Concrete Compressive Strength?? 1

[USFEngineer]

I have a project that we are currently working on in the construction phase. Several concrete compressive strength tests have come back well below our specified at 28 days. Most of them achieved the desired strength at or near 56 days and a couple of them never achieved design strength. For example at 28 days the average strength was about 1800 psi - 2200 psi. Specified was 3000 psi at 28 days.

My first question and probably the most important is what would be the proper way of handling such a situation? I have been asked if the concrete needs to be "ripped out". This seems a little harsh, are there any possible solutions?

Thanks in advance.

Andrew

 

[concretemasonry]

The very low strengths of 1800-2200 psi are obviously far below specs. Before tearing out, I would take a close look at the areas where the concrete was used. It could be it still may be strong enough to let it remain if you can identify the areas.

In place determination of the strength is possible, but the results are always difficult to interpret.

I would also take a close look at the preparation, handling and curing of the cylinders. It is really difficult to get strengths that low. Since you have 56 day breaks, it seems to rule out freezing of the samples, but poor handling (contractor, driver, testing lab, etc.) could be a possibility. A green cylinder is easy to damage.

In the interim, I would suggest requiring making more cylinders and making both 7 and 14 day tests in addition to the 28 day. No one can object to early testing since there is a proven problem based on previous test results. Any good ready-mix supplier should have some sort of time-strength history. In any case, it would put the supplier, contractor and laboratory on notice and give everyone the necessary warning of the problem. Is the R/M plant certified by any association or qualified by any recognized group? Is the testing lab certified?

Have you taken a good look at he mix designs and the admixtures? It is always possible to get incompatible admixtures if specialized properties are desired. Admixture companies are always will to jump in and help if only to protect their image.

Cement is so cheap that a supplier usually has little to gain by cheating on the content, but there appears to be a problem somewhere in the supply/testing sequence.

In an earlier life, I was a concrete inspector and realize that the person making a cylinder can have a great effect (10 to 30%) on the strength. The same goes for handling, curing and sample preparation. A testing machine can be out of calibration or there could be a problem with the platten for applying an axial the load. - There are many possibilities, but usually a lab can/should spot these quickly.

In college we could get 50 to over 100% cylinder strength variations by using either sugar or rebar.

Dick

 

[apsix]

I agree, consider the problem carefully before condemning.

If the concrete is exposed externally you will have to consider durability. I'm not sure, but I would think just because the concrete reaches strength at 56, or even 117 days doesn't mean it provides the same degree of protection to the rebar as designed. Coatings can be applied to improve durability.

Are you in a position to take a core or 2 to test actual insitu strength?

 

[henri2]

As concretemasonry has pointed out, there are a host of factors that can lead to low compressive strength test results.

Depending on how the cylinders are handled from fabrication right through testing, results obtained, may not even provide a proper indication of the in-situ strength.

A few questions for you.

1. Can you please post the mix design? Weights of material per cubic yards, including name of admix and dosage rate would suffice for now.

2. How has this mix performed in the past under similar conditions and around the same time of year? You should be ble to get this information from the ready mix supplier and or testing lab.

3. Over what period of time have the results been low...a couple of days, weeks..?

4. Were any specimens cast for 7-day results...and if so, were these results lower than expected?

5. Were the cylinders fabricated by ACI certified techs?

6. Do you have any evidence that the initial curing of the cylinders was done in accordance with ASTM C 31? Check Section 9.2.1 and Note 6 for editions of C31 prior to 2006. In the 2006 edition, check Sec 10.1.2 and Note 5

7. Typically how long were the cylinders out in the field before they were picked up by the lab? You should be able to get this information from the lab. In cold weather, if they are left out too long without adequate protection/curing, the results will most likely be lower than expected...same thing can happen during hot weather.

 

[JAE]

There have been some other threads in the past that have dealt with low concrete strengths - and with some good answers by BigH and Ron. Do an Eng-tips search.

