!Hl MI

105
680
:1TH Sr

 

, mmmvwwm

WNW

M for a» boom d B. 8-
mm .STATE‘ COLLEG!
W115. Crydm

‘ 19.48..

.__ ——-v‘_ gm“ "“

 

THESIS

Full Scale Study of Guard Rail Anchors

A Thesis Submitted to

The Faculty of
MICHIGAN STATE COLLEGE
of
AGRICULTURE AND éPPLIED SCIENCE

by
Samuel F. Crvderman
h‘.

Candidate for the Degree of

Bachelor of Science

December 1948

THEQ‘S

 

PREFACE

The work of this thesis is a continuation of H. I.
Bacon's "Experimental Investigation of Michigan State
Highway Guard Rail Anchorages" and J. F. Bovill's ”Guard
Rail Anchorage Investigation". The procedure of the
experiments is similar to Mr. Bacon's and.Mr. Bovill's, but
it was carried on with full scale anchors under actual field
conditions.

I would like to acknowledge and thank Mr. E.A. Finney
and Mr. L.D. Childs of the Testing and Research Division,
Michigan State Highway Department, for their COOperation

and advice in this project.

216960

mam

FULL SCALE STUDY OF GUARD RAIL ANCHORS

Guard rail anchorage was suggested to me as an
excellent tOpic for a thesis. This work is now under in-
vestigation by the Michigan State Highway Department Research
Laboratory, and as I wanted to do something of practical

value for future reference, I selected this subject.
PURPOSE

The guard rail anchorage investigation is now in the
third and final stage, full scale pullouts. The purpose of
this investigation is to verify or to disclaim, in full
scale, the eXperimental stages carried on by Mr. Bacon and
Mr. Bovill. It is also to determine if the pull out re-
sistance of special anchors is of a greater value so that
it is more economical to use them in place of the common
deadman, a wooden post, now used by the Michigan State High-

way Department.
HISTORY

Many articles have been written on various phases of
guard rail anchorage; but there is still a great deal of
work to be done on this subject. Dr. Wilhelm Buchholz, a
German experimentalist, wrote an article entitled "Soil
Resistance of Anchor Plates". His work consisted of testing
different sizes of square plates buried in sand at depths of

l to 4 feet with a horizontal pull through a wall.

D. C. Hubbard, A.B. Change Company engineer, worked
on full scale metal anchors. His various anchors, the screw,
expanding, rock, pyramid cone types, were buried from 3 to 7
feet invarious type soils. the anchors were pulled at a 45°
angle by means of a pole, block and tackle, Spring scales and
a caterpiller tractor. The pull out values did not vary
lineally with the depth of burial and a 15% variation due to

the different types of soil conditions, was found in the results.

R. T. Haggerstrom did some more investigation on
anchors for his thesis entitled "An Experimental Study of
Special Anchors in Sand and Clay". The anchors were buried at
8 inches and 4 inches in sand and clay. They were pulled 45°
and 600 angles. An angle of 450 was found to have the most
resistance to pull out. Clay was found to be more resistive
to impact loads than the sand.

The thesis, "An EXperimental Investigation of Michigan
State Highway Guard Rail Anchorages", by H. I. Bacon is on
the primary or eXperimental stage of guard rail anchors. The
anchors were square, round, oval, and log type buried at a
depth of 12 inches in sand. The cable was placed over a
pulley and weights were loaded in a bucket at the end of the
cable, till failure occured in the sand. The oval anchor was
found to have the most resistance to pull out.

J. F. Bovill also eXperimented with anchors and wrote
a thesis entitled "Guard Rail Anchorage Investigation". A

number of curves were drawn from the data compiled by Mr.

-2-

Bovill. These curves were entitled "Effect of Anchor Area
on Pullout Value", "Effect of Depth of Anchor on Pullout
Value", and the "Effect of Moisture Content on Pullout

Vallue". This entire work was done on the oval anchor.

With this experimental work completed; the Specific
purpose of this investigation is to do full scale work

and varify or disclaim the findings of the primary eXperiments.
MATERIALS

Figures 2,3, and 4 show the type of concrete anchors
used in this investigation. A common wooden post, shown in
Figure 7, was also used. All Anchors are shown as they were

placed in the trench in Figure 7.

The anchor Specifications are as follows:

 

 

Type Dimension Area
inches Square inches
1 Circular - Concrete 27.1 dia x 6 576.5
2 Oval - Concrete 26.6 x 15 dia 575.7
x 6
3 Square - Concrete 24x 24 x 6 576.0
4 Log - Wood 72 x 8 dia 576.0

Anchor (3) and (4) are models of a chors as specified by the
Michigan State Highway Department.
The anchors were buried in sand found near the Michigan

State Highway Maintenance Garage at Charlotte, Michigan.

 

Figure 1. Pull out apparatus

 

Figure 2. Oval Concrete

 

Figure 3. Circular Concrete

u

v. 5. “"7." A ' h I
“12%”. ,JW'A‘

v _*n‘,‘ ’3" if

 

Figure 4. Square Concrete

EQUIPKERT

The anchors were pulled out with the apparatus as
shown in Figure 1. This piece of equipment was made from
two 9 inch channels and some old lengths of grader blades

welded tOgether in a right angle with a 45° oar between

them. As the hydraulic jack was pumped, it moved the 45°
angle bar at right angles to the cable. With this mechanical

advantage the anchors were pulled until failure occurred.

