Coupling beams - diagonal reinforcement

[kostast88]

I have a RC building with with short coupling beams in coupled wall system.

1. The short coupling beams designed with diagonal reinforcement as seen on the image on the left.
2. Due to a number of site reasons in many cases they end up on site more like the right image below. They are slightly rotated.

Reasons include:

• Imperfections on diagonals during cut and bent
• Elevation adjsutment of the MEP sleeve forces them to rotate
• General rebar positioning tolerances (pushing stuff around)
• None of the above are good reasons, but they do happen

Q: Is there any impact on structural behaviour because of that? Judging from Eurocode 8 this looks more like a basic principles problem rather than a code-based approach.

What I understand is the diagonals take the shear, and if properly anchored they transmit the shear to the wall. If anything else, I would guess this only affects the shear on the coupled walls. Any opinions?

 

[KootK]

1) I don't feel that there would be a substantial difference in performance between the two conditions that you show.

2) Fundamentally, I don't feel that it's a good idea to bend coupling beam diagonals at all. I feel that they should be run straight as shown below. In a high seismic application, coupling beam diagonal bars need to yield deep into the plastic range cyclically. Under such conditions, I would expect the concrete surrounding the bar bends to crush in compression pretty badly. Perhaps your case is a low seismic situation not requiring great ductility?

What I understand is the diagonals take the shear

In the context of North American, high seismic design, the diagonals technically take both the shear and the moment.

 

[kostast88]

Hi Kootk, Ductility requirement is Medium in this case - as defined in Eurocode 8.
Meaning a behaviour factor (Load reduction factor in US) around 3.5 so ductility is required.

I did find something searching in NEHRP Seismic Design Technical Brief No. 6

Figure 5-14 shows typical details for a coupling beam
reinforced with two intersecting groups of diagonally placed
bars symmetrical about the midspan. Each group of diagonal
bars consists of a minimum of four bars provided in two or more
layers. The diagonal bars are required to extend straight into
the wall a distance at least 1.25 times the development length
for fy in tension. A challenge is avoiding interference between
the diagonal bars and the boundary element transverse and
longitudinal reinforcement. If an adjacent wall opening or
edge (for example, at the top of the wall) requires the diagonal
bar extension to be bent, additional reinforcement is required
to resist the unbalanced force resulting from the change in
reinforcement direction, similar to the requirement for offset
bars in columns (ACI 318 § 7.8.1.3). This detail should be
avoided where practicable. The minimum wall thickness to
accommodate both wall and coupling beam reinforcement is
around 14 inches, although 16 to 18 inches is more practical.

This makes sense in my head: the tensile force from the rebar would be gradually dissipated by the bond stress with concrete. If you change the direction where this happens then there is an unbalanced force which causes the shear wall to develop stresses NOT in the direction of the wall reinforcement (which is rectangular: verticals & horizontals). This is something that the wall wouldn't be designed for. So the issue would really be not with the beam, but with the wall and what it was designed for.

I agree with the bad compression scenario you mention. I suppose the above could make things even worse.
And yes - the diagonals should take both.

 

[KootK]

So the issue would really be not with the beam, but with the wall and what it was designed for.

Agreed, at least in the local sense. This wouldn't change the global moments and shears that your wall is designed for. I would describe it as a connection issue between the beam and the wall.