5 ' 37 . g M 2 REMOTE STORAGE IS ^'1 Note. — T his paper is sent to you that you may examine it in advance of the eeting, and prepare any discussion of it which you may wish to present. It is issued to the membership in confidence, and w ith the distinct understand- ing that it is not to be given to the press or to the public until after it has been presented at the meeting. As there will be no supply of extra copies there, and papers are liable to be read by abstract only, preserve this copy for your use, and BEING THIS COPY WITH YOU TO THE MEETING. ( Subject to Revision.) DCCLXVU NOTES ON RATING ELECTRIC POWER PLANTS UPON THE HEAT-UNIT STANDARD. (SECOND PAPER.) BY WM. S. ALDRICH, MORGANTOWN, W. VA. (Member of the Society.) These notes refer to a paper t on this subject read by the author before the Hartford meeting (May, 1897) of this Society. No comparative data were given in the original paper. It was known that the Committee on Data of the National Electric Light Association had a report in preparation for the Niagara Falls convention, June 8, 1897. This was the fourth and probably the last of such valuable reports. It was deemed ex- pedient, rather than refer to the previous reports, to await the publication of the 1897 report for the data needed in discussion of the author’s paper. The earlier reports of the above Committee on Data have been full of instructive information relating to many types of steam-power electric plants. In all of the cases finally reported, their rating has been based on watt-hours per pound of coal. Over three years ago Mr. F. M. Eites, at the Montreal meeting of our Society, discussed the data of the Washington conven- tion of the association (1894). His remarks X at that time were b * To be presented at tbe New York meeting (December, J897) of the American | Society of Mechanical Engineers, and forming part of Volume XIX. of the | Transactions. f Transactions of the American Society of Mechanical Engineers, vol. xviii., | No. 738, “On Rating Electric Power Plants upon the Heat-Unit Standard,” fry William S. Aldrich. X Transactions of the American Society of Mechanical Engineers, vol. xv., No. I 509, “ A New Method of Compound Steam Distribution,” by F. M. Rites. 2 RATING ELECTRIC POWER PLANTS I HEAT-UNIT STANDARD. so pertinent to tlie whole question of the economy of electric power plants that it will not be amiss to quote them here. “ It is impossible that competent engineering ability should be confined exclusively to the manufacturing industries. “ It cannot be assumed that the average intelligence of the designers and operators of electric light stations is inferior to that displayed in establishments of different character, and yet the enormous discrepancy between the actual results and those which should be realized surely deserves some attempt at explanation. “It is but proper to note that the Committee has chosen the record of a very high duty as a basis of comparison, and that the nature of the exacting service of electric light and street railway plants precludes the possibility of a close approximation to the highest economy under more favorable conditions ; but these figures are entirely unexpected and incidentally somewhat ridic- ulous, considering the energy with which the last per cent, of efficiency of the electric apparatus is insisted on by its users. “ Possibly some reason for such a remarkable state of things may be found in the miscellaneous engineering errors which usually follow an ignorantly wasteful policy, but these are as frequently met in other power plants. Perhaps, also, stations improperly proportioned and generally unfitted for economic competition may be found in the list, but these are far from sufficient to account for such universal failure to realize even a moderate degree of efficiency. “ There seems'to be but one general explanation applicable to electric light or railway stations which can account with any degree of probability for such extravagant fuel consumption, and that is the excessive wastefulness of the steam-engine under varying conditions of load.” Tt is proposed in the present paper (I.) to discuss briefly the progress shown during the past four years in the economic per- formance of steam-power electric plants as summarized by the Committee on Data of the National Electric Light Association ; (II.) to show that the very low economy of such electric-light and railway stations is not entirely due to uneconomical engines and variable loads ; (III.) to present a few notes upon a further consideration of the heat-unit rating for such plants. This treatment will have specially in view the necessity for some standard rating by means of which the design, installation, test- RATING ELECTRIC POWER PLANTS I HEAT-UNIT STANDARD. 