Down to Earth

We often hear statements like “we did a CAPWAP and now we know the bearing capacity of the foundation”. From our follow-up questions, it is clear to us that some who make such statements do not really know what was done, and therefore don't appreciate that their conclusion may not be correct.

So what has taken place when people say they "did a CAPWAP"? A so-called dynamic load test (DLT), or, to use the official name for that test, a High Strain Dynamic Test (HSDT) was performed. This test is defined in ASTM D4945, and involves applying a load to the foundation with either an impact hammer or a drop hammer. The generated compression wave travels down the foundation and reflects from the bottom upward. This reflected wave contains information about the friction along the foundation (the shaft friction) and the toe resistance. Using accelerometers and strain gauges mounted at the top of the foundation, this reflected wave is recorded. That is in a nut shell what happened during the test itself.

After the test the recorded signals are analyzed using a signal matching program. What this program does is adjust the model of the surrounding soil originally entered by the person analyzing the data until the calculated results for the test match those measured. In other words, the model is adjusted until the signals match, and the program used for this analysis can be CAPWAP or DLTWAVE or similar programs. All these programs have one thing in common: they help the person analyzing the test results in developing a possible solution. It is important to remember that the outcome is a possible solution (and not the solution), because there is no unique solution for this process, and therefore two people working with exactly the same data may come up with different results. For that reason it is very important that an experienced person, fully familiar with the site conditions, is used to analyze the test data.

But the fact that there is no unique solution is not the only reason why this test does not necessarily provide you with the bearing capacity of the foundation (irrespective of what others in the industry may claim). If you assume that the Static Load Test (SLT) of a foundation provides you with the actual bearing capacity, a DLT sometimes over-predicts the capacity and sometimes it under-predicts. That is why it is more appropriate to say that a DLT is a quick method to evaluate the bearing capacity of a foundation. In addition there are certain limitations when it comes to using a DLT on cast in-situ piles. A good overview of these limitations may be found in an article I wrote a couple of years ago (http://www.profound-usa.com/downloads/articles/ArticleHSDT-RLT_DFI2006.pdf).

So what can be done to get the most accurate results from this test. Obviously the bearing capacity of the foundations depends on the soil conditions, and therefore a more detailed soil model will provide better results. This is where the various signal matching programs differ, e.g. the soil model in DLTWAVE can be very basic (just like that in CAPWAP) or very detailed (an option not offered by CAPWAP). The DLTWAVE program allows the user to enter the very basic SPT or CPT site data and let the program use default values to characterize that soil profile, or enter a very detailed set of soil parameters. By using the latter option, the user can really define the soil conditions, which will generate a better indication of the foundation's bearing capacity.
Another thing to consider when trying to get more accurate results is the approach during signal matching: assuming that an accurate soil model is entered into the program, the soil model is a boundary condition while trying to match the measured signals with those calculated. Some in the industry advocate that the main objective of signal matching is to find the best signal match. But this approach could lead to the situation that the outcome is based on a soil model that is substantially different from that originally entered into the program, and you have to wonder about the accuracy of results derived on that basis.

The bottom line is this: CAPWAP and DLTWAVE are programs used to analyze the data obtained during a Dynamic Load Test, and they provide an indication of the bearing capacity. The accuracy of that indication depends on the experience of the person analyzing the test results and the models used for that analysis (the more detailed the better). And hopefully this blog will provide a better understanding of what really happens when they "do a CAPWAP".