Q:
I'm running an earthquake analysis using the dynamics module. I have, though, two questions for which I could find no answers in the manual:
1. How can you look at output from intermediate steps?
2. How can you view animated output? The manual describes how to create animation, but not how to view it.
A:
1.) The analysis of intermediate steps is explained in the faq, bulletin number 9.
2.) 2.) After selecting the option 'create animation' from the 'view' menu a window appears in which the phases of the calculation are shown and can be selected for making an animation.
Select the phases of interest and press 'OK'.
A small window will appear that shows the amount of time left until completion of the animation.
Upon completion, the animation is automatically started.
The animation file is created with the extension .AVI and is placed in the .DTA directory of your problem.
Q:
I have a typical excavation problem, soil layers separated by a vertical wall. On one side excavation is going on with water table lowering 2 ft below at each step, on the other side the water table remain constant. My questions is when perform the first step of excavation, in order to generate water pressure via phreatic level, what is the "correct" way to do it near the wall? (Due to unknown permeability parameters, we don't want to use Groundwater flow to generate water pressure.)
Should my phreatic level follow the wall excactly (a vertical phreatic level? which seems to be theoretically not correct), or should I intentionally make it slightly slanted? ( 1 ft off laterally or make it 89 degree instead of 90 degree)
From my experiment, these two methods produces roughly the same results anywhere. However, in the water pressure input window, after you draw a vertical phreatic line,
some external water pressure is shown on both the wall side and the bottom of excavation. They both take the triangular shapes.
What do these pressure diagram mean? Where are these pressure applied to? The Wall or the soil behind the wall?
One funny thing is the pressure applied vertically downward the excavation. On the point near the wall it shows a water pressure based on water head of the unexcavated side. On the point away from the wall it shows a water pressure based on the 2-ft-below-excavation. This external water pressure (according to Plaxis manual is only defined at two end points) is a significant number. I assume this load only could make a huge difference. However, there is almost no difference in two analyses.
Also if I specify my phreatic line slightly slanted, the external pressure on the wall also disappeared? So what is going on with this "external pressure diagrams"
To make it clear:1st way of phreatic line:(-200, 0)---(0,0)--(0,-31)--(50,-31)
2nd way of phreatic line:(-200, 0)---(-1,0)--(0,-31)--(50,-31) Wall (0,6)--(0,-55)
A:
I was wondering whether you have considered few important points while generating external water pressure through General Phreatic Level option. They are
1. If the phreatic level crosses the boundary in a non-existing geometry point, the external water pressures cannot be calculated accurately. (refer to Page 3-61, Reference Manual, Plaxis 2D-Version 8).
2. Thus, while creating geometry model introduce extra geometry points for phreatic level boundary. Then, define the phreatic level by using these geometry points.
3. Then generate external pressure using different phreatic lines (slanted or vertical following excavation boundary) and you will see the difference.
4. Always keep it in mind that the phreatic level is nothing but the level where the water pressure is zero and below its the water pressure increases linearly with depth according to the specified water weight (i.e. the pressure variation is assumed to be hydrostatic).
This water pressure is taken into consideration while calculating stress in the soil body or any structure, thus exist in the model.
The disipation of water pressure to the load applied depends on the drained or undrained condition. In case of undrained condition (general general calculation practise in fine grained soils) excess water pressure generated while in case of drained condition, the water excess water pressure is generally compensated with the settlement (or deformation) in the soil body.
5. If you are already aware of the above things and the problem still persist in your model then please revert back with the exact input condition of your model.
A:
" 4. Always keep it in mind that the phreatic level is nothing but the level where the water pressure is zero and below its the water pressure increases linearly with depth according to the specified water weight (i.e. the pressure variation is assumed to be hydrostatic). "
I think that is an important point. I don't know how plaxis is coded. But to me, if a vertical phreatic line is specified, it seems that the program is confused about which point to use as the zero-pressure-point for calculation of water pressure along that vertical line.
And it is confused with vertical phreatic line as I observed the pore pressures along the line.
I have been able to circumvent the problem by specifying a discontinuous phreatic line via cluster phreatic line. It worked well. But a little bit more time consuming.
Another thing I observed with vertical phreatic line (or slightly slanted phreatic line) is that it generate a large gradient of water pressure either upstream or downstream. I suspect that may leads to numerical problems. However if no numerical issues ever happen at last, then it still provide a good approximation.
I also noticed that it is suggest that a phreatic line should pass through an existing point/line to obtain desirable accuracy. I have tried to avoid using additional geometry lines, since that generates an additional layer. I put some additional geometry points instead.
Still I wonder if the phreatic line does not pass the exact geometry point/line, how much the error is. If the error is propotional to the element size of the element crossed by the phreatic line, then probably it is still not too bad. If the error is propotional to the distance between the two closes existing geometry point, it may be very large. But I suspect that
the error should be decided by the element size.