3 ing, and management, as well as specifications and contracts for these plants, may be reduced to a satisfactory basis for advanc- ing this industry along engineering lines. I. — PERFORMANCE OF ELECTRIC POWER PLANTS ON A COAL BASIS. From the several reports of the Committee on Data of the National Electric Light Association, Tables I. and II. have been compiled. An inspection of these will show what little progress has been made during the last four years in such installations. In fact, the expectations of the (1894) committee seem not to have been realized in that they looked for much better values in the reports of subsequent years. It is a matter with which the mechanical engineer is most directly concerned. His work in the design and installation of even the most recent central station is open to criticism that cannot be applied to the elec- trical features of the same. Electrical engineers themselves acknowledge that the efficiency of the modern dynamo has prac- tically reached the limit set by structural and economic consid- erations. Mr. Kites’ remarks, quoted above, are singularly ap- plicable to the conditions existing at the present day. In the light of what he has said, the following comparative data should be carefully studied by the mechanical engineer. TABLE I. Showing Results of Four Years’ Progress in the Economic Per- formance of Steam-power Electric Plants. Year. Convention of the Nat. Electric Light Assoc., at— No. of Stations Watt-hours per Pound of Coau. Reported. Maximum. Minimum. Average. 1894. . Washington 65 208 25 91.7 1895. .1 Cleveland 24 262 36 128 1896. . New York 81 237 33 108 1897.. Buffalo 14 269.5 98.7 156 4 EATING ELEGTEIC TOWEE PLANTS I HEAT-UNIT STANDAED. TABLE II. Equipment of the Stations Given in Table I., Showing the Maximum and Minimum Economy. Year. Economy. Boilers. Engines. Dynamos. Daily Out- put, Watt- hours. Fuel. r 1894 \ l Maximum 1 Arc, Power, and Incandescent. Arc and Incan- 7,971,600 80,670 Coal, Hard (208) Minimum ! | Screenings. Coal. (25) ! descent. f 1895 \ l Maximum (262) Minimum (36) Horizontal Water Tube. Horizontal Tubu- lar. Trip. Exp. Con- D i rect Con- densing. 1 nected. High S p e e d Belted Direct. Non-cond’g. I 22,967,952 2,790,565 i Soft Coal, i Hd. Scn en’gs. Bitumin. Pea. r 1896 -j l Maximum (23 7) Minimum (33) Horizontal Tubu- lar. Horizontal Tubu- High Speed Belted Direct. Comp. Cond’g High S peed Belted Direct. Condensing. 1 3,270,392 203,555 Bitumin. Lump. Anthracite Buckwheat. f Maximum Water Tube. Vertical Couid. Dir. Connected. 1837 1 (2:19.5) Minimum Horiz. Tubular. . . Cond’g 4-Valve Comp. Cond’g. Comp. Cond’g. Incandescent. Belted to Dyn., Belted to C.-Sh. Screenings. l (98. 7 ) and Water Tube. A comparison of tlie above with some of the best results ob- tained in modern mill engines has been noted by the commit- tee, as follows : In the 1894 report compared to the performance of the engine of the Chelsea Jute Mills, Brooklyn, N. Y., showing a coal con- sumption of 1.482 pounds per indicated horse-power per hour ? with the load varying from 495.21 to 764.96 horse-power. If such a performance were possible in the central station, it should result in over 409 watt-hours per pound of coal, on the basis of the committee’s assumption of 90 per cent, for the mechanical efficiency of the engine and for the same efficiency in the dynamo. In the 1896 report the committee called attention to the then world’s record for steam economy as shown by the Chest- nut Hill pumping station engine at Boston — a steam con- sumption of 11.22 pounds per horse-power per hour, or an effec- tive pump liorse-power per hour on 1.34 pounds of coal. If the efficiency of the direct-connected electric generators should compare favorably with that of the pumps of this engine, with no allowance for variation in load, anthracite coal used in the plant with the same economy of installation should produce 557 watt-hours per pound of coal. In the 1897 report of the committee, Mr. F. R. Low, member of our Society, very fully discussed the several sources of loss EATIN’G ELECTRIC POWER PLANTS : HEAT-UNIT STANDARD. 5 in tlie electric power plant, stating the discrepancies would be made up mainly from the following items : (1) Decreased boiler efficiency ; (2) lesser normal efficiency of engine ; (3) impaired conditions of engines ; (4) unfavorable engine load ; (5) leak- age ; (6) condensation; (7) auxiliaries; (8) heating. It is not our purpose to discuss this admirable report ; but there is no reason why every feature of installation of an electric power plant should not be as fully considered in design and construc- tion as in the case of the modern high-duty pumping station. In fact, it has been repeatedly pointed out that the chief dif- ferences are those due to the running conditions of electric plants, and not entirely due to sudden and wide variations of load. Messrs. A. G. Pierce and R S. Hale report * of the per- formance of the Boston stations of the Edison companies : “ In our test we have finally found the variation due to causes which we first thought negligible, to be more than the variation due to the change of load.” In this connection it is important to notice the results of Mr. H. A. Foster’s analysis of the tests of twenty-two different power plants, f These included manufacturing establishments 3 electric-light stations, pumping engines, etc. Plants above 200 horse-power show a remarkable uniformity in fixed charges ; namely, interest on first cost, depreciation, taxes, and insurance. The operating expenses gradually decrease in plants from 200 to 1,000 horse-power, above which capacity the operating ex- penses seem to remain remarkably uniform and quite irrespec- tive of load variations. The large electric stations supplying many smaller industries are scarcely affected by the instan- taneous load changes in one or more of these particulars. The conclusions to be drawn from all of the preceding clearly indie ate that the economy of the modern high-duty pumping-engine plant is due to a refinement of design and economic arrangement of the installation that has not yet been reached in the electric power plant. Anything which tends to advance the latter industry along the lines which have been so clearly marked out in the development of the former, merit the * Quoted by Mr. F. R. Low, rcpo*t of Committee on Data, National Electric Light Association, Buffalo meeting, June 8, 1897. •f Transactions of the American Institute of Electrical Engineers, vol. xiv., “ V ariations in the Cost of Steam Power,” by Mr. H. A. Foster. Paper pre- sented at the annual convention, July 28, 1897. 