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

bolt modeling

  • Thread starter Thread starter Luca Carciofi
  • Start date Start date

Luca Carciofi

Guest
Hey, guys.

I need a little help to shape a bolt in a plate.
I have to model the bolt shaft with the link11 element in ansys.
then I have to connect the two knot pairs to the surface of the plate so that they go to simulate the two ends of the bolt that coming into contact with the plate create the pretension.

How can I combine the two knots of the link11 element with the surface of the plate as it was done in the photo below?

I'll put a design on you because there's someone on the web who's already done it, so you understand better.

The purple part is the link11 element and the green parts are the elements I should create.

hello guys and thank you so much!!
 

Attachments

  • prova.webp
    prova.webp
    59.2 KB · Views: 25
to do what you posted in the attachment you have to create the link11 element.
then for each of the 2 knots of extremities create "coupled sets".
this means that the knots of the upper surface will be numerically bound to have the same degrees of freedom (you end as: ux, uy, uz or all) of the top node of the link11.
This way of modeling bolts dates back to the 1980s. now you have the possibility to explicitly draw the bolt and introduce the "bolt-pretension".
 
Oh thanks stefano, you are always punctual and precise.

you are completely right, but when analyzing large assemblies it is impossible to model the bolt with the "classic solid modeling" for reasons of cpu time.
do you know other methods of modeling?

As for the method I presented, do you think it's over? do you know if this method provides good simulation results?

to create this type of bolt I have to create a link11 element that acts as a pin, then later I connect the two knots of the link 11 element (extreme) to the surface of the plate with the coupled sets element, right?
 
The use of the coupled sets along with the link11 is actually an elegant and lightweight way to handle matches with many joints. if this is your need then the road is right.

Yes, you create two knots, you piss a link11 and then go to couple the degrees of freedom.
 
Hi, stefano, I have to ask you one last question.

how do I select the knots that are located on one surface and a single knot that is in another surface?
Can you do that? to select the nodes of a surface I have no problem but when I go to re-check the individual additional node of the other surface the others are unchecked.

I hope I've been clear.

hello and thanks again
 
study the tutorials.
in selecting any item there are 4 options (when I write item I mean item, node, keypoint, line, area, volume)
1-select
select one or more elements from the whole set of elements, whether they have previously been selected, or not. Previous selections are erased.
It's like creating a new set.
2-reselect
select one or more elements from the set of the previously selected elements, either no. previous selections are reduced.
It's like extracting a subaxeme from a previous set.
3-alse select
select one or more elements from the whole set of elements and add it to those already selected.
4-unselect
deseleziono one or more elements from an already selected set.
 
I'm sorry if I break you again.
I finished the model. I should be there. It's a simple plate with two bolts represented by the elements we talked about above.
in the plate I should calculate stiffness by applying a bond in one side of the plate and a shift of 0.00015m in the other side. with formula f=k*(deltax)
The plate was divided into two for symmetry.
I've launched the solution, but I've got something weird, I don't understand, it doesn't make me wrong, but the solution is really weird.
I'll send you the link file, if you can take a look at it, I'd really appreciate it.
ps: what do I pay you for these advices?

link file http://dl.dropbox.com/u/18523175/couple bolt model ansys classic.zip
 
hi stefano, I ask you one last favor, you could take a look at this file.
I am super stuck and I can't move on, I managed to shape the bolt with the emtento link180 (not the link11) then I applied a load to the sides of the two plates and the other side of the plate I bound it.
I also applied the pretension on the two bolts.

But the solution makes me mistake, you could take a look at it, if you don't have any way you don't worry enough that you tell me you can't open it so I ask someone else.

Thank you very much anyway.

hello stefano good end
 
wrong to create the pretension section.
You do that.
1-preprocessor-sections-delete section and deletes the 4 sections you created
2-preprocessor-sections-pretension-pretensn mesh-elements in line
3-select the line of the first bolt, then select a knot about half the line. opens a window that asks you secid section number and put 1, as name you put what you want and as direction select kdir y-axis. click OK.
4-fold as above for the second line, secid number 2.

then.. .

