Material assigment in HFSS

[wlcsp]

Let's say I have an object A which is assigned as silicon. And another object B which is located completely inside the object A. I assign object B as vacuum. Without doing any boolean operation, what would be the material for object B? Sorry for the rather silly question. I think B would be vacuum. Am I right????

 

[flyhigh]

Hi,

I am not HFSS expert but I did some simulations. I also had the case of one object inside the other. However, I couldn't get HFSS to simulate the structure unless I did the boolean operation of subtracting the inner object from the outher. It might be the case that HFSS will not let you have one object inside the other, so it is irelevant what the program would think of material of embedded object, it will not accept it anyway.

flyhigh

 

[asdfaaa]

Let's say I have an object A which is assigned as silicon. And another object B which is located completely inside the object A. I assign object B as vacuum. Without doing any boolean operation, what would be the material for object B? Sorry for the rather silly question. I think B would be vacuum. Am I right????

Yes, B would be vacuum, but if there is any other boundary condition involved, you must carefully check the boundary condition priority.

 

[limate]

I think boolean operation is necessary, otherwise the confusing boundary will cause problem for simulation solver.

 

[smithchart]

I guess B should remain to be vacuum if the software do not issue any warning or error for a boolean operation. However, I would suggest you do a quick check with a very small and simple object for two cases: a: inner object is assigned as vacuum; b: a large DK value, say 4.5. Try it out then you should know the results with 100% confidence

 

[dirtypig]

a and b will be overlapped

 

[itaifrenkel]

Hi,

Since both materials are dielectirc the material of the inner box is vacuum. No need to subtract.

HFSS will report an error only if these objects overlap. Only then the user has to resolve the overlapping area - by subtracting the vacum from the silicon.



Itai

 

[svarun]

Hi Itai,
Can you please confirm the order of priority in HFSS for material assignment ? I want to know the order for PEC,General dielectric and Vacuum. Thanks

-svarun

 

[tw20053010]

Yes, it is vacuum
wave pot cannot be defined inner, expect for lumped port.
If you want the port inner, you can put a pec behing the port.

Regards,

 

[itaifrenkel]

The heirarchy depends on the geometry and not on the material assignment of the object.

There is some heirarchy in boundary conditions. For example, a waveport overides a radiation boundary.

Itai

 

[Feri4030]

Hi Itai,
Can you please confirm the order of priority in HFSS for material assignment ? I want to know the order for PEC,General dielectric and Vacuum. Thanks

-svarun

here is a excerpt from the help of HFSS v11
"Object Overlap Settings for Complicated Models HFSS>Set Material Override

Complicated geometries often have small object overlaps. This setting will allow overlaps between dielectrics and metals. In the overlap region, the metal will locally take priority over the dielectric, as if this part of the dielectric has been subtracted. Overlaps between two dielectrics and overlaps between two metals are still not allowed."

It also seems that vacuum overrides by dielectric, in when two dielectric overlap the answer should be wrong, note that HFSS may not return an error message, see the attached simulation

 

[mah@sh]

Let's say I have an object A which is assigned as silicon. And another object B which is located completely inside the object A. I assign object B as vacuum. Without doing any boolean operation, what would be the material for object B? Sorry for the rather silly question. I think B would be vacuum. Am I right????

while doing modelling , nothing happens.
But when run the model , u will face a error
You must do some boolean operation

 

[simuboss]

it should be vacuum, since later definitions override previous ones. But it doesn't hurt if u do the operator...

 

[qqpost]

I don't think the order matters in this case, which means there is nothing to do with later or previous. I agree with Feri4030 that "the metal will locally take priority over the dielectric". Definitely, It will be more precise if do the boolean operation.

 

[simuboss]

I don't think the order matters in this case, which means there is nothing to do with later or previous. I agree with Feri4030 that "the metal will locally take priority over the dielectric". Definitely, It will be more precise if do the boolean operation.

disagree! so if you create 2 identical spheres at the same location, first one from metal, and second one from air, u think that the properties of metal is kept at that location?! that's wrong...the order matters.

- - - Updated - - -

I may add, if you have two metals, then what? the rules are in favor of order not material type...i hope this is helping

 

[qqpost]

disagree! so if you create 2 identical spheres at the same location, first one from metal, and second one from air, u think that the properties of metal is kept at that location?! that's wrong...the order matters.

- - - Updated - - -

I may add, if you have two metals, then what? the rules are in favor of order not material type...i hope this is helping

I'm using HFSS v14. If you see the HFSS->Design Setting->Set Material Override, it says:

" This option allows some intersections to be resolved automatically in the mesh. If metal intersects dielectric, the metal will override the dielectric in the region of overlap. If objects with the same material intersect, the smaller object will override the larger. All other intersections will be treated as errors. Note that dielectrics contained within metals will be completely overridden."

You can choose whether enable material override or not. But as far as I am concerned, the order doesn't matter.

I'm glad to hear more discussion from you. Thank you.