Hello,

Below are my suggestions:

1. Is there any specific reason for the using 3-D model ? . why not use Axis-symmetic 2-D model using shell elements

2. Rather than measuring any specific element(s) , why not create a node set and then history output its displacment.

Cheers,

Prasanna

Mr.Gagnon,

I am getting some problem while doing FE analysis on Dog bone specimen ..can you help me ??

In fact, I cannot use any other type of model since no contact algorithm was yet well defined for axisymetric model in abaqus explicit (the version I used) and later the aluminum dog bone is to be replaced with composite shell, which canot be modeled adequately in my application.

As for displacement measurement, it is a good idea. I already as nodeset which measure those value, i should have thought of that. My problem is actually a lack of experimental data to compare with, but thank you for the anwsere.

I might be able to help for your dog bone analysis

Thank you Philippe,

My problem is:Dog bone model is failing in the fillet area (am getting max.stress in fillet region) but it should be in test region ...

am fixing my model one side and another side I am giving axial load over the surface of circular gripping region..

Fillet radius is 96 mm ..I am thinking that,this problem is because of meshing in fillet region...

pls suggest me some clues about the meshing of this type of fillets.

what type of material is used? (if you can give a clue) In fact if you modeled the fillet, you will have some problem because of the stress concentrations (you may have avoided rounds in the fillet geometry). For the fillet part, i suggest that you relly on machine component design. For typical fillet size, you can determine allowed forces considering the tensile stress area (ref. juvinall fundamentals of machine element design). When you have determined a basic thread diameter , model your specimen without the treads and apply a tie contact and/or a boundary condition, depending of an assembled model or not, on the side (since you know how strong is your specimen in the thread region). You will now be able to optimise your gage length and radius to have uniform deformations in the gage length section.

I hope I helped

My specimen is not having any threads ,I am going to use plain gripping section to hold in gripper..

The Fillet which I mentioned is between test section and gripping section..

I hope you got it now !!

Gage length should be according to Standards,so we can n't change it ..and I want to take exact dimensions of my experimental test specimen.

sorry for the misunderstanding (in french a thread is called fillet ...). Which standard do you use. I tested dog bone shaped composite specimen and found that breaking tend to occur near the fillet since shear stress are introduced as well. For isotropic material, this effect may be less important. So depending on the type of material, it may be a better idea to concider rectangular specimens.

it is ok Mr.Gagnon

my material is steel so I have to use Circular specimen..

my problem is while doing FE simulation..exactly like yours mine is also failing near fillet..

In the case of composites, a fillet cause shear stress, due to material orientation. I used to model dog bone shaped cylinder to use with tensile test of aluminum specimens. The trick is to use a fast fillet and a small cross section in the gage length. You can use a ratio between the big and small section of around 3. You can also use a small section decrease inside the gage length : ex: gage length D = 10 mm and center D = 10 - 0.1*D, with a slow regular slope. This will ensure a breaking in the middle. I hope it works. I did that and was working perfectly

Thank you Mr.Philippe

Re:You can also use a small section decrease inside the gage length?

Can you elaborate it some clearly?

for example: you decide to have a general section in the gage lenght of 10 mm diameter. At both end (near the fillets) of the gage length, you put a diameter of 10 but at the center of the gage length section you put a diameter of 9.9 (the center section is a bit smaller).

So how do we determine element characteristic length in Abaqus?!

Hi Philippe,

Did you figure out the answer of your questions? I also have the same problem of how to determine the characteristic length?

Hello,

Can anyone share how to find the element characteristic length in ABAQUS/Explicit?

It would be really helpful

Thanks and regards

Dhanraj Rajaraman