Making the grade - The required compressive strength in concrete design

A few years back, I discussed the merits of the NPCA Plant Certification program with the owner of a small precast plant.  He did not feel the need to become certified.  His exact statement was something like: “I’ve made 5,000 psi concrete for 30 years.”  That is a strong statement, and one that hopefully was supported by fact.  In a non-threatening manner, I responded with a question: “do you have one piece of paper that documents the evidence of this claim?”  He was silent.

A quality system will not magically transform the average precaster into super precaster.  And, the strength of concrete is only one part of making a quality product.  As I think about the comments made by this precaster, I wonder just how many folks understand the compressive strength requirements.  Jay Shilstone (2012), in his blog “Missed it by that much” – Concrete tests and f’c, got me to thinking about the importance of this quality principle. 

The product design calculations, or possibly the customer, will state a minimum compressive strength requirement.  The desire is that no concrete compressive test result will be lower than the specified strength.  All processes have variation, and concrete is no different.  A population set of data will be distributed in such a manner that when plotted by a curved line, the data points will form a bell shaped curve.  In a normal distribution curve, the apex of the curve will represent the mean or average.  Theoretically, 50% of the data will be to the left, and 50% will be to the right of the mean.

ACI 318 defines the standard method for determining the target compressive strength required to assure that 99% of the time the compressive strength will be greater than the specified strength.  According to ACI 318 section 5.1.1, the average strength is called the required strength or f’_cr.  The specified strength is noted as f’_c.  Section 5.3.2.1 provides the formulas for establishing the required strength when the specified strength and sample standard deviation are known.  In the formula below, s_s is the sample standard deviation.

Specified compressive strength, psi	Required average compressive strength, psi
f’c ≤ 5000	Use the larger value computed from Eq. (5-1) and (5-2) f’cr = f’c + 1.34ss                  (5-1) f’cr = f’c+2.33ss – 500             (5-2)
f’c ≥ 5000	Use the larger value computed from Eq. (5-1) and (5-2) f’cr = f’c + 1.34ss                  (5-1) f’cr = 0.90f’c+2.33ss              (5-2)

                              Table 5.3.2.1 from ACI 318

The standard sample deviation is calculated from “30 consecutive tests or two groups of consecutive tests totaling 30 tests” (ACI, 2005).  A modification factor is allowed when the number of tests is less than 30 but greater than or equal to 15.  When the number of consecutive tests is less than 15, the required average strength will be [f’_cr = f’_c + 1200] when the specified strength is ≥3,000 psi and ≤5,000psi.  For a specified concrete compressive strength of >5,000 psi, the required average strength will be [f’_cr = 1.10f’_c + 700].

For the precast producer, making consistent concrete with less variability will reduce the sample standard deviation.  This will allow the precaster to produce a concrete design with a required average strength that is lower, while still meeting the over design requirements of ACI 318.  This might result in a more economical concrete mix, and it will also provide a better batch-to-batch consistency for concrete products.  So, the next time you tell someone that “you make 5,000 psi concrete”, consider the statistical variation.  Maybe you do, and maybe you don’t.

References:

ACI. 2005. Building Code Requirements for Structural Concrete (ACI 318-05). American Concrete Institute: Farmington Hills, MI.

Shilstone, Jay. 2012. “Missed it by that much – Concrete Tests and f’c. Accessed on September 27, 2012 from http://www.commandalkonconnect.com/2012/09/26/missed-it-by-that-much-concrete-tests-and-fc/.

Concrete Cracking

I heard the statement one time that "all concrete cracks."  It is true that concrete is a composite that is very strong in compression, but weak in tension.  This makes it susceptible to cracking.  And, reinforced concrete is actually designed for cracking.  The reinforcement is engaged as the concrete section bends and yes, cracks.  But the statement above is not all its cracked up to be.

There are several different types of cracks, and most of them can be avoided with a proper mix design and best production practices.  The following is a list of types of cracks that are typical in concrete:

• Plastic Shrinkage
• Dry Shrinkage
• Alkali-Aggregate Reaction
• Alkali-Silica Reaction
• Corrosion
• Structural Overload

For this blog, I want to focus on Dry Shrinkage cracking.

Concrete is typically made with more water than is necessary for hydration.  The excess water evaporates or "bleeds" out.  As this occurs, there is noting to fill the volume left behind.  As the concrete drys it will naturally shrink in volume.  But, if the concrete is constrained by the formwork or reinforcement, then the concrete cannot shrink which causes internal stressing.  

Some times drying shrinkage occurs where the concrete is exposed to the air while still in the form.  The water evaporates more rapidly on exposed surfaces, and therefore the precast element contains small micro-cracks.  In reinforced concrete, due to the constraint of the steel, the cracks will appear somewhat equally spaced along an edge.  If the concrete is fiber reinforced, it may still crack, but instead of a 0.002" crack every 6", the cracks are not visible, and there may be dozens within a 6" span.

If you are a precaster who manufacturers septic tanks, you may notice drying shrinkage more in a certain season.  This is easily explained by the theory of drying shrinkage.  When the dew points are low, and the day time high is 30+ degrees above it, then the relative humidity is low.  New concrete contains moisture that is necessary for hydration, and as long as the water is present, the concrete hydration cycle will continue for weeks.  New concrete placed on the yard in storage will dryout where the surface is exposed to air.  A small septic tank is usually stored as a complete, sealed unit, ready for installation.  The relative humidity inside is high, but the outside surface is losing moisture.  This internal stress will lead to hairline cracking on the outside.

Sometime a crack is misdiagnosed as a dry shrinkage crack.  Here is a scenario from my past that will illustrate this.  A 1,250 gallon mid-seam septic tank was developing a "smiley" crack after 3-7 days in the yard.  We had two sets of forms to make this same size tank, and the crack only appeared on the castings from one of these forms.  We thoroughly inspected the castings after they were stripped.  No cracks.  Since we did not see a crack until the casting was about a week old, we were certain that it was caused by drying shrinkage.  But why did it only affect the castings off one of the two forms?

I realized that any cracks in the castings were most apparent after a rain.  As the concrete dried, the moisture retained in the crack longer, making it visible.  With this knowledge, I established a new QC procedure: using a wet rag, the QC inspector would wipe the casting walls prior to removal from the plant.  Sure enough, we found the "smiley" crack that had eluded us for several weeks.  It was there all along, and was not caused by drying shrinkage.  The actual cause was discovered to be a faulty air regulator on the air cylinder for the collapsing core of the form.

If you are seeing cracks your concrete products, perform a root cause analysis to identify the primary and contributing causes.  One or more of these will be the root cause of the cracking.  Don't just accept the statement that "all concrete cracks."  Many of the causes can be eliminated with minor, inexpensive changes.  The customer perception of the quality of your precast products will improve as a result.