Cancel
Showing results for 
Search instead for 
Did you mean: 

Buckling with pre-stressing Gravity LOAD (Dead Load)

Phenom
Phenom

Hello Experts,

 

I want to analyze a buckling situation of a complex structure loaded with gravity AND additional working loads.

 

I ask myself is there an opportunity in NX Simcenter to create the model in that way that only the additional working loads are influenced/scaled by Load scaling factor and the gravity load remains with 1g without scaling.

I guess there is a way to do so, because in thread

 

CAE-Simulation-Femap-Forum/Buckling-with-Body-Loads/m-p/535433#M8323

 

it is pointed out that NASTRAN can do it. My problem is,

- that the dat-file provided there is not working in SimCenter (it's interpreted as linear Static Solution SOL101),

- and the thread comes from FEMAP forum.

 

So can someone explain what shall I define in NX Simcenter Pre-Processor?

What solution do I have to use? - Linear buckling SOL105 or SOL601?

Where do I have to place Gravity load?

Where do I have to place additional working load?

 

2019-01-16_094045.png

 

 

In the second Load block (in Subcase - Buckling Method) it's possible to define pre-stressed  Bolt-PreLoads. But that's not what I'm looking for. I want to extract 1g in Z- direction from buckling loads an put it into the still unknown "pre-stress zone". - Who can help?

 

Every suggestion is welcome. Best wishes. Michael

| Production: NX10; Development: VB, TCL/TK, FORTRAN; Testing: NX11 | engelke engineering art GmbH, Germany
| Kudos for good posts! And if my post answers your question, please mark it as an "Accepted Solution".

Kick Off (2015-09-14)On the Map (2016-10-21)1st Gear 2018 (2018-02-19)To the Rescue (25, ‎2018-02-24)5th Birthday! (2018-02-25)Popular (100, ‎2018-10-30)Prolific (100, 2019-02-27)Philosopher (500, ‎2019-02-28)

8 REPLIES 8

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Gears Phenom Gears Phenom
Gears Phenom

I use SOL 105, make two subcases, buckling and prestress load must be separated in two subcases.

Then I edit .dat file, I add third subcase (1001) where I mark one of subcase as buckling and other as prestress.

STATSUB(PRELOAD) = 1
STATSUB(BUCKLING) = 2

Снимок.JPG

And I know that it is possible to make NL buckling in sol 601 by using ramped/sloped load functions for backing/prestress loads. But that is not as fun as working with Nastran input file)

Maybe ti is possible in SOL106 too, but I don’t know how to make it, because load cases in sol 106 applied sequentially and loads don’t have functions.

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Phenom
Phenom

Hi Karachun,

thank you for answering. It looks like a lot of handicraft, but, it's a way.

I will reproduce it with my model and will post the results here.

Best wishes, Michael

| Production: NX10; Development: VB, TCL/TK, FORTRAN; Testing: NX11 | engelke engineering art GmbH, Germany
| Kudos for good posts! And if my post answers your question, please mark it as an "Accepted Solution".

Kick Off (2015-09-14)On the Map (2016-10-21)1st Gear 2018 (2018-02-19)To the Rescue (25, ‎2018-02-24)5th Birthday! (2018-02-25)Popular (100, ‎2018-10-30)Prolific (100, 2019-02-27)Philosopher (500, ‎2019-02-28)

Highlighted

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Gears Phenom Gears Phenom
Gears Phenom

Only this part should be created manualy.Снимок.JPG

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Phenom
Phenom

Hi Karachun,

thank you for your contributions. The procedure which you described works very well.

 

Only one addition by working in NX SimCenter (NX10):

By starting with Static Solution (SOL101) I have to correct two things furthermore:

1. Modify Solution from 101 to 105 (SEBUCKL)

 

$*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$*
ID,NASTRAN,buckling_preload-grav_prt1_fem1_sim1-buckling_01
SOL 105
CEND

After having a new LoadCase with METHOD=102 (in my case)

SUBCASE 3
LABEL = Subcase - Artifical Buckling LoadCase with Pre-Load and Buckling Load
STATSUB(PRELOAD) = 1
STATSUB(BUCKLING) = 2
METHOD = 102

 

 

2. Insert a definition for METHOD = 102 (ID_EIGRL) in BULK DATA SECTION

 

$
$* NX Modeling Object: Buckling_Lanczos_10_max EIGRL 102 10 0 7 MAX

 

Nevertheless I wonder why a common loadcase for Preload is not usable in Simcenter directly? - It would be easy to define and to handle it in SimCenter but really complicated to do the modifications everytime in DAT-File. The risk of overwriting the file is huge!