The trench was dug by a small dragline and the sand
was shoveled back in layers and tamped by an air hammer as
shown in Figure 8. The pullout indicator (Figure 5) was
made from a common 1 inch steel bar, bent to form inch
circles at each end to hook cables to. Strain gauges were
put on the bar and the pullout bar was calibrated in a
universal testing machine.’ The strain which was placed on

the bar was recorded by the S.R.4 portable strain indicator

(Figure 6).

n4.-

 

Figure 5. Pull out bar.

:5:

' ’0-

r"' f0 " 24f 2f

t
n

     

3.16 . ‘
u-

.

Figure 0.
S.R.4. Portable
Strain Indicator

" ;;;

PROCEDURE

A trench was dug and the anchors were placed in such
a manner that the eye bolts would be at a 45° angle (Figure 7).
Long cables were then placed on all the anchors to be con-
nected to the pull out bar. Layers of sand of approximately
1 foot were added. Each layer was compacted with an air
hammer until the center of gravity of all the anchors was
covered with 4 feet of sand.

The anchors were then ready to be pulled out of the
trench. The apparatus was set up as shown in Figure l. with
the cable from the anchor attached to the pull out bar and
the bar attached to the pulling mechanism.

The following data recorded:

1. Force needed to cause ground failure

2. Movement of sand (recorded photographically)

Figure 7.
Anchors as
placed in
the trench.

 

~
‘ 7"“ f o
O

- s ‘T‘L' "13K!“

\

- raw-alibi” “’23“

 

Figure 8. Trench digging and
tamping apparatus.

DATA

Anchor Tested

 

Log.

Conditions

 

Longitudinal axis horizontal and at 900 to cable.

Pounds Pull
Required for

 

fig; Failure
1 21,650
2 19,950
3 2.9..922

Mean 20,540

Anchor Tested

 

Oval Concrete.

Conditions

 

Longitudinal axis horizontal and at 90° to cable.

Pounds Pull-
Required for

 

33; Failure
1 20,700
2 23,850
3 92am

Mean 22,520

Figure 9.
Ground failure
caused by

oval anchor.

 

:7
A” -v 0 ‘I
' m

 

I'.‘
*1“ " \. .~'
s .",.-l ~ ' ‘.‘ \ -

L A" if" P" 3.06:? ‘7 '\ i;

 

Figure 10. Ground failure caused by log.

DATA (continued)

Anchor Tested

Circular Concrete.
Conditions

Circular face 900 to cable.

Pounds Pull
Required for

 

No, Failure
1 19,880
2 21,150
3 12.199

Mean 20 ,110

Anchor Tested
Square Concrete.

Conditions

 

Axis horizontal; face at 900 to cable.

Pounds Pull
Required for

 

__;§g& Failure
1 19,990
2 18,870
3 22.329.

Mean l9,780

Movement of sand recorded in Figures 9, 10, 11 and 12 for

each anchor.

Figure 11.
Ground failure
caused by
circular anchor.

 

 

Figure 12. Ground failure caused by
square anchor.

 

 

 

 

 

 

SO
77...- C
RE
AT

 

////////////////////////////////////,

E
D.
Y
T.
//../ T

 

mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm

 

 

PULLOUT VALUES
07‘ DIFFERENT TYPES a»; ANCHORS

 

 

DATA (continued)

The oval anchor, which had the greatest resistance

to pull out, was selected and further tests were made.

Purpose

To determine the relationships between the depth
of burial of the oval anchor and the pull out value.

Conditions

 

0
Longitudinal axis horizontal and at 90 to cable.

Pounds Pull

 

 

Burial Depths Required for
in Feet Failure

2 10,250 #
17,180 #

4 ~ 22,520 #

Graph of Depth Vs. Pull Out Figure 12.

Fl _L1."E:’ 14

”WT” " ' "“—T—“"‘T—“' T“ T"—T

a; f
., ,/

 

 

 

 

 

 

IO

 

PULL OUT FORCE IN THOUSAND OF POUNDS

 

 

 

 

 

 

 

 

 

2 3 4 5
DEPTH' OF ANCHOR IN FEET

“—
m

RELATION a; DEPTH a; OVAL ANCHOR
m PULL OUT VALUE

 

SUKKARY

The holding power of the various types of anchors
given in the data and shown in the bar graph (Figure 15)
are very much in agreement with the scale model work. As
this investigation was undertaken to varify the work done by
H. I. Bacon; the results on this phase of eXperimentation
can be called satisfactory. The oval anchor was found to
have the most resistance to pull out, in full scale as it
did in the primary investigation. Also the order of pull
out values of the remaining anchors varied in accordance
with data compiled by Mr. Bacon.

The Depth Vs. Pull Out graph (Figure 14) drawn from
the data of this work was also in agreement with the graph
of Mr. Bovill. In both graphs the deeper the anchor was
buried, the greater resistance to pull out it had, as would
be expected” 'Mr. Bovill's graph was a straight line; the
graph from the full scale eXperiment was very nearly a
straight line.

The small difference between the pull out values of
the special anchors and of the 10g does not constitute a
change in present highway practice. At first glance one
could say the oval anchor should be used in place of the 10g,
because of its greater pull out resistance. Further invest-
igation, because of the cost of materials, time of curing,
and great weight, proves that it is not economical to use the

Special anchors in place of the common wooden deadman.

-9-

.00” [NE ONLY

 

 

’1‘I'JI, r,‘ . ‘1;
‘- \

v ‘
;'.

.
.\‘

; "‘4

.,..
. mm.

..
.
—
f -
—
. -
‘ ......
.
-
p - .
-
, o

 

 

 

MICHI BAN bYATE UNlJVERSlT( LIBRARIES

llll | IN

65