for operating dampers. The thermostat is drawn relatively to a very large scale. The temperature regulator as a whole consists first of an air compressor, as shown in Fig. 225, or one of similar construction, and arranged so as to maintain a constant pressure in air reservoir R or in the pipes of the building.

The principle of operation of the thermostat is illustrated by the diagram, although the details of construction of the actual instrument are quite different. Compressed air from the reservoir or air-pump passes through the pipe A to the chamber B, thence, if the double valve ab is open, it will pass out through the pipe C to the chamber V above the diaphragm. Its pressure then causes the end X' of the lever X'X to move downward. This lever is connected to the damper in such a manner as to close off the supply of heat when in the lowest position. If the room becomes too cold, mechanism to be hereafter described moves the valve ab into such a position as to close the communication to the compressed air in the chamber B and open communication with the atmosphere at b. This permits the air to escape from the chamber V, through the pipe C and opening b, into the air, the diaphragm in the lower part of the chamber V being moved upward by a spring or weight not shown in the sketch. Thus it is seen that by moving the double valve ab the chamber Vis put in communication with the compressed air and the damper moved to close off the heat, or with the outside air, in which case the pressure in the chamber Vis lessened and the damper is moved by action of a weight or a spring so as to admit the warm air.

The mechanism for moving the valve ab consists of a thermostat T, which may be made of any two materials having a different rate of expansion, as rubber and brass, zinc and brass, etc. Connected to the thermostatic strip is a small valve K, so adjusted that when the room is too warm the valve will be opened and when too cold it will be closed by the expansion and contraction of the thermostatic strip. Suppose the room too warm and the valve K open, air then flows through the chamber B, through the filtering cotton in the lower part of B', thence through the small tube d and the valve K to the air. The small tube d connects with an expansible chamber D and opens back of a small diaphragm. When the

valve K is open the spring S forces the diaphragm into the contracted or collapsed position, causing the lever GF to move the valve ab so as to put the chamber B in communication with chamber and permit the air-pressure to close the damper connected to the lever X'X. If, however, the room becomes. too cold, the thermostat T moves so as to close the valve K; this stops the escape of air from the pipe d and causes sufficient pressure to accumulate under the diaphragm at D to move the lever FG, so as to move ab to the left, thus cutting off the supply of compressed air from the chamber V and permitting the air to escape at b. It will be noted that air is continually escaping at K during the time the room is too hot, but this is a very short interval as compared with the entire time, and moreover the orifice at K is exceedingly small, so that the loss of air is quite insignificant. It will also be noted that with this apparatus the damper is quickly moved from a position fully open to shut, or vice versa, and that it will not stand in an intermediate position fully open or fully shut.

The manufacturers of the Johnson thermostat have quite recently designed an instrument which will move the adjusting damper connected to the line XX' slowly and will hold it in any intermediate position as desired. This is considered an advantage for systems of ventilation in which it is always desired to admit the same volume of air, but in which the relative amounts of hot and cold air are varied to maintain the desired temperature.

CHAPTER XVI.

SPECIFICATION PROPOSALS AND BUSINESS SUGGESTIONS.

170. General Business Methods. - Nearly all heatingplants are constructed by contractors, who agree for a specified sum to install a heating-plant in accordance with certain specifications, or, in absence of specifiations, one which is guaranteed to fulfil certain stipulations as to warming and ventilating in any stress of weather. Specifications are prepared either by a disinterested third party who is thoroughly familiar with the subject, or by the party submitting the proposal. The first method, although not common except in the case of large buildings, is, when the specifications are properly drawn, satisfactory both to the owner and the contractor. With proper specifications estimates can be obtained from different bidders on work of the same class and quantity, and this is likely to result in a better quality of work, and often in lower prices. Where each contractor bids on his own specifications and arranges for apparatus in accordance with his own judgment, there will be a very great difference in the quality and method of construction proposed, which is likely to result to the advantage of an unscrupulous bidder, who would, if possible, use cheap material and the least possible quantity of heating and radiating surface. It is for these reasons to the advantage of all concerned that full and complete specifications should be provided which will show, accurately, the character, amount and quality of the required work.

The specifications may be written as a part of the tender for the work, or as an independent document to which reference is made in the proposals.

The specifications are often accompanied with drawings which show the location of all the principal parts of the heating apparatus and frequently many details of construction; the

drawings are considered in every case a portion of the specifications and are equally binding on the contractor.

After the bid has been accepted a contract is drawn which should contain a full statement of the agreement between contractor and owner, and of all conditions relating to the method of payment, penalties, time of completion of work, etc.

J. J. Blackmore and J. G. Dudley, New York, acting as a committee appointed by the National Association of Manufacturers of Heating Apparatus, have given the matter relating to uniform specifications much study, and we are indebted to them for the following discussion, and also for the copy of the uniform proposals here submitted.

171. General Requirements.*" It is not within the scope of a work such as this, nor have the trade conditions in the heating business advanced to such a point, that all the details of any or every system can be provided for. The following proposed form for uniform standard specifications, however, covers the ground as fully as can be done at this time, as is shown by the recommendation by the National Association of Manufacturers of Heating Apparatus, and if generally accepted by heating contractors, manufacturers, architects, investors, and the laymen installing steam or hot-water heating apparatus, would result in a higher standard of excellence. Much trouble now exists in securing best results, due to ignorance on part of owner, architect, or contractor, as well as to unfair competition or unauthorized substitutions of 'cheap' materials.

"Any specification should set forth unequivocally and in detail (as far as feasible) all that the contractor is to furnish and exactly what is to be accomplished by his guarantee, which should embody a standard of economy as well as one of efficiency. The function of the owner or architect is to stipulate what results must be accomplished according to standards in accepted use, and to give the consulting engineer (when character of heating-plant demands one) or the contractor proper latitude as to methods to be pursued. Further than this, it is the office of owner or architect, in justice to himself and to competing bidders, as well as to the successful contractor, to

* Written for this work by J. J. Blackmore and J. G. Dudley.

see that the provisions of the specifications are carried out, and that the quantity and character of material agreed upon are actually furnished and used. Certificates to that end should be demanded and given, if it is deemed necessary, since much injury is done to a legitimate and beneficial calling by what is termed skinning the job,' that is, agreeing to furnish certain things and then by taking advantage of lay' ignorance substituting inferior goods or omitting them outright.

"As already shown, the attainment of certain results follows from, and is accomplished by, scientific and mathematical processes, whether actually figured and reasoned out, or arrived at by 'rule of thumb,' as many really excellent contractors are known to do.

"In illustration, imagine a country residence in course of erection after plans by, and under supervision of, a competent architect, and note how a proper heating-plant is installed. To begin with, the owner should learn from his architect or from any other properly informed person that the desired efficiency, sufficiency, and results to be procured by the heating system depends more on amount of investment than on anything else. For instance, the same results can be achieved by employing either steam or water. The first cost, however, is less with steam, while, it is contended by many, the running and ultimate cost is less with water. The reason for this is that with the hot-water system as usually installed, with an open tank for expansion of water, the temperature of the heating medium ranges from 150° to 200° F., while with steam it ranges from 212° to 240° F.; as a consequence more radiating surface is needed for the former than for the latter.

"To continue the illustration, let the owner select steam, and also suppose that he elects to have indirect heating on ground-floor, to obtain extra ventilation (for be it understood that some ventilation, accidental or otherwise, is absolutely necessary to obtain right heating results), while on the upper floors he chooses direct heating. This done, it then devolves on the engineer, contractor, or architect to determine the respective amounts of heating surfaces required to warm the several rooms to the indicated temperature according to an accepted standard. Much harm at present results from de