Here's a short thread with a sequence of steps to take in accordance with Chapter 5 of ACI. You may be familiar with it but I thought it would be an appropriate link here:

thread167-72649

 

[USFEngineer]

Thanks for the replies.

I have a call into the testing lab and concrete supplier.

I will get the mix info as soon as possible and post. Thanks.

I too have concerns about durability and think coring the concrete would be a good idea.

 

[civilperson]

Cylinders are a measure of the strength of the concrete IN THE CYLINDERS. Take cores and use Swedish impact hammer to identify locations in the structure where strength seems less than specifications. Run calcs to see if the location can survive with the measured compressive strengths. Ask for a extended warranty or a price reduction for the portion left in place with low strength. Remove and replace where necessary.

 

[structuralaggie]

If the strength was reached at 56 days, isn't that acceptable? I know 28 days was speced, but the required compressive strength would have been reached.

 

[Lion06]

I've always wondered about the cylinder break results. Aren't these cylinders taken back to the lab and allowed to cure IN WATER until it is time to break?
If that is the case, how can that possibly reflect the actual curing conditions of the in-place concrete that is not kept consistently moist?
It seems to me that the in-place strength should always be lower than the cylinders that have "ideal" curing conditions.
I used to work at a pre-caster that had a curing system that was able to adjust the temperature, as well as some other conditions, based on the information sent to it from wires embedded in the concrete beds. This was just to ensure that it reached the required release strength. Once the pieces were stripped from the beds, the cylinders went in the water and the pieces went out in the yard. It rained occasionally, but those pieces were certainly not kept constantly wet.

 

[UcfSE]

Make sure you discuss this with your client. If you approve the low strengths, you accept the liability if there's a problem down the road, such as severe cracking. Even if after all your research you approve the low concrete strength, make sure that you discuss the ramifications with your client and let that person go to the owner. The owner, after all, is paying for 3000 psi and deserves to have a hand in the decision-making process along with your client.

Also check out ACI 318 1.1.1, if applicable to your area. Going with less than the code minimum does not sound like a good idea. I would make sure to do what tests you can, but if it needs to come out, so be it. Some one else's mistake is not worth losing your high-knee

 

[golf39]

I have read the posts and I have a couple of comments that may help as we just went through this on four buildings where 25% the column strengths were less than the specified 5000 psi. One 56 day test result is not considered a test as two cylinders are required to constitute a test. Therefore, it is up to the Structural Engineer of Record if he wants to accept one 56 day test as an indication that the concrete has reached its design strength. We did not, as they were also low and we required testing in accordance with ACI 318-02, Section 5.6.5 - Investigation of low Strength Test Results, ACI 214.4R-03 Guide for Obtaining Cores and Interpreting Compressive Strength Results and ASTM C 42/C 42M-99 Standard Test Method for Obraining and Testing Drilled Cores and Sawed Beams of Concrete. The contractor & concrete supplier wanted to use a Swiss Hammer or Winsor Probe, but we rejected these due to the unrealibility of these test procedures to accurately determine the in place concrete strength. Three 3 3/4" cores at each deficient column were taken, they were corrected for length as some of them did not meet the height of 2 times the diameter, and they were tested, and all but three cores failed to meet the
85% of F'c or 4250 psi required for cores by ACI. The Structural Engineer of Record ploted Interaction Curves of the Column Axial Loads vs Moments, showing the reduced concrete strength could be accepted. However; all the cost for coring including engineering time to determine acceptance was passed to the contractor/supplier, along with a extended 5 year warranty to repair or replace any of the colums that develop any future structural distress. While I think all this work was worthwhile from a life safety viewpoint, it should be pointed out that usually the cores I have tested in the past usually meet the 85% F'c. Hope this helps, of course it is always best to resolve these issues as soon as they occur, but unfortunately that is not always the case.

 

[boffintech]

StructuralEIT,

To answer your question I'll paraphrase Ron from an earlier thread: The purpose of taking cylinder specimens under standardized procedures, curing them under standardized procedures, and testing them under standardized procedures is to see what the MIX DESIGN is capable of doing.

