Topic: Define Bond slip law for CFRP and concrete

Dear Sir,

In Atena 2D, it is not possible to capture debonding/delamination of CFRP and concrete. And, if not considering a perfect connection between the 2 elements, a bond slip law can be defined (referring to previous post in this forum)

1. Is defining the bond slip law similar to the captured of debonding and delamination of CFRP and concrete?

2. May I know at which part/section to define the bond slip law in ATENA 2D?

Thank you.

Re: Define Bond slip law for CFRP and concrete

Dear Irene,

1. Bond slip only models damage of the connection between the bar and the volume in the bar direction - it only allows slip along the bar axis. The bond model has been developed for bond of steel bars and cables in concrete, however, this modelling may also be good enough in many cases of FRP bands/wraps and similar.
Modelling the interface between concrete and wrap with GAP elements implies using 2D elements for the wrap in ATENA 2D, which is typically not feasible (very small elements required -> extreme memory consumption and analysis running times).

2. Please see section 4.4.3.6 Bond of reinforcement of the ATENA 2D User's Manual

Regards,

Re: Define Bond slip law for CFRP and concrete

I have a few questions on defining bond slip law for frp-concrete interface in ATENA 2D. To define some bond failures in frp-concrete, i have applied the max. shear stress, relative displacement and max. relative displacement in 'User Defined' in the " Bond for reinforcement" .However, when i analyze the model, similar results (ult. load and deflection) were given compared to the model without 'bond of reinforcement'.

1. Is this the correct way to define a bond slip in FRP-concrete? Or do i miss out any other steps?

2. When I defined the bond slip law for FRP-concrete in> Material>Bond for Reinforcement> User Defined, does ATENA recognized it? which is for FRP-concrete?

3. After defining the Bond slip in >BOnd for reinforcement>User defined, do i need to assign to the interface between FRP-concrete? A discrete element (a line) representing the CFRP in my model on the concrete material.

Please advise.

Re: Define Bond slip law for CFRP and concrete

Have you used the bond law in the reinforcement bar definition? On the right side of the reinforcement bar creation/edit dialog, switch from "perfect connection" to "bond model", select the Bond material you have defined, and set the remaining options.

Re: Define Bond slip law for CFRP and concrete

dpryl wrote:

Have you used the bond law in the reinforcement bar definition? On the right side of the reinforcement bar creation/edit dialog, switch from "perfect connection" to "bond model", select the Bond material you have defined, and set the remaining options.

Thank you for the above information. Now I managed to model the FRP/Concrete with the bond slip law defined.

Re: Define Bond slip law for CFRP and concrete

The results demonstrated in ATENA 2D with RC beams strengthened with CFRP laminates, as mentioned in this topic using a 'user defined' bond slip law.

The results can be obtained in ATENA is:
1. Load versus displacement graph
2. Crack pattern of beam in the window
3. Stress/Strain

I understand that failure of CFRP e.g. debonding/rupture could not be captured when using the 'bond for reinforcement'. Can ATENA give the failure mode result of a RC Beam with and without CFRP? How to determine the failure?

Re: Define Bond slip law for CFRP and concrete

With the "bar with bond" model, debonding can only be calculated in the fibre direction (i.e., no opening between the CFRP and the concrete), but you can of course get CFRP rupture - just define the Reinforcement law according to the CFRP properties.

To compare the response with and without strengthening, you have to prepare and analyze 2 models (one without CFRP, one including the strengthening).

To determine failure, monitor the L-D response and the convergence criteria.

8 (edited by irene 2012-01-03 15:10:04)

Re: Define Bond slip law for CFRP and concrete

With the "bar with bond" model, debonding can only be calculated in the fibre direction (i.e., no opening between the CFRP and the concrete), but you can of course get CFRP rupture - just define the Reinforcement law according to the CFRP properties.

Is there any specific calculation for the debonding in the fiber direction? is it mentioned in the ATENA user manual?

And, to get the CFRP rupture, can u briefly explain how to define the reinforcement law accordingto the CFRP properties? is it mentioned in the manual as well?

thank you.

Re: Define Bond slip law for CFRP and concrete

For bond definition, please see section "4.4.3.6 Bond of reinforcement" in the ATENA Engineering 2D User's Manual.
For the multilinear reinforcement material law, you simply choose several siginificant points from the CFRP working diagram, e.g., (0, 0), (e_lin_end; sigma_lin_end), (e_lim, sigma_max), and (1.01 * e_lim, 0.001 * sigma_max), and define them in the reinf. material dialog.

Re: Define Bond slip law for CFRP and concrete

Hello!  I am a new user of ATENA. Based on above discussion I would like to ask which is the best option 2D or 3D for modelling RC beam strengthen using FRP? Plus I want to apply CFRP full wrap, U-shape wrap and combine of U-shape wrap (Flexible CFRP wrap) with bottom wrap (Rigid CFRP Wrap). In this combination of flexible and rigid wraps which option I have to use to get perfect results.

Re: Define Bond slip law for CFRP and concrete

Another issue is I want to apply eccentric load on the beam not longitudinally means if I have a section width of 200mm I want to apply load on 160mm. Is it possible in 2D?

Re: Define Bond slip law for CFRP and concrete

Hello Muhammad, please see ATENA Troubleshooting, 2.1.8 How to model carbon or glass fibre wrap (CFRP, GFRP) strengthening in ATENA?

With respect to the eccentric load, I am not sure if I understand correctly - you have a beam of some length, some height, and 200mm width, which is loaded 40mm from one of the long edges (along the whole beam length or just in a single point)? If the error introduced by replacing your load with a distributed load at the whole top surface (or a line across the full 200mm width in case of point load) is too large, I think you need to prepare a 3-dimensional model.