Determination of Preconsolidation Pressure

[Geo69]

Good evening all,

I have a question regarding preconsolidation pressure. I was taught, like many of you, the Casagrande method of determining preconsolidation pressure which includes drawing a straight line along the virgin compression curve and determining where it intersects the angle bisector line constructed at the point of maximum curvature.

So, here is the question: review comments for a recent settlement analysis indicate the consolidation test should be extended out to a loading pressure of 50 ksf or more (my test used 10 ksf). I don't anticipate this would change the preconsolidation pressure, as long as I'm into the virgin compression portion of the curve and have enough of this part of the curve to draw a straight line.

Any input on recommended maximum loads and determining preconsolidation pressure would be appreciated.

geo69

 

[GeoPaveTraffic]

There is no "magic" value. The consolidation test should be ran to a high enough pressure to fully develope the consolidation curve. After that you are just gathering data.

Sounds like a reviewer who doesn't really understand what they are looking at.

Mike Lambert

 

[BigH]

There have been a number of posts on this and you should search the forums - several suggest that you plot log(e) or log(strain) vs Log (p') - this generally gives two straight lines and their intersection is taken as the preconsolidation pressure. I've related this before, but Casagrande came up to our company back in the early 70s (I wasn't working there then - was still in school) - and one young engineer showed him a number of test results faithfully following his method of construction - Casagrande hemmed and hawed on a few saying it should be here or there. "But, I followed your method." - "Yes, but the reason the method came about is that I had to teach my students how to approximately obtain it." - to accurately determine the "staight" line after the preconsolidation pressure you will need a number of points - not just one or two. I've had cases where the "straight" line is actually curved - and this wasn't an anomoly.