but ... If you have a lot of local volumetric shrinkage, and the orientation of the material 'random' than you can assume that the shrinkage is uniformly distributed. If you have a lot of fiber orientation in one direction, then the volumetric shrinkage will be resisted most in the fiber direction (fibers prevent shrinkage in that direction) and it will have to be absorbed mostly by in the non-dominant directions. If you pack your part well, there is little volumetric shrinkage and there effectively is not going to be too much shrinkage happening perpendicular to the fiber direction, but if you have a lot of shrinkage perpendicular to the locally dominant fiber direction.

Though I cannot fully follow what you are trying to say with the large paragraph, at a high level, this is consistent with what I would consider to be expected.

If you need further help, it's probably worth connecting with the support team.

Best regards,

Hanno

[takafumiaoyama]

Dear Mr. Hanno

Thank you.

Defflection of orientation direction are checked and its show same situation of its all direction.

I will try to check more situation that have different packing pressure about 50to70MPa which has pressure at freeze over 10MPa.

And I would ask support team why 2 situation have almost same fiber orientation  but have different orientation effects.

Sincerely yours

[raalteh · Accepted solution]

Hi,

This goes back the point I was trying to make earlier. The fiber orientation is more or less fixed, but the shrinkage of the plastic is different. The fibers will resist shrinkage in the (dominant) fiber direction, but the volumetric shrinkage will still have to be distributed somehow and is taken up primarily perpendicular to the (dominant) fiber direction.  THAT distribution will be different between a high and low shrinkage case.

I'm slightly simplifying it as there if as the thickness shrinkage is typically a smaller factor because by nature, the plastic wants to come off the mold wall (thickness shrinkage) but plastic coming through the core during packing will push it back to the mold wall. So a lot of the packing will normally compensate thickness shrinkage more than proportional.