UniSettle

UniSettle is the most versatile program available in the market for calculating total and differential settlement using conventional principles. The settlement can result from an increase in effective stress due to loads, fills, and pore pressure changes. Effect of excavations can be included. The calculations accurately model the conditions at a site with a multitude of soil layers-cohesive as well as non-cohesive-where several loads and excavations act simultaneously. The stress and settlement calculations can be performed using stress distribution according to 2:1, Boussinesq, or Westergaard methods. The influence of a changed groundwater table and/or non-hydrostatic pore pressure distribution can easily be included in the calculations.

UniSettle is the only program that enables a geotechnical engineer to calculate stress changes and total and differential settlements from a number of imposed uniform as well as non-uniform loads, excavations, and groundwater table changes, using Boussinesq or Westergaard stress distribution. And do it in mere minutes including printing out the results in tables and diagrams.

Do you remember the classical Newmark's Influence diagram?:

Investing a bit of effort and time, the diagram will help you determine the stress from a loaded area acting at one depth below the ground surface. How would you like to determine the stresses at any number of points from the ground surface and down and plot the diagram?

To produce the above graph took less time than it would take to even get started drawing the footprint onto the Newmark's diagram. This is not all, UniSettle will determine the settlement caused by this load at this point and the differential settlement between this point and another arbitrarily located point. Are there more than one loaded area? UniSettle will do it for all. And, in a shorter time than it took to absorb this text.

The groundwater table is by definition the level (to be exact, the first level) where the pore water pressure is zero. If the pore pressure distribution is hydrostatic, there is no problem with any calculation of effective stress. However, UniSettle can easily consider non-hydrostatic pore water pressure, be the gradient upward or downward, even artesian pressure.

UniSettle always assumes that no consolidation occurs at the initial condition and that the consolidation is completed at final condition. For calculation of settlement when a layer has only achieved partial consolidation, say, a certain percentage, the user simply adjusts the modulus number of that layer to that same percentage of the actual number and the settlement calculated will be for the particular condition.

A common message from Users is "why does UniSettle and UniPile use the Janbu method? I prefer to use E-modulus or cc-e0 approach". Note, UniSettle and UniPile do use the E-modulus or cc-e0 approach. The User can input soil compressibility in the form of modulus numbers, as well as E and cc-e0, as preferred. There is a simple direct relation between the Janbu modulus number m and the parameters E and cc-e0: If stress exponent is equal to unity, E-modulus can be input. If the stress exponent is equal to zero, cc-e0 is input. The transfer from the conventional parameters is automatic. In addition, UniSettle offers the Janbu method for silts and sands, which have stress exponents closer to 0.5. Try it. If you hesitate the slightest, press on {F1} or click on "Help" and the program provides the needed explanation. Very quickly, you will notice that you will classify soil compressibility with the single modulus number m, rather than by an E-modulus or cc-value (the latter is ambiguous because of the need to combine it with the e0-value).

The screen-print below shows the soil data input for UniPile when the cursor rests on the input for modulus number of the soil layer. The window that has pooped up asks fro input of void ratio and compression coefficient of compression (e0 and Cc). The a slash between the input values separates the value at the upper and lower boundary of the soil layer, which values are used to interpolate when calculating settlement for the sublayers (sublayer thickness is indicated by the "step").

UniBear

UniBear was produced to meet the need for a program capable of performing the rather time-consuming calculations necessary when combining bearing capacity and earth pressure in the design of bridge abutments and bridge piers. In addition to producing calculation result for the conventional time-proven "working stress design", the program also determines the design results for computations per the AASHTO specifications as well as other LRFD methods and partial factor of safety methods. Has anyone out there actually tried a hand calculation according to AASHTO for anything but the simplest case? Well, possibly "tried" it, but completed it? Imagine a sloping wall with a sloping ground, add a few line loads behind the wall and throw in a few tie-backs or struts, then, even a conventional design becomes a time-consuming effort. UniBear makes the most complicated combination of conditions very easy and produces rapid results and "what-if" insights. The most time-consuming aspect is to figure out what the particular code requirements are.

