atmosphere without. If the barometer be at sea-level and the temperature of the mercury column 32°F., the height of mercury

FIG. 18. The principle of the atmospheric barometer, the condenser vacuum and the measurement of pressures above the atmosphere.

will now measure exactly 29.921 inches for such standard conditions.

Vacuum Pressures. It has already been pointed out that measurement of pressure by means of the steam gage indicates a pressure over and above that exerted by the atmosphere and consequently to ascertain the true absolute pressure of the fluid under measurement we must add to the gage reading the atmospheric pressure of the day. And so in the measuring of the pressure of a condenser, unavoidably there has grown up a similar but opposite custom in which the pressure is measured down from the atmosphere. Such a reading is known a vacuum pressure. In order then to ascertain the absolute pressure P under which a condenser is operating it is necessary to subtract the vacuum pressure reading P. from the atmospheric pressure reading P.. Thus

as

Thus if a condenser is operating under 28.5 in. of vacuum and the atmospheric pressure is 29.92 in., we mean that the actual air and steam still undisposed of in the condenser exert an abso

lute pressure equivalent to the difference between 29.92 and 28.50 which is 1.42 in. of mercury.

Confusion in Pressure Units.-We now see that readings in inches of mercury for low pressure and pounds pressure per sq. in. for high pressure are expressions that are not at all comparable to each other and hence their interrelation becomes an endless source of confusion.

Relationship of Pressure Units. By careful measurement of the atmosphere at sea-level, scientists have established that the height of a mercury column with the mercury at 32°F. in temperature is 29.921 in. Such a column of mercury one square inch in cross-section weighs 14.696 lb. This gives us at once a method by which we may transfer inches of mercury Im into pounds of pressure per square inch P.

Thus

29.921

P 14.696

(3)

Inches of Water and Pounds Pressure per Square Inch.Very slight pressures are often measured in inches of water above or below atmospheric pressures. Thus, in determining the draft of a chimney, a "U" tube is inserted into the chimney, and the height of the unbalanced portion of the water column indicates the draft in the chimney in inches of water. Since a column of water 1728 in. high and one square inch in cross-section at 100°F. weighs exactly 62 lb., the inches of water I may be converted into lb. pressure per sq. in. P by the formula

The Thirty Inch Vacuum.-In engineering practice a thirty inch mercury vacuum is considered to be the point of absolute zero in pressure. This is not strictly true, however, for we have just seen that such an absolute zero point is reached under a vacuum pressure of 29.921 in. of mercury. The reading of the column of mercury in this case is taken when the mercury is at a temperature of 32°F., which is the standard temperature for scientific measurement. If, however, we change our standard to that of 58.4°F. the same weight or pressure of mercury now measures just 30.0 in. This temperature is more nearly that of the condenser room where atmospheric pressures are read and since it makes a column of even thirty inches in height, we shall adopt such a reading at 58.4°F. as standard for absolute vacuum meas

FIG. 20.-5,000 k.w. steam turbine station, oil burning, Pacific Gas and Electric Company, Sacramento, California. Standby plants of this nature viewed in connection with hydro-electric network afford unusual opportunities for æsthetic design in exterior, as shown in this view.

urement. We shall, however, bear in mind that the same column at 32°F. would stand at 29.921 in.

The Practical Formula for Conversion of Pressures. Since we have thus established an even unit for the standard vacuum, we may also consider 14.7 lb. pressure per sq. in. as its equivalent instead of the cumbersome figure of 14.696 as stated above. This involves an error of four points in fifteen thousand which is negligible. Our formula for reduction on the thirty inch vacuum becomes

To Reduce Barometer Readings to the Standard Thirty Inch Vacuum. Although 58.4° is nearer the condenser room temperature than is the 32°F. basis, still for accurate measurement the actual temperature of the medium surrounding the mercury column should be ascertained and thus a correction must be made to reduce the height of the mercury column to what it would read if at a temperature of 58.4°F.

This is best illustrated by taking a concrete example. Let us suppose that the mercury column inserted into the condenser of a turbine reads 28.56 in. when the mercury temperature is 82°F. and that a barometer in the vicinity indicates the atmospheric pressure in the condenser room to be 30.08 in. of mercury when its mercury column is at 78°F.

The first thing to be done in the solution of this problem is to ascertain what the two mercury columns would have read had their respective mercury columns been at 58.4°F. Scientific investigation indicates that the expansion of mercury is according to the following equation in which I, is the height in inches. of mercury at t°F. and Im at 58.4°F.

It = Im [1.0026 + .000104 (t-
[1.0026.000104 58.4)]

Ι.

(6)

Hence, to ascertain the true vacuum reading in inches of mercury we find by substitution

28.56 Im [1.0026 + .000104 (82

..Im

=

= 28.415.

58.4)]

Similarly to compute the corrected barometer reading of the day, we find by substitution that

The net absolute pressure will now be the difference between the corrected atmospheric barometer reading and the corrected. vacuum reading for the condenser, which according to equation. (2) is

I p

=

29.942 28.415 =

1.527 in. of mercury.

Since all standard vacuums in engineering practice are now measured on a 30 in. vacuum basis, we find that the corrected vacuum reading I, for a condenser is

со

30 1.527 = 28.473 in. (vacuum).

су

(7)

This corrected vacuum reading I., which in this case is 28.473 is commonly spoken of as the vacuum referred to a 30 in. barometer.

For delicate scientific work this reading should be carried to still further refinements by making a correction for the expansion of the brass on the barometer scale and also for a variation in gravity at the particular place of measurement. At high altitudes and extreme northern and southern latitudes such a correction is essential.

Corrections for the Brass Scale of a Barometer.-Professor Marks in his computation of steam tables for condenser work published by the Wheeler Condenser and Engineering Company has ably discussed the correction for relative expansion of mercury and the brass scale of the barometer as follows:

The linear expansion of brass is about one-tenth that of the apparent linear expansion of mercury exerting a constant pressure. Where a mercury column has a brass scale extending its whole height which is free to expand with changes in temperature, the readings on the brass scale of the height of the mercury column must be corrected for the relative expansion of the mercury and the brass scale. The following table is taken from table 99 of the Smithsonian physical tables and gives the constants for various barometer heights by which to multiply the temperature correction in order to obtain the corrections of the mercury column.

Example. Reading of barometer 29.84, temperature of barometer 79°F. . In the foregoing table the nearest figure to 29.84 is 29.8 opposite which the correction factor is .0027. If it is desired to reduce the barometer to a 58.4°F. standard, the change in temperature is from 79° to 58.4° 20.6° and multiplying .0027

=