........... I h .5 IL" A STUDY OF THE ECONOMIC VALUE OF SIPHON SETTINGS FOR Low>H£AD HYDRO-ELECTRIC. POWER PLANTS ,- . .J'Hgsus Fon'DEGREE’Or c. I. - .C. 1. MC [JAN - - .E x 19.32 . ‘ l.\. '. QQ‘.H\‘.‘5‘. - '.'.°0;‘ ‘ ..... ' ' p 031033914: 0 c o o n c o o a I r “a” r IJJJJLDJJJJ 0111444144444‘114 aM’ THESIS ‘ . 7 "WM“: bklr end/WA)“ Wed-sax —~ a»: tuiv . ; .~ I .‘ A STUDY OF THE ECONOMIC VALUE OF SIPHON SETTINGS FOR LOW - HEAD HYDRO - ELECTRIC POWER PLANTS A Thesis Submitted to the Faculty of MICHIGAN STATE COLLEGE AT East Lansing, Michigan. BY 00011 John glean B. S. and M. 8. in Eng. Candidate for the Degree or CIVIL ENGINEER JUNE 1938 THESIB List or Illustrations Plate No. Description Page 1. Interior and exterior view of plant 14 at Sterling, Illinois with bevel mortis gear drive. 2. Comparison of old and new plants at 15 Cheboygan, Michigan. 3. Cross-section of the plant or the 18 Concord Electric Company at Sewalls Falls, New HampShire. Vertical Type Triplex turbine built in 1905. 4. Cross-section of the plant or the 19 Edison Sault Electric Company at Sault Ste. Marie, Michigan. First direct connected single run- ner plant in the'United States. 5. Cross-section of the plant or the 22 Rock River Light and Power Compamy at Sterling, Illinois. 6. Plant of the Rock River Light am 25 Power Company, Sterling, Illinois, Construction view showing draft tube forms. 7. Pages from.catalogue or Stout Mills 27 and Temple Showing early siphon settings. 8. Pages from catalogue of Stout Mills 28 and Temple showing early siphon settings. 9. Seotion.or Plainlell.(flucht) plant 53 using new type of high speed turbine 3e 10. Cross-section of plant at Oregon, 35 Illinois with siphon setting 1311113 in 1931s 11. Cross-section of Green Island, New 56 York plant of Henry Ford and Son. List of Illustrations ( 2 ) Plate No. Description Page 12. Curve s owing Kaximum values of 42 Specific speed attained at various periods during the past 70 years. 15. PrOpeller Type turbine installed at 56 Dixon, Illinois. 14. Cross-section of the plant at Dixan, 51 Illinois. 15. View of the turbine setting for the 55 plant at Dixon, Illinois. 16. View of the turbine setting for the 57 plant at Sterling, Illinois (New unit ) e 17. Cross-section of the plant at 58 Sterling, Illinois (New unit). 18. View inside the Sterling plant Show- 60 ing old and new generators. 19. Efficiency curves for units of 64. various specific speeds. 20. Cross-section of the plant at 72 Rockton, Illinois. 21. Views of the plant at Otsego, 85 Michigan. 22. Cross-section of the plant at 92 Appleton, Wisconsin. 25. Cross-section of the plant under 96 construction at Vergon, Sweden. 24. Interior View of I—Iydro Plant at 106 ' Dixon, Illinois. 25. Exterior View of Hydro Plant at 107 Dixon, Illinois. 26. map oi'the United States showing 125 location of principal hydro- electric power plants using siphon settings. _ Page No. l mamucrrcw ' The developments in the design of hydro-elec- tric power plants and the advances made in the design of the necessary machinery have been very rapid during the last fifteen years. Many plants operating today while not obsolete in the sense that they cannot com- pete in production with newer plants are nevertheless far from modern when compared with the latest develop- ments. New ideas are deve10ped with each new plant de- signed and while it is not the duty of the engineer to keep plants in style with the latest fashion, it is his duty to keep pace with the modern developmts so that the plants which he designs may be as efficient and productive as it is possible for them to be. The large investment necessary for a hydro-electric plant demands that the maximum production possible be obtained. The steady downward trend of costs of current in large steam generating stations makes it necessary that hydro- electric costs also be reduced to the absolute minimum. The time was not so long ago that hydro power represent- ed the only really cheap power available. Those cities or industries blessed with this power in the necessary quantities were able to reduce costs of producing manu- factured goods and so obtained an advantage over their neighbors. This present age has brought us low cost steam power but more valuable still, electric power from central genera ting stations which means low cost power Page No. 2 to the small as well as the large consumer. Under these new conditions the designer of hydro-electric power plants must take advantage of every modern means to in- crease efficiency and reduce the investment in the plants which he designs. There are no two hydro plants that are exactly alike and each development presents its own special problems. The high-head hydro-electric plant uses a small quantity of water under high presmre. This pre- sents problems to the designer which must be met all solved. The low-head hydro-electric plant must utilise large quantities of water and this condition presents its problems equally subject to serious study in order that the loss of head may be reduced to a minimum and that the acceleration and deceleration of so large a mass of water may be accomplished during load changes without serious injury to the plant or equipment. The terms low-head and high-head are relative. Plants with heads as high as eighty (80) to one hundred (100) feet may be and are considered as low-head develop- ments. It is easy to see how the operator of a plant with a head of eight hundred (800) feet might consider a plant with a head of eighty (80) feet as a low-head plant. However, to the Operator of a plant with a head of eight (8) feet the plant with the head of eighty (80) feet really appears to be a high-head plant. In order that this subject may be properly limited to those P389 N00 3 plants where the siphon setting is applicable, the term "Low-Head hydro-electric power plant” as used in this thesis shall be taken to mean.on1y those plants operat- ing withta head of twenty (20) feet or less. While this type of plant is not the kind that challenges the imagination.of the public mind it nevertheless is a very prevalent type of installation in this country and Canada. These plants may be found along mny of our streams and they are usually designed to take what water is available as it comes as no storage is available as a rule. H. E. M. Kensit writing in the Electrical Times, London, on.the subject of ”Water Power in Great Britain”, states that in Canada there are about thirty (50) central station plants Operating with a head of ten (10) feet or less. This does not include that multitude of plants developed by private industries to provide power in their own factories. So Canada with its abundant water- power resources also sees value in the low-head hydro- electric plant. In.this country the number of this type of plants is even larger and the extent to which they are developed is no doubt due to the early use of hydro power by the people who settled.and developed the great central territory of our country. The higher heads have been more rapidly developed since the generation or electricity by water power has made possible their utilization. In.a report of’the water power resources of Illinois published in 1915 over 50 power sites were Page No. 4 listed in the state. most of these were operating small mills or factories at that time and only three (5) of the plants listed had heads over fifteen (15) feet. many of these plants have since been converted to modems hydro-electric plants. The tens "Siphon setting" as used in the title to this thesis refers to that design of the turbine pit or wheel pit which permits the water surrounding the turbine to be raised above the level of the normal head water by a siphon or vacuum action to provide the re- quired submergence of the turbine. It differs from.the cpen flume setting which is the type of