Ok Eric, here is my brush with this issue. Back in my bridge construction days (1970's) I often wondered about this also. The state DOT designed bridges using f'c = 3000 psi, then required Contractors to use "Class A" concrete (6 bags of cement per cubic yard - usually resulting in better than 4000 psi @ 28 days).

...then came the day when we (the Contractor) had "problems" with concrete placement of a bridge deck. The DOT inspector closely followed the test cylinder breaks for that day. They did not reach 3000 psi at 28 days, but at 56 days they just "squeaked" over the requirement.

During that 56 day "waiting period", we could not, and did not wait - construction of the bridge continued on schedule. I still shutter to think what would have happened is we had to tear out 56 days of work because the concrete design had been worked out to "a nat's eyelash". The differential cost of stronger concrete is cheap insurance on some (but not all) projects.

www.SlideRuleEra.net

I hope the answer can be found in Section 1701.5 of the 1997 UBC or 1704.4 of the 2000 IBC.  (I do not have later IBC's...)

To avoid special inspection during placement of concrete, one may design based on 2500 psi but specify a higher strength concrete.  If special inspection is to be provided, there is no need to assume a weak concrete for design.

I think it is similar to allowing only half strength for expansion anchors or masonry shear when special inspection is not provided.

whyun, the item number in UBC 1701.5 is 1-exception 2. In the IBC it is still under 1704.4 exception 2.3

Let us suppose that for footings a design strength of 2500 psi is sufficient for strength considerations but for durability f'c of 4,000 psi is required. Under these circumstances, will special inspection of the concrete and rebar for footings be required?

SlideRuleEra gave a nice practical take on the issue.  One other one is that the 3000 psi concrete is the characteristic strength.  Pretty much by definition, there is a 1 in 20 chance that a strength will fall below this value.  This is taken into account in the AASHTO Bridge code in that the strength of a "sample" (meaning the average of 2 cylinders at 28-d) may be lower than the characteristic strength but only so far as it is now lower than 500 psi (3.5 MPa) - along with a few other caveats.  It seems reasonable, then, taking into account the probability that you will have some low strengths to design for the value that is characteristic less a certain value (say, for this instance 500 psi).
    henri2 makes a good point in that strength of specified concrete is not always chosen on the basis of required structural strength but insofar as strength is a good measure of so many other concrete properties, strengths might be chosen to match these other properties.  Instance - for a concrete lined side road ditch in a mountainous area, the strength of the concrete from a structural viewpoint needn't be more than, say 1500 psi if that - no loads other than a foot of water or so.  But, given the velocity of the water flow in high gradiant drains, the water would eat away such low strength concrete and you would specify a higher strength, as henri2 points, out for the durability - erosion aspects.  

As henri alluded, if you design concrete for durability, strength is essentially incidental.  There are numerous applications where strength is not the issue, but durability is necessary.  2500 psi concrete does not have good durability in exposed applications.

Somewhat anecdotal:
A situation very similar to the scenario mention by SlideRuleEra occurred at a very large pour for a hospital here.  The pour continued, even after some initial tests raised questions about the mix, and nothing really happened until after the 28d tests.  This caused every core drill in this part of country to on the job site, still no good.  The project was bit by the 1 in 20 rule mention by BigH as all the 56d test samples and additional core samples failed using 3 different labs.  Every air compressor and jack hammer you find ended up on this job.  

According to the Geotechnical people the second time around the mix exceeded the specified requirements by at least 1000 psi.  There were two GT inspectors on site at all times during the second pour.  Sure glad this is not for profit hospital.         

I tend to agree with Ron's post, but I would say it is more normal in this case to specify a minimum cement content which would satisfy the durability requirements by providing a dense concrete and would cause an incidental increase in strength.

The other arguement about inspection requirements is interesting but I think we cannot assume that any concrete would have to be broken out due to low strength. It is a requirement that all steps are taken during mix design, concrete production, placement and sampling that the correct grade of concrete is used in the permanent works - this is not really a design issue.



 

I thought that, in this day and age of admixtures and pozzolans that the minimum cement content concept was going out the door especially given the design w/c ratio? . . . and how do you determine if the specified minimum cement content is sufficient for the property in question?

I'm not a fan of the minimum cement content approach...mostly because it tends to cause overdesign for the application and promotes excessive shrinkage, thus increasing crack potential.

It is my opinion that mix designs should be carefully tailored to the application...if you need strength and durability for an application, then consider them both.  If strength is the only consideration, then design for that, but the whole mix design is important, from cement content, to w-c ratio, to aggregate size, to aggregate ratios....they all have an effect on the performance.

Remember, good concrete is made with water, cement, and aggregates.....bad concrete is made with the same ingredients.

Hopefully, not at the same time.

Not using traditional concrete acceptance criteria sounds great given our recent knowledge, but it is difficult for me to envision how we will perform QC on these new ideas.  There are too many players to point the finger for us to just say "We'll check the cylinders at 28-days."  By then it is too late.  It's all we can do to check the slump before the concrete is flowing out the truck.

Another instance where it is prudent to use a higher strength than what is determined by design is for drilled shaft or auger-cast pile foundations.

It is relatively common that the concrete strength is reduced due to segregation or mixing of soil with the concrete.

Field conditions often vary from lab settings, water is added, less then optimal cure conditions, ect.  Here in NW Florida the cost is only a few dollars per yard (2500 vs 3000 psi).

as well described already, the specifying of concrete needs to take into account not only the strength requirements, but also the durability and construction requirements.

Many, if not most concrete repairs are due to either poor onsite workmanship such as insufficient cover to reinforcement or poor quality concrete from a durability standpoint.

in relative terms, an increase in concrete grade often does cause great financial pains but gives benefits in reduced covers (or greater net durability), slightly better young's modulus (thus deflection), flexibility during construction (ie striking slabs or loading slabs early) and possibly lower amounts of reinforcement (particularly if shear is a governing criteria).

Such are the benefits of increase concrete grade during construction that i find many contractors propose it themselves to allow for greater onsite flexibility and thus programme advantages!

all in all, i wouldnt regularly recommend on skimping on concrete grade!


blihp

ps for clarity i should add to my previous post that im not involved in concrete supply at all! (im actually a structural engineer consultant...)