6 EATING ELECTRIC POWER PLANTS : HEAT-UNIT STANDARD. attention of those having such work in hand. It is believed that the standard heat-unit specifications aud the subsequent contract trials of pumping plants upon this basis have combined to develop this industry to an unprecedented degree. It is not therefore too much to expect that similar standard heat-unit specifications and contract trials of electric power plants will advance this industry also along the same engineer- ing lines. At least the efficiencies, economies, guarantees, and contracts now being realized in pumping stations should be much more nearly approached by the modern electric power plant. II. — PLANT ECONOMY AS RELATED TO ENGINE ECONOMY AND VARI- ABLE LOADS. This seems to be the chief feature of Mr. Rites’ explanation of the very low economy of the electric power plants reported upon up to the time of his paper previously referred to. We certainly do not wish to be misunderstood as taking issue with this explanation ; but it is apparent, from a careful study of the last four years’ record of the Committee on Data, that some very uneconomical engines have produced remarkable results in point of economy of operation and efficiency of installation. The best of engines may be poorly operated on the one hand, and the whole plant badly arranged on the other hand. We enter as strong a plea as any one for the most economical engine in electric power plants ; but we wish to add a further requirement, that there should be economic installation and efficient operation to produce the best all-round results in the course of a day, a month, or a year. We think these two feat- ures may possibly have produced the lowest cost of steam power yet recorded ; namely, $11.55 per year of 3,070 working hours, reported by Dr. R. H. Thurston,* member of our Society. The plant is at the Warren Steam Cotton Mill, Providence, R. I. The 1,950 horse-power “ Allis ” cross-compound condensing engine (cylinders : 32 and 68 inches by 5 feet stroke, 74 revolu- tions per minute), with Heine water-tube boilers, at 155 pounds steam pressure, show an economic performance of 1.35 pounds coal per horse-power per hour. The question will continue to be asked : Why are not such Reported by Dr. R. H. Thurston, in Science , October 1, 1897. FATING ELECTRIC POWER PLANTS : HEAT-UNIT STANDARD. 7 results obtainable in electric power plants with similar units ? Low cost of steam power or of electric power is not due entirely to multiple-expansion engines of the greatest individual econ- omy ; for in the last noted instance, as reported by Dr. Thurs- ton, the cross-compound condensing engine replaced a quadru- ple-expansion engine. Concerning the effect of variation of load upon modern elec- tric power plant engines, it is further interesting to note that Messrs. A. G-. Pierce and It. S. Hale state of the Boston Edison stations in the report previously noted : “As a matter of fact, the steam per indicated liorse-power in our two 200 units holds within 12 per cent, over a range from | up to full out-put.” Along the same line it is to be noted that throughout quite a wide range of load variation the compound engines reported by Mr. A. K. Mansfield,* at the recent Hartford meeting, show a remarkably uniform rate of steam consumption. The ques- tion naturally arises whether such uniformity under wide varia- tions of load is not more of a characteristic of compound engines than formerly considered by those who lay all the blame upon the steam engine for the poor showing in the economy of elec- tric power stations. With dynamos which electrical engineers now design and build, maintaining an efficiency over 93 per cent, from about J load to 20 per cent, over load, directly coupled to cross-com- pound condensing engines (let us say) which mechanical engi- neers are designing and building, having a characteristic low range of steam consumption between similar limits of light load and over load, the question will naturally arise : Why cannot the two units be more economically put together, installed, and operated ? III.— NOTES ON HEAT-UNIT RATING. In the discussion which followed the presentation of the author’s paper on this subject before the Hartford meeting, as well as in conference with members and other electrical and mechanical engineers since, it will appear that the following points have been brought out : That the heat-unit, as a basis for such ratings, is both rational * Transactions of tlie American Society of Mechanical Engineers, vol. xviii., No. 727, “The Best Load for the Compound Steam Engine/’ by Mr. A. K. Mansfield. 