modal analysis

[ygna]

I explain briefly how I shaped the church (in mm)
structure:
- Young module: 180 m
- polka dot coefficient: 0.25
- density: 1.8^-09 ton/mm3
element solid65
cover:
- module of young in parallel direction to the axis e1 = 11700 mpa
- module of young in orthogonal direction to the axis e2 = e3 = 830 mpa
- cutting module g12 = 700 mpa; g13 = g23 = 70 mpa
- coefficient of poisson υ13 = υ12 = 0.42; υ23 = 0.35
- density ρ: 0.4^-09 ton/mm3
solid element
constraints to the base define loads– apply – structural –displacement – on nodes null shift in all directions.
for coupling/ceqn beams – couple dofs for each beam I selected six faces (maybe I had to insert the trolleys in all areas in contact with the structure).
for the analysis freqb 0 ; freqe 0 ; nrmkey to mass
the null frequencies are much but much more than 6..
I don't know how to do

 

[luigi.paiano]

I explain briefly how I shaped the church (in mm)
structure:
- Young module: 180 m
- polka dot coefficient: 0.25
- density: 1.8^-09 ton/mm3
element solid65
cover:
- module of young in parallel direction to the axis e1 = 11700 mpa
- module of young in orthogonal direction to the axis e2 = e3 = 830 mpa
- cutting module g12 = 700 mpa; g13 = g23 = 70 mpa
- coefficient of poisson υ13 = υ12 = 0.42; υ23 = 0.35
- density ρ: 0.4^-09 ton/mm3
solid element
constraints to the base define loads– apply – structural –displacement – on nodes null shift in all directions.
for coupling/ceqn beams – couple dofs for each beam I selected six faces (maybe I had to insert the trolleys in all areas in contact with the structure).
for the analysis freqb 0 ; freqe 0 ; nrmkey to mass
the null frequencies are much but much more than 6..
I don't know how to do

Listen and these couple dofs why did you put them?

 

[ygna]

as a bond between structure and cover (of trolleys in the three dimensions)

 

[ygna]

*** index of date sets on results file *****

set time/freq load step substep cumulative
1 0.0000 1 1 1
2 0.0000 1 2
3 0.0000 1 3
4 0.0000 1 4
5 0.0000 1 5 5
6 0.0000 1 6
7 0.0000 1 7
8 0.0000 1 8
9 0.0000 1 9
10 0.0000 1 10
11 0.0000 1 11
12 0.0000 1 12
13 0.0000 1 13
14 0.0000 1 14
15 0.0000 1 15
16 0.0000 1 16
17 0.0000 1 17
18 0.0000 1 18
19
20 0.0000 1 20 20
21 0.0000 1 21
22
23 0.0000 1 23
24 0.0000 1 24
25 0.0000 1 25
26 0.0000 1 26
27
28 0.0000
29 0.0000 1 29
30 0.0000 1 30
31 0.0000 1 31
32 0.0000 1 32 32
33
34 0.0000 1 34
35 0.0000 1 35
36 0.0000 1 36
37 0.0000 1 37
38 0.0000 1 38
39 0.0000 1 39
40 0.0000 1 40
41
42 0.0000 1 42
43 0.0000 1 43
44 0.0000 1 44
45 0.0000 1 45
46
47
48 0.0000 1 48
49
50 0.0000 1 50
51
52
53 0.0000 1 53
54 0.0000 1 54
55 0.0000 1 55
56 0.0000 1 56
57
58 0.0000 1 58
59 0.0000 1 59
60 0.0000 1 60 60
61
62 0.0000 1 62
63
64
65
66 0.0000 1 66
67 0.17291e-04 1 67 67
68 0.25503-04 1 68
69
70 0.26596e-04
71 0.32192e-04 1 71 71
72 0.33448e-04 1 72 72
73 0.41445e-04 1 73
74 0.41651e-04 1 74
75 0.45474e-04 1 75
76 0.46767-04 1 76
77 0.50656e-04 1 77
78 0.50666e-04 1 78
79 0.59264e-04 1 79 79
80 0.61966e-04 1 80
81 0.62320-04 1 81
82 0.67904-04 1 82 82
83 0.75094e-04
84 0.75327e-04 1 84
85 0.77494-04 1 85
86 0.77828-04 1 86
87 0.84389e-04 1 87 87
88 0.84586e-04 1 88 88
89 0.10050e-03 1 89 89
90 0.10796e-03 1 90 90

ecco cosa mi attend su results summary

 

[luigi.paiano]

as a bond between structure and cover (of trolleys in the three dimensions)

Can't you put classical constraints without using coupled dofs? (maybe that's why, in modal analysis they can give problems)

 

[Onda]

without doubt the problem lies in the constraints between the components of your structure.
explain to me how materials you're modeling?
what has a stiffness of 180mpa and a density of 1800kg/m^3?
and a stiffness of 11700mpa and a density of 400kg/m^3?

the easiest way to combine different parts is to meshare with matching knots. eliminate all couple dofs and solve the problem.


(if you are able to make a congruent mesh! ! )

 

[ygna]

stone and wood as cover I am sorry but meshare to coincidence knots I do not know what it is but if I do so I do not change the behavior of the structure?

