• This forum is the machine-generated translation of www.cad3d.it/forum1 - the Italian design community. Several terms are not translated correctly.

ansys joint glued

alessandro genova

Guest
Hello.
I am a new user of this forum, my name is alessandro and I am a student of the last year of mechanical engineering, in recent years I have worked enough with proe and matlab and lately I am facing ansys.
work with ansys 14 in classic version and I find problems in the following exercise:

a joint glued by simple overlap is formed by two aluminum foils (identical among them) glued for a length l. then I provided the data of the thickness of the foil and glue, tangential elastic module g glue and elasticity and material. These two plates are drawn by a force f.
the purpose of the exercise is to calculate the tangential and normal tension in the medium layer of glue.

I tried to define the problem with the target and contact elements, my problem at this point are the constraints, if I put the bonds between glue and aluminum the tension turns out zero, if I do not put them infinite shift.

Does anyone have any suggestions?

Thank you for your attention

alessandro
 
Bye.
personalment I had to deal with a problem similar to yours.

I shaped the glue as an elastic foundation ie as a bed of springs between coin knots without using "strange" elements as targets and similar.

I had to do with two panels (modeled as shell181) that were glued along a common area.

Now I don't know if this is your case.

If I can help you well!
 
thanks for the idea, but at that point I do not know how to find the tension in the middle layer of the glue (which would be modeled as springs. . )

Anyway, I think about it.
 
I would not say baggianate but at the end the course of the glue should be symmetrical compared to the half-works, maybe there is nothing to do with what you have to do, but it is an input!!!
 
you can't make three solid elements: two for plates and one inside for glue that you will then cut with a special plan to analyze the performance of tangential tau?
with springs how do you find an equivalent stiffness?
 
So you say you leave the way to define aluminum as target(169) and glue as contact(171), but to shape the adhesive as defined in the link you sent me and impose constraints between the rigid element of metal and the rigid element of glue? Right?

Thanks again
 
nicnoc said:
you can't make three solid elements: two for plates and one inside for glue that you will then cut with a special plan to analyze the performance of tangential tau?
with springs how do you find an equivalent stiffness?
Click to expand...
In fact it is what I tried to do, defining the glue element as contact and aluminum as target, but then when I apply the force opens the system, (infinite ux shift) and if the bond is zero.

If you want to train the .txt

end
/Cycle

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

f=40000!
s=4e-3!
c=0.25e-3! thick glue (m)
l=40e-3!gluing length (m)
b=100e-3!width glued part (m)
ee=70e9!module of young aluminium (pa)
gg=1.4e9!module g of glue (pa)
polka dots=0.37!
thensall=0.33!

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

ec=2*gg*(1+poiscol)! young glue module (pa)
ainc=l*b!gluing area (m^2)
sez=s*b!lamine section (m^2)
pre=f/sez!pression applied on the sections of the lamines
eps = 1 and 6

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

!
!
(b*sinh(w*l))
!

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

/prep7
et,1,42,,3
et,2,42,,3

r,2,b
r,1,b

mp,ex,1
mp,ex,2,ec
mp,nuxy,1,poisall
mp,nuxy,2,poiscol

-----------mesh----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

type,1
No.1
2/2,2
2/2.1.2
Lesi,7,,200,20
Lesi,5,200,1/20
Lesi,1,200,20
lesi,3,,200,1/20
Lesi,6,, 20,20
Lesi,4,, 20,20

mshk,1
amesh,all

type,2
6/2,6
Lesi,10,,4
Lesi,9,,320

mshk,1
amesh,all

!--------------- Target Elements

et,4,169,1
type,2
real,1
lsel,s,line,5
I'm sorry.
♪

et,4,169,1
type,2
real,3
lsel,s,line,3
I'm sorry.
♪


...-------------------- contact information.

et,3,171
type,3
keyo,3,12.5

real,2
♪
No, no, no, no.
♪
esurfing
♪

real,3
No, no, no.
No, no, no, no.
♪
esurfing
♪


-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

lsel,s,line,2$ lsel,a,line,8
♪
d,all,uy
♪

nsel,s,loc,y-c/2-eps,-c/2+eps
No, no, no, no.
d,all,all!
♪

nsel,s,loc,y,c/2-eps,c/2+eps
No, no, no, no.
d,all,all
♪


!--------force application------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

lsel,s,line,2$ lsel,a,line,8
♪
the internal pressure is positive
/psf,pres,,2! shows pressure with arrows
eplot
♪

end
/solu

No, 20!
autots,off! turn off the "automatic time-stepping"
Nlgeo,off! large shifts activated
outre,all,all! solution saved in all substeps


