ply stresses in analisi random

[boraborazz]

hi to everyone, from a few days I've been trying to solve a problem using patran-nastran but I don't understand how to do it.
the problem is just this: strength ratio in a random patran-nastran analysis. are at the basic level therefore I ask you forgiveness for any banality.

I did a random analysis using composite materials and got the various accelerations of rms, given a certain input psd.
My problem is to find the maximum tensions: in patran random analysis gives the possibility to get and display stress, but not for the various layers of the composite, but only for z1/z2. I would like to get stress for each layer, so I thought I would do a static analysis but I can't figure out how to do it: I thought I would apply the accelerations resulting from random analysis on a static model, but how should I do it? the bond conditions remain the same as the random analysis (i.e. 23456 at the base, leaving free x) or do I have to bind differently?
In this case I thought of applying the resulting accelerations, but perhaps I could apply the movements resulting from random analysis if you think it is easier.
I thought about the accelerations because in the .bdf is present the command param,nocomps,-1 and I do not want the movements to be calculated from the stresses (in doubt I used the accelerations).

in the quick reference guide of nastran I found some remarks useful to evaluate the tensions that I attach, although I must say I have not understood well with "composite lamina stress/strain are available in random analysis by xyplot command."

n.b: obviously if you change the command to param,nocomps,1 you could get stress for the various layers of composite, but in this case it is not possible to apply the input psd to finalize the random analysis, so it is not possible to do this.

 

[Onda]

nastran solves for travel, then calculates stress.
If you take a model, and apply a field of travel, then get the corresponding stress field. so the easiest thing to do is to apply the field of moving the first solution on a static.
p.s. why do you use triangular elements? are the worst in nastran, too rigid, you should avoid them

 

[boraborazz]

nastran solves for travel, then calculates stress.
If you take a model, and apply a field of travel, then get the corresponding stress field. so the easiest thing to do is to apply the field of moving the first solution on a static.
p.s. why do you use triangular elements? are the worst in nastran, too rigid, you should avoid them

Thank you for the answer.
I use triangular elements because to make mesh automatically in the contact zone between the two beams worked only with triangular elements I think. I will try to use other elements, thank you in the meantime.
I did as you said, applying the field of travel, and I got the stress field.
But I don't know if the procedure I followed is correct: I took off the disaster and applied the movements to the whole structure (even where there was the disaster). the maximum voltage I obtained following the analysis with a random stress along the x direction therefore is the following. Could it make sense to you?

n.b. I have two concentrated masses: one at the top end of the vertical beam, and one at the end of the horizontal beam.

 

[Onda]

if I understood the model well is in meters and kg so the results in pascal. so you have a peak of 1000 mpa. Maybe too much. .

 

[boraborazz]

if I understood the model well is in meters and kg so the results in pascal. so you have a peak of 1000 mpa. Maybe too much. .

I think I'm too big.
Perhaps the problem is to apply the field of travel where I have the masses concentrated at the end of the beams.
in random analysis I applied mpc rbe2 at two ends and to independent nodes I created two concentrated masses.
in static analysis I have different shifts for the dependent nodes, so I apply the field of shifting also to the dependent nodes, and I eliminate the mpc otherwise it gives me error.
I thought I'd split the concentrated mass along the knots at the end but I don't know if it's right, do you have any ideas?
Thank you.

 

[Onda]

I don't understand how you do static.
If the goal is to replicate the other analysis, you must download the field of shifting to the knots of all knots (except one, otherwise it does not turn) and apply these shifts, induced, to the whole model. In this way, you replise the tension state. the masses you must remove them if applied to rbe because you still have induced movements. Of course you have to take off the rbe. In this way the whole model is behaved as in the previous analysis and you can plot stress.