A:
While I fully agree with what Mr S K Panda has written, I would like to add the following points.
You said "Due to unknown permeability parameters, we don't want to use Groundwater flow to generate water pressure". My comments are:
1. before we carry out an analysis we should at least have an idea what the soil is, being sandy soil or clayey soil. Although we know that the permeability of a soil can vary by orders of magnitude, the typical range of permeability of sandy and clayey soils can still be found from a text book.
2. the Groundwater flow function in PAXIS Professional is to establish the steady state pore pressure distribution. If the soil is very sandy then the Groundwater flow function is applicable to your case, where steady state flownet can be established quickly within the excavation. Therefore, a high pore pressure gradient is built up beneath the excavation base which may impose a boiling/uplift problem. On the other hand, if the soil is very clayey then the steady state condition may not be reached in short time. But you have to specify the clay to undrained behaviour where excess pore pressure can be generated within the excavation. If the excavation is to be carried out for a sufficient long time, then a Consolidation type calculation has to be carried out to model the swelling of the clay within the excavation (i.e. dissipation of the excess pore pressure which is suction). For the Consolidation type calculation you have to do the Groundwater flow calculation to let the analysis know towards what steady state pore pressure condition it is consolidating to.
If I am not mistaken you had drawn the phreatic line in the fashion “horizontal (retained ground surface) – vertical (along the wall face) – horizontal (excavation base). Then you generated the pore pressure by the “Phreatic line” option. The pore pressure distribution that you had generated was theoretically wrong, as the pore pressure distribution shows a steep gradient across the toe of the wall. The pore pressure distribution within the excavation tends to be low, which may underestimate the boiling problem. You can manually calculate the pore pressure at the wall toe using the “linear seepage approximation”, for which the formula can be easily found from a standard soil mechanics text book. Then you can specify the manually calculated pore pressure distribution using the “User defined pore pressure distribution” function in PLAXIS, which is very straightforward.
Please bear in mind that your previous method of specifying the Phreatic line “horizontal – vertical – horizontal” could be detrimental to your design for deep excavation, as you may seriously underestimate the high pore pressure gradient (i.e. boiling problem) within the excavation.
The problem associated with the specification of a vertical phreatic line is common to FE software packages. In PLAXIS, you can specify the line slightly slanted (such as 89.99 degree) to eliminate this tedious problem. If you are still not sure with the pore pressure generated then you can always draw a cross-section line immediately next to the wall to check the pore pressure acting behind and in front of the wall.
Last comment. Why don’t we just do a hand calculation first before jumping into more sophisticated FE modelling, considering that we still don’t understand what the soil is.
A:
The problem I am working on seems to be different than what you are assuming. It's a slurry wall to bottom of rock with water
cutoff. There is little water flow if not perfectly impervious. While the permeability pamameter of the Rock layer is not available.
Instead of grabbing some textbook permeabilities, I think its more conservative to use phreatic line to generate pore water pressure.
My current way of generationg pore pressure is to specify different phreatic lines for each cluster. Which generates no gradient (exactly horizontal active pore pressure contour in both the excavation side or the retained soil side.
Also from the general methodology point of view; about the following points you mentioned, I have some interesting observations.
" If I am not mistaken you had drawn the phreatic line in the fashion “horizontal (retained ground surface) – vertical (along the wall face) – horizontal (excavation base). Then you generated the pore pressure by the “Phreatic line” option. The pore pressure distribution that you had generated was theoretically wrong..."
" In PLAXIS, you can specify the line slightly slanted (such as 89.99 degree) to eliminate this tedious problem."
The first way (horizontal-vertical-horizontal phreatic line) leads to a large gradient at the excavation side. The 2nd way (89.99 degree) leads to a large gradient at the retained soil side. This seems to be different from what you as pointed out, instead, the 1st way is more likely associated with soil boiling problems.
Try the 2nd way (89.99 degree), and I bet you won't see the water pressure gradient downstream (or the excavation side).
In general I agree with you that a steady state flow model is more realistic. But we'd better have enough good parameters to do that.
Hand calculation, hmm, what do you suggest we shall check for? The water pressure or something else? For more complicated geometry, its gonna be a little bit difficult. Again I agree with you that we can't rely on the computer program too much.
A:
It seemed to me that I have misunderstood your typical excavation problem. I thought there is a retaining wall for which the wall toe is not founded on rockhead. This means there is flow around the toe of the wall.
In your case if you are confident that the flow beneath the wall toe could be reasonably cut off then I fully agree with you that Groundwater flow calculation is not necessary. Your previous method of specifying the pore presssure is reasonable. The boiling problem is unlikely in your case.