Cylinders are a comparison to the mix design not to the in-place concrete. Probably why ACI-318 requires cores from in-place concrete to be at least 85% of f'c.

 

[dik]

Some issues are being overlooked...

The concrete is outside specifications! what is the contractor going to do to remedy this? It really matters little why a particular mix was spec'd.

Can the contractor get an engineering report to the effect that the concrete is adequate as placed?

The purpose of casting cylinders is to have a uniformity of strength. I'm not good enough to know, by looking at a deposited mix what the characterists are... I do know that by spec'ing 25 MPa concrete that it has characterists of a 25 MPa mix that I spec'd 10 years ago and that the characteristics will be similar in 10 years and I have come to expect that... I'm not so sure about a mix that is only 70% of this...

In most cases, except for shear, the amount of reinforcing is small enough that the concrete strength has little effect on the flexural strength... I spec and design 25MPa generally just because I'm comfortable with it and it behaves as I expect it should...

 

[DaveVikingPE]

OK...

1. You spec'd a particular f'c. right?
2. You didn't get it, if I read your thread correctly.
3. Someone - maybe you, maybe not - wants a workaround. An "ok, that's good enough" so you don't have to rip out the work and repout, a costly alternative, no doubt. You want to avoid the added cost, lost time, etc.

Are the codes, your own specifications, and everything else just loose guidelines? Are cylinder break strengths really nothing more than something to satisfy someone's idea of "good enough"?

Why should we do any testing at all? Everything's good enough, right? It will probably be OK, right? 56-day strength is good enough and if that doesn't satisfy, just do it at 117 days, it's good enough, right?

After all, we don't want to cost the contractor money.
And we don't want to delay the schedule.

You called for f'c = 3000 at 28 days and didn't get it. Once you get the mix design... if the water content isn't in compliance with what you specified... if the strength isn't what you specified...

You have every right and a duty to reject it, to charge liquidated damages for your extra calcs (if needed), to sue the contractor to comply, etc. etc. Don't settle. A win-win for the contractor and the owner isn't a win-win for the engineer necessarily and certainly not for anyone in the building, should it fall down.

 

[JAE]

DaveVikingPE - I agree with you in principle, but the steps laid out in ACI are a bit more sequential:

1. Review what cylinder tests (2 cylinders = 1 test) have come up below the required strength per 5.6.3.3 (this is where your 500 psi comes in). If you are within the 500psi limit (5.6.3.3(b)) but are below the required f'c (5.6.3.3(a)) then what you are required to do is take steps to increase the concrete strength. This is outlined in the commentary of 5.6.3.4. You do NOT have to take cores, or do anything else...just take steps so that future concrete comes up better.

2. IF you DO come in below the 500 psi limit, you have to take steps to assure the capacity of the structure. (see 5.6.5.1).

3. These steps would first be to take the NEW, LOWER f'c that comes out of the original cylinder tests and go back to your calculations and recheck the area of the affected structure. This may surprise you as f'c doesn't have a huge impact on flexure, but may have on shear. If the various parts of the design come up OK based on your original required Live Loads, etc. then you are OK - no other steps are necessary.

4. IF the lower f'c causes problems in your design capacity, then you go to the cores. The coring takes place in threes - the average of three cores must be > 85% of f'c and no single core can be less than 75% f'c. (see 5.6.5.4). If the cores meet the above, then you are OK, no other steps are required.

5. IF the cores don't meet the required levels, then I would take the next step of either more cores, perform a load test (ACI Chapter 20), strengthen the floor with supplemental structural members, or reject the concrete and do over - but ONLY strengthen for those areas deemed understrength per step 3 above.

All of the above is laid out in 5.6.5 of the ACI Code.

 

[concretemasonry]

All of the above comments are based on the assumption that the cylinder tests are reliable and accurate. This should be verified.

