Pile Side Friction in Tension Pile Load Test 1

[yuanjiaxian]

Hi
Several days ago I did a tension pile test. The max tension load was 110t, which was twice the design load.Steel strain Gauges were installed in different depth of the pile.(1m,3m,6m,11 metre...) The pileside friction between 1m and 3m could be calculated with the margin of two tension (f=N1-N3).The section tension was used to be counted in the following formula: N=e*Es*A (1)
e--strain measured by gauges
Es--Young elastic modulus, the composite modulus of steel bar and concrete
A--Pile section area

But after the test, I found the Strain abnormal in the upper part of the pile. Some e exceeded 60e-5, when the permitted maximum strain of the concrete we used is 5e-5 to 8e-5. Some section tension are even larger than the load we applied.
So, in this case, what happened to the pile, How to deduce the real side friction?
Thanks for your reply.

 

[GeoPaveTraffic]

Was the top strain gauge at 1m below the surface?

How were the strain gauges mounted to the pile?

How much care was taken to ensure that the gauges were mounted coaxially with the pile?

How did you figure Es for the pile? There is likely some error in Es in each of the calculations. Remember that Es is not liniear, however, you will likely have to assume that it is and will therefore introduce some error.

Is it likely that the pile gained significant load in the upper 1 meter, i.e. the distance to the first strain gauge? If not, then I would back calculate Es based on that gauge, the pile area, and the applied load. Using that Es caluclate the load at each of the other gauge locations.

If you had a maximum strain at a gauge other than the top gauge, something is wrong. The only time I have seen this situation without a bad gauge or reading is when I was unloading a pile. During unloading the gauges in the middle tend to show a strain set, i.e. strain at no load, due to the pile locking in some of the movement that it experienced.

Give me some more information and I may be able to give you some better advise.

 

[yuanjiaxian]

The top strain gauge was at 1m below the surface.

The strain gauges were installed in the steel bars,which were welded inside the reinforcement cage

The procedure of piling: first_ boring then_grouting last_inserting the reinforcement cage
No Extra care, after droping the cage, we just move the cage head to the center of the bore.So, maybe the lower part of the cage was not in absolute verticality.

Es is not linear, but considering it to be linear is the only way that I could figure out the calculation of section tension. from most of the strain could I reckon their Es to be linear? Because the strain show that both the cage and the concrete are in their linear phrase.(e.g. 1e-6 for steel and concrete) if not, even I know the true strain of steel and concrete individually, how to count their tension?

From the surface to the bottom of the pile ,the strain were minishing, which is quite conventional. Thanks god!If they were not, I think I would be carzy.!

 

[GeoPaveTraffic]

What I would do is back calculate Es based on the upper strain gauge. Then I would use that Es to calculate the load at each other point.

 

[yuanjiaxian]

Thank you very much, GeoPaveTraffic
Still I have a doubt if the concrete near the surface was shattered, So that's why the strain increased so rapidly.

Es=(E1*A1+E2*A2)/(A1+A2) E1-Steel Es E2-Concrete Es
A1,A2 their area in pile section
I used this formula to calculate Es. In your opinion, how would I back calculaye Es based on the strain?

 

[GeoPaveTraffic]

The way you caluculated Es is correct in theory. However, it is difficult to acurately predict E2 and A2 in auger cast piles.

The way you back calculate Es is to take the applied load divided by the strain gauge reading. This will give you the load per gauge reading. Using this number and the other gauge readings you can determine the load at each point in the pile.

 

[yuanjiaxian]

Thanks again!
Now I am back Calculating...

 

[bogard]

If you plot the raw strain gage readings (in strain units) on the x-axis versus the pile head load on the y-axis, the slope of the curve will give the axial stiffness AE directly.

The graphs should initially be concave, becoming parallel when the shear transferred to the soil above each level becomes constant. The vertical distance between any two levels represents the load taken out by the soil between the two depths.

If the relationships do not become linear, then either (1) the shear transfer does not become constant above that level (unlikely at shallow depths) or (2) the axial stiffness (AE) is nonlinear. By looking at the data in this fashion, you can quickly judge whether further manipulation of the data will be fruitful.

Due to uncertainties in both the area and the stiffness of the drilled shaft, it is quite likely that you will never attain sufficient accuracy to differentiate shear transfer. Calculate the difference in load between adjacent depths by the average load at the midpoint, based on your normal pile capacity calculations. At shallow depths, this difference is only a few percent of the total load, indicating that an accuracy much better than 1% is needed to differentiate the loads with any degree of precision.

 

[yuanjiaxian]

Thank you for your reply, Bogard.
I ploted the strain and load and recaculated the axial stiffness AE. It was linear 5m below the surface , but at shallow depths(e.g. 1m, 3m),the graph are parallel to the deep ones at first,then almost become horizontal , and slope again.

could I paste the result graph on this site? Or I could email it to you if you have interest.

By the way, I wonder whether the displacement of the pile head and the pole end are equal. Someone told me the latter is much smaller. I calculated it with the following formula and found they are almost the same.
U'=U-(0+S1)*L1-(S1+S2)*L2-...-(Si-1 + Si)*Li
U' pile end displacement
U pile head displacement
Si strain measured in section i
Li the pile length between section i-1 and section i