Weekly Puzzle #5

JMB365 · 2010-12-04 01:37:54

This week's puzzle involves using the geometry/meshes we have developed in the past to do a Non-linear FEA analysis so we can explore more of CodeAster.

Use the meshed pop / soda can (our mock chemical pressure vessel for acid storage!) of Weekly Puzzle #2 and perform a Non-Linear Static Analysis using an internal pressure of 2[atm] (2*1.01325e+5 [Pa]) while the 'Fixed' surface has DX=DY=DZ=0.

Find the maximum values of VonMises stress and displacement values (whereever they occur) from the analysis. Attach your *.py, *.hdf, *.comm and *.export files (in a zip or tar.gz format) along with a brief explantion of your methodology please.

Additional info: Most aluminum cola cans made in North America contain two different aluminum alloys. The can body is composed of aluminum alloy 3004 or 3104, whereas the top end of the can is aluminum alloy 5182. Aluminum 3004 contains approximately 97.8% aluminum, 1.2% manganese, and 1.0% magnesium (aluminum 3104 is very similar in composition). Aluminum 5182 contains 95.2% aluminum, 4.5% magnesium, and 0.35% manganese.

Some quick measurements with a digital calliper (double checked with a screw micrometer) show:

Side Wall thickness:    0.10 ~ 0.12 [mm] depending on location.
Bottom dome thickness:  0.30 ~ 0.35 [mm] usually seems to be thicker in the center
Top cap thickness:      0.22 ~ 0.27 [mm] usually seems to be thicker in the center

Use a median value for the thicknesses for the 3 zones.

You possibly can obtain the stress-strain curve for the Al can material from: http://hal.archives-ouvertes.fr/docs/00 … 03C739.pdf, or elsewhere.

If possible try to create an export file that will run unaltered in the /tmp directory, so it makes it easy for any of us to run studies for easy comparison. Thanks. Good Luck! Enjoy...

Regards, JMB

Last edited by JMB365 (2010-12-13 15:39:37)

rcbsimoes · 2010-12-13 15:21:28

Hi,

believe you meant to say 2 atm = 2 x 1.01e5 Pa.

rui

JMB365 · 2010-12-13 15:39:19

rcbsimoes wrote:

Hi,believe you meant to say 2 atm = 2 x 1.01e5 Pa.rui

Hello rcbsimoes, Yes you are right, I have corrected it. Thank you. Regards, JMB

Claws · 2010-12-16 08:10:22

I thought I'd post in the spirit of keeping the Weekly Puzzle theme alive: The weekend you posted the puzzle, I gave a valiant attempt at solving it, but I had to throw in the towel.

Applying 2 Atm internally to my version of the acidic fluid container created a while back, yields too big of a strain to possibly use shell elements. It is in particular the top surface that acts like a circular plate and has great deformation and strain.

Convergence cannot be reached without a large strain formulation like Simo/Miehe for shells. PETIT_REAC with small time steps wont converge either.

So Im keen to see a solution of other users - possibly just a small change in geometry should help the problem converge.

Regards,

Claus

Image: it is pretty obvious that it is the top surface that generates the greatest deformation - taken from the linear simulation.

Attachments:
linear depl.png, Size: 54.41 KiB, Downloads: 451

rcbsimoes · 2010-12-16 11:16:47

Hi,

i've been wating for a post too...

I cant get Stat_non_linear to work.
still i did a linear study and found same places for max stress and strain as Mr Claws.

bottom line, i'm posting to show my attention too.

rui

rcbsimoes · 2010-12-18 16:56:53

Hi all,

i'm trying to upload my file (as tar ou tar.gz or..) but I cant...

is there a size limitation?? is 2.6M too much? don't understand.

any way,
i haven't reach a valid solution (i think). looks like the can was blown apart.

used the SI (meters,kg, Newtons...and Pa) and the results were:
VALEUR MAXIMALE DE VMIS EST 2.30231E+10 EN
VALEUR MAXIMALE DE DY EST 1.81847E+01 EN

rui

ps- the files here:
http://caelinux.com/CMS/index.php?optio … 32&catid=4

Last edited by rcbsimoes (2010-12-18 17:09:49)

JMB365 · 2010-12-18 17:39:25

rcbsimoes wrote:

ps- the files here:
http://caelinux.com/CMS/index.php?optio … 32&catid=4

Hello rcbsimoes,

I am glad to see you attempting a solution. I tried to download your file but there is no meaningful file it. It may be corrupted since it is only 11091 bytes long as shown below:

File name: puzzle5-bc05456ffeb9635c085317945253a0ba.gz
File size:11091 bytes

Would you try uploading again please?
Regards, JMB

rcbsimoes · 2010-12-18 17:42:42

one more try.

Attachments:
puzzle5.tar.gz, Size: 10.83 KiB, Downloads: 637

rcbsimoes · 2010-12-18 17:44:31

i've only uploaded the mess, resu and comm.

now i am going (try) to upload the mesh.

rui

.... I cant!
can i send it to an email?

Last edited by rcbsimoes (2010-12-18 17:45:52)

JMB365 · 2010-12-18 19:21:38

rcbsimoes wrote:

i've only uploaded the mess, resu and comm.

Hello rcbsimoes,

Is it possible to zip (or tar+gzip) just the med file and upload it here or CAELinux? How big is the *.med file after it is *.tar.gz (with compress turned on)? I would like everybody to be able to access it, so sending it by email would not help others. I was able to successfully download the 3 other files. So give it another try. Good Luck. If you are still having trouble let us know.

Regards, JMB

JMB365 · 2010-12-18 19:33:48

Hello all,

Here are some interesting facts about the internal pressure in a soft-drink (soda/pop) can that makes interesting reading and gives us a reference for pressures and the reason(s) for the carbonation.

See: http://hypertextbook.com/facts/2000/SeemaMeraj.shtml

Enjoy! JMB

rcbsimoes · 2010-12-19 00:12:49

hi,

after compression (.tar.gz) still has 9.8M, so..

i'll try in caelinux.

rui

rcbsimoes · 2010-12-19 00:20:54

hi,

it didnt work in caelinux either.

for those who may want to take a look, just drop me an email.

rui

JMB365 · 2010-12-21 00:10:40

Hello rcbsimoes,

I received your emails with all the files attached. I will try to review your study as I get the time. I don't know if I will be able to do so this year or not. Thank you for your submissions.

Regards, JMB

Last edited by JMB365 (2010-12-21 00:10:56)

#1 2010-12-04 01:37:54

Weekly Puzzle #5

#2 2010-12-13 15:21:28

Re: Weekly Puzzle #5

#3 2010-12-13 15:39:19

Re: Weekly Puzzle #5

#4 2010-12-16 08:10:22

Re: Weekly Puzzle #5

#5 2010-12-16 11:16:47

Re: Weekly Puzzle #5

#6 2010-12-18 16:56:53

Re: Weekly Puzzle #5

#7 2010-12-18 17:39:25

Re: Weekly Puzzle #5

#8 2010-12-18 17:42:42

Re: Weekly Puzzle #5

#9 2010-12-18 17:44:31

Re: Weekly Puzzle #5

#10 2010-12-18 19:21:38

Re: Weekly Puzzle #5

#11 2010-12-18 19:33:48

Re: Weekly Puzzle #5

#12 2010-12-19 00:12:49

Re: Weekly Puzzle #5

#13 2010-12-19 00:20:54

Re: Weekly Puzzle #5

#14 2010-12-21 00:10:40

Re: Weekly Puzzle #5

Board footer