design which precceded it, mainly by reason of the fact that the walls and ceiling of the pit are air tight and at the intake or entrance, the upstream wall extends below the surface of the head water to form.a lip. This lip pre- vents air being drawn into the turbine pit while the unit is operating. A more detailed description will be given later when describing actual installations in operation. While this simple principle is old its general application to powerhouse design is rather re- cent. No study can be made of any one feature of power house design without also considering the part which the improvements in machinery and equipment have played in its deveIOpment. So long as power was re- quired in small units and.used direct in factories _ Page No. 5 there was no demand for large capacity high speed tur- bines. When necessity demanded it the turbine designer produced the high specific speed turbine which is now built in such large capacities that for low-head hydro- electric power plants the siphon setting becomes an ab- solute economic necessity. A brief review of the his- tory of the development of water power is necessary for this study of the economic value of siphon settings for low-head hydro-electric power plants. Beside the Nile, the Euphrates and the Yellow Rivers thousands of years ago primitive Hydraulic Ingin- eers planned and constructed their simple forms of current wheels. These consisted simply of a wheel with paddles attached to the rim and so set that the paddles, dipping below the water surface were moved by the current and die wheel kept in motion. They utilized this energy which they took from the stream to raise water for irrigation and so transformed the otherwise barren land adjacent to the streams into gardens of plenty. In some remote sections of China there can still be seen in operation irrigation works which are almost as primitive but which are nevertheless working and supply their owners with all the power they need. is civilisation developed man found ways of developing power from the more swiftly moving water at rapids and falls, at first with breast wheels and later with the overshot water wheels. The energy thus obtained was Page No. 6 applied to the grinding of grain as well as pumping and as long as man was able to provide the power necessary for his immediate needs and relieve him.of the hard manual labor incident to grinding and pumping for.irri- gation he was satisfied and little progress was made for thousands of years. By the time of the American Revolu- tion these water wheels were developed to a state that made then.of considerable value wherever a dan.cou1d be built or where nature had provided the necessary fall. They were generally used for the grinding of iced and flour and such other manufacturing as had at that time been developed to a state where power could be applied to the process. The mills were built on the banks of the streams and the power was taken.directly from.the waterwheel shaft either by belts or crude gears. The development of these water power sites in New England was the beginning of the industrial development of this part of our country. As the central part of our country was settled, power sites along the streals were develop- ed for grinding feed and flour that were later destined to be redeveloped as hydra-electric generating plants to provide more universal use of this waterpower. The plants at these dens are today in still another stage of redevelopment as modern hydro-electric plants which are interconnected in large systems supplying power to many for removed from the banks of the stream. The old type breast and overshot water wheels Page No. 7 served for many centuries as the principle means of power development but in.1843 the hydraulic turbine was intro- duced into this country by Ellwood.Mbrris ofiPennsylvanins This turbine was later developed.end.brought to popular attention largely through the inventions cf'Uriah.A. Boyden. The great advantages clailed for the water tur- bine over the old style water wheel were (1) it occupied a smaller space, (2) it operated at a higher speed, (5) it would work submerged, (4) it could he built in much greater capacity, (5) it could be used in cold climates as it was more readily protected from.ice. All of the above features had become increasingly important with the more widespread application of power to manufacturb ing processes. Here again the turbine was a product born of man‘s necessity. About 1849 James B. Francis, Hydraulic Engineer, connected with the hydraulic plants at Lowell, Massa- chnsetts, designed endinward flow turbine which had many improvements and.a higher efficiency than the turbines then in use. This type of water turbine, mneh.improved today and new manufactured by many water wheel manu- facturers still bears the name of the Francis type runmr. These water turbines rapidly replaced the older less efficient overshot and breast wheels and they are to be found in practically all sections of the country today. On the smaller stream the power was usually developed by a single individual or company but on the r) Page in. 8 larger streams more power was available than could profitably be used in one factory and in such cases comp panics were formed to make the develOpment and the power was allotted to the several owners in proportion to the acount of money they had put into the company for the building of the dam.and canals. This are of development lasted from.about 1845 until near the end of the last century. This period was marked by a wide spread appli- cation of waterpower to industry and many of our present large industrial institutions had their beginnings in this period when the factory with available water power had a considerable advantage over its competitor. These industries as a rule used small turbines as turbines are rated today and the development of the low heads which were available on most streams presented no very serious problem. When more power was required than could be developed with one turbine as many turbines as were requir— ed were installed and their combined output utilized by gearing them to a common shaft. There were many examples of this type of development along the Rock River in northern Illinois and the heads available at these dams were usually very low. In 1879 Edison invented the electric lamp and the following year installed the first generating station for producing electric power with steam. In 1882 there was installed at Appleton, Wisconsin.the first hydro- electric power'plant. This plant contained one of the Page No. 9 early Edison bi-polar generators of 250 lamp capacity. The single turbine was geared to a horizontal shaft and a belt from.this shaft turned the generator. This plant was crude indeed when compared with the modern plants of today, but the idea developed in this o:iginal plant was used in many small electric generating plants for years until the increased use of electricity and development of the steam turbine provided the urge for further development in the hydro-electric plant. The result is the large capacity high-speed hydraulic turbine which today replaces several small turbines and a: combines in a single compact unit what before was an inefficient collection of gears, bearings, shafting and in many cases long expensive belts. The modern type of hydro-electric plant with its concrete foundations and substantial superstructure of brick or concrete is so different from.the early type of hydro-electric plant of 40 or 50 years ago as hardly to be recognized as a decendent. In place of the tur- bine operating in an open flume usually floored over with planks soaked with grease from frequent applica- tions of oil and gear dope the modern turbine is entire- ly enclosed in an.air tight chamber and operates without vibration