Not usual behavior of sleeve friction during PCPTU 1

[MaksimR]

Short description of behavior:
"Often a noticeable decrease in friction (ratio) occurs at 1 metre
intervals. This happens when the force on the rods/cone is reduced in
order to add another length of pushrod (1m). When the test is continued,
the friction has to "build up" to attain the previous level (end of
previous push). Full clipping op this data will result in data
loss. A similar effect occurs with the cone resistance, but the
"recovery" is much faster and barely noticeable after clipping"

Could it be "usual" behavior of silty sand, or it could be connected with water content and type of soil particles?

Any comments are welcome. I could send a sample by e-mail.
Best Regards,
Maxim

 

[Robert168]

Did you have your cone calibrated before and after the sounding?

What do you mean by unusual behaviour? Is the friction value increased or decreased, or there are too many spikes?

 

[MaksimR]

I have a calibration before the tests. We used 2 cones and result is same-same.
As i understand it depends from the tested soil but i dont understand the process of this.
Unusual - friction require more time for increase up to level before break
Show me your e-mail and i will send you a sample.

Best Regards and thanks for reply

MaksimR

 

[Robert168]

My e-mail address is robertswan65@hotmail.com

 

[Robert168]

The spikes you circled seem normal to me as they often occur in sands. I don't know its mechanism.

Why did you say they were unusual?

 

[MaksimR]

I have not so big experience in CPT testing, but i did not noted the effect before. As i understand it is not so usual. It could be local effect in this area.

I asked the question to find out this. Maybe someone more experience then me will say me......Hey, don`t worry. It`s normal. It is always take place in silty sand.

Thank you for your reply.
Merry Christmas for you, Robert118

 

[BigRedGeo]

We have had similar experiences in getting the low spikes at 1-meter intervals where the rods are changed. I'm not sure of your CPT setup, but we have attributed this to equipment deficiencies when pushing in dense/hard soils. Depending on your equipment, this may not be the case.

We use a Geotech wireless piezocone (

http://www.geotech.se/

)pushed by standard auger rigs. With this system, a "stringpot" is used for measuring depth and push speed. The stringpot is interfaced with the microphone for the cone and a computer to record the resistance parameters at user-defined depth intervals (we use 2-cm). The string from the stringpot is attached to the drill head, so the string moves in and out with the movement of the drill head. On a hard push, the drill will lift slightly off the truck due to some play in the skid. This lift will cause the stringpot to move, even though the push rods did not, and will result in the data spike.

On a CPT rig, or a different CPT unit, this may not be the case. I would take a close look at the equipment when it is operating to see if you can see anything happening which may cause this. If you've done this, then maybe there is some soi-related phenomena......I'd be interested to find out!

 

[ashjun]

Hi,
Please allow me to chip in, although a bit late here. I believe this phenomenon is not unreal and is a part of mechanics that occurs during pile or cone drive in.

When the cone is driven into the ground, a pressure bulb is generated in front of the cone tip. This causes an over expansion of the soil ahead of the cone. And as the soil is passed by the cone (or the pile tip), the soil collapses back to the shaft; that is, there is a reduction in the radial stress (radial stress relates to the sleeve friction). If the pushing is slowed down or stopped altogether this phenomenon is apparent. Furthermore, if the test is stopped for a period, you may notice that it may be difficult to push further (sometimes?). This occurs because of the increase in strength of the disturbed soil around the shaft (called Thixotropy or something) and reconsolidation...but that’s another story.

Back to this thread (sorry) during the continuous pushing (or driving in), this "reduction" is probably not apparent because the cone passes through the next layer of the soil and the sleeve friction is continuously been added.

This is true for all kinds of normally consolidated clays and loose to medium sands. I do not recall what happens in over consolidated clays and dense sands (your point BigRedGeo); I need to check that. But if you look at some of the papers by Vesic, Randolph, Puolos and Baligh, you may find much literature on the subject.

Regards

 

[ashjun]

I have given you a star to your question..an excellent point!

Regards

 

[coneboy]

Based on my experience observing CPT tests a small drop in the tip and/or friction sleeve can occur at the rod connection depths in some of the more dense soils. Ashjun previously covered one possibly reason for this but I have found this usually happens when the rig or pushing unit really has to work hard to continue the pushing operations after each rod addition. Purpose built cone trucks don't usually have this problem due to the dead weight and pushing power.

This issue could possibly show its self is if the CPT operator starts the data collection computer too quickly at each rod addition.....this is an operator error. I think most CPT operator review the data and correct these little error before the data is given to the client. On other occasions I have had the CPT operator tell me where he had to correct for a rod break.


A final note: The vast majority of the cone data I have reviewed has been seamless, that is you could not tell the depth at which the rod connection was made. In some soils the pore pressures can indicate the rod addition depths....but that is usefull info regarding the nature of the soil.

Coneboy

 

[BigH]

I don't have that much experience with actually doing piezecones - etc. other than some in middle 80s. But, could this be a result of a bit of relaxation during the push? Then, when you start to push again, you have to take up the slack - then go into your normal push. I have seen this on Sprague and Hensen (sp??) vane test when the lads would turn the crank 10 times, then stop, take the reading and then do the next 10 times - rather than continually turning the crank and taking on the fly readings. Also, you see this in pile driving at times.

 

[Robert168]

With automatic recording of data, the spikes could result from the build-break mechanism (Ashjun) of sands layer as well as from the release of pressure during rod change. When recording with a pen and a paper, an operator can wait for a second untill the erratic figure is gone.

If the spikes do result from the change of rods, which repeat at equal intervals, the erractic figures should be removed from graph and calculation.

I had a look at the graphs sent by MaksimR, the spikes, as I interpret, should be the result of the build-break mechanism (Spikes occur between the change of rods).