Non destructive testing of concrete structures
How risky is it to estimate the value of compressive strength of concrete using a rebound hammer test (RHT)?
The concrete testing failed when testing the concrete cylinders for one of our pile caps. One cyclinder failed early in shear and the other was a machine error. This means we do not have the 56 day compressive test strength for the cylinders to indicate the strength of the pile cap. We require 80MPa. At 7 days we had 60MPa, at 28 days we had 69 MPa and we don’t have a value for it at 56 days. The concrete company came out on site and conducted the RHT at 12 locations on the pile cap in question, and 12 locations on a similar pile cap as a comparison. They estimate from this test the compressive strength is 86.2MPa so comfortably above the 80MPa required. This doesn’t necessarily sit easy with me. The RHT is a test to establish the surface hardness of the concrete structure so I don’t 100% understand how this can be compared to compressive strength unless through empirical data maybe. The fact that we went from 60 at 7 days to 69 at 28 days doesn’t fill me with confidence that we will hit 80 at 56 on that current trend.
So what…..core it and test the core? or, does this pile cap actually need 80MPa therefore can we take the risk on it being borderline? – it is one of the mega piles which attaches into the core as part of the stability system though!!
I know the fact we have a bit of paper from the concrete manufacturers giving an estimate of 86.2MPa will probably suffice for BMC – someone has put their signature to it – but hardness vs strength doesn’t quite compute in my head. Have we mitigated the risk appropriately or not? Thoughts welcome.
Having sought advice….
The RHT (Schmidt) compares concrete thereby giving an indirect assessment of concrete strength. In this scenario, the test can be conducted on the pile cap in question and another similar pile cap for which we have a positive crush strength, and the results compared to estimate the strength.
BS EN 12504-2 details the procedure for doing this (thanks John). Having cast my eye over this short document there are a couple of points which are important to abide by to get reliable results:
- The surface must be smooth to get an accurate result so any rough surfaces must be ground down first.
- Test locations between comparative structures must be tested under similar conditions and they should be tested at similar positions to reduce any possible effects due to differences in rigidity and uniformity of concrete.
- Moisture condition of the surface should be consistent throughout the testing. Dry surfaces are preferred.
- The coefficient of variation of individual readings within one test should be in the order of 10%. It should decrease with an increase in strength.
You now know what you can do, just in case any of you end up in a similar position!!
Holdfast Training Services Trial Blog 
Jo,
I assume that only one set of cubes was taken owing to the fact that the total volume was below 75m3 or there about meaning you only took one sample set. We have been taking a spare cube per test for this reason.
Do you have much confidence in the quality of concrete or type. If you have already poured 100’s of m3 of the same concrete and it was fine you could show similar trends or compare the 28 day strength for other pours.
Just thinking If this happened here. In the past I suspect that we could have argued the above case (trends etc). However since we messed up and poured the wrong concrete a few weeks ago confidence has taken a bit of a hit and if it happened now, I suspect we would be looking at the coring option.
Johnny, the pile cap was the only concrete structure poured that day. It measures 2700 x 2700 x 1700 so has a volume of 12.4m3. Only one sample was taken. We’ve only poured the pile caps and the capping beam to date and so far everything has reached the 80MPa we needed so this gives us confidence. However, I had my concerns for the following reasons: 69MPa at 28 days was a little lower than others had reached – most hit the required 80MPa at 28 days; this pile cap sits underneath a mega column transferring 111MN of axial load and a moment of 16.5MNm into the foundations – its one of the largest and most important ones on the project; it also plays a key role in the main stability design for the building – the floor acts as a large diaphragm linking the mega column pile caps back to the core and this pile cap sits in the middle of this system.
I didn’t really know much about this test, how it was used, and what risk was posed, and owing to the risk associated with the potential failure of this pile cap, I didn’t feel comfortable to take someones word that all is good when that person knows even less than I do. Having read the BS EN and checked the testing procedure by the concrete subcontractors I feel confident that the pile cap has sufficient strength and is structurally sound.
Jo, this is a common issue so great to bring it up. Within the codes and standards there is an allowance for a variation in results. I think plus/minus 4 MPa in a certain number of tests. This might not be applicable because you only have the one result but perhaps one for the future.
Risk. I had a number of instances on site where ‘actual’ (or tested) compressive strength was below design strength. This was packaged up in a non-conformity report and sent to the designer to conduct checks to confirm whether the deviations were acceptable. Handily they were. I suspect this pile cap forms one part of the spring stiffness resistance to the structural system as a whole. Therefore if this cap is less stiff, the load might be able to be re-distributed – something the designers can check. (possible my strength to stiffness relationship is widely out!)
Does the risk of a failed strength gain sit with you (contractor), the designer or the supplier? I.e. who pays to fix the problem?
Testing. A cube test is a useful indicator if mimicking identical conditions to the pile cap. Are you curing cubes in a water tank at 20 degrees? Are you curing the cubes in the same way you did the pile cap? Another option for the next pour, might be to use temperature matured curing – this ensures like for like conditions.
Alternatively, thermocouples placed in the pour and the test cubes can be used to equate in-situ concrete temperature to strength. Will probably be too late to start for pile caps, but might be worth doing if you have lots more in-situ pours. Issue is thermocouples are quite expensive as is the data logger so might not be so easy to persuade the PM you need one!
Trials. Did the concrete supplier do any trials on this mix design – I assume you will have more pours of the same mix to do? If so, did they get something similar?
Random other question – did you see any water while excavating to formation level?
Damo, all really good questions however I am going to struggle to answer all of them.
The ground works subcontractor are responsible for constructing the pile caps as per the design specification. They source the concrete and take full responsibility for it so the responsibility of understrength concrete lies with them. The concrete manufacturers have done the concrete mix design for Delta and they also do the testing so I can’t tell you how they are curing the cyclinders – they are taken away from site and a piece of paper arrives 30 days later with the test results. As for thermocouples – not a chance!!! So long as BM get a piece of paper from the subby’s saying the concrete has reached the design strength, they are happy How they test the concrete is up to them. We do have significant amounts of insitu concrete to do, but this is done by a different subcontractor with a different concrete manufacture. My thesis is related to concrete properties and admixtures so I’ll be taking a much greater interest into this sort of thing – in the ‘planning stage’ at the moment to work out what I actually intend to do.
As for water, we have seen a little but I am quite impressed by the low levels of water ingress considering the rain we have had recently. The walls are pretty much dry and the couple of leaks we have had were subcontractor non-conformances where they excavated too much of the capping beam/jet grout interface. The base of the excavation appears to do be doing its job as well – it is a little soggy in the base of the excavation but thats mainly because, a) we’ve had a lot of rain into a waterproofed hole and b) we’ve been coring 144 1500mm holes into the king piles for the raft so they have needed substantial amounts of water. I’ve got some pics which I’ll up load onto a separate blog later on so you can see. The whole concept of the grout strut is very interesting, especially as it appears to be working quite well, but there has been a big payoff with regard to the design of the structure. Watch this space.