8 RATING ELECTRIC TOWER PLANTS : HEAT-UNIT STANDARD. and scientific. It is, however, not in consequence the most satisfactory standard for use by builders, contractors, and prac- tical engineers dealing with this class of motive-power machin- ery ; namely, steam engines and dynamos. That great differences of opinion exist as to the proper defini- tion of the heat-unit required for such a standard. There are at least four different heat-units commonly employed. That the present extensive and satisfactory use of the lieat- unit for steam-pumping installations is as it should be and is all right in that place ; but this is no argument for its introduc- tion and use in a similar manner in the rating of steam electric plants. That the present way of stating the performance of electric power plants, however unsatisfactory, is easily understood by all parties interested. Chief of these, of course, is the capitalist ; he can readily comprehend rating based on the coal bill. That the load factor, after all, has not so much to do with the fuel economy of the plant, as such, however much the varying loads may individually affect any of the units of the installation, such as the steam-engine. Therefore, in the large city and suburban steam-power and electric plants now being installed, there is not the necessity for such strict adherence to eco- nomic load factors as in the case of plants with smaller units. That the watt expresses the activity or rate of the electrical output, in joules per second. In this respect it is analogous to the horse-power rating of mechanical output. Hence, the standard rating should be in kilowatt-hours per 1,000,000 B.T.U. supplied to the steam used in the whole plant. That if the heat-unit basis is considered as the proper stand- ard for the steam electric plant, the whole heat supplied to the plant should be as carefully determined, and in the same man- ner, as now in vogue for similar standard ratings and contract trials for steam -pumping plants. That the boiler should be in evidence in all cases in which plant performance is mentioned. In the electric plant it is economy of installation that is desired quite as much as in the case of pumping plants. Why should the boiler perfor- mance be urged into consideration in the former case and not in the latter ? If it is a good thing to introduce it in either case it would seem proper to do so in both cases. That the common rating of performance of pumping plants in RATING ELECTRIC POWER PLANTS : HEAT-GNIT STANDARD. 9 foot-pounds per 1,000 pounds steam would be amply sufficient for all purposes of rating electric plants for which the heat-unit basis is advocated. This seems particularly plausible on ac- count of the small variation in the total heat of one pound of steam for quite a wide range of pressures now used in modern electric plants. Taking the standard temperature of feed water at 212 degrees Fahr. exactly 1,000 B.T.U. are required to raise the temperature and evaporate one pound of feed water into steam at 77.3 pounds per gauge (92 pounds abs.). Taking this as suitable for a simple non-condensing engine, we may compare it with that of 150.3 pounds gauge (165 pounds abs.), in which 1 013.5 B.T.U. are required to raise the temperature from feed water (212 degrees Fahr.) and evaporate it into steam at the given pressure.. In this case, therefore, if we adopt 1,000 pounds steam instead of the 1,000,000 B.T.U., we make an error of only 1.3 per cent. It is claimed that this is within the usual allowable errors of observation and measurements in power- plant tests, and that there is not enough difference to warrant the trouble required to obtain the performance reduced to a B.T.U. standard. In this connection it is interesting to note the progress shown by the committee reports on data made to the National Electric Light Association. In eleven out of the fourteen cases noted in the report presented at the Buffalo meeting of that associa- tion, the water per kilowatt-hour at best efficiency of the en- gine is noted. As t the average temperature of the feed water in the best stations reported is from 208 to 212 degrees Fahr., and the best results are shown by the compound condensing en- gines, we may conclude that the comparative water ratings are within about 1 per cent, of what such comparative ratings would be if based on the B.T.U. standard. It is a question whether mechanical engineers will remain satisfied with results even within this close degree of approx- imation. The fact has been repeatedly pointed out by elec- trical engineers that their system of units is altogether unique, is thoroughly scientific (being based upon the C. G. S. system and is the only system of engineering units universally adopted. Mistakes are said to be occasionally avoided in the sister pro- fession of mechanical engineering by the insistence on the ac- curate use of terms and of units in electrical engineering. Dealers in electrical stocks and capitalists exploiting electrical 10 EATING ELECTRIC POWER PLANTS : HEAT-UNIT STANDARD. enterprises generally have an appreciative insight into the meaning of volts, amperes, and kilowatts. Why should the time-honored heat-unit be so difficult of comprehension by the same class of interested citizens ? A quarter of a century ago Maxwell wrote : “ The conse- quences of this demand for electrical knowledge and of these experimental opportunities for acquiring it have been already very great both in stimulating the energies of the advanced electrician and in diffusing among practical men a degree of accurate knowledge which is likely to conduce to the general scientific progress of the whole engineering profession.”