Low strain Pile Integrity Test

[mbilal89]

Hello everyone,

I have a few questions regarding interpretation of low strain pile integrity test graphs. Attached with my post are the results of four piles tested using PDI's Pile integrity tester. All that I have read on the internet and in books regarding the analysis process is that we have to apply a low pass filter not more than one tenth of the length, high pass filter not less than three times the pile length, the magnification delay is set at 20% by default and changing the wavelength removes the noise in the results. What I don't know is how much should I magnify the waves in order to identify if a toe reflection is achieved.

I have tried to analyse the raw results, in pile 1 and 4, I certainly have noise most probably due to the protruding dowels, which I am not able to remove even by applying wavelet analysis. But I am confused in case of pile 2 and 3, whether it is actually the toe reflection or noise. It cannot be noise since I don't see a source for it. But how can I be certain??

 

[jdonville]

mbilal,

From what I understand, this type of analysis needs to be performed by a really experienced geotech - and no, it's not me. If you need someone to look at your data, I can suggest a respected practitioner for you to contact.

JD

 

[mbilal89]

Thank you JD,

Actually I did contact an expert, Garland Likins of Pile dynamics Inc. To be very honest, I wasn't expecting a reply from him, but fortunately to my disappointment he did reply, which was very helpful. His interpretation is as follows:

'The protruding dowels are likely the cause for pile 4, and perhaps to lesser extent on pile 1. On pile 1, you might convince yourself of a toe reflection around 25 m, although that length could be longer if the wavespeed were higher (WS 4200 would give about 27 m).

For pile 2, there is a large negative around 8 m, and this is likely a large bulge and it is large enough that any interpretation below that point is doubtful.

For pile 3, it is similar to pile 1 except the noise is a absent from dowel bars. Pile length is perhaps only 25 m (or maybe even 24), although as previously mentioned the indicated length also depends on the wavespeed assumption.

Magnification is used in whatever amount is appropriate (depends on hammer size and pile size and length, and also so strength), so no specific value can be given. However, I would have preferred to also see the data without any HI pass filter . LO pass filter was used and no wavelet is used (although I doubt these had much effect).

Garland Likins'

MBY