Welcome to the forums. Please post in English or French.

You are not logged in.

#1 2019-02-25 12:29:06

nirmaljoshi
Member
From: Japan
Registered: 2018-10-12
Posts: 152

modelling bond-slip due to corrosion?

May be some of you have already  modelled bond-slip between steel and concrete. I am doing some work that require slip due to corrosion and calculate the mechanical strength of a beam. I want your guidance to know the followings:
1. Is it necessary to make separate mesh for concrete, steel and rust and perform contact analysis or it can be done without such?
2. Which element types should I use?

As a first step, I want my analysis to be as simple as possible but not too simpler.

Last edited by nirmaljoshi (2019-02-26 02:58:06)

Offline

#2 2019-02-26 12:20:17

nirmaljoshi
Member
From: Japan
Registered: 2018-10-12
Posts: 152

Re: modelling bond-slip due to corrosion?

Update:
I have made a simple mesh for steel and concrete separately and used LIAISON_MAIL to combine the meshes. See attachment.

The analysis runs fine. But I am unable to prescribe rusting.

Perhaps there should be a way to define the characteristics of the contact surface between concrete and steel and provide some material parameters to it. Any guidance?

Last edited by nirmaljoshi (2019-02-26 12:21:53)


Attachments:
pullout.zip, Size: 41.2 KiB, Downloads: 84

Offline

#3 2019-02-26 21:04:58

mecour
Member
From: Ostrava (Czech)
Registered: 2011-04-04
Posts: 143

Re: modelling bond-slip due to corrosion?

Hello,

a year ago I did model of steel concrete connection in 3D with BETON_DOUBLE_DP forc concrete and VMIS_ISOT_LINE for steel but only for tensile test not for corrosion.  Here you can download my med file and attached you can found comm file.

drive.google.com/open?id=1J0KreUaewBnv2YenSKhfZTEuK5AwwTMU


For concrete and steel connection you can use cohesive elements CZM_LAB_MIX, but the usage of cohesive elements requires more experiences ... I don't know your level of knowledge.
And also for reinforcement corrosion there is exit the material model CORR_ACIER.


Attachments:
B500B_20mm_DP.comm, Size: 3.96 KiB, Downloads: 107

Offline

#4 2019-03-17 13:14:49

nirmaljoshi
Member
From: Japan
Registered: 2018-10-12
Posts: 152

Re: modelling bond-slip due to corrosion?

I experimented with verification manual SSNL127 and tried to understand the reference manual R7.01.20 which mostly refers to works by Almusalla and few other authors.

Now I am confused.

As steel corrodes, its actual section area decreases. But the manual mentions decrease in area is not considered (R7.01.20, page 7) and has to be provided in the command file using element properties. This means, the effect of corrosion is only to decrease the ductility (i.e. strain at the rupture point will be reduced).

BUT, in the case of reinforced beam it is well-known that, when the steel content decreases (i.e. undereinforced section), the beam becomes more ductile (as steel yields completely prior to failure). So, for a corroded RC beam, how can we model the ultimate effect of corrosion to couple the both effects?

I am not sure if my question is correct.

Offline

#5 2019-03-20 08:54:04

sb1966
Member
Registered: 2015-03-16
Posts: 124

Re: modelling bond-slip due to corrosion?

Hi,
I have not performed any contact analysis in code-aster yet and cannot comment on the elements/materials available in CA for the problem in hand.

However, regarding your question about effect of effective steel area on ductility I can say that the increased ductility (in 'under-reinforced beam') due to decrease in steel area is a true statement for the RCC as a composite material.

But when we model the beam as contact pair between concrete and steel material, (i.e. two different materials and their interaction) the above fact may not (should not?) guide the modelling. May be you can decrease the bond strength by playing with the friction coefficient between steel and concrete at the interface to approximately take into account the effect of corrosion. Some comparison with experimental/numerical results already available should be helpful to fine tune such coefficient(s).

I hope it helps in someway to move you ahead.

Regards and best wishes.

Offline

#6 2019-09-20 17:34:00

praful
Member
From: Chengdu, China
Registered: 2016-01-07
Posts: 47

Re: modelling bond-slip due to corrosion?

In order to consider rusting, its better to reduce the strength of reinforcing bar to consider the decrease in area of bar.
I read a paper using this modelling methodology but I could not find the link. the bar is used as truss element

whereas for bond-slip effect, I dont know in code-aster but in abaqus I used Clough-steel model for reinforcing bar (for cyclic loading)
if you are modeling bar as a solid element and play with a friction coefficient, it might be too difficult.

Last edited by praful (2019-09-20 17:45:33)


Prafulla Malla

Offline

#7 2019-09-23 22:34:46

sb1966
Member
Registered: 2015-03-16
Posts: 124

Re: modelling bond-slip due to corrosion?

whereas for bond-slip effect, I dont know in code-aster but in abaqus I used Clough-steel model for reinforcing bar (for cyclic loading)
if you are modeling bar as a solid element and play with a friction coefficient, it might be too difficult.

Hi,
Thanks for your reply. In FEM, there are various steel material models, with kinematic and/or isotropic hardening and degraded stiffness/ductility in case of cyclic loading. For static loading usually bilinear models are used. In my earlier post I was referring to static (monotonic) loading only.

In cyclic loading, I do agree that the approach with friction coefficient will be too difficult. What I have experienced is that there are unresolved convergence issues in dynamic analyses with contact.

Using reduced strength to model area decrease due to corrosion, I believe, might be an approximate and ad-hoc, but a practically  viable way  of modelling the bond-slip problem.

Regards.

Sukumar

Offline