If possible please send the current picture of the mesh after motion and the various parameters which you have used.

I would suggest to use axisymetric model.

The stress is tensile or compressive?

Hi Mr. Niazi, 

The model is also axisymmetric and the force is compressive.

Im not sure but it seems as element locking problem. Did you try quadratic elements (with mid side nodes)?

The internal and external threads have same material properties? What are the axial B.Cs. An axial load is applied on the edge of internal threads and the edge of external threads are constrained OR is it something else?

I also tried the quadratic elements but nothing has changed. The triangular elements has midside nodes.

No, the external thread has 20 GPa and the internal screw thread has 100 GPa so its different material.

Axial load is applied to the  screw head (it's an edge as it's 2D). 

The interface has frictional and symmetric contact type.

Thank you very much for the responses by the way.

I see something peculiar in the mesh. The nodes of screw and thread are coinciding at the interface. How did you make the two bodies and how did you mesh these? The two bodies must not share nodes at the interface.

The surface bodies are created with NX and the mesh is generated in ANSYS Workbench 16.2. The mesh has no special features but the refinement on the screw. 

But why must not the bodies share the same nodes?

For the bodies to move relative to each other, the bodies must not have common nodes at the interface. It is possible to have two nodes on top of each other; one node belonging to first body while thd other to second. Furthermore in standard threads there is always a clearance between internal and external threads which is not present in your case and thd thread shape is also not standard. It is very difficult to make/manufacture such threaded connections in real life.

I will try to seperate those nodes Mr. Niazi, thanks for valuable advices. I see your point for clereance but in our model we really don't have clearance due to content of project. And yes you are right it is hard to manufacture it, but it is just a prototype to see the problems and solve them in a quick way. Normally the threads will be helix and not in this shape. Thanks again for allocating time. 

So finally, if I can achieve to seperate the nodes and extinguish the singularity problem, I will definetly inform you. 

Ok. But be careful, changing the shape of threads will change the location and magnitude of stress concentration. You have tapered threaded connection, this also influence the stress concentration location and magnitude.

Mr. Niazi, here are the good news. By seperating the nodes on the bodies, I soon overcame the singularity problem. Thanks for the suggestion. It will help a lot for our project. I am sending the new images.

So the shared nodes are not proper for static structural analysis am I right ? Maybe they can be preferred for CFD analysis ?

Im glad to hear that. CFD is Eulerian approach where the mesh remains fixed and material flows through it. Solid mechanics is (mostly) Lagrangian where the mesh moves with the material.

There is also another approach called ALE which is partly Eulerian and partly Lagrangian.

Dear Sir,

If possible please send the current picture of the mesh after motion and the various parameters which you have used.

I recommend using the hybrid mesh instead of auto meshing because hybrid mesh approach will give more control on the creating nodes in the critical regions where different surface parts are joining. 

If possible please send the current picture of the mesh after motion and the various parameters which you have used.