Insitu Direct Shear Test (concrete block on rock)

[cmorgh]

Hi.
To investigate the friction between concrete and rock in a project insitu direct shear test was performed which consisted of 3 blocks and each block was assigned to a specific normal load. The concrete blocks were cast in place as 70 x 70 cm in plan and 30 cm in height. The normal loads were 5, 10 and 15 metric tones, respectively. Upon completion of the tests, three different shear failure modes occurred:
1- At 5 T: Shear (sliding) plane passed through the concrete block.
2- At 10 t: Shear plane passed through the interface of the concrete block and rock.
3- At 15 t: Shear plane passed through the rock itself.
Questions:
a) Is such a result (various sliding mechanism) quite normal? As I expected mostly the sliding planes passing through the concrete and rock interface.
b) As the failure mechanism occurred in different materials each time, hence plotting the corresponding normal and shear stresses for each block and sketching the regression line (to get the friction angel and the cohesion) should not be a proper analysis?
c) In such a case, how to evaluate the friction angel between the rock and the concrete having the said results?

 

[GeoPaveTraffic]

I would start by plotting all of the data together, as you indicated in b). While it may not be "proper"; it may be telling. It is possible that the strength of the concrete and rock are about the same, and that the strength of the bond between the two is close enough to the other two to get lost in the scatter.

Would you be willing to share the data with the rest of us?

 

[TDAA]

I agree with GeoPave Traffic. Plot it all together, and see how it looks. I would draw various lines through it (depending on how the data lines up). These would include “connect the dots”, a best fit, and from zero to each point. Look at the data compared to the known failure and see what makes sense.

We tend to concentrate on the cohesion and friction angle in these tests, when often what we are really looking for is the shear force vs. the normal load, and the actual friction angle numbers do not matter. What you may have are differing failure modes, with three different angles; however, as a system, you may be able to show that at a certain normal load, you have a certain shear resistance. Do you care if the angle is low, if the shear resistance is high enough, and you don’t anticipate loosing that "cohesive" bond? What you may want to do is obtain data between the points you have, and see if it forms a non-linear relationship.

What type of rock was the concrete cast against? How rough is the surface? How comparable is the surface at the three locations?

 

[cmorgh]

Dear GeoPaveTraffic and dear TDAA,
Thank you very much for your cooperation. I have the same idea, but the point is that by sketching the three points, you may fit a line fairly suitable for the 3 points. The point is that the line shows negative cohesion (all the 3 points fairly coincide on the line fitted).
I won't be available for some days (going on duty for some projects). Will share the data as soon as I comeback.

Regards
cmorgh