News: HyperSizer.com has a Community Board and Customer Support System. Submit a ticket at http://hypersizer.com/ticket

Recent Posts

Pages: 1 2 3 [4] 5 6 ... 10
31
Analyzing & Optimizing Stiffened Panels / Re: FEA Loads - Peak Element Filtered
« Last post by James on July 08, 2021, 07:45:30 AM »
I'd like to add that the peak element processing is repeated for every unique thermal set and/or assigned reference temperature. So if you have 2 unique thermal sets, then you get two blocks of processed peak element forces. During sizing, both blocks are checked.

I hope that's helpful.

-James
32
Hello oyama,

you can use the "Stiffness Requirements File" in Express by defining the file path in the: Analysis Criteria -> Stiffness Criteria dialog.

This way, you can define a minimum required stiffness (ABD matrix diagonal) for every individual element in your model.

The file format is as follows:
___________________________________
-1
#of Entries (i.e. number of elements you want to define a stiffness target for)
ElId, [dummy], A11, A22, A33, D11, D22, D33

-1
____________________________________

A simple example file defining the minimum (required) stiffness for elements with ID 1 through 10 is attached to this post.

Now, during element sizing optimization, these minimum stiffness requirements will be fulfilled (if your design restrictions allows for it; e.g. if max gauge is high enough for your chosen material).

I hope this helps!
-- Bertram
33
Hi,
How can I define "Stiffness Requirements File"?

I use Express for sizing composite structure, and I try to optimize with respect to the stiffness(Nodal deformation), but I don't want to use iteration. Then, I try to define "Stiffness Requirements File".

Any help would be welcome.

Sincerely,
34
Analyzing & Optimizing Stiffened Panels / Re: FEA Loads - Peak Element Filtered
« Last post by Brent on May 17, 2021, 10:55:19 AM »
Hi Mike,

You bring up a good point. HyperSizer does not account for the reduced allowable when filtering cases based on peak element metrics.

One suggestion to check if you are missing any critical cases is to run an analysis with peak element and then another analysis with element based. Make sure you are not sizing to keep runtimes down to a minimum. Then compare the critical load cases from the element based run to see if any "Hot" conditions were flagged that were not flagged in the peak element run.

Hope this helps!

Brent
35
Analyzing & Optimizing Stiffened Panels / FEA Loads - Peak Element Filtered
« Last post by mjlafrenier on May 14, 2021, 03:08:52 PM »
All,

Does HyperSizer give any consideration to thermal loading when developing the peak elements when peak element filtered is selected (for large models/many load cases)?

The concern is that until final sizing is done for all elements (element based), then its possible that a cold day/room temperature case with a larger load could be hiding a more critical but lower load hot day case with a reduced allowable.

Thanks,
Mike
36
Is there method of creating a new component similar to how a new assembly can be created with AssemblyCol.Create()?  Neither classes (Component or ComponentCol) have a "New()" or ".Create" method.  Is there a trick to it or is it impossible to do via scripting?

I'm wanting to create components similar to the way you can use the FEM Viewer GUI by selecting menu items Component -> Create Using FEM Viewer (by element).  I would like to use Excel to create components by using [Component Name, Component ID, Element ID List].  The FEM Viewer allows a user to define a component with nothing but these three variables.

Thank you.
37
In that case, I'd recommend modeling them in and using HyperSizer + Nastran Sol15 to size for buckling.

-James
38
Thanks James,

The cutouts would be bound by stiffeners and integrated into the orthogrid, but would be larger and span at least several bays. The shape will likely be a large slotted shape or circular.

Best,
Mike
39
The smeared sizing in HyperSizer assumes scenario 1 (0.5*S,S,S,0.5*S).
I understand that you'd prefer not to have 1/2 bay on each side.
I recommend to update the FEM with the stringers meshed-in, with bar elements, using the same spacing that HyperSizer determined as optimal. So model it as (SSSS). Then you can use the stringer segment (tech2) approach to size each skin pocket and stringer.

What is the size and shape of the cutout?
Is it within a pocket, bounded on all sides by stringers ?

Geometry rule 40 is a suggestion, to ensure that the smeared approach is more valid. If the panel is bounded on all sides by supporting structure, e.g. other panels, then I think it's okay to turn that rule off and size the smeared panel with fewer than 3 stringers.

40
Analysis approach: HS Orthogrid smeared stiffened panel
FE Modeling: Webs (CQUADS), Outer Cap (BAR elements)

1) Panel stiffener spacing and relationship with FE modeling:
- Example: Orthogrid (S_0&90 = 5", t_skin = 0.1", t_stiff_0&90 = 0.1", h_stiff_0&90 = 1", X_span = Y_span = 15"), Minimum unit for Geometry Rule 40 (3S)
- The panel is modeled with outer caps (outer flange, etc) using bar elements.
- Option 1: 3 stiffeners (Spacing: 1/2S,S,S,1/2S). If the smeared stiffened approach leaves a 0.5*S skin segment on either side, then I'm left with a row of half bay webs next to the outer cap. This is not ideal for fastener installation and may require us to carry extra weight as those 1/2*S bays are the same thickness as full sized interior bays, but are also supported by the outer caps.
- Option 2: 4 stiffeners (Spacing: S,S,S). The other alternative is to assume full bays and then the outer smeared stiffeners are coincident with the location of the bar element outer caps. This bay sizing is more ideal for design.
- See attachment. Do you have any experience or guidance with this scenario?


2) Panels with cutouts
- If a larger cutout is modeled in the FEM, is HyperSizer able to accommodate the cutout for global panel buckling, or would the procedure need to be to follow the smeared approach for a uniform panel and then analyze the cutout detail for stability using other methods (NASTRAN SOL105, etc.)?

3) High Aspect Ratio Panels:
- Since the geometry rule 40 requires width/spacing >= 3, is there a HS method to analyze an orthogrid panel with less than 3 stiffener widths in only one direction?
- Currently it seems as though one would have to either discretely model it and deviate from the smeared (type 1) approach, or analyze it as uniaxial stiffened panel along the long dimension.
Pages: 1 2 3 [4] 5 6 ... 10