curve directly, i.e., at 1 hour; see figure 4-55 for details.

The foundation soil temperature is nearly constant and is

the same for the model as for the foundation during the

critical period of the year (defined above). The test is

performed in in-situ soil that the structure foundation is to

be founded on and that the same soil conditions extend

at least to a depth equal to the smaller dimension of the

entire foundation. The model dimensions are large

enough to minimize the edge and side effects.

k. The various test requirements may make

this approach difficult to employ; the soil temperature

requirements, for example, may substantially restrict the

time of year within which the field tests may be

performed.

I.

It must be emphasized that the methods

and ideas presented here for predicting creep of frozen

soils are still under investigation.

m. Where the tolerable foundation movement

is very small, a special investigation for the determination

of creep deformation may be required.

An illustration of the use of this method is given in

paragraph 4-7b(2).

a. General.

h. The third and probably most accurate

(1) Foundations supported on frozen

method of predicting creep is to run a field test on a

ground, ice or snow may be affected by high stress type

prototype or large size model of the foundation under

dynamic loadings such as shock loadings from high yield

consideration and apply equation 6. The field test should

explosions, by lower stress pulse type loadings as from

be performed on the model using the same

earthquakes or impacts, or by relatively low stress,

configuration, soil pressure and soil temperatures as for

relatively low frequency, steady-state vibrations. In

the foundation to be constructed, and on the same

general, the same design procedures used for non-

frozen soil. The design stress should be applied to the

frozen soil conditions are applicable to frozen soils.

model as nearly instantaneously as possible but without

Design criteria are given in TM 5-809-10/AFM 88-3,

impact. (One method of applying the load is to release

3

18

20

Chapter 13 , TM 5-856-4 , and EM 1110-345-310 .

the hydraulic pressure from jacks in a manner so as to

These manuals also contain references to sources of

quickly transfer a dead weight load from the jacks to the

data on the general behavior and properties of non-

model foundation.) After the full load is applied, the

frozen soils under dynamic load and discuss types of

deformation of the model should be recorded at frequent

laboratory and field tests available. However, design

intervals to define the time vs. deformation curve for a

criteria, test techniques and methods of analysis are not

period of eight hours. The elevation should be recorded

yet firmly established for engineering problems of

before loading and immediately after loading. (The

dynamic loading of foundations.

Therefore, HQDA

difference between these two readings gives an estimate

(DAEN-ECE-G), WASH DC 20314 or HQUSAF/PREE,

of the instantaneous deformation that occurs during load,

WASH, DC should be notified upon initiation of design

i.e. eo.) Using the time after load application as time

and should participate in establishing criteria and

zero, then deformation readings should be taken at times

approach and in planning field and/or laboratory tests.

1, 2, 3, 4, 5, 10, 20, 30 minutes, 1 hour, and every hour,

(2) All

design

approaches

require

until 8 hours have elapsed.

knowledge of the response characteristics of the

i.

Using the data obtained from the model

foundation materials, frozen or non-frozen, under the

tests, the values of e, and M in equation 6 can be

particular load involved. As dynamic loadings occur in a

determined graphically. One technique is to use the

range of stresses, frequencies and types (shock, pulse,

slopes of the tangents to the deformation vs. time curve

steady-state vibrations, etc.) and the response of the soil

on arithmetic coordinates at times of 1/2 hour and 1 hour

varies depending upon the load characteristics, the

after stress application (see figure 4-54) for details). A

required data must be obtained from tests that produce

second technique is to determine the rate of deformation

the same responses as the actual load. Different design

of the foundation at several times and plot log e vs. log l/t

criteria are used for the

curve for the 8 hours of the test. The slope of this curve

is the value of M. The value of ε1 can be read from the