Transient coupled mechanism and structure

[ccz90]

I want to analyze a system combine flexible structure with mechanism. The displacement of mechanism might be several order higher than structure. I know Algor supports static stress, MES, fatigue, ……,and thermal problem analyses. But how about transient coupled mechanism and structure? Can I use Algor to analyze the case in dynamic? Thanks!            

                                                                                                                                                                Ccz90

[xli]

Algor MES can do coupled analysis with transient thermal analysis results. The coupling is one-way though, i.e. temperature distribution changes do join into MES simulation as you MES mode temperature dependent.materials. But since thermal analysis is previous obtained, it does not count model reformation there. And MES job requires thermal analysis using same time steps. If deformation in your model is significant and that is what your interested, you might consider doing a sequence of such coupling analysis with manual update of your model deformations.

-xli

[S.LI]

Actually, the same load step is not necessary here. Also, after Algor 2011, the coupling simulation works even different meshes are used by thermal and MES.

[bjorn_fallqvist]

Hi, ccz90

What is it exactly that you mean by a coupled transient mechanism and structure analysis? Do you not mean a transient structure/thermal analysis, as mentioned by the posts above? I'm not quite sure what you are asking about.

Regards

Björn

[ccz90]

Let me try to describe my question in detail.

Assume a system combine flexible structure with mechanism. Some designers will neglect the coupling effect between the structure and mechanism. It might be the displacement of mechanism is several order higher than structure, and how to let the whole model simulation converge is not easy. The analysis processes are:

(Step1)Assume the structure has no deformation and just analyze the motion of mechanism. Find out if the design of the mechanism is ok or not.

(Step2) Assume the mechanism has no deformation , then use the reaction forces in step1 as boundary condition to analyze the structure parts. Find out if the design of the structure parts is suitable.

In fact, both the structure and mechanism have deformation and displacement when the system are in active. The motion of the mechanism will influence the deformation of the structure. Similarly, the deformation of the structure will also influence the motion of the mechanism. It’s a complex mechanism and structure coupling transient problem.

I want to know can I use Algor to analyze the whole system in dynamic just one model and one step?Thanks﹗

ccz90

[bjorn_fallqvist]

Hm, as far as I understand, what you want is some kind of coupled analysis where the structure and mechanism are analysed one after the other, with the results being transferred back and forth during the analysis?

I think an Algor employee might be better suited to answer this question, but as far as I have seen, there is no such function in Algor, unless you do it manually through several restart-analyses. I would guess MES is the best way to deal with this problem. Is there contact involved?

[John_Holtz]

ccz90,

MES stands for Mechanical Event Simulation. So for a majority of systems, it does not matter what the motions and deflections are. You simply create the model of the system (structure and mechanism in your case) and apply the known loads or motion.

[S.LI]

In Algor, there is an 3D element type called kinematic brick, which is designed to simulate rigid bodies.

Not sure if this element type could simulate your mechanism part.

If yes, you can handle the whole problem in one model.

Also,  Björn asked a very good question, is there any contact? In these kinds of coupling models, too many contact surfaces will cause converge issue.

[xli]

As my understanding, the original question is about thermal analysis coupled MES analysis. Algor does them through two relative independent packages so that there is NO two-way or integrated coupling between structural and thermal analysis. Period. The coupling mechanism is based on interpolating thermal results of "same model" for MES analysis. (I am sorry I forgot recently we removed some limit of same time step and exact same mesh etc. in interpolation in my previous post, thanks S.Li made up for me there).

To care of deformation of mesh caused changes in thermal analysis since both are time-dependent, you have to do a sequence of analysis with updated mesh. Several of us had mentioned this above. However, Algor currently does not have an integrated way of mesh updating. You have to do it by hands through some features in programs. And the update is not easy to be back to original geometry so that doing a re-mesh is difficult now. But if we consider mesh is not distorted, its geometry can be easily updated and start next step of analysis in the sequence. As far as I know, most industrial level package do practical simulations using similar way, some may have better tools to assistant updating and interpolation etc. There maybe some packages has fully integrated way to do such coupled analysis but probably they will lost many special features that rely on mesh in thermal alone or nonlinear analysis alone. Like MES contacts etc.

OK, I see the topic in this thread turned to MES modeling, which is about deformation vs motion. Like John mentioned, MES is designed to deal them together. Because this, MES is particular superior to deal models consider these two parts in same order. When you know your model/parts go large deformation or motion only, you can simplify model based on the fact. For example, as S.li mentioned, you may use kinematic elements (2D/3D) for the parts you can ignore deformation, that will greatly reduce your model size, whole part will have only 6 DOFs joined into analysis! This is another great feature for MES, not many other package offered.

Hope it helps,

-xli