the driver speed, a reduction from percentage slip in terms of the driven to that of the driver had to be made. In case of differences in the diameters of the driven and driving pulleys, a slip would also be indicated, even though no power was being transmitted. This is the case in Figure 6; there was a negative slip indicated corresponding to a difference of 0.209 per cent. in the diameters of the driving and driven pulleys.

FIGURE 6.-SLIP BRAKE LOAD FOR CAST IRON PULLEYS.

This required the addition of 0.209 per cent. to each slip observation given in column 5. With this correction and the reduction above referred to, the slips from column 5 were transferred to column 6 of the table.

Had the initial tension placed in the belt at the beginning of each test remained constant throughout the test, the work of calculating the results would have been comparatively simple but it was rarely found that the tension as measured at the end of the test was exactly the same as that at the beginning, the difference depending upon the stretch of the belt when subjected to load, its lag in recovering its original length after being stretched and the tendency of the pulleys to expand as the temperature rises due to the friction of the belt when the slip is

T1—T=1.936 W.

TABLE 3.

CAST IRON PULLEYS.-Continued.

80

82.2 2.890

Initial Tension, Net Scale Reading.

great. The effect of the former is to produce a lower tension and that of the latter a higher tension at the end of the test than at the beginning. These two influences had the effect of gradually changing the tension as the brake loads increased, from that intended and placed on the belt at the start. This rendered it necessary to make a correction in the initial tension for all brake loads after the first.

CAST IRON PULLEYS.

This correction was made as illustrated in Figure 7; the initial tensions being plotted against brake loads. The tension at the beginning of the test was plotted against the first or twenty pound brake load and the tension at the end of the test against the brake load next higher than that actually applied. Through these points straight lines were drawn, representing as nearly as possible average values. From this sheet the initial tensions corresponding to the respective brake loads were lifted and recorded in the second column of Table 3. It will be observed for the cast iron pulleys that in both the nominal tensions of 37.5 and 75 pounds per square inch, the two tensions in which the strain on the belt was the least, the tension at the end of the test was greater

than at the beginning. This is accounted for by the fact that a rise of say 40 degrees in temperature, which was no doubt reached in many cases, would expand the pulleys and increase the length of the belt about 0.016 inches, at the same time probably setting up a shrinkage or contraction of the leather itself. For the higher tensions, in which the stress in the belt was great, a decided lag was produced, resulting in the drop in tension shown.

These changes in tension also rendered it necessary to make a further correction of the slip observations as given in column 6. This was done by plotting the slip values, as corrected for difference in diameters and reduced to terms of driver speed (see column 6), against initial tensions (see column 2).

37/2

(INITIAL TENSION-LBS. PER $Q. INCH)

40

60

80

του

120

18792

140

160

- INITIAL TENSION SLIP CURVES OF CAST IRON PULLEYS.

This is shown in Figure 8, in which the numbers on the curves represent the various brake loads. The values of slip were then taken for the exact tensions of 37.5, 75, etc., pounds per square inch and recorded in the last column of Table 3.

The sum of the tensions in the two sides of the belt, or the running tension, was observed twice for each brake load and the average of all observations for each nominal initial tension