jaa. 24,1916 Effect of Variables on Asphalt Penetration Test 
817 
ute and usually under a load of 200 gm. This is done in most o° or 4 0 C. 
tests. If a material is normally very soft or becomes very soft at 46° a 
1-second test under a load of 50 gm. may be used. The time of test may 
be controlled by means of a swinging pendulum, a second clock, or 
metronome. The last is to be preferred because it leaves the eye free 
to watch the test itself and at the same time incurs less chance of error. 
In order to determine the effect of variations in time upon the pene¬ 
tration test, samples of all 12 asphalt cements were prepared and tested 
at 25 0 C. under a load of 100 gm. applied for periods ranging from 1 to 10 
seconds. The results of these tests are given in Table VII, in which every 
value recorded represents an average of a number of determinations made 
directly for the intervals of time stated. 
Table; VII .—Effect of variations in time on penetration of asphalt cements , 25 0 C., 
100 gm. 
Time. 
California. 
Mexican. 
Blended. 
Blown. 
8961 
8962 
8963 
8948 
8949 
8950 
8994 
8995 
8996 
8956 
8957 
8958 
Seconds . 
50 
62 
26 
48 
77 
35 
53 
80 
33 
63 
9 i 
3 ? 
63 
84 
33 
62 
96 
45 
67 
105 
3 <S 
73 
108 
38 
74 
104 
37 
73 
116 
S 3 
76 
124 
39 
80 
117 
42 
84 
118 
44 
81 
132 
56 
84 
136 
42 
85 
126 
4K . 
45 
88 
126 
45 
84 
140 
58 
88 
148 
43 
88 
131 
47 
93 
132 
48 
89 
145 
60 
9 i 
155 
44 
90 
135 
sK . 
49 
98 
137 
50 
9 i 
150 
62 
94 
x6o 
45 
9 i 
138 
64 
130 
183 
61 
xi6 
192 
74 
«5 
210 
49 
104 
159 
Upon reviewing these results it will be noted that for any material a 
greater number of points penetration is recorded for the first second than 
for any other one second. In general, upon the basis of a 5-second test 
it will be found that about 50 per cent of the penetration occurs during 
the first second for all but the blown type. With this type, owing 
probably to less surface tension and adhesion, considerably more than 
50 per cent of the total 5-second penetration occurs during the first 
second. After the first second there is a decided tendency for the pene¬ 
tration to become less and less for each succeeding second. But with 
the softer grades of material a difference of one-half second from the 
5-second test may make as much as 7 points difference in penetration. 
It is evident, therefore, that for accurate work in the 5-second test the 
time of penetration should be controlled to within less than half-second 
variations. From numerous tests it appears that if a metronome is used, 
the time of penetration may be controlled by any careful operator to 
within a maximum variation of one-fifth second from the selected time of 
test, and this is believed to be sufficient for all practical purposes. 
