ACI-318 and interesting core results 4

[techmaximus]

10k PSI concrete in a 15” x 24” column with 8 # 7 verts.

The mix =
Type I = 495
Slag = 270
Flyash = 135
Natural sand = 992
Stone # 7 (3/8” max) = 1930
33 gal water
153 oz high range water reducer
slump 6 to 8 post SP
air 1 to 3
W/C = .31

One set of 6 test cylinders (6”x12”)
Two 7 day breaks = 6110 and 6190
Two 28 day breaks = 9140 and 9190
Two 56 day breaks = 9860 and 9920

3 Cores tested at 73 days = 9730 and 8180 and 7230
average 8380 psi and 84% of fc
one single core was 72% of fc

Technically the core results failed because the average of the 3 cores was under 85% of fc and one single core was under 75% of fc. Additionally, I think I recall seeing something in the EOR specs about requiring cores to be 90% of fc for fc mixes over 6000psi or 7000psi. I’ll have to look that up.

The point of this post:

ONE core 14” long was drilled from the 15” face. The core was subsequently sawed into THREE pieces to get the ACI-318 required 3 compression tests.
The core nearest the outside of the column was the highest break and the core nearest the inside of the column was the lowest break. Upon coring, I had expected the cores breaks to be within a hundred psi of each other or so, but obviously this did not happen.

ACI-318 5.6.5.2 — If the likelihood of low-strength concrete is confirmed and calculations indicate that load-carrying capacity is significantly reduced, tests of cores drilled from the area in question in accordance with “Method of Obtaining and Testing Drilled Cores and Sawed Beams of Concrete” (ASTM C 42) shall be permitted. In such cases, three cores shall be taken for each strength test that falls below the values given in 5.6.3.3(b).

Two questions:

Does the taking of ONE long core and then sawing it up to get three compression tests satisfy the intent of ACI-318-5.6.5.2.

Also, how does one interpret the core results? The cylinder breaks appear to be dead on the money with those tight numbers but the cores had a difference of more than 2000 psi. Could this be a phenomenon of the vibrating technique used during concrete placement, of a moisture gradient between the exterior and the interior of the column, or of a thermal condition in the interior of the column causing micro cracking at the aggregate?


Techmaximus

 

[Rjeffery]

Techmaximus,

What was the Diameter of the cores, the L/D ratio and the condition of the cores when tested? (i.e. wet or dried?)

 

[techmaximus]

The cores were 3" nominal diamter. A correction factor was used to determine the strengths. I believe that the core results are accurate.

Techmaximus

 

[Ron]

Techmaximus....there are lots of reasons for discrepancies between the cores and the cylinders. Curing temperature can have a significant affect on the strength gain, so if your in-situ temperatures were lower than the lab curing temperature, there's one issue.

The one core as three specimens is a stretchy argument. The intent is to use discrete specimens from various locations representing the same placement, but not essentially the same piece cut into 3 pieces. That can badly skew your evaluation. Suppose you had three specimens from different locations and the one in the area that you cut into 3 pieces was the only low one.

Don't expect the in-situ strengths to be the same as the cylinder strengths. That's one reason for the 85% acceptance criterion.

 

[techmaximus]

My question was not related to the discrepancies between the strengths of the cores and the cylinders.

The core breaks were:
3 Cores tested at 73 days = 9730 and 8180 and 7230

The cores had a wide difference of more than 2000 psi and this is the discrepancy I am interested in. Could this be a phenomenon of the vibrating technique used during concrete placement, of a moisture gradient between the exterior and the interior of the column, or of a thermal condition in the interior of the column causing micro cracking at the aggregate?




Techmaximus

 

[jheidt2543]

techmaximus:

Just a few thoughts to consider:

First, I would send a copy of your first post to the ACI Technical Services Dept. and ask for their opinion, afterall they wrote the code.

Secondly, I seem to recall, but haven't looked yet, that the direction of the core has some influence on the results. Test cylinders are poured vertical and tested vertical, while your core was horizontal. This is common in core testing of columns, but I do believe I've read that there is an effect to be expected.

