Nguyenngocba

Minimum anchorage length in tension and compression

9 posts in this topic

Under clause 8.4.4 (1) of EC2 the minimum anchorage length shall be:

- in tension: lb,min >= max(0.3lb,rqd; 10*phi, 100mm)

- in compression: lb,min >= max(0.6lb,rqd; 10*phi, 100mm)

Can anyone enlighten me why the anchorage length in tension is generally shorter than in compression.

Many thanks

Share this post


Link to post
Share on other sites

I have never found a full explanation for this but the opinion that I have formed is that it is because of the different failure mechanisms that apply in the case of tension and compression.

There are several things acting: Firstly, the failure of anchorage bond must occur by the bar ribs forcing apart the concrete (as the ribs are not likely to shear off) - the work done by the concrete displacement and tensile related failure is what is required to be overcome by the work performed by pulling out of the bar. If you have a tension bar, then it is likely to remain straight but a compression bar may enjoy buckling restraint from the concrete, which would decrease significantly during failure. Also, if the bar has a tendency to buckle, it would tend to exacerbate the concrete splitting forces. These effects may be complicated further at laps where there could be an eccentricity between the forces in the bars, especially if the bars are cranked. Arguably, bars in compression at a lap might have a greater tendency to move apart than tension bars.

Such considerations as these will be even more important in seismic design and detailing than in non-seismic cases. Even in static compression failures, such as columns, bar buckling can precipitate failure whereby the compression bars buckle and burst the cover between any confining links, if the latter are insufficient. Dynamic loads, such as due to vertical seismic acceleration, can make the structure even more vulnerable to this kind of failure.

Previous UK codes, such as BS8110, produced lesser anchorage lengths for compression than tension. In the EN 1992 method (and I speculate that the intended application area of the code must include countries where seismic design is more routine than in the UK area), taking a straight bar as an example, one may compute the same lbd value for compression and tension anchorage but the application of the alpha factors will generally result in a larger length for compression anchorage. The main influence will probably come from alpha2 (the one pertaining to cover and spacing etc).

If anyone else has better knowledge of this, I would be interested to hear.

Edited by AndyBeeton
Spelling!

Share this post


Link to post
Share on other sites

Thank you Andy Beeton for the explanation. However I am not convinced by the effect of buckling of rebar in compression to the anchorage length because there are requirements on the link/tie spacing to control the rebar buckling, especially in seismic design. Moreover buckling only occurs when there is a chance of cover spalling, normally outside anchorage area, so in case column rebar anchored into pad foundation or pilecap it would not happen.

Share this post


Link to post
Share on other sites

I don't claim to have made a detailed study of this, but I would say that it is necessary to consider the failure mechanism in all cases. In the case of anchorage failure, the mechanism involves the moving apart of the concrete containing the bar. That requires some energy and that is provided mainly by the force in the bar and its axial translation. If the bar is compressed, there may be sufficient additional energy available on the supply side to balance the dissipation in the concrete which is fracturing. The point here is that the concrete is modelled with a fracture or crack that is being mobilised by the anchorage failure regardless of the fact that the bars are undergoing some kind of lateral displacement, such as buckling. However, when the bars are no longer restrained by the concrete at this ULS, they may buckle or displace and thereby mobilise that energy and participate in the failure mechanism proposed.

I have no better explanation at the moment and would be very happy to hear of one. If I get time, I will try to look around as it seems there must be some information somewhere on it.

Share this post


Link to post
Share on other sites

Dear AndyBeeton,

I have contacted ICE and they said they would look for an Eurocode panel member to explain this. I just feel it is quit a waste with EC2 when we put much longer compression rebar anchorage length than designed with BS and there is no case of failure due to bond in compression reported so far with BS design. In my tests with 32 short columns under compression at NTU Singapore with different amount of confining tie, although it is not particularly focusing on the bond failure I recognize buckling of longitudinal rebar only occurs with large space transverse reinforcement, other cases it only happens after the failure of concrete in compression.

Share this post


Link to post
Share on other sites

Do please let us know if you get a satisfactory reply. The issue is very interesting.

One must not rule out the possibility that it it is a mistake but I would not do so without good reason.

For a fascinating discourse on bond failure, with various theories and references to other studies and tests, I would recommend Section 9 of M P Nielsen's 'Limit Analysis and Concrete Plasticity'. That deals only with tensile anchorages but there may be material there that is relevant to compressive anchorages.

Other questions that I reflect on are:

- In a compression anchorage, one might suppose that the surrounding concrete is also in compression along the direction of the bar. That compression will induce tensions (and crack development) in other directions - i.e. generally perpendicular to the bar. What effect has that on the bond resistance, given that the resistance appears to be significantly dependent on the tensile strength of the concrete? Does this imply a different failure mechanism to a tensile anchorage, where the greater crack development might be initially in a different orientation? We know that transverse pressure improves bond so transverse tension must do the opposite - what is the effect, can it be quantified?

- In the tests, was the longitudinal steel yielding, or the concrete, both or neither?

Share this post


Link to post
Share on other sites
Dear AndyBeeton,

I have contacted ICE and they said they would look for an Eurocode panel member to explain this. I just feel it is quit a waste with EC2 when we put much longer compression rebar anchorage length than designed with BS and there is no case of failure due to bond in compression reported so far with BS design. In my tests with 32 short columns under compression at NTU Singapore with different amount of confining tie, although it is not particularly focusing on the bond failure I recognize buckling of longitudinal rebar only occurs with large space transverse reinforcement, other cases it only happens after the failure of concrete in compression.

ACI also provides lower compression anchorage length. I'm also interested in any follow up on this matter.

Share this post


Link to post
Share on other sites

Hi

 

What eventually ICE said about it?

 

Best regards

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Similar Content

    • By Edu90
      Hi all,

      Is anyone aware of a requirement to control crack width in a reinforced concrete corbel to eurocode 2?  Could I just take the area of the tension strut and check the minimum rebar controls cracking?

      Thanks for your thoughts!
    • By TonyB
      What I'm trying to do (for completeness in my software) is to provide an analysis of a simply supported RC beam at intervals of span/20 - BM, SF, depth of concrete compression zone, lever arm and force in tension steel. For a singly reinforced beam it's a trivial exercise since there can be only one solution at any point.
       
      What I can't seem to see, despite consulting several books, is the assumed distribution of forces in a doubly reinforced beam: obviously some of the moment is carried by opposing compressive and tension bars with the rest being taken by the compression zone concrete and tension steel. At a point where (say) the applied moment is one third of the maximum design moment it's probably the case that it could be resisted by the reinforcement only or by the section without the compression reinforcement, but presumably the correct answer is some of each. Help appreciated.
    • By Sjefen
      Hey,
      I have been trying to find solutions to composite column design, that is for example a circular column filled with concrete.
      Is there any quick approx. (rule of thumbs) solution like design tolerance of the steel column multiplied by any X number that gives a result within 10% of correct solution.
      I know there are two way´s to calculate this type of columns to EC, that is the official solution and the simplified solution. What i am looking for is an even more simplified solution
      If any one could help me with this or give me a hint on where I can find a spreadsheet or some freeware where I can design this type of problem that would be much appreciated !
      (My English is probably not 100% sorry for that)
      Best wishes,
      Sjefen
    • By a_geni
      The attached link is for Worked examples to Eurocode 2, hope that you find it useful.
      http://www.ermco.eu/documents/ermco-documents/worked-examples-ec2-def080723.pdf