or'noise and where oil or grease is required it is applied with such efficiency that there is no sur— plus remaining to soak the floor or detract from the cleanliness of the place. In place of the numerous Page No. 10 gears, shafting and bearing boxes requiring constant attention and maintenance a single rugged shaft extends vertically through a substantial water lubricated bear- ing. In place of the large pulley and long belt or end- less rope drive a few strong bolts connect the turbine shaft and the generator shaft together and the two pieces of machinery operate as a single unit. The low-head hydro plants which a few years ago were turning factory or mill wheels and which later were converted into electric generating stations by belting generators to the turbine shafts are now being economically converted into modern hydro-shactric plants with the generators directly connected to the turbines. The machinery developed for this purpose requires the application of the principle of the siphon setting for low heads but these plants have already proved their worth in many cases by several years of successful operation. Considerable progress has already been made in this redevelopment of the old power sites but there are many more waiting to be modernized. In addition to those sites which were pre- viously developed as sources of power for factories and which are now being changed to modern.hydro-electric plants there are many sites which are not utilized. These sites are equally valuable and many of them.mey now be exonomically developed. Some of these sites are being utilized but as these plants must of necessity be Page No. 11 run-of-river plants they can.be most economically utilized as'a part of interconnected systems where steam.reserves are already available to supply the demand in seasons of deficient flow. Steam.pover and hydro power have now Joined hands in interconnected systems to supply power for home and industry in a way which was impossible a few years ago. modern Low-head Hydro-electric Power Plants with Siphon Settings. The energy of falling water is converted into electricity, a more useful and mere readily controlled form.of energy, by means of hydro-electric power plants. Following the installation of the first hydro-electric- plant at Appleton, Wisconsin in 1882, many of the power sites which were furnishing power to grist mills and small industries were changed to electric generating plants by belting the generators to the horizontal shafts of the hydraulic plant. Prior to 1890 synchronous speeds were not a problem.in water turbine design as nest in- stalhrtions were provided with either belt or gear drive between the prime mover and the generator. Turbines were built with either vertical or horizontal shafts depending on the application but for the very low-head plants vertical turbines with bevel mortis gears and horizontal shafts were generally used. These early plants consisted of an Open box or flume in which.the turbine was set. The upstream.side was open and was protected with rack Page No. 12 bars to keep out drift wood and ice. Beneath the turbine was a similar area opening down stream, The draft tubes used with these early turbines consisted of short conical sections which discharged vertically into the draft pit. These early draft tubes were not very efficient as they were short and the discharge velocity was high. Another source of loss in the extremely low-head plants was the air which was drawn into the turbine by the vertexes or small whirlpools. In order to prevent the formation of these vortexes and reduce the loss in power from this source it was considered necessary to install the turbine so that the t0p of the gates were from one to one and a quarter times the turbine diameter below the surface of the later. This rule of submergence, given in Professor Head‘s text on Water Power, was a limiting factor in the size of the turbines for low-head plants and in order to reduce the cost of the generator and improve its efficiency, several turbines geared to a horizontal shaft through bevel mortis gears were usually employed to drive the generator. A typical installation of this kind built in 1908 is still in cperation at Oregon, Illinois on the Rock River. In this plant there are three Leffel Samson turbines and two low American tur- bines driving a horizontal shaft through bevel mortis gears. The turbines operate at 66 8/5 r.p.m. and the shaft revolves at 200 r.p.m. Another typical plant of thiskind designed Page No. 13 by Professor Mead and built in 1904 is still in opera- tion at Sterling, Illinois. Plate No. 1 shows an inter- ior and an exterior view of this plant. The number of turbines employed to drive one generator varied and de- pended usually on.the water available. The type of drive also varied, some generators being driven by gears, some were belted and others were direct connected to the horizontal shaft. Improvements in turbine design and methods of manufacture made possible some developments along on- tirely different lines and in the period from.l900 to 1910 a great many multiple runner units were built where the head was sufficiently high. These usually conshsted of either a double, triple, or quadruple runner unit on one shaft and built so that the thrust on the runners was balanced. The shafts were usually set horizontal and this type of plant, illustrated on plate No. a, utilized larger capacity generators Operating at higher speeds which not only reduced the cost of the direct-connected generator but also increased its efficiency; With this type of installation the length of the plant was reduced as two or four units were placed in a flume not much wider than would have been required by a single runner of the same diameter but the distance from the intake to the downstream side of the power house was greatly in- creased. This type of'runner could be used only with such heads as would provide submergence of the turbines Page No. 14 Plate No. 1 Interior view of plant with bevel mortis gear drive .‘ at Sterling, Illinois. Exterior view of the plant at Sterling Illinois. 15 .nmmfinofiz .nmwhonono pa madman son one 0H0 mo nomanwmaoo age No. .C . is: a: 63.5.: as? 92.232 59:35: 3 cease—:5 59: 3:31;. :=-.5>o up; dots—SE: 3c 9: 5:3 ceases—cc ma mmEZesfiEOo M539? £52550 nelsom a. Eu: 2.5.5.”.— .Eu>.c._e:O 9: no “:5: arsed—canvas Bo: 535.2: :23?» e m: Eu: .2: :C 3.12 éuuifieso .hseQEoO nelson an EMS arsed—m cauzcsczo on. no :23 .._.:.E_e-c.::.; Be first: .523; a :32? E :2 m5 :0 Jarmz 2:8 HART; HZ... 7:35... ...=.... s ... 3...... 7.1.2.... as: 39.9.1.8 75...)... .2. :3 5.75:... 551.5... c... ... 2.5.... 5...: .5333... :3. the: .123... :35... 1.2:... wit... .2. 7.....2: .5: 1.3:... .52... 3.1.9.: o... a: .=.. 2.... canon 98...... PEERSOO 6.330932. .2... .2: s. .22.... :2... a... :e... 9.11.2.1»; 11.... 1.. ...... ..........u=.c..¢ 3.72....— ..:s ....:......7.:.l. 9...... a 1.3.5: 2...... 5...... 3...... .c :9. 75...»... E. z...‘ 1.32.... .75.... a... 3...... ... .5: 1...: .z .52... e... 3...... 3 1. ..........u:..L..e 5.3.1... .. . 4......— ... .:. L333 .2 225:... .2... a. 1.9.33.5.» a... A... L... .7321...» .33.... ... .::::.u ..... ... 1.3.2.3 .355 a... 3...... c. ML... 1.. .. 5:5 .6 .55.. 9...: .s 71.. 0.. as... ....x. b. an: .e... .1. .u....=..w....su 325.13.... 3.... E... :5... 9:27 3... :. 7:57.... 0.... 1.3....» .5». :9...” 4...... .3... a: a..." 9...... as... .155... ...... 17...: c. 3...... £39.»... ...:..::.::...E. o... :3. o... a. 1.9.3....» a... .1... 5.3. t... .... .53.... 72.1.5.7. z... .1. .39....» 2C... 9:: :3... «2.2.2.1.??— 0.5:. 3...... :2. 1.75.... .12.... :3. 3:95.. .1. ..../. z... z. :32... 5. .3 .94.... 379...». ..... c: :52... .3 7.25:... 5.5.... .: :5. s ..:...:.=:....: at 3.9... 3:. .31.... 5...... a... .o .398 a... o. :9... 3... .. :5: :2. a... ....s.: 3.2:... 