#### Topic: Negative crack width

While analyzing the crack behavior, I am getting negative crack width values for cod1 of my specimen.
What does this negative crack width mean ? Does it have any physical significance?

#### Re: Negative crack width

Dear arjun.vithalkar,
closed cracks can have negative opening, because the numerical stiffness acting against overlapping while closing can not be infinite.
Regards.

#### Re: Negative crack width

Thanks for the clarification.

I need to ask one more thing about the interface material model.

-Is the thickness (specified in the connection input of line properties), related to the thickness of interface in y-direction or into the plane of model (along z-axis) (If I'm modelling in 2D ATENA and x-axis is my longitudinal axis)?

I ask this because, if I'm not wrong, while defining the macroelements, the thickness refers to the width of the specimen (that is thickness into the plane of model, i.e, along z-axis), as dimensions for the macroelements are assigned separately while defining the number of joints.

Therefore, these two values of thicknesses should be always different, because interface thickness will be very less as compared to width of adjacent macroelement.

#### Re: Negative crack width

Dear arjun.vithalkar,
the Thickness in the 2D Interface is the same as in all other 2D material models/Macroelements, i.e., the dimension in the 3rd direction - perpendicular to the screen.

Regards.

#### Re: Negative crack width

So, does that mean, if the thickness of interface is smaller than the adjacent macroelement (say interface th. = 100mm and macroelement th. = 400mm), it denotes that only a small percentage of the two macroelement layers are connected to each other (in this case only 25%) ?

If yes, then what should be the optimum percentage of interface thickness in relation with the adjacent macroelement thickness?

#### Re: Negative crack width

Dear arjun.vithalkar,
yes. For your second question, I am not sure if I understand, but the most typical setting is to make the interface thickness equal to to the smaller thickness of the 2 MEs being connected with it.

Regards.

#### Re: Negative crack width

Thanks a lot for your explanations