Third, as Ron stated and you may already know, there is a real difference to be expected between cylinder tests and core tests. Cylinder tests are really more of a check on mix consistancy than actual field strength.

Fourth, I've never heard of taking a long core and sawing it into the three tests. I also agree with Ron that the intent of the code is three independant cores.

Lastly, I would expect strength variation thru the thickness of a member as well as along it's length, particulaly in a column where the lower levels are more consolidated than the upper. Although, I have to admit 2,500 psi high to low seems like a lot!

 

[concreteguru]

To try to explain why three parts of the same coring may exhibit wide variation, I would suggest that in-situ strength is a function of time and temperature. Since mass traps heat and heat cures concrete, one would expect higher peak heat closer to the core of the concrete member and this would translate into a higher compressive strength. You would also expect that moisture would be better retained in the core, so again, closer to ideal humidity in the core vs exterior, hence higher strengths.

We put computerized loggers in concrete to measure maturity and estimate strength. The most conservative estimates always come from concrete closest to the surface. See engius.com for detailed explanations.

Fred J. Croen, RSM
Engius, LLC
Boston, MA
www: engius.com

 

[techmaximus]

concreteguru,

3 Cores tested at 73 days = 9730 and 8180 and 7230

ONE core 14” long was drilled and then cut into 3 pieces for 3 compression tests.

The core nearest the outside of the column was the highest break and the core nearest the inside of the column was the lowest break.

It appears that my core achieved the reverse of what you would have expected as far as in-situ strength being a function of time and temperature.

BTW, the core was from the mid-height on the column. What is your opinion on the one core being sawed into 3 pieces for 3 compression tests?


Techmaximus

 

[JAE]

A couple of things from ASTM C42:

1. Taking core specimens: "A core specimen taken perpendicular to a horizontal surface shall be located, when possible, so that its axis is perpendicular to teh bed of concrete as originally placed and not near formed joints or obvious edges of a unit of deposit. A specimen taken perpendicular to a vertical surface, or perpendicular to a surface with a batter, shall be taken from near the middle of a unit of deposit when possible and not near formed joints or obvious edges of a unit of deposit."

2. From section 6.9.1: "..results of two properly conducted tests of single cores by the same operator on the same sample of material should not differ from each other by more than 9% of their average."

3. There is nothing stated in C42 about using a single core and cutting it. However, it "seems" as though cutting a single core might go against the inherent assumption that you are getting three discreet samples to average from across a sample area of your element. One long core at one location doesn't really accomplish this.

 

[concreteguru]

It is my opinion that taking one core and cutting it into three specimens does not best represent an average of the in-situ strength of the concrete member. Your objective would be better served by random sampling core locations and then averaging the results. As to why the actual strengths were opposite to my expectations only serves to prove that "Murphy's Law" is alive and well. Concrete is one of the few construction products where final production takes place off site from the manufacturing facility. Responsibility for final quality belongs to both the contractor placing, consolidating and curing the concrete and with the readymix concrete supplier. Good luck with your project.

Fred J. Croen, RSM
Engius, LLC
Boston, MA
www: engius.com

 

[BigH]

Was your horizontal core hole still actually straight or did it show a bending downwards towards the end of the core? Interesting discussion thus far; should read Neville's article of a few years back on core testing.

 

[techmaximus]

A bending downwards? Not sure on this. Can you explain why you asked this? Where can one find Neville's article on core testing?

Techmaximus

 

[techmaximus]

Oh, I get it. The longer the core barrel and the higher the strength of the concrete the more difficult it is to keep the core straight during drilling. The answer is no, in this case we didn't find any evidence of this; the core appeared to be straight.

Based on responses, I’m going to recommend that long cores cut up in to multiple compression tests not be performed. I am thinking that this is a function determined by the driller, a subcontractor of the contractor or the concrete supplier. The driller gets paid by the inch, not by the hour. Obviously more inches can be obtained in fewer hours and much less effort with one long core rather than with three separate cores of the correct length.


Techmaximus

 

[BigH]

Neville's paper was in Concrete International - about 2001 or 2002, I believe - I have a copy and will look it up tomorrow.