5...... .3... .s... S. .2... .........._.o.=.s s .3 hit... t... .2. :5. .53... z... 32...... o. .1... a. u. ....:.m .99....» 3...... a... so he... .8 ............._.x. s .9. .5.......a¢. a. .33.... c... 2...; 3:29.... .o 3.15... 2...... a a. £25. m... :. 15:5... 5 71.: 0.. >..:: so... 2...... .5: 5.5.... 93...... .x. c. 2...... ¢.. 1...... c... 3...... a....: a... c. 7...: R. at... .52. 2.5.... mitts... e... a. 7.1.3.: 1...... .3... .1... .01.... o... .3 ac. o... c. 3.... .c: :..r 3...... o... ..: Damage: 5...»...— s. 9.1.7.7.... a... ..._B in... a... .: ..:. 9... 5:30:95 on... e... 55:8... .....7. a. 5...... a... he... a... :2... as 75...: 3...... a... .c 8:... a... smug... a... 2.3 .7...“ e... .5: .35.... s h... .1237... .x. z...’ 3.... .2552... o: J. 30.0.. .00. 93. ES... .22.... a... .3 3:7... 0... v.2. .5: 3.93-...8: o... 5.... .93.. a. .25.... e... .3 ac. a... .a... 9. 73.3... 9:. £92.... a... .: .m.....&5:§.i .53.. 9. E... omen 3.1.1.... a... so .5: a. 2:8... ......=.&.:Ea as .95... has... a. 5.... o... .238 E... dates... 0... ..w=c...3~ d: s... 7:52:55...— bifi: .5... so. 3...... £00....» 13.53:... .c 832...... a... 3...... .839... a n. .939... 8...... 3.7.9:... o... .8 Se 2.... dSHETBoA486ns.JQSEB.isnefihfim swam 8.8.30 23 no «so 23 mo .8393 I I‘ll! 1.6.5:...5. a. mains Fir—0.. c 1‘ )9 MIMFQMH... . \I’... NEW AMERICAN TURBINE MOUNTED ABOVE HEAD-WATER LINEI’ag-e No, 27 Put-ant uppflod for. Plate'No. Design N o. 19. 7 Pages from cataIOgue of Stout Mills and Temple showing early siphon settings. (J .32.... .3 :2. r... as... o... 5.22.... 1.... .3... :21 0.2 35-13;... S... u... o... .1. 3.22.... .53.... a. 3.1.3 o... 3 75.5.... D 99...... 2.5-1.5. 1252.13. .3 9.... z. 2...... ..........u.:....... 1.12 a. .3. 5...... 13.2.15... 1.. ma? 2...... o... :. .awfivucfiuc E... :. 12.33... P... .m .oz :33; :. 19.2.1.1 1.... :32... «Susan a... ..< 5......» .3 .1; 1.. 9...... a... 2...: {oasis L... a... 1. 51:13:... a... .31.... ..u::.:2 .1... w... z. .55.... 9.... .23.... a. a; 2.... .221 o... 1.... :5... v.2 .c ..:. .9... 3.95:... a... £1. 2.... .1... 2.3.7.... 5. ......:..:. OH. .95... a .3 3.15:... .2. o... z. 89...... .. 1.. on... a... .c a... a... 5 $2.... :. 11:0. .2... .1. a... .15.... a... ..u..:.:.. .1. a... .1. 9.1.1.... «:2 .1: 1... o. y... a... :5... o... a. .73.... 5... .51...» a... .......5. .75....» 13:32.... 3.5.1.... 2... a. 5. .25 L... a... -559... a... PE... 1.... .15.... 3...... a... 1. 19...... a... ..u=:.:2 E... 1.... .318. w... a. 7.... .9... .7...— ....$ .. .91.. .15. ..... z. .9... 9... 1...... .1 .. .319... o... c. ..= .. .15.. on... .1... 2.... .221 o... s. ..= .51.... a... 3...... .1... 2...... 91...... 2:3 2.. E....:....... .....1...... 1.... 2.... 1...... o... .: 9.2.25.2... o... .: ...w..o ....:o z. 5......» 3.21:5... ...u2-....a 9.. 3.2.7:... :14... 15...... 5...... 1.... 91..-...E1 a... $2.7. ....:=.. a .: 9.... 9.2.5.. 9... z. .51.... .. a: C... a... ..... 7.1.5.5: 9.5... a... :1 :. 2.... a... z. 1...... .5... .31.... 2.... 5.13-151 a... .c .2: .2. w... 12...... 1.2.... .22 .....::.:2 2.15.... .51.? o... 75...: .....::s....... .0351... o... a. w... .1... 3.2-1.2.. .22 .: 1:. h... a... 5...... 2.9:... 1...... 6...: 1...... a... .232... z... 3 5...? a... 3...... 1.... 0...... 1.... :2... a? .75....» ..... 2...: ..... .31... .. .2...» .5139... x. 9.... 1.... 1.... 1.... .22.. .12.. v... .21.. a... 3.5:... 1.... 9...... a... 5............. ........s 9.... .22.... a m. 9...... 1:2 1... o... a. 1.52:: m... .1... .. ”3:38 m...3:::. a... .3 1.... .........E:....:. 9.5 .15...» w... 2...: :. m... Cm .51.... .2...» 1...... no.3... v... .8. Q. 1.... 6...... 1.... 15....» 5......» 2.. .1. .2: .1. 2.. ...m c. m. .9155... 13.9%.... >15 2.... .2....7....=. 021.133.. .3 2.1.15... .22 :5... :3... a. 1.5.3... 9...... z. .. 5.... saga... a a. E... 11...... .92.? .51.... .. 1...... ... :5... 1.... :. 13.5.... m. 3...... o... .21.... .....=o.:.. 52...... a .o .155... o... 1. ....:..$ 5...... ...2 w._........ 5 1.1.1.2.... .. .31.... 1...... .1. 0.... o........:..... 59...... .51.... 3.2.5:... 1. 122...: .5: 13...... 2...... .............: a... 2: 1... .:.... £8.63. 1.8m mo 2.5 o... oboe-4 0.293.... $6.384 30.7.. o... 9.356: Page No. 29 duplex unit of this type in 1904 for the Empire Sulphite Pulp Company of Carthage, New Yerk which had a capacity . of 120 h.p. and operated at 87 r.p.m. This was a hori- ‘ zontal unit and was belted to the generator. The idea seems next to have gone abroad and while engineers like Mr. Gardner S. Williams were working on the vertical single unit direct connected type of plant, the European engineers who still favored the horizontal shaft princi- ple were increasing the size of these units by an appli- cation of the principle of the siphon setting. In an article published in the magazine "Die Turbine" Vol. II page 80, December 5, 1912 mr. H. Keller describes several plants in Europe utilizing this principle. The first described was a plant with three quadruple: units of 650 h.p. each at 15.4 feet head installed in 1908 at'Unter- brfich, Germany. In 1909 two units were installed at St. Mortier, France producing 945 h.p. each under a head of 23 feet. In 1910 a smaller quadrupleX'unit operating under a 10.5 feet head was installed at Kennelbaoh in Austria which produced 250 h.p. A similar unit was in- stalled at Barcelona, Spain in 1911. No doubt there were many other units of this type placed in service before the war broke out in 1914. All of the above plants consisted of multiple turbines on a horizontal shaft similar to the arrangement shown on.Plate No. 2. In order that the size might be as large as possible 'under the low heads available the turbine pit was sealed Page No. 50 by extending the upstream.wa11 of the power house to a point below the surface of the head water and in this way an air tight chamber was formed into which the water might be-raised by exhausting the air from.above the turbine either by means of'an.eJector or by connecting the turbine pit to the throat of the draft tube. The Escher Wyss Company of Zurich were evidently thepdoneers in developing these units in Europe but tests of units in siphon settings were made later in the testing flame of the J. M; VOith Company in Heidenheim, Germany and these tests show that there may be a drop of about one percent in turbine efficiency where siphon settings are used. With present day refinements in power house de- sign it is doubtful if tests today would show this same drop in turbine efficiency by the use of this principle. In this country the vertical shaft direct- connected unit became so popular*that it practically replaced the multiple unit horizontal shaft type of development long before the war. The horizontal shaft units that have been installed since that time have usually been instalLed in Wheelpits built for horizon- tal units where it is desired to keep the costs of re- development down by utilizing the old wheel pit. Plate No. 8 shows a case of this kind.where a quadruplex unit at Cheboygan, Michigan'was replaced with a sing1e runner [horizontal unit 72 inches in diameter driving the orig- inal 500.K.V.A. generator. This plant was designed in Page No. 31 1917 by Gardner 8. Williams who had pioneered the verti- cal single runner direct connected unit principle as mentioned above. The replacing of the quadruple: unit with a larger diameter sing1e runner turbim» where tie head was only 17 feet reduced the clearance above the runner so much that it was necessary to employ the si- phon principle and seal the turbine pit. This was one of the first applications of this principle to modern installations in this country and very definitely demon» strated the economic value of this principle. An un- usual feature of this design was the placing also of a part of the draft tube chamber above the head water