Shrinkage and creep strain in analysis 1

[SobreroPeet]

I have started a new job recently, and the engineers do things a bit differently.
The following comments are regarding concrete structures and Australian codes.
I was taught that shrinkage and creep effects should be considered in the serviceability limit state and not the ultimate limit state.
I used to check SLS: 1.0G+ψ*1.0Q+Shrinkage and creep restraint(S).

The new company throws shrinkage restraint into every load case:
ULS:
1.2G+1.5Q+S
1.2G+wind+ψQ+S
G+Earthquake+ψQ+S
G+fire+S
etc

This seems very conservative. I have never seen so much rebar in a design before. My thinking is that the structure will settle, crack and release restraint when SLS loads are exceeded.
It just feels like designing yourself a quick way out of the industry. Have I been doing it incorrectly since the start? Is this industry standard? How is this an economical design approach?
Any insight or comments against or for this will be appreciated.

 

[centondollar]

"This seems very conservative. I have never seen so much rebar in a design before. My thinking is that the structure will settle, crack and release restraint when SLS loads are exceeded. "
This "release of restraint" you refer to does not actually happen to the extent that some designers like to believe. Shrinkage causes tension in the cross-section, and thus it will alter the location of the neutral axis (make it smaller, i.e. located closer to extreme compressive fibre), while creep does the opposite (increases steel stress and increases depth of N.A.) - no matter the loading, SLS or ULS. However, shrinkage has a larger effect on concrete stress than creep, so the most conservative way to design is to ignore creep and account for:

1. Drying shrinkage
2. Autogeneous shrinkage

If you want to be very precise, you will also account for heat of hydration (which causes cracks if the degree of external restraint, e.g., SOG or rigidly connected wall, is high and the cross-section is massive) and the associated temperature differential (short-term and long-term ambient temperature compared to heat of hydration at casting), but this is very seldom done.

"It just feels like designing yourself a quick way out of the industry. Is this industry standard? "
You should not concern yourself with feelings - you are an engineer, and engineers use physics, mathematics and experimental results to validate their design. What the "industry standard" is does not necessarily reflect reality.

" Have I been doing it incorrectly since the start?"
The effect of ignoring shrinkage, creep and heat of hydration is small if you design mostly "ordinary" structures where the degree of restraint is not massive (e.g., (thin) walls, columns, beams), so I wouldn't worry about that if I were you.

"How is this an economical design approach?"
The important question is whether or not it is the right approach, and from the engineering standpoint, ignoring shrinkage and creep (and, to a lesser extent, heat of hydration) is incorrect. Concrete always shrinks and creeps, and these effects (while counteracting eachother) cause stresses in the reinforced concrete member in all "limit states".

PS. For some structures, the minimal amount of reinforcement is governed by the cracks induced by shrinkage and heat of hydration, and not by cracks induced from mechanical loading.

 

[Sheer Force Engineer]

Hi SobreroPeet, curious as to what your design approach is. Are you analysing concrete structures by hand or a type of spreadsheet based approach?

Shrinkage restraint can be very important for strength calculations if you are designing a post-tensioned slab system. If your slab is restrained against shrinkage this will prevent your installed post-tensioning from being fully effective in the slab and take some of the "P/A" out of it (for example a basement slab situation with all 4 walls fully restrained without a movement joint).

If I were to design the example outlined above, I would check using a modelling software like RAM Concept and model the shear walls and restraints properly which would account for these effects which are occurring in the slab. If I were to do the same using a hand based approach I would use the method you have outlined in your original post.

The aim should be to detail your connections and framing arrangement so that restraint is not too much of a concern (temporary movement joints, etc.)

For the latest insights into the art of Structural Engineering

http://www.sheerforceeng.com

 

[SobreroPeet]

@centondollar

I have never thought of the counteracting effect between shrinkage strain and creep strain. Thank you for that.

I do agree on all fronts. Let me make it clear, I do design for strain. Both drying shrinkage and autogenous shrinkage are addressed explicitly in the AS 3600 code, minimum steel and crack requirement concepts are also well defined.

I am designing a framed structure with high live loads (25kPa). We know shrinkage will be there, the shrinkage and dead loads are the only givens. So let me rephrase, I am struggling with the idea of having fully factored live loads and including shrinkage in ultimate limit state design. I am referring specifically to case 1.2G+1.5Q+S. I would argue that 1.0G+1.0Q+S is conservative but reasonable, I think 1.0G+0.6Q+S is a good estimate for high loads. Why not apply shor-term factor to Q? or the combination factor, 1.2G+ψc*1.5Q+S

@Sheer Force Engineer

We use Inducta software + some spreadsheets/handcalcs to validate. I'm not happy with the layout, it is a large slab with high loads and is an unbraced frame with PT slab. PT is done in collaboration with D&S, thanks for the insight on restraint concerns with P/A. I am definitely going to query it. The outer columns are getting nailed by the strain restraint. I am intrigued with the temporary movement joints, do you mean lockable dowel joints for the slabs or column slip joint?