SECRETARY DUNBAR: The report, as I understand, is in the hands of the Committee to be returned.

MR. DOHERTY: Yes.

STANDARD METHOD OF TESTING FUEL GAS APPLIANCES.

Your committee wishes to recommend to the association that its report of 1903 be amended by substituting a new report, which we think should be in the form of a detailed explanation and should be published in pamphlet form, and supplied to all gas companies and appliance manufacturers. Considerable interest has been evidenced in the work of this committee, and, on the other hand, many appliance manufacturers and gas companies have seemed to oppose it. Others seem to think that it is the wish of this committee that economy of operation should be considered of paramount importance to all other desired features in gas appliances. This is not the position of your committee, for we recognize that economy alone does not determine the value of an appliance. The degree of satisfaction of results is of prime importance, and unsatisfactory results cannot in any way be compensated for by economy. We presume, however, that no gas man will disregard in the future, or has disregarded in the past, the consideration of satisfactory results secured from appliances, their artistic appearance, workmanship, ease of repairs, ease of cleaning and durability.

Many factors which go to make up the most desirable range from a commercial standpoint are matters of opinion, to be determined by visual inspection. Economy is not so easily determined, and some of the foremost appliance manufacturers confess that they do not know the exact economy of the gas appliances they manufacture, the extent of the losses, or the division of these losses. High economy should not in any way interfere with other desirable features of gas appliances, and, as a rule, should contribute to, rather than detract from, these desirable features. Your committee assumes that every manager believes it to be to his company's interest to supply

the appliance which shows the highest efficiency consistent with other desirable features.

For the benefit of those who are not entirely familiar with this work, it may be well to state that until our 1903 convention no attempt had ever been made to establish standard methods for testing fuel gas appliances. Miscellaneous methods were in use which could only result in conclusions which failed to agree. The purpose of this work is to reduce the testing of fuel gas appliances to a condition where the same conclusions will be reached, no matter where, when, or under what conditions the investigation is made. It is also the purpose of this work to determine what portion of available heat in the gas is made to do useful work and what portion is lost. Further, it is the intention of this work that these losses shall be subdivided into their determining causes to enable the designer and the purchaser to curtail these losses, or, if possible, to eliminate them. The simplest means possible are recommended for this work, but your committee realizes that we are probably far from the best solution of these various problems. We are simply doing what we can, and asking for all the help which can be secured.

The field for gas fuel is practically unlimited, and the extent to which gas is used depends upon the aggressiveness of the management of the gas companies, the price of gas, and the efficiency and desirability of the appliances available. It is possible to materially increase the efficiency of all gas-burning appliances, and your committee believes that it is possible to open up more extensive fields for the use of additional gas in this way than could be secured by a reduction of 25% in the average selling prices of gas.

The apparatus for conducting these tests is inexpensive, but sometimes an expense must be incurred by each gas company which wishes to be thoroughly informed regarding the desirability of various gas appliances. This expense, however, is insignificant compared with the great benefit which can be secured by this knowledge. Compare, if you will, the expense of research work on this problem with a material reduction in the price you are charging for gas, and the insignificance of this expense will be made apparent.

With this explanation we will take up the detailed methods of doing this work, and it will be the aim of your committee to outline the methods it recommends in a way intended to be intelligible to the man who does not enjoy a technical education and in a way which will enable this work to be taken up by an ordinary workman without demanding more than minimum attention from the manager or engineer wishing to adopt these standard methods.

GAS SUPPLY.

The first essential is a gas of known calorific value, and as this value is apt to vary considerably, great care must be exercised or a serious error will be introduced right at the start.

It will be found that the calorific value of the street gas varies more at certain hours of the day than at others and the tests should be run at the time when the variation is found to be the least.

The second essential is that the supply be of constant pressure, and as two inches of water is generally accepted as the desired pressure at which gas should be supplied to consumers' premises, your committee recommends that the-supply of gas for testing purposes should be at this pressure. It is not, however, so important that the exact pressure recommended should be used as it is necessary that whatever pressure is used should be a constant and not a variable one.

Some misapprehension seems to exist regarding the efficiency of Bunsen burners supplied at a higher or lower pressure. Generally speaking, best results can be obtained at the higher pressures. This is due to the increased potential energy dependent on this higher pressure, which secures higher inductive effects and a more intimate mixture of gas and air, also the condition necessary for a high flame temperature. To secure these results, however, the gas must be checked only by the size and nature of the gas orifice, and where a fixed gas orifice is used and the pressure reduced by a stop-cock, no gain is secured by the higher pressure. No matter what the pressure may be behind the stop-cock, the velocity or rate

of flow of the gas through the orifice is fixed only by the excess of pressure immediately back of the orifice over that of the atmosphere, and if it is desired to reduce the rate of flow from say 20 to 5 feet, it is accomplished by reducing the pressure of gas on the orifice by partially closing the stop-cock. The velocity of the gas issuing from the orifice would be the same (neglecting the effect of eddy currents) whether this reduction of pressure is secured at the stop-cock immediately behind the orifice or at the works several miles away. Assuming that the orifice will be of constant size, the effect

FLUSH TANK.

Fig. 1.

of high pressure is therefore only manifested in the maximum velocity of the gas which will be passed by the orifice.

The space occupied by a given weight of gas will vary as the absolute pressure. Therefore the pressure to which the gas is subjected during measurement would introduce a slight error. At sea level the pressure of the gas would be 14.7 pounds per square inch, or 480 inches of water, so that a variation of one inch in pressure would only affect the measurement to about .2%, which is negligible. But the same variation in pressure would make a considerable difference in the

amount of gas passed by the orifice, that is, an appliance which is furnished with an orifice intended to pass say 35 feet per hour under two inches water pressure would, with a slight variation of this pressure, pass several feet more or less in that time, causing an increase or decrease in the temperature of the appliance under consideration, and as it invariably requires the lapse of some time to obtain constant conditions, which are necessary for a test, it is imperative that the rate of flow should remain constant. As it is seldom possible to get constant pressure unless tests are conducted at the works, it is absolutely essential that a good and reliable gas pressure regulator should be used, and for uniformity this regulator should be adjusted to deliver gas at a pressure of two inches of water.

WATER SUPPLY.

It is equally important that constancy of water supply, which is used for absorbing the heat of the gas, should be observed. It is to be desired that the water shall be supplied at constant pressure and constant temperature. The temperature of the water from city mains is usually approximately constant, provided it is not conducted through heated rooms, and even then it can generally be secured at a constant temperature by allowing it to flow for some time at a constant rate, and the variation in temperature will thus be reduced to a negligible quantity or the effect of this change in temperature can be established with reasonable accuracy.

If, however, the pressure is not constant, the requisite temperature conditions cannot be obtained, for as velocity of water flowing through an orifice being theoretically equal to the square root of 2GH, any variation in the head H, or pressure, will cause a proportionate change in the rate of flow. As a change of velocity or quantity will change the differential temperature through the absorbing chamber of the appliance under test, this change may cause an appreciable error, for there is always more or less capacity to hold heat, and unless the test is started and stopped with the same temperature conditions prevailing, error is occasioned to the extent of the thermal capacity of the apparatus. A simple and very