I elevation. A similar redevelopment was made by the Fargo Engineering Company of Jackson, Nuchigan under the direction of mr. A. Streiff in 1921. This plant was for the Indiana and Michigan Electric Company at Elkart, Indiana. In this plant, which has a head of 18 feet, mr. Streiff followed the examples of the Eureopean de- signers and placed a quadruple: horizontal unit of 2400 h.p. capacity in a vacuum.f1ume. While European.example was followed in this design.the size was much.larger than previous installations. The flume was 22 feet wide and '71 feet long and the headwater was raised four and one-half (4%) feet.by means of ejectors and this re- quired the removal of over 7000 cubic feet of air to completely exhaust the air from the wheel pit. Page No. 32 The first application of the principle of the siphon setting to the modern vertical type unit in this country so far as the writer has been able to learn was at Plainlell, Michigan. This plant was also designed by Cr. Streiff of the Fargo Engineering Company and a Negler high speed propeller type turbine was used. This plant replaced an older plant with timber flumes containing seven old type turbines geared to shaft and coupled to one 750 K.W. generator. A fire destroyed the old plant on.May 20, 1919. The new plant shown on.Plate No. 9 went into operation on January 28, 1920 and contained three direct connected units in siphon settings, each with a capacity of 400 K.V.A. Within the same space limita- tions a plant of practically 50 percent increased capaci- ty had been installed. This increased capacity and high speed of 164 r.p.m. was possible due to the newly devel- oped Nagler runner but runners of the size installed wound not have had the required submergence in the conventional open flume. The siphon setting therefore made possible the use of the larger sized turbines in an installation, that would normally have been made with much smaller turbines because of the head available. This plant was described in the Electrical World for July 17, 1920 by the. Streiff and Plate No. 9 is from this paper. The Plainwell plant was equipped with the hydreucone type of draft tubes but the turbine pit was the conventional cpen flume type of setting with only Page No. 35 Plate No. 9 Rifle]: S“ Q ,GOVémord'H/mp D :" § 7 3‘ i a * e A ’10 no t 3"" ‘; 3‘35ré‘ii. ' " . "O'- U“; i ’s m. § ! "L toy-gs. r; -' _. ,- w 9;. - - 'n' .' ~.- . -' '.-. “‘. . .i'i" I I III I 'I o iii-«3,59%. '-f-;..- . .- .. '. :a Q . (,3 C’ 72?” 'fi dflxjs a Tm», 3717,35...“ .fi‘ Ififilhdnvs k1 .. 2.; . head Gale --. -/5 ’6 "fiend -.ii°.0‘~.'-.-.9;.’- . : ‘. - I. e - ' 'Ifit [WV v lag " fl— .— fl —‘ —‘ 51‘0 Grim/e a“ 8' x? .- . . . ., . . . fl" 9-u-..._H':- . has '5 “11.99,; ‘W. a '0’“. . fl. 0"... ..., ‘ 3"-.-' '0). “)2 _ ‘\ \ SECTION OF PLAINWELL (MICH.) PLANT USING NEW TYPE OF HIGH-SPEED TURBINES. The so-called vacuum-turbine setting is used, being sealed off by the curtain wall. Air is removed by the 2-in. ejector. Page Nb. 34 the additioh of air tight roof and an upstream.wall ex- tending below the surface of the headwater. A similar though smaller unit in a siphon setting was installed about this time at Oregon, Illinois for the Illinois Northern'Utilities Company. This unit occupied an idle wheel pit that had been built for an additional unit to be geared to a horizontal shaft driving a horizontal generator. The five units adjacent to this siphon flume unit generate about 450 K.W. or about 90 K.W. per tur- bine under a head of 8 feet. The direct connected unit which is shown on Plate No. 10 generates 125 K.W. or nearly 40 i more and is installed in the same sized flume. By use of the higher speed Leffel "Z” turbine a speed of 80 r.p.m. was obtained. The adjacent units cperated at 66-2/5. r.p.m. While these two plants were being built, Stone and Webster were designing for Henry Ford and Son a vacuum.flume plant with four units which was installed and placed in cperation in 1922 at Green Island, New YOrk on the Hudson River. This plant is shown on.P1ate No. 11. It will be noted from this illustration.that in this plant a scroll type of flux» was used and in addi- tion to extending the upstream.wall below the water sur- face it was curved and extended toward the roof of the pit near the turbine. This lip arrangement tends to re- duce eddies and increases plant efficiency. The Green Island plant, while one of'the early plants of this type, wet-=- 2., {)1} GE /~ 1 In \ FL A 7 .L.‘ A; " /0 f;::::;-:°- _;___-___— - _.____._ __ _ _ --- _‘_ d ,4/ a; ire - - — - 22 0 - - --» P? '% , . a i v f} fl G E Ge neruflaf 5 {,3 fiz—.:__:___. . r, 54: _;_ -__ - r f’sad q If? 91 'de .4: :“"‘i [*‘7 ;- t H4 9 4 r ’ / A H A i '. Y " :3 -_ y A ‘1 (I "’7’ >V‘ u ‘m. ._ :- _ ..' .. \ I’l'df’ ’rm 3. ' "9 T h ‘(I c.“ 48.2, Home .o ' 5 . - .~ . ' racks F Tar-bine ‘ ’ Head Wa+er 5/6» H I l k_ #1.th s um " o; " , . i A h . . ql/ Wafer " . 7 n. .- mng-fl- Q; A .‘ D L ‘ I. *7 ’—* 1---- v ' «4:. 3 .94: , . skins ., i b a: . “"339 z. ‘ ‘ .- ‘- :7 .3; . ' L _ ‘x ’ " ' \ _-_ fl - C :den’” .y . . /'/‘.\ \ ONCRET’E d_"_-'A ‘s- 0",:,-:_ ‘. DRAFT 72/545 . ~- 33., , ..\ J‘u. .'.\/ ..' \ .M . Caess SECTION oF' PLANT AT OREGON ILL/NO/S mrH .S/PHON SETT/NG BUILT /N /92/ Page NO. 36 11 Plate No. has: Swan «:0 'II'I"|'|I'II' O 52.! 8 elk E5? plant Cross-section of Green Island New York of Henry Ford and Son. Page No. 57 is the largest in h.p. capacity built so far in the United States. The turbines are rated at 2000 h.p. and operate at a speed of 80 r.p.m. under a head of 15 feet. After the step had once been taken the number of plants operating with siphon settings increased rapidly. In Appendex No. II is given a chronilogical tabulation of such plants of this type as the author has been able to collect data on. The principle has been applied to a large variety of plants with heads from 7 to 18 feet and with capacities from 200 h.p. to 2000 h.p. As has already been pointed out the applica- tion of this principle has made possible the installa- tion of turbines of large size by providing the neces- sary submergence without placing the turbine so low that the cost of excavation is prohibitive and excessive operating and maintenance costs result. The large pro- peller type turbines now being used for low-head in- stallations represent real economies over the smaller units formerly offered by manufacturers of power plant equipment and their use in the larger sizes is only possible by the use of siphon settings. The placing of these turbines at the higher elevations possible 'with the vacuum.flume and the use of the hydraucone or spreading type of draft tube has reduced the excava- ‘tions necessary for most installations. The placing Page No. 38 of larger units in the plants has reduced the number of units necessary and while the flumes may be larger the net result is a shorter plant than one of equal capacity built with smaller units without the use of siphon sett— ings. These features have tended to reduce the invest~ ment required per h.p. and with the reduced excavations and shorter power house design the construction costs for cofferdams and excavations have also been.matetially reduced. Still further reductions have been made in power costs by reductions in operating expenses. It is not the purpose of this paper to compare costs of various projects in an effort to prove the economic value of siphon settings but rather to show by example, where savings have been.made in plants that have siphon settings. No two projects are alike and direct comparisons of cost if they were available would not tell the story of economy. Labor costs vary from place to place, construction materials vary in price depending on their proximity to the project and the foundation conditions may affect materially the cost of the installation. Two plants built from the same