Non-Ductile Concrete Members 3

[Gousey]

Hi all,

In regards to non-ductile (over-reinforced) concrete members, I usually steer away from it all together as I would only design concrete members that are ductile.
I am currently designing a RC beams that supports dead loading as well as traffic moving loads.
As there is limited room for the beam (due to offsets to services and height limitations) the beam is required to contain additional reinforcement.
The beam is currently designed with 75% flexual utilisation (25% spare capacity) for the governing load combination.

My question is; at what stage do I reach the ultimate failure (beam fails catastrophically. Does the beam need to be loaded to 100% utilisation for this to occur? Or can this failure occur before this point?

Thanks

 

[WARose]

In general, a over-reinforced beam will fail by concrete crushing before the steel yields.

I've run across a few in practice......you determine the capacity (IIRC) by a strain compatibility method that is limited by the concrete failure strain.

 

[Gousey]

WARose how would I be able to best determine the point at which the concrete crushing occurs?

I havent got access to FEA software - any guidance is much appreciated.

 

[WARose]

WARose how would I be able to best determine the point at which the concrete crushing occurs?

You have to assume it. Typically we are talking a strain of about 0.003. IIRC, that is covered in Section 10.3 of ACI 318.

If you want to read a good treatment of over-reinforced beams (including capacity calculation).....see Appendix C of 'Design of Reinforced Concrete', 3rd Edition, by: Jack McCormac. (1993). I've seen it treated in some other texts as well.

By the way, if this is a design situation (as opposed to an existing beam you are trying to evaluate)......you probably shouldn't be coming out with an over-reinforced beam to begin with. (IIRC, the code prohibits them.) It's not good practice.

 

[MotorCity]

You should be designing with a reinforcement ratio below rho-balanced.

 

[Gousey]

WARose
Thank you for the reference, in regards to the design - I am currently at concept stage and I need to determine if this method of design is suitable for the project. I personally wouldn't allow for over-reinforced sections - however as it is not my field, I thought to reach out on some guidance.
This clears it up..........stay clear from the over-reinforced concrete sections.

 

[KootK]

If the goal here is the design of a new beam, and you're stuck on your parameters (depth etc), you've got a couple of options:

1) Let the beam be over reinforced and adjust your material safety factors to compensate for the reduced reliability. Some codes have procedures for this. Basically, the more over reinforced your are the lower your material safety factors get. The result is an over reinforced beam with the same degree of reliability that normal beam would have, albeit a scarier failure mode.

2) Add compression reinforcement to push your design back to no longer being over reinforced. This is what I would pursue first.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[WARose]

Thank you for the reference,...

You are welcome. Be advised: I don't think they have that Appendix in later editions.

This clears it up..........stay clear from the over-reinforced concrete sections.

Always a good idea. KootK (as usual) has some excellent advice as well.

 

[KootK]

Thanks for the plug WARose.

Just to be clear, while there are ways to design over-reinforced beams, I also steer clear of it unless I'm desperate (including the compression steel option). The more likely problem with them will tend to be excessive deflection unless that's studied carefully in design.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[WARose]

Just to be clear, while there are ways to design over-reinforced beams, I also steer clear of it unless I'm desperate (including the compression steel option). The more likely problem with them will tend to be excessive deflection unless that's studied carefully in design.

Yeah, I forgot to mention the deflection issue. Good point.

As far as flexural capacity capacity goes......most times I've had to evaluate a existing over-reinforced beam (with tension steel only), the capacity turned out to be about what it would be if the beam was reinforced to the limits of the code. Except that the capacity reduction factor (normally taken as 0.9) is taken as 0.7 "to account for the brittle nature of overreinforced beams" (to quote a line from the reference I mentioned above).

 

[rapt]

There should never be a need for an over-reinforced member in a building. Problems are

- Most buildings rely on formation of cracks, plastic hinges and moment redistribution. We analyse them as elastic but they never are. The highest loaded section cracks first and redistribution commences, normally at less than full service loads, not overload. Then another section cracks and further redistribution occurs. There is continuous changing in redistribution in a continuous member throughout the load range. Then under overload plastic hinges start to form.
Over reinforced members do not handle redistribution well. That is why planned redistribution is limited by design codes.

- In over reinforced sections, the reinforcement is not yielded, so you cannot assume full steel yield in the calculations. As the section becomes more and more over reinforced, the stress in the steel continues to reduce making the section even more over reinforced. It is not economical! The only way to reduce this is to add compression reinforcement. Again a waste of money.

- We do not allow for a lot of the real world load effects that occur, normally caused by restraint to shortening. While creep actually helps reduce effects and is actually good, stresses caused by restraint to shrinkage and temperature movement are often larger than those caused by vertical loading. The members need to be ductile enough to absorb these effects and allow the redistribution they require for the building to survive them.

- If there is a requirement for earthquake actions, ductility is essential. You probably should be designing closer to .5 Pbal than the current code limits which are closer to .7 - .75 Pbal.

In conclusion, keep them ductile and well under the code maximum reinforcement limits and you will sleep a lot better.

 

[KootK]

Cudos to rapt. Excellent summary of the finer points.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[hokie66]

Yes, rapt has summed it up well. In addition to using code prescriptions to design buildings, we have to try to think like the structure. Structures will always do things we don't expect, but with experience, we can develop good concrete heads...if we want to.

 

[Gousey]

Fantastic replies by all! Thankyou everyone. I plan to sleep well at night so I will keep my options limited to du

http://www.eng-tips.com/ctile

members only.

 

[rapt]

Hokie,

That is an interesting point you brought up. Design codes do not advise on good design, in many areas they set the limits on trying to avoid bad design. If you are designing at those limits, you are on the limit of bad design.

An Australian professor I refer to quite often, when I was discussing the code ductility limits with him, could not understand why I wanted them looked at because he could not understand why anyone was designing near or at the limits. As far as he was concerned they should generally be designing members far more ductile than the code limits, not deliberately designing everything at the limits.

 

[KootK]

An Australian professor...could not understand why anyone was designing near or at the limits.

All he`d have to do is ask the nearest practitioner. The reasons being:

1) Trying to retain architects as clients
2) Economics of floor to floor heights
3) Mechanical systems
4) Being stuck with schematic design errors



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[WARose]

I kind of see how it could happen......space restrictions, development issues, etc.

Almost every time I've encountered an existing one.....I think (in some cases: knew) it was because it was done by working stress design.

 

[rapt]

Koot,

You missed the over-riding one, if I do not do it someone else who does not understand why I am not doing it will. So I may as well get paid to do it, even though I think it is wrong.

Unfortunately, it tends to be done more often on the most important members, such as transfer beams, because they tend to cause most headroom problems!

 

[TehMightyEngineer]

Yep, us crazy precast engineers who design things to 99% will come in and steal your jorbs. Gotta keep the weight down!

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

http://www.americanconcrete.com