AASHTO specifies Caquot-Kerisel bearing capacity coefficients. Most texts prefer the Terzaghi-Meyerhof coefficients, however, and the results differ significantly. UniBear lets you use either approach.

The Working Stress Design is the standard textbook approach to the analysis of footings and retaining walls. Some texts prefer to use the Meyerhof and Terzaghi bearing capacity, shape, and inclination factors. Others prefer the Caquot-Kerisel and Vesic factors. The former is required by the Canadian codes and recommended in many text books. The latter is required in the AASHTO Specs. UniBear allows the use of all and even allows for toggling between the methods for the same structure without new input.

The above figure shows an inclined wall loaded with both vertical and a horizontal loads retaining an inclined ground surface that is subjected to a triangular surface load of limited horizontal extent. The groundwater surface is different on the inside ("inboard") and the outside ("outboard") sides.

The computation can be by the method and code preferred by the User and employ SI or English units. UniBear then computes the bearing capacity of the footing, the overturning ratio, the sliding, and the resultant location (checking for eccentricity), as well as the bending moments and the shear stress in the three sections at the stem connection to the footing. Input takes no more than a minute or two and calculations only few seconds.

Many are surprised to learn that the bearing capacity coefficients normally used in North American design of foundations for bridges and buildings are the Terzaghi and Meyerhof factors, while AASHTO Specs require the Caquot-Kerisel factors. This is important. A factor of safety of 3.0 on the one set is only about 2.3 on the other, depending on the configurations of the case. UniBear allows the User to employ either set of coefficients. It also does away with the need for all nomograms otherwise necessary to use when including line load, sloping walls, etc. and allows the computations to consider the actual wall conditions as quickly as the simplest case.

Note a wall on the footing is not necessary. UniBear can equally well calculate the bearing capacity of a footing. Neither is a footing, for that matter. UniBear can calculated the earth pressure against a sheet pile wall. And, calculate the effect of a row or more of tie backs on the wall.

Final Words

Amongst the frequent questions to our technical support is "what beta-coefficient should I use; I am designing a foundation in silt?" or "what is the best cc value to go with my e0-value, I have no oedometer test data?" These questions probably come from that same person who combines one set of data published from tests in a quartz sand with another set from a micaceous sand and applies the concoction to a design in a calcareous sand. Although, the UniSoft computer programs conform to calculation and design principles of general validity, the user must choose the input parameters from results of field and laboratory testing at the site of the structure to be designed. A good way to obtain reference values is to back-calculate observations. For example, matching the results of a static loading test to a UniPile computation of capacity. Most geotechnical engineers know of structures that settled a certain amount. When designing a new structure on the same or similar soil, why not use UniSettle to calculate the settlement of the reference structure and find what input of compressibility parameters that produces calculated settlement matching observed values. When these calibrated conditions are used in an analysis of the new structure, the results will be much more reliable than those using input from a text book.

A recent paper compared uplift resistance between anchors in clay at two different sites and gave rather hypothetical reasons for why the resistances were different although the clays were very similar. The "analysis" applied total stress. The anchors were installed at different depths and it should have been obvious that were the resistances related to effective stress, they would be almost equal. Naturally! But why do so many persist in using total stress?

Another question received was how in UniSettle to input the loading from an ore fill in the shape of a truncated cone with a square footprint causing increased differential settlement between footings which support a series of columns at a building some distance away from the ore fill, a problem not easily handled by hand calculation. The load from the ore fill can be input to UniSettle as four sloping embankments and one central square fill. The program then calculates the differential settlement of the columns using Boussinesq (or Westergaard) stress distribution. Input and results are completed in a few minutes. - One User was grateful to learn that if the pore pressure distribution is non-hydrostatic, effective stresses cannot be determined using buoyant unit weights (both UniPile and UniSettle calculate effective stress by subtracting pore pressure from total stress).

Return to main menu