set of plans at different localities might vary in cost to a considerable extent. It is rather the abject of this thesis to show that by the use of the principle of the siphon in the design of lowhhead plants real economies have resulted. It will be shown that larger sized, high speed turbines are possible in siphon settings which has Page No. 59 resulted in lower cost per h.p. Larger sized units have made it possible to equip plants with fewer generators and less auxilliary apparatus. The equipment being more compact, smaller power houses are required which has contributed to reduced construction costs. The above economies have not only resulted in lower fixed charges but also have made possible considerable reductions in Operating and maintenance costs. Page No. 40 Reduction in Investment by the USe of Larger Sized Turbines Operating at Higher Speeds. In the outline of the development of power plant design given above it will be noted that through the years engineers were constantly striving for in- creased size and speed of turbines as a means of simpli- fying the power plant layout and thereby reducing the plant investment. The progress made so far is due to a large extent to the efforts of turbine designers who have been working not only toward the improvement of turbine efficiency but also toward the increase of tur- bine speeds and capacities. The history of water tur- bine development as we know water turbines today is covered by the developments of the past thirty or forty years. This means that improvement in turbine design has gone'hand in hand with the growth of the electrical industry which alone has made possible the utilization of high capacity units operating at high speeds. In the matter of speeds, electrical generator'development has permitted of higher limits-of r.p.m. than water turbine designers have been able to reach.under low-head and large capacity conditions. At the time the first hydro-electric plant 'was placed in operation in 1882, turbine designers had ‘been able to build turbines that had efficiencies of nearly 90 76 when tested at the Holyoke testing flame. Demands for high speed and large capacity were not made Page No. 41 upon the designers and there were many turbines on the market under various trade names which gave good results when operating under conditions that required 810! speed in r.p.m. In order to compare turbines of this early period with those manufactured today it is necessary to set up some standard of comparison so that their charac- teristics may be compared under identical conditions. Water-power sites are far from standard and no two of them are alike as to head, quantity of water available or foundation conditions. To compare any two runners cf the same period or of different dates of manufacture some common ground is necessary. This can best be done by a comparison of "specific speeds" or "characteristh speeds” of the turbines. This term is defined in most text books on the subject of water power. Professor Mead in his text on "Water Power Engineering" defines it as: ”The number of revolutions per minute of a wheel of such size that it will produce one h.p. under a head of one foot.“ Plate No. 12‘ shows graphically how this value has increased through the years as improvements have been made in turbine design. At the time of connecting the first generator to a water turbine in 1882 the value of this constant was only a little above 80 for the highest speed wheels then installed and this value was twice that of the early Francis turbines of 184.9. During the period of the multiple runner developments speed was men, 2'4." {80 I70 [(30 I40 CT... to - B :1 ‘5 G O 0 0 O E: O SPEC/Fm 5PEED (M) \J O O .50 .40 ' tour/55 OF C CU F? Vii \JHUVV lN'G ”Cf/1 X/lVf U/‘V.’ v - -— C~~ J~ 5 ,— A’LUEé Of J.“£C/.’fC K F’hil) AIT'A.’NED AT VARIOUS pars/cue 9 .~— / DURING THI.‘ F3457” {"0 >£AQ5 :50 .2340 ’tI f- 15180 __ ‘ __/¢?;9c_7__ _ 1.900 . . 1’ 91A,“, Page No. 43 increased so that specific speeds of 75 to 80 were possi- ble. Through the efforts of Professor Zowski of the University of Michigan and others, further increases in specific speed were made in the period from 1905 to 1915 and this increase in specific speed made possible the single runner direct connected plants for low heads. In the plant at Sault Ste. Marie, Michigan and the plant at Superior near Ypsilanti, Michigan designed by Mr. Gardner S. Williams the specific speed of the runners was about 74. The direct connected plants at Watertown, New York and at Sterling, Illinois have turbines with specific speeds of 93.4. In 1914 Professor Zowski published an article in which he described tests of a unit with a specific speed of 102 and with an efficiency of over 90 %. The high values of turbine efficiencies now possible leaves very little room for improvement in this direction but increases of even fractions of a percent are welcomed by plant designers. The work of Lerner, Houdy, and Zowski was directed principally toward the improvement of the Francis type or mixed flow turbine. Improvements in specific speed of this type of turbine were accompanied by decreases in the top diameter of the runner and by a cutting back of the intake edge. This cutting back was carried to the ultima te in the Nagler or prepeller type turbine which is virtually an axial flow mrbine with specific speeds as hia as 175 to 200. This turbine was Page No. 44 a radical departure from the accepted standards of the day but has become today the recognized standard for, low heads and has been built for heads as high as 66 feet in this country and for heads as high as '75 feet in Europe. In low-head plants where a large amount of power is to be developed it is desirable to sake the units as large as possible. As an increase in size is accompanied by a decrease in turbine speed the lower limit of generator speeds becomes a limiting factor in turbine capacity and the short space between upper and lower pool levels places still further limitations on the also of units. The first limitation has now been raised by the increase in the specific speed and the spa cc limitation has been practically removed by the application of the principle of the siphon setting which raises the water inside the wheel pit above the head water level before it enters the turbine. Mr. Forrest Nagler first presented a descrip- tion of this new type of turbine to The American Society of Mechanical Engineers at its annual meeting in Decem- ber 1919. Several of these runners had previously been built and placed in successful cperation. Mr. Nagler in his paper presented a history of the development of this runner and clearly set forth its hydraulic characteris- tics. The commercial advantages which he claimed for it were as follows: a. Lower generator cost due to- an increased Page No. 45 speed of 50 73 and over. This saving varies fran 15 to :55 percent of the generator cost depending on the size and speed. b. Lower turbine cost due to simpler runner. This averages around 10 percent of the turbine cost, the weight being about one-third the weight of the corresponding mixed flow mrbine and much easier to build. 