 

[Onda]

stone and wood as cover I am sorry but meshare to coincidence knots I do not know what it is but if I do so I do not change the behavior of the structure?

Meshare to matching knots means that the knots of the wooden beam in the contact points with the stone correspond to the knots of the stone.
to do so I think you have to divide the surface of the stone and make it common to both solids. I'm sorry but I don't use ansys and I can't tell you more.

with regard to your model, meshando to simple coincidence knots consider perfectly solid the beam with the stone, and it is about the same thing that use coupling dofs, only simpler, since these are the cause of your error.

the model must behave like a unique body, and to see from the results does not.

first thing, check if the weight of the structure returns.
then delete all beams and see how they come out. without beams the values must be low but not null.
If they were still null it would mean that you have problems in the masonry mesh, otherwise the problems are only in the connection between masonry and wooden beams.

a question, since you entered the wood as an orthotropic material, did you make a reference system for each beam and define a property for each beam?

 

[ygna]

No... now I feel what you said

 

[Onda]

No... now I feel what you said

and how did you give the orthotropic properties to the beams so oriented in space?

 

[ygna]

I didn't! What a mess but it's not automatically excused but it's the first time I've been doing such an analysis and more with that program. .

 

[Onda]

I didn't! What a mess but it's not automatically excused but it's the first time I've been doing such an analysis and more with that program. .

Automatic! ! !

But nobody's following you? ! ?

you must define for each beam the correct properties with the correct orientation of a reference system for each beam, or beam group if they have similar orientation.

However, for the moment, you can also give it homogeneous property isotrope to wood, on the behavior of a beam, it changes little. just the correct cutting module.
for the moment you have to run the model trying to remove lability.

 

[ygna]

no the prof who should follow me for the thesis does not know how to use the program!!! Now I will try to solve these problems thanks so much to all of you!!!

 

[luigi.paiano]

no the prof who should follow me for the thesis does not know how to use the program!!! Now I will try to solve these problems thanks so much to all of you!!!

cmq I would remove beams from modal analysis (to make a first test).
As for the problem raised by the engineer unfortunately contacts are not as simple as you might think.for this problem it takes stefano.
However structures like this I would do them with ansys wb that allows to set contacts in a much easier way.
You still want my opinion? I would treat the structure without beams, then I would do another analysis only with beams with appropriate constraints. It is already something. .

 

[ygna]

I tried without beams and the deformed are more truemili cmq I have to find a solution with the cover because the prof wants it now I try for the umpteenth time to re-enter the couple dof and I see...ps the areas and therefore the knots to be selected must be escusively belonging to the beams just? because in the model I have seen that there are overlapping areas (a mistake of those who made the 3d in autocad)

 

[Onda]

cmq I would remove beams from modal analysis (to make a first test).
As for the problem raised by the engineer unfortunately contacts are not as simple as you might think.for this problem it takes stefano.
However structures like this I would do them with ansys wb that allows to set contacts in a much easier way.
You still want my opinion? I would treat the structure without beams, then I would do another analysis only with beams with appropriate constraints. It is already something. .

I would be careful to use contacts in a modal analysis.
contacts are by definition non-linear and cannot be resolved in a modal analysis.
some software (including ansys) allow, however, to use the "glued" contacts in modal analysis. you need to be sure of what you do and the settings you use. Otherwise we risk introducing new elements of error.
the cleanest method is meshare to congruent knots in the contact point. with this system you have no problem.

 

[luigi.paiano]

Why don't you do it in ansys workbench? you just need to draw the structure with a cad (cad, solidworks). results are good and save time.

 

[luigi.paiano]

I would be careful to use contacts in a modal analysis.
contacts are by definition non-linear and cannot be resolved in a modal analysis.
some software (including ansys) allow, however, to use the "glued" contacts in modal analysis. you need to be sure of what you do and the settings you use. Otherwise we risk introducing new elements of error.
the cleanest method is meshare to congruent knots in the contact point. with this system you have no problem.

Yeah, that's why I was talking about ansys wb. recognizes the contact between two surfaces, creates an appropriate mesh automatically and the modal analysis proceeds smoothly. I hardly use classical ansys in these problems (except if the problem is relatively simple) and I can assure you that the results are also excellent for complex structures (I have made a thesis on a very complex robotic arm and experimentally things coincided)

 

[ygna]

another girl was working with wb and the prof let her pass everything on that classic... you see that she doesn't know what she's talking about!. I'm sorry, but I'm very slow because I have a nine-month-old baby who takes me most of the day.

 

[Onda]

another girl was working with wb and the prof let her pass everything on that classic... you see that she doesn't know what she's talking about!. I'm sorry, but I'm very slow because I have a nine-month-old baby who takes me most of the day.

Actually, it's not like the professor doesn't know what he says.
wb is just a user interface that eases the use of fem.
the solutor is exactly the same between wb and classical. In fact, with the classic you have greater control of the solutor but at the cost of setup problems of the analysis.
Of course, from the educational point of view, it is much better to use classical, because you understand better what the fem does and for what reason.