Welcome to the forums. Please post in English or French.

You are not logged in.

#1 2019-03-04 14:04:34

jeanpierreaubry
Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 3,616

crazy results with EFGE_ELNO on plate elements

hello

if i try to calculate EFGE_ELNO on plate element

i get no warning in CALC_CHAMP

when printing in a .resu file with

 _F(
            RESULTAT=stat,NOM_CHAM=('EFGE_ELNO',),
            INST=(4,),GROUP_MA=('plat',),
            FORMAT_R='1PE12.3',VALE_MAX='OUI',VALE_MIN='OUI',
        ),  

i get about 150000 lines of output with crazy values in an  non understandable encoding

what is wrong ?

jean pierre aubry

Offline

#2 2019-03-04 19:12:08

GXA_William
Member
Registered: 2016-04-30
Posts: 44

Re: crazy results with EFGE_ELNO on plate elements

Hello jean pierre aubry.

I cannot know for sure if the attached setup is similar enough to your case (to assess any errors) but it seems to work fine for DKT elements using AsterStudy, v13.6.

-William


Attachments:
DKT_EfgeElno.tar.gz, Size: 58.03 KiB, Downloads: 81

Offline

#3 2019-03-04 21:25:08

mecour
Member
From: Ostrava (Czech)
Registered: 2011-04-04
Posts: 132

Re: crazy results with EFGE_ELNO on plate elements

hello,

on a very simple plate I obtain results without errors. Attached files.
I tried on aster 13.4

mecour


Attachments:
Deska_pokus.zip, Size: 3.11 KiB, Downloads: 75

Offline

#4 2019-03-05 07:59:11

jeanpierreaubry
Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 3,616

Re: crazy results with EFGE_ELNO on plate elements

it looks like the trouble comes from mixing EFGE_ELNO for 1D and 2D element in the same result concept

if it has absolutely to be done that way this would be worth a warning
is not it?

this gives the trouble

stat=CALC_CHAMP(
	reuse =stat,RESULTAT=stat, 
	#some groups of 1D element
	GROUP_MA=('lxbot','lxtop','cvxz','vz','vzt','ly','cvyz','pied','atty',),
	CONTRAINTE=('SIPM_ELNO','SIPO_ELNO','EFGE_ELNO',),
);

stat=CALC_CHAMP(
	reuse =stat,RESULTAT=stat, 
	GROUP_MA=('plat',), #group of plate
	CONTRAINTE=('EFGE_ELNO',),
);

while this works nicely

stat=CALC_CHAMP(
	reuse =stat,RESULTAT=stat, 
	#some groups of 1D element
	GROUP_MA=('lxbot','lxtop','cvxz','vz','vzt','ly','cvyz','pied','atty',),
	CONTRAINTE=('SIPM_ELNO','SIPO_ELNO','EFGE_ELNO',),
);

statplt=CALC_CHAMP(
	RESULTAT=statplt, 
	GROUP_MA=('plat',), #group of plate
	CONTRAINTE=('EFGE_NOEU',),
);

Last edited by jeanpierreaubry (2019-03-05 09:55:28)

Offline

#5 2019-03-05 08:52:46

mecour
Member
From: Ostrava (Czech)
Registered: 2011-04-04
Posts: 132

Re: crazy results with EFGE_ELNO on plate elements

Is it possible to use twice the CALC_CHAMP  with the same resu and resultat  name? I think the second calc_champ have to overwrite the first one.

And my question. Why did you compute the results in two chalc_champ. When I combine the plate and beam I can compute results in one chalc_camp.

RESU=CALC_CHAMP(reuse =RESU,
                RESULTAT=RESU,
                CONTRAINTE=('SIPM_ELNO','SIPO_ELNO','EFGE_ELNO','EFGE_NOEU',),);

mecour


Attachments:
deska.zip, Size: 8.71 KiB, Downloads: 68

Offline

#6 2019-03-05 09:57:18

jeanpierreaubry
Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 3,616

Re: crazy results with EFGE_ELNO on plate elements

Is it possible to use twice the CALC_CHAMP  with the same resu and resultat  name? I think the second calc_champ have to overwrite the first one.

yes it is possible with the keyword reuse

And my question. Why did you compute the results in two chalc_champ. When I combine the plate and beam I can compute results in one chalc_camp.

this is my way of working
i like to write different CALC_CHAMP for different element types

obviously this makes troubles with EFGE_ELNO which is strange to me

Offline

#7 2019-03-05 10:26:25

mecour
Member
From: Ostrava (Czech)
Registered: 2011-04-04
Posts: 132

Re: crazy results with EFGE_ELNO on plate elements

jeanpierreaubry wrote:

yes it is possible with the keyword reuse

My mistake, it is obvious

When I try your way of two CALC_CHAMP in my structure I obtain the alarm

 <A> <CALCCHAMP_1>                                                 !
   !                                                                   !
   !  Le champ EFGE_ELNO est déjà présent dans la structure de données !
   !  à tous les instants demandés.                                    !
   !  Aucun calcul ne sera donc réalisé pour cette option.             !
   !                                                                   !
   ! Conseil :                                                         !
   !  Si vous souhaitez réellement calculer à nouveau cette option,    !
   !  créez une nouvelle structure de données.                         !
   !                                                                   !
   !                                                                   !
   ! Ceci est une alarme. Si vous ne comprenez pas le sens de cette    !
   ! alarme, vous pouvez obtenir des résultats inattendus !   

If I understand well, there is problem that it is not possible to write second EFGE_ELNO field to result structure, because there is the first one. So it is obviously necessary to use another result structure as you did in you second post.
This behavior is interesting from point of view of "reuse" keyword. And I will have to try if this behavior is the same also for another type of elements (2D and 3D for example)

Offline