0. Smaller genera tor diameter which means a smaller power house. d. Higher generator efficiency due to better design possible with higher speeds. a. Greater turbine flexibility which permits the plant to give more power under flood conditions when the head is greatly reduced. All of these advantages have been proved in practice as will be shown by examples of actual install- ctions. These high speed runners are now designed by most turbine manufacturers and improvements in design have raised the efficiency of these turbines to a value above 90 percent. New types of draft tubes have also been devel- oped with these high specific speed turbines which have aided materially in improving efficiency and operating characteristics. As will be shown later these new type draft tubes have also helped considerably in reducing plant costs by a reduction in excavation necessary for Page No. 46 foundations and the draft pit. When the speed in r.p.mw and the maximum.h.p. which a unit will develop under a given head are known, the specific speed can be calculated by the formula _ R.P.M; x‘j H.P. )% Specific Speed - ( Head )5/4 From.this formula it will be noted that for a given specific speed any increase in capacity, which means an increase in the physical dimensions of the unit, must of necessity mean a decrease in speed. low specific speed units were built which would operate at speeds above 60 r.p.m, but the capacity was so low that generators of this size were impractical. Increases in specific speed were therefore an absolute necessity before the direct connected units at Sault Ste. Marie and Ypsilanti were possible. With propeller type units it was possible to increase both the speed and horsepower capacity of low- head units because of their high specific speed but for heads below 15 feet units of:maximum.size had already been attained due to the limitation of submergence re- quired. While increases in speed and h.p. would result in decreased cost and increased efficiency larger units could not be installed in open flumes as they would not have sufficient submergence. To overcome this difficulty engineers revived the principle of'the vacuum.flume ‘which, as has been stated above, was used by some manu- facturers of water turbines previous to 1900. Siphon Page No. 47 settings therefore have made possible the use of high specific speed propeller type turbines in larger sizes than would have been possible in open flumes and are now almost universally used for heads below 15 feet. Some examples of the economies that have been effected by the use of these turbines will serve to demonstrate the economic value of the siphon setting which makes possi- ble their use when the head is so low that sufficient submergence cannot be obtained in an open.flume. The first modern vertical unit with a siphon setting, as mentioned above, was at Plainwell, Michigan. The rephncing of 7 turbines by 5 propeller type turbines represented not only a considerable reduction in the number of moving parts to be carefully machined and assembled but also because of the simplicity of the runner a considerable saving in the weight of the meter- ials used. This saving as estimated by Mr. Negler is about 10 % in the cost of the turbines. The higher speeds possible with the new runner at Plainwell made it possible to install direct connected generators of 400 k.v.a. capacity operating at 164 r.p.m. The saving in generator costs and the elimination of the cumber- some harness work and gearing all helped in reducing the investment in this installation. With the runner diameter of 6 feet the top of the turbine was too high for proper submergence and the siphon setting was used to raise the head-water inside the turbine pit to provide sufficient Page No. 48 cover over the top of the wicket gates. The specific speed of these turbines was 145 and with.a speed of 164 r.p.m. much smaller generators were necessary than.were usually employed. This reduction in generator size not only improved the appearance of the plant but also pro- vides more space between the units. A considerable number of turbines of this type have been installed along the Rock River in Illinois and Wisconsin. The unit installed at Oregon, Illinois the year following the Plainwell installation was much smaller in capacity and the head being only 7 feet the use of’the siphon setting was even more essential for a successful direct connected vertical unit. The Oregon plant in 1920 contained 5 turbines geared to a horizontal shaftv driving a horizontal type generator direct. The speed of the shaft and generator was 200 r.p.ms but the tur- bines operated through gears with.a three to one ratio so that their speed was about 66-2/5 r.p.m. Six wheel pits had been installed but only five of them were in ,use in 1921. To provide additional capacity and also partly as an emperimental installation a unit with a siphon setting was installed in this idle wheel pit in 1921. The original turbines had specific speeds of approximately 60. The new turbine installed was one of the Zowski turbines manufactured by the James Leffel And Company. It had a specific speed of 80 and operated at 80 r.p.m. This unit had its top set at head water eleva- Page we. 49 vation and the roof of the wheel pit westbout six and one-half feet above the head-water elevation. The small size of this unit made its cost per h.p. higher than some larger units but its satisfactory operation had demonstrated its economy. The next installation on the Rock River was at Rockford, Illinois. Two units were installed for the Rockford Electric Company in a plant that originally contained four units. In the interests of economy as much or the original foundations were used as possible. The units were designed for operation at 100 r.p.m. and developed 750 h.p. under a head of 9 feet. The specific speed was 175. The rephacing of four*units by two units represented a considerable saving and the increase in speed made further savings possible in the cost of the electrical equipment. The plant at Dixon was started in July 1924 and finished in June 1925. A view of'the prOpeller type runner built by the I. P. Mbrris Department of the Wm. Cramp and Sons Ship and Engine Building Company and now a division of the Baldwin-Southwark Corporation is shown on Plate No. 13. The physical dimensions are compara- tively large and the wide areas between blades permit the passage of large quantities of water. This type of unit has the further advantage of permitting a wider rack spacing and results in a reduction in.rack:losses. Plate Nb. 14 shows a cross section of the Dixon plant 50 Page NO- , . m.-. Maw. _ m/ ale-II. Ill. ,¢- Show. DMOTI ‘ 1L ‘ ‘- I. . IXOH HY WPATWT; PrOpeller type turbine installed at 'i i Illinois. Dixon, 14 Plate NO. I . «t//\/\Z v l/RX >L :r‘ i I:' ///hmm¥ILI/V .. . . ...-......u. ... . on a.....-.P'>...i’as V... A J‘- . ll fl on.» E 5.351 £163go, A ‘ \ m I I-77 \ Cuuss-&CHON or m: PLANT A r Due/v. ILL/~01: Page No. 52 which has wheel pits 45 feet wide. At the time this plant was built the average head was between 8 and 9 feet but dredging of the river below the plant has in- creased the head so that at tie present time the average head is between 9 and 10 feet. The runner diameter is 125 inches or just equal to the maximum.head on the plant. 'Units of this size would have been impossible at this site without the siphon settings in which they are installed. The top of the turbine is above the normal head water elevation.and a view of the turbine setting is shown on Plate No. 15. The bottom of the runner is Just above normal tail water elevation so that it can be inspected periodically a feature which is important if turbines are to be properly maintained. The Dixon units are rated at 800 h.p. when operating at 80 r.p.m. under a head of 8 feet. The specific speed is about 1'70. The Sterling plant mentioned above, located about 14 miles down stream from Dixon, was built Just 10 years before the Dixon.plant. These turbines are in cpen flume settings and are about the maximum.in size which can be installed in an open flume when the head is only about 9 feet. Some comparisons of these two plants will illustrate the economy of the high specific speed turbine with siphon settings. The Sterling units have a specific speed of 95.4 while the units at Dixon have a specific speed of 170 or_almost twice as high. Both plants contained five units each until December 1950. Plate No. 15 59-4‘ I w |!