5-solution-define loads-apply-structural-pretnsn sectn. a green window will open. select everything (2 rows), enter the pretension force value (force) or preload in terms of movement (relative motion) and click ok.

so it should go.
 
There's something wrong because when I launch the solution tells me that the shift of a bolt knot in the direction of force is too high ( error=the value ux is "too high value") and here I stop the solution:

I have some questions because I want to know where I'm wrong:

1) in the window that appears to me of celestial color, when I have to insert the value of the precarious I have inserted the following things:

loading 1
apply at ls=1
force = 7300
lock at 2

are the values that I entered after the same? in this way should first apply the value of the preload then external loads? If I enter only the force tells me I can't do it.

2) when I go to define the set of coupled dofs command asks me to write:

n set reference number= what should I insert here?
lab degree of freedom label

I would like the shifts of the bolt end node to be equal to all other nodes connected. Okay?
what properties have the lines connecting the knots? Are they rigid lines?

3) on real costant ho:

set 1
set 5
set 6

with all the same elements? Why do I have set five and six?

4) in the real constants of the link180 element the window appears to me:

element type reference n2
real constant
cross sectional area= 19 mm (bolt area)
add mass/lenght=0.000000273 ton/mm4(I split the fract bolt mass its length
tension only

--------

Are there any mistakes?
What are you talking about?
 
1) in the window that appears to me of celestial color, when I have to insert the value of the precarious I have inserted the following things:

loading 1
apply at ls=1
force = 7300
lock at 2

are the values that I entered after the same? in this way should first apply the value of the preload then external loads? If I enter only the force tells me I can't do it.
Click to expand...
split your analysis into 2 step. in the first step precarise bolts, in the second apply strength to the plate. the values that are indicated for the preload (lock at 2) are fine.
2) when I go to define the set of coupled dofs command asks me to write:

n set reference number= what should I insert here?
lab degree of freedom label
Click to expand...
each coupled set has a reference ID. in your case will have to be 4 (id=1, 2, 3, 4), each for each face of the screws/bulbs.
I would like the shifts of the bolt end node to be equal to all other nodes connected. Okay?
what properties have the lines connecting the knots? Are they rigid lines?
Click to expand...
you are infinitely rigid "traves".
3) on real costant ho:

set 1
set 5
set 6

with all the same elements? Why do I have set five and six?
Click to expand...
study the tutorials.
4) in the real constants of the link180 element the window appears to me:

element type reference n2
real constant
cross sectional area= 19 mm (bolt area)
add mass/lenght=0.000000273 ton/mm4(I split the fract bolt mass its length
tension only
Click to expand...
Why tension only? Are you shaping a cable or a beam?
 
In theory, I'm modeling a beam because I'm putting bullion in high blood pressure, so it's like it's live.
 
Guys still have a few questions. .

I can't change section of my link 180 element, ie when I go to see the section I show it square, I would like to round it, can this do this thing or for the link element remains square? ps the link element does not curve, does it? while the beam element can be curved or is it wrong?

another small problem, I have to connect the end node of an element link along with the surface of a shell element, I practically have to bind the end node of the element link in the surface of the plate, you know with what command to do it and how?

a last question..I would like to use a spring(molla) element that had a rigidity matrix formed by 6 components (long translation x,y,z and long rotation x,y,z). in msc.patran this element is called cbush, know what element to use in ansys?

thank you so many guys, I wait for your advice
 
then the link element is type rod, then compression-traction, while the beam4 element can be flexed.

As for the spring element I am using a combin14 element similar to a bed of springs.
 
I would need a single spring, with a 6x6 rigidity matrix (kx,ky,kz,k rot x,k rot y,k rot z), with conbin14 do I have a chance to use this matrix?

I've heard that there's a chance to use conbin14 as a spring and matrix27 for the rigidity matrix. .know how they use together?
 

Forum statistics

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

Members online

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