If no other projects supplier by the supplier are experiencing similar problems, I would suggest determining if the batching of the concrete was in agreement with the design mixes approved. There should be batching records for a paper trail.

As civilperson said, a cylinder test is a measure of the strength of the concrete in the cylinder. It is very easy to make cylinders improperly and to mis-handle them during the critical days on site and during transportation to the laboratory. One man making cylinders improperly can cause havoc. It is also necessary for the laboratory to handle, cure, prepare and test the cylinders according to ASTM proceedures.

Is the laboratory certified? Who selected and approved the laboratory? Even in a certified laboratory, it is possible to get faulty results. In an earlier life, while working as an inspector and employee at a testing laboratory, I saw dozens of cylinders from a project tipped over in the curing room. Management neve knew and was puzzled why there was such an erratic difference between two projects supplied by the same supplier at the same period.

Often, the lower end employees (sometimes temporary during the busy seasons) test cylinders, the accuracy required for preparation and placement in the testing machine can effect the test results.

The testing laboratory may appreciate being informed of your concerns. This is especially true if all further actions are based on the results of may persons actions. Often, a good laboratory can sit back and look at the big picture and spot areas of concern if they are informed of the situation. They see the results by tesing numerous products and the materials from different sources that go into them. I have had testing laboratories call me when they recognized improper sampling, labeling or handling of test specimins.

This should give you a good basis for your further actions, since there are well thought out actions and proceedures based on accurate testing of representative samples.

Dick

 

[golf39]

While everone has some valid points, I agree with JAE. He outlined the steps you need to follow per ACI and that is essenttially what we did in the reply that I gave. While most everyone would like to take the non-compromising position to tear it out, you must follow these procedures first. Then if the concrete is truly insufficient and a life safety problem exist, you will have fully documented you case. Litigation could still result, but you are in a much better position to justify the rejection, other than that it did not comply with the specifications.

 

[BigH]

A good discussion and many valid points and views. As has been indicated in the thread and other similar threads, the cylinders are cast and cured in such a manner as to determine the concrete's potential strength - and so that different laboratories have a very good presumption of obtaining the same results.
The 28-day test strength as been specified as the "acceptance criteria" (not the 56 day or the 112 day). The test strength as as been rightly pointed out is the average of the two (or in some countries three) cylinders (or cubes). There is usually (see AASHTO) a caveat in that the results of one of the cylinder/cube breaks can be discarded if there is evidence that it was improperly cast, cured, handled, etc. Then you can, by rights, base your acceptance on the single remaining result.
You have breaks that are below the acceptance criteria. These are, in fact, non-conformances and should be reported as such. If the non-conformance meets the criteria by ACI that permits a single set to fall 500 psi lower than specified so long as the running average of any three sets in which it is part is above the specified value, you can close the NCR in this manner. If the NCR cannot be closed, then the contractor has the obligation of proposing his remedial action. The Engineer of record may accept or not the contractor's way of going forward. If not, then the Engineer of Record would determine whether or not (based on many of the responses given in this thread) to accept the concrete. You may core or carry out other tests to provide a suitable level of comfort - costs usually born by the contractor.
Remember, the contractor has stipulated that he would provide conforming concrete - it really puts the ball in his court, (does it not?) to do the work in proving that the concrete is okay or not? Concurrently, the EOR should be prepared to be able to back up his decisions.
If the concfrete is non-conforming, the EOR does not have to accept - no matter if the Contractor thinks the concrete is good or not. Any relaxation of the contractual requirements should be made with eyes open and with a view of increasing warranty period, decrease in unit rate, etc. even if the concrete doesn't have, in the EOR's point of view, any negative effect - the contractor provided a non-conforming product.
This is how I view it.

 

[epoxybot]

Fly ash will delay the strength gain if used in quantiy.

Regardless, if you are still uncomfortable with the strength gain but are considering some exception or allowance look at the Modulus of Elasticity of the concrete and or the Poisson’s Ratio first.

My 2 cents.