Long Term Deflection Estimates 1

[Trenno]

Hi all,

I know full well the ins and outs of long term deflection - creep, shrinkage, cracking, f'c, mix design, restraint, P/A, compression reo and the list goes on...

But what I'm asking here are people's 'back of the envelope' deflection estimates.

For example, I've seen people get a quick figure by factoring up the elastic uncracked deflections - 3G + 1.5Q. This can be back calculated from various code equations.

I've also seen a RAM Concept file with the load combo - 3.35G + 1.64Q (note the creep factor being set to 3.35).

Anyone have any reference material that quotes some of these elastic uncracked deflection multipliers, purely just to get ballpark figures?

Or do people more opt for span/depth ratios to give them a feel for a particular design?

 

[hokie66]

Interesting discussion, and it gave me a few chuckles as well. I applaud the research by Gilbert and others, but the main defence against deflection will continue to be depth.

Taking this in a different direction (sorry, Trenno), I wonder if those here who have knowledge of in service deflection problems can share what types of structural systems were involved, and what caused the deflection issues. I don't have a lot of anecdotal examples, just things I have heard and tried to avoid on my own jobs.

I suspect that most deflection issues in recent times have been caused, at least in part, by pressures to cheapen construction. The main type system to which this applies is flat plates. It used to be that when the thickness of a flat plate was selected, the first criteria was to satisfy punching shear. When punching shear was good, there was generally enough depth that deflections were satisfactory, provided the design was respected during construction. Now the desire for thinner flat plates, along with the use of those accursed studrails, means this insurance is no longer there.

Another important factor, which was addressed in Doug's paper, is the use of higher strength reinforcement. Because ultimate flexural strength has been allowed to drive designs, the reduction in amount of reinforcement has adversely affected serviceability, i.e. deflections and cracking. I don't think we needed 500 MPa steel when 400 worked just fine.

Lack of attention to restraint, both in design and construction, is a big contributor to direct tension cracking, leading then to greater deflection.

Poor placement of top reinforcement is a problem that will never go away without adequate inspection. Many contractors and site workers don't even know that top steel needs to be at the top, rather than just with minimum cover, so they don't use the right height chairs. Another construction problem is speed...too early stripping, haphazard and premature removal of back propping, etc.

 

[rapt]

Hokie,

Most of the cases I have heard of have been flat plates/flat slabs as you suggest, but many of the Australian ones were designed in the 1960/70's, well before stud rails were invented. Design was typically based on L/D ratios and often took advantage of compression reinforcement rules to reduce the slab depth.
Spans were often 8*8m or 9*9m and final deflections on the bay were up to 80-90mm. In the mid 1970' I had consultants who were putting out RC flat slabs with L/D rations of 35 arguing that they did not need PT because their slabs worked (according to their L/D calculations). These same consultants 15 years later were telling me that they never do RC flat slabs any more because the deflections are far worse in practice than is being predicted by the code rules. I then proceeded to show them that we were predicting 80-90mm long term deflections for these slabs, similar to what they were getting in practice!

A lot of the later cases I have heard of tend to be designed using FEM software with automated design routines attached. I have seen wondrous deflection predictions by consultants using this type of software and they have proceeded to build the buildings based on these designs, not realizing that the software they are using is predicting short term uncracked deflections while the codes are talking about long term cracked deflections. This does not apply to all FEM software and does not mean that the FEM software is wrong, the designer simply has to understand what type of numbers the software is giving in each case and treat them accordingly. But if the designer does not understand the difference then there will be problems. A lot of designers from a steel design background will accept the deflections by well know 3D building analysis/design software because it is pretty good for steel buildings, not realizing that it is not allowing for cracking and long term effects in its RC design. It is doing exactly the same elastic design for concrete as it does for steel. Unfortunately, concrete cracks, creeps and shrinks, unlike most steel members.

In the simplest case, if the deflection is short term uncracked for RC slabs, multiple it by about 6!

I think the most problems I have seen is in buildings designed by engineers used to working with design codes that are predominately based on L/D ratios who then try to optimise the design by moving to FEM, and simply not understanding what results they are getting.

This has been happening with PT flat slabs over about the last 15 years also, where a lot of the early PT FEM software was also predicting short term uncracked deflections and/or still using RC multipliers to predict long term deflections in PT slabs. Contacts in PT companies are forever telling me about super thin slabs they are being asked to build and the design is always based on a FEM analysis that has been misinterpreted by the designer.