Ultrasonic concrete testing

[iLiConsult]

We are currently sounding concrete with a swiss-made ultrasonic instrument call TICO. Literature says that the pulse velocity for indirect or surface transmissio is 5 to 20% lower than the velocity of direct transmission.
In our tests we find the pulse velocity of indirect transmission 50 to 100% lower than the one for direct transmission.
Is anybody out there with experience in ultrasonic concrete testing?

 

[LittleWheels]

Are those measured velocity differences fairly consistent for various concrete types/pours/strengths or is only on one concrete type/pour?

 

[iLiConsult]

Thanks for the reply.

The difference is fairly consistent considering the scatter one has to expect for these readings. Please have a look at attached spread sheet. (I guess we do not have enough readings.) The sounding was done on concrete of different pours. The mix, however is supposed to be the same for each poor.

 

[LittleWheels]

I've not had time to do more than skim your results. I have found significant differences between the surface layer and deeper depths within the same concrete pour when measuring indirectly.

Are you following the multiple measurement procedure recommended in the TICO manual? e.g. BS1881 Part 203 NBN B 15.229 as per

http://papworths.com.au/docs/8/Manuals/TICO Ultrasonics Manual.pdf

 

[concretemasonry]

Do you know if the different areas had similar actual placement.

You are measuring in-situ conditions than can be greater the the "classic" cylinder tests conducted only to measure the properties if the concrete before placement and not recognizing actual placement and curing conditions?

differences in the layer could easily be due to placement/vibration and curing conditions.

I would imagine a greater scatter compared to lab or equipment test results.

 

[iLiConsult]

Yes, we think we follow the multiple measurement procedure. Could you have a look at the two files? We do four readings per location. We avoid rebars by locating them beforehand. The diagrams of the readings suggest that there are little hiccups in most locations.
From your link I figure that your are based in Australia. The site we are talking about is an Australian investment in Phnom Penh, Cambodia. The contractor is Leighton. Leighton is struggleing with the concrete quality as the local mixing plants are not yet up to the task yet. Could the phenomenon we are taling about be concrete mix related. I have the impression that the concrete mix is a bit gap-graded consisting mainly of course agreegates and sand.

http://www.ili-consult.com/003 Ultrasonic Instrument Test Data_b.xls

http://www.ili-consult.com/002 Ultrasonic Instrument Test Data_b.xls

 

[iLiConsult]

We sounded a few (12) concrete cores before they were cruhed in the lab. The scatter was that high that one could not see any relation between pulse velocity and concrete strength.

 

[concretemasonry]

Did you look at the failure modes in detail after the cores were tested for compressive strength?

Dick

 

[LittleWheels]

I think concretemasonry is on the right track with differences in placement, etc and am doubtful that the mix alone is the problem. Obviously some mixes are more difficult to place.

 

[iLiConsult]

Many thanks for your replies.
The contractor had not only a problem with controlling the mixing plants. Placement, compaction and grout loss were issues, too. This is what kick off the investigation. I'm not supprise that we encounter concrete of different density and strength, even when the cylinders results are good.
My question is more about the difference in pulse velocity between direct transmission and indirect transmission. We learnd from the manuals and books that the ultrasonic signal is much lower for indirect transmission. Therefore the puls velocity is expected to be 5% to 20% lower. On our site the pulse velosity for indirect transmission is only about 60% of the velocity of direct transmission (means 40% lower).
If somebody could tell me that this is possible and can be explained by ... I would be happy.

 

[LittleWheels]

I have had indirect results that were more than 20% slower than direct results but not as slow or as frequently as you've had. I have some thoughts as to why this may happen but don't actually know.

 

[iLiConsult]

Dave, could you share your thoughts?
Could shrinkage cracks close to the surface be the cause?

 

[LittleWheels]

Surface or near-surface cracks can give slower indirect velocity results by requiring a longer path length be traversed than is assumed. I'm not sure that drying shrinkage cracks are deep enough to create a significant path length difference but it is possible, particularly if worsened by thermal or other effects.

Significant voiding or porosity can also result in longer path lengths but I'm not certain how much difference it makes between direct and indirect measurements. Usually a lack of signal strength dominates.

 

[goncetra]

usually, high w/c ratio when concrete is overfinished can lead to this results.

It may be possible that you are not catching the P-wave and the unit detects the R-wave, which is 60% slower than the P-wave.

 

[iLiConsult]

Goncetra,

Thanks for the replay.
I think overfinishing is not the case as we are talking about walls that were not treated after the formwork was taken off.
If the unit detects only the P-wave and not the R-wave, would this in your opinion be an instrument issue? Or could this also be concrete mix and/or w/c-ratio related?

 

[goncetra]

Sometimes it is not easy to detect the P-wave because these waves carry little energy, most of the ultrasonic energy is propagated as R-waves.

You can avoid this problem by using a through transmition configuration.