Longitudinal reo in torsion punching reo

[blihpandgeorge]

Hi
When providing punching reo via torsion stirrups, AS3600-18 9.3.6(c) clearly requires a longitudinal bar in the corner of the stirrup, but gives no further comment on how to size. For torsion to be resolved it makes sense to have this bar and in the beam shear section there is clear requirements for this longitudinal bar 8.2.5.5(b). (The previous As3600-2009 was similar albeit the beam torsion section had a simpler equation for the longitudinal bar 8.3.6). I am finding that these longitudinal bars can be very significant and is leading to congestion.

Looking at a few examples and text books, with the exception of one, none seem to address sizing the corner longitudinal bars for the torsion punching.

My question is how should this corner bar be sized? (ie using the beam approach or another way)

 

[sdz]

My understanding is that in beams torsion caused tension in the longitudinal bars but in the torsion strip in a slab this tension is for practical purposes restrained in the rest of the slab so there is no size requirement; it's only to locate and help anchor the torsion reinforcement. In that case I would use whatever bars are being used for the top and bottom slab reinforcement and this can also count as part of the flexural reinforcement.

 

[Agent666]

If it's of any help, in NZS3101 it notes that for torsion stirrups the corner bars in members should be sized as follows (this isn't specific to punching shear though, just general torsion requirements):-
(note s is the closed stirrup spacing)

 

[blihpandgeorge]

thanks both for comments - basically sounds like there isn't a specific 'sizing'.

FYI I attach an example from a textbook that shows the sizing of longitudinal reo from the beam section and places this reo in either the flexural tension zone (not the corners). But this is the only example i have found, and primarily what stirred my query

 

[Agent666]

Maybe it's in the missing AS3600 commentary...

That example looks like they put the required torsion longitudinal reinforcement in the corners, so I'm not following your comments regarding that they were not put in the corners?

 

[Agent666]

If in doubt regarding AS3600 provisions, just follow NZS3101 requirements or another international code. You're allowed to use your judgement if no specific guidance exists, and as you've realised its something that should have some limits. NZS3101 seems like the natural contender here to adopt.

 

[rapt]

AS3600 has always said all in corners, but as was discussed in another thread last year, everyone else basically says evenly distributed around the torsion ties.

Canadian code gives a minimum corner bar size based on torsion tie spacing in clause 11.2.7 as a minimum of s/16! Other than that, the amount required is the same as AS3600-2018.

 

[blihpandgeorge]

thanks Agent666 and Rapt for comments, will look at other standards as well

Agent666 - my comment about the reo was badly worded on re-reading. The example calculates an area of steel (352mm) but distributes it across the top of the ties, ie placing a N16 (200mm2) in each corner.

As RAPT notes it appears the intention was to place this area in the corners which would be 2N16 in corner, but appears in practice it is being done differently. Being distributed across the ties would only requier an N12 in each corner.

In summary, and using the example numbers, it appears AS3600 would be placing 2N16 in each corner, the example places 1N16 and industry is possibly placing 1N12!

 

[Agent666]

I've definitely taken the approach in the past to assume you need to firstly satisfy the corner bar size limitations. Then the remainder is uniformly distributed around the perimeter (NZS3101 has limitations on spacing around the perimeter, inferring it is requiring distributed reinforcement, rather than it all being lumped in the corners as Rapt noted AS code infers/requires).

It's usual to use bar area for torsion being considered on the basis of part of the total bar area being used for flexure, and part for torsion.

 

[rapt]

Agent666,

Interesting. All codes I have used require it as extra above what is required for flexure.

blihpandgeorge
We realised this difference from other codes after 2018 was released. I think you will find the next Australian code (not the next amendment) will follow the practice in other codes and spread it around the perimeter with a minimum corner bar size. Might even get it into that mythical commentary!

 

[Agent666]

Interesting. All codes I have used require it as extra above what is required for flexure.

Well true in terms of how it is worded in the code, but after a bar area is provided some of the bar is used in flexure up to the load it is required to carry in flexure and some in torsion. To be clear, I'm not saying you are double counting the area for torsion and flexure which is perhaps what you might have assumed by your reply, that is definitely not what I meant for clarity!

Usually you have additional flexural capacity typically was I guess what I was getting at, that extra bit you are not using in flexure is available to carry the torsion in addition to any other additional bars on the tension face. Loads don't know this is a torsion bar or a flexural bar around the perimeter, loads don't read your calculations and behave accordingly.