The
end

/posts
plns,s,xy

♪
nfin=10047
ntot=(nfin-nin)/3)+1
path,useless,ntot

*do,i,nin,nfin,3
(i-9087)/3
ppath,j,i
*end

pdef,tauxy,s
pdef,tennorm,s,y

/grid,1
/grop,view,1
/plo,frame,off
/plo,leg1,off
/plo,leg2,off
/axlab,x,length(m)
/axlab,y,tension (pa)
/title, tang voltages (pa)
plpa,tauxy,tennorm
 
If you do it in workbench, you're very short... I am not very skilled with ansys classic.. .
 
if you model the two plates with matching knots you don't have to put any rigid element but only insert the 3 stiffnesses kx,ky,kz along the 3 directions.

the rigid bond you see in the link only serves in case the two plates are not coinciding knots.
If you, originally, create the flat knot plates (with an offset equal to the thickness of the plates) you can tmit only 3 stiffnesses.

I don't know if I was clear.

ps:I should also have a model created with ansys 12.1 where there is the case of two aluminum plates glued along a common area for some thickness. If you care, let me know... But first try:)
 
nicnoc said:
If you do it in workbench, you're very short... I am not very skilled with ansys classic.. .
Click to expand...
I know, but unfortunately I don't have ansys workbench and then we were asked to solve it with the classic... :
 
volaff said:
if you model the two plates with matching knots you don't have to put any rigid element but only insert the 3 stiffnesses kx,ky,kz along the 3 directions.

the rigid bond you see in the link only serves in case the two plates are not coinciding knots.
If you, originally, create the flat knot plates (with an offset equal to the thickness of the plates) you can tmit only 3 stiffnesses.

I don't know if I was clear.

ps:I should also have a model created with ansys 12.1 where there is the case of two aluminum plates glued along a common area for some thickness. If you care, let me know... But first try:)
Click to expand...
ni, i.e. ok I can model the plates with the matching knots, and until they are there, but then how do I define the values of the three rigidities in ansys classic having only as data the module g and the ratio of poisson of the glue?

thanks again to all
 
just follow the calculations here:
http://femci.gsfc.nasa.gov/adhesive/of course you will have a long rigidity x and y (cut-shear, if you models the plates in the x-y plane) and a long stiffness z (peeling).

If you don't understand, I have to send you an extract from my thesis:-)
 
volaff said:
just follow the calculations here:
http://femci.gsfc.nasa.gov/adhesive/of course you will have a long rigidity x and y (cut-shear, if you models the plates in the x-y plane) and a long stiffness z (peeling).

If you don't understand, I have to send you an extract from my thesis:-)
Click to expand...
Okay thank you so much now I am, I understood how to shape the glue element through the use of springs, but then how do I calculate the average tension on the glue plane? (always with classic ansys)

Thanks again
 
Good question:-)
I don't remember now that my "committee" came out of this.
I think you should go through the coincidence knots that describe the behavior of glue
 
My passionate advice is to use a nice "bonded" contact.
you have maximum freedom of mesh, the setup of the contact is easy, the results, in terms of tangential (shear) and normal (peeling) efforts, are easily graphicable. I add more than, writing a few lines in apdl, you can simulate the phenomenon of debonding.
Alternatively you can use interface elements.
I would avoid the springs.. .
 
stefano.garbin said:
My passionate advice is to use a nice "bonded" contact.
you have maximum freedom of mesh, the setup of the contact is easy, the results, in terms of tangential (shear) and normal (peeling) efforts, are easily graphicable. I add more than, writing a few lines in apdl, you can simulate the phenomenon of debonding.
Alternatively you can use interface elements.
I would avoid the springs.. .
Click to expand...
with bonded contact, you mean for example a 171 count with keyopt(12)=3 or 5, right?
I before trying to do it with the springs I tried to do it exactly so, but then the problem arises when I apply the force that opens the whole system and gives me how error an infinite shift... How can I do it for you?

Thank you.
 
Let's start from afar.
1st question: are the aluminium sheets represented by two solids or two surfaces?
if they are surfaces do well attention to how normal are oriented: must be directed towards each other.
2nd question: are you using the "contact tool"?
 
stefano.garbin said:
Let's start from afar.
1st question: are the aluminium sheets represented by two solids or two surfaces?
if they are surfaces do well attention to how normal are oriented: must be directed towards each other.
2nd question: are you using the "contact tool"?
Click to expand...
1 sheets are two surfaces with the normal facings one towards the other
2 no I am not using the contact tool but input with the .txt
 
Good morning.

I'm new in the use of ansys and I wanted to ask if there are any quick ways to simulate whole laminates.

Thank you.
 

Forum statistics

Threads
44,997
Messages
339,767
Members
4
Latest member
ibt

Members online

No members online now.
Total: 12 (members: 0, guests: 12)
Back
Top