 

But it works, it's a workaround. Maybe it's better in NX12, I only use NX10 and can test it in NX11.

Best wishes, Michael

 

| Production: NX10; Development: VB, TCL/TK, FORTRAN; Testing: NX11 | engelke engineering art GmbH, Germany
| Kudos for good posts! And if my post answers your question, please mark it as an "Accepted Solution".

Kick Off (2015-09-14)On the Map (2016-10-21)1st Gear 2018 (2018-02-19)To the Rescue (25, ‎2018-02-24)5th Birthday! (2018-02-25)Popular (100, ‎2018-10-30)Prolific (100, 2019-02-27)Philosopher (500, ‎2019-02-28)

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Gears Phenom Gears Phenom
Gears Phenom

In Femap I start with solution 105, here I can select two subcases and then only create third (main) case by hand.

Снимок.JPG

At current time, I make some rotordynamics calculation in Femap. It is much faster and easier to prepare one .dat file with rotor model, variants of bearings properties (one bearing may have different clearance and therefore stiffness and damping) and other stuff. Then I simply copy this file and comment/uncomment properties of current bearing or add "-" to phase of load and run all files using .bat file.

So I learned to stop worrying and love the Nastran input fileSmiley LOL

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Phenom
Phenom

Karachun,

 

oh yes, I remember the times when I was using FEMAP as Pre-Processor, and FEMAP and I - we had UPs and DOWNs.

But time is changing, so NX Simcenter is now for me the tool to deal with.

 

Thank you for your answers.

Best wishes, Michael

| Production: NX10; Development: VB, TCL/TK, FORTRAN; Testing: NX11 | engelke engineering art GmbH, Germany
| Kudos for good posts! And if my post answers your question, please mark it as an "Accepted Solution".

Kick Off (2015-09-14)On the Map (2016-10-21)1st Gear 2018 (2018-02-19)To the Rescue (25, ‎2018-02-24)5th Birthday! (2018-02-25)Popular (100, ‎2018-10-30)Prolific (100, 2019-02-27)Philosopher (500, ‎2019-02-28)

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Experimenter
Experimenter

Hi Michael,

 

it is very easy within Simcenter (NX 10) to setup your required preloaded buckling solution.

Only two lines need to be added to the solver input manually, but can be done via a Modelling Object in SC.

 

1. Create solution 105

2. Clone "Subcase - Buckling Loads"

3. Reorder subcases. Both Buckling load subcases before "Subcase - Buckling Method"

4. Apply the defined preload (gravity) to the 1st buckling load subcase

5. Apply the "working load" to the 2nd buckling load subcase

6. "Edit" the "Subcase - Buckling Method". Create new Modelling Object at "Case Control User Defined Text". The Keyin Text is as follows:

STATSUB(PRELOAD) = 1

STATSUB(BUCKLING) = 2

6. Apply BCs

7. Now you are ready to solve.

 

If the two lines are defined via the Modelling Object, no further modifications are required in the .dat file!

 

Kind regards.

 

Re: Buckling with pre-stressing Gravity LOAD (Dead Load)

Phenom
Phenom

aae, you saved my day!!!!

 

The solution you provided is the best I've ever seen in different ways.

1. It gives me the "illusion" that NX Simcenter can deal with problems that NASTRAN can solve but NX can't  control automatically. Now I assume that's not the only position in Solution were I can define customer declararions or control entries for the DAT file. I'll keep my eyes wide open next time when I look for a way to give NX additional key words!

 

2. That is  a very fine way to define a preload for buckling only one time and keep it for ever!

 

Thank you, aae and

@Karachun: Don't be disappointed for loosing the "Solution" button on your article. Your suggestion works also but not in that comfortable way. Nevertheless, thank you  once again.

 

Best wishes, Michael

| Production: NX10; Development: VB, TCL/TK, FORTRAN; Testing: NX11 | engelke engineering art GmbH, Germany
| Kudos for good posts! And if my post answers your question, please mark it as an "Accepted Solution".

Kick Off (2015-09-14)On the Map (2016-10-21)1st Gear 2018 (2018-02-19)To the Rescue (25, ‎2018-02-24)5th Birthday! (2018-02-25)Popular (100, ‎2018-10-30)Prolific (100, 2019-02-27)Philosopher (500, ‎2019-02-28)