\ lung,” "Y ‘4'“: l "1 ‘Wi’ of the turbine setting for the plant at Dixon, Illinois. Page No . 54 The units at sterling have a capacity of 550 h.p. under a head of 8.5 feet. The Dixon units have a capacity of .75 h.p. under a similar head which is a 60 percent greater capacity. It would have required eight turbines of the capacity of the Sterling units to equal the capacity of the Dixon plant. The Sterling turbines are the Francis type mixed flow runners and as stated above are more expensive to build. The high speed character- istics of the Dixon units makes it possible to operate at 80 r.p.m. while the Sterling units operate at 60 r.p.m. This higher speed at Dixon reduces the cost of generators as well as their size. The Sterling units are 18 feet 9% inches in diameter and the Dixon units are 15 feet 4 inches in diameter. The higher speeds not only reduce the generator costs but also the turbine costs due to the fact that for a given runner diameter a larger quantity of water can be passed at a higher velocity and consequently a greater h.p. is produced. The runner diameter of the Sterling units is 122 inches or only three inches less than the Dixon units. The Dixon units, however, have about a 33—1/5 % greater’water capacity and about a 60 percent higher h.p. rating. The Sterling units were purchased prior to the World War and any comparison of costs must be adjusted to values such as would exist 1: the units were made at the same period of time. In 1925 the wholesale commodity index was 150 as compared to Page No. 55 1914. In other words the Sterling units might have cost 50 % more than they did coat if they had been furnished in 1925. The combined cost of generators and turbines at Sterling in 1914 was about 56 dollars per h.p. includ- ing the erecting. Based on 1925 price levels this cost would have been 55 dollars per h.p. The cost of the Dixon units in 1925 was only 47 dollars per h.p. including erection based on their capacity at an 8.5 foot head. The relative lower cost of'the Dixon units is due to two things. First the units are of larger capacity and second the prOpeller type units are lighter and cheaper to build. Units of the size of the Dixon units are only possible for such a low head when installed in siphon settings. The actual savings at the Dixon plant were about 10 % of the Engineers' estimates on.the turbines and nearly 30 % on the generators. The cost of'the equipment was over $50,000.00 less than the estimates and this saving is a rather large percent of the total cost of the project. Still further comparisons as to economy can be made at the Sterling plant for in 1950 a sixth unit was installed in an idle wheel pit. The original units were installed in Wheel pits much too small for their capacity. In low-head plants water velocities must be kept as low as possible to reduce the drop in head in the wheel pit. Flume velocities at Sterling are too high for high efficiency operation and at times they Page No. 56 reach values as high as 5 feet per second. The new unit was selected with a capacity which would keep this value to about 2 feet per second. The water capacity for nor- mal operation was therefore limited to 500 cubic feet per second. The normal capacity of the original units with a 9‘0" head is about 800 c.f.s. The new unit was installed in a siphon setting with the top of the run- ner 2 feet above head water. Plate No. 16 shows the turbine setting for this unit and should be compared with Plate No. 15 showing the setting for the turbines at the Dixon plant. Plate No. 17 shows a cross section of the new unit and it is to be noted that there is no large air chamber above the unit as at Dixon. While this new unit is installed in a siphon setting it is of such size that it might have been installed without the siphon setting by placing it lower. Draft tube design however, raised the setting so high that a vacuum.flume was necessary. The draft tube for the larger units hav- ing been installed in 1914 when the plant was built economy required that it be utilized by extending the upper part of the tube until its diameter conformed to the-diameter of the turbine speed ring. Had units of this size been.installed at Sterling originally it wouhi have required 9 of them to have equalled the capacity of the Dixon plant. A comparison of the cost of this equipment with that at Dixon shows that the equipment for the Dixon plant with fewer units of larger capacity 16 View of the turbine setting for the 11’ 3’30 plant at Sterling, Illinois. I. Riv-draws, Centraclor 1m). and“, War. Plant (New unit) Ilate NO. '1 ~n “Wuizumggme ‘ ui' ' {W : l . I !!!:!"'°! 93;:1'29L Lyn-eels: fen Sufism“? 19% F ’ q V"‘\~D" :“"~"-l' \ :u‘.‘ b' ”-1. “19L". Jar ..St‘vr; .. _; .1132 ._' u \\ _:-:A'[ v— ..j- , ‘E, ; ‘r 1 f ‘ ngjhx-‘q‘qs— .. . .,- were, 7. \ - ‘g\ :s - - ~ J y "’4- 101$ l ‘ “ L ' I l ‘ / - - xkr' -DC \./ .. "- r 2. L- ’1 : :- P ‘ " x I .--, . 4 ..v‘. . . ‘ C y 9'- Ag r . 5 4 «r; . “I it‘ ‘ . G a I: 5 Q a -. 1. ‘ \ ‘ - .3: “E: v I V . ‘ l-‘-’xz. 1; ”LE. aLCC" . J I ‘ 4 L I”, " RC>‘A'_~ ’ “ x \ ) N ’5 ~ - he ‘ /( . _ 4 1 ‘ ' ’ A“ nu, “‘ ' “H'I'H' II 0 HI ' I . y .A .. o; no I I \ U r - __1 _ -.—‘ IE. . Qu- « / 1;. I T -‘, Q :3 . . __-‘_‘ t;_,,l . - u‘ - . - . — i _ ¢f-;£:§-.__§_ “ /4 , $3.;Al- AC3.Q\‘S:_I .af. , I ,u / y. F D 1 ES ._ x. E 4‘66 I ’I ‘ . - -_ .. / . v 1’ ’ “ I ‘ - _. LEI»! Tut-J er r h -’ ’ ‘JQC. P,',? ( (r , . ‘.‘ \c-‘I 5 23‘ I I \\ I \ ‘ I w / \“\ l ' l l _, I “a ' ' \ /’ ,7., I. _-‘_‘ \-_’ .——"/ ‘\\\ I l .\\ In I '\ ‘HH ‘IIH ““"HH 'H“ ' .\- "\\\‘ Cross-section of the plant at . J E sterling, illinois. {38“ unit) Page No. 59 cost less per h.p. Plate No. 18 is a view inside the Sterling plant and shows the difference in the size of the generators for the old and new units. The new unit has a specific speed of 175 and Operates at 128% r.p.m. and has the further advantage of having turbine blades which are manually adjustable for varying conditions of head and quantity of water available. The water capacity of this turbine is about 50 at that of the Dixon turbines and its h.p. rating at 9 feet head is 450 as compared to 950 h.p. for the Dixon units at the same head. Due to the larger hub diameter required by the adjustable blade feature the turbine diameter is slightly larger than half that of the Dixon turbines. Its throat diameter is 81% inches. The cost of this unit for generator and a fixed blade propeller type runner would have been about 48 dollars per h.p. The adjustable blade feature added about 10 ddllars per h.p. To the cost but this increased investment is justified by the additional kilowatt hour output possible from the turbine with adjustable blades. While the Keokuh plant does not have a siphon setting some interesting facts as to the economy of the high specific speed turbines can be shown by comparing this plant with the Ohio Falls plant at Louisville, Kentucky which has propeller type turbines. The 10,000 h.p. turbines at Keokuk are 168 inches in diameter and Operate at 60 r.p.m. The runners weigh 150,000 pounds. Page No. 60 Plate No. 18 31 i h m mm. ”I" ‘3 k v . i 15 PM _ Vie" insi ’e the Sterling plant ' . O .‘ . 1 shoving 013 an” new generators. } i “ M Page No. 61 At the Ohio Falls plant the 15,500 h.p. propeller type turbines are 180 inches in diameter and cperate at 100 r.p.m. The runners in this plant, however weigh onLy about one-third as much.as do the runners at Keokuk or about 4C,000 pounds. The decrease in turbine weight means economy in the cost of the turbines and the in- creased capacity possible means that today a plant the size of Keokuk could be built with fewer units and the higher speed generators would not only cost less but would be much smaller in diameter and more efficient. These same comparisons are true with almost any low- head instalh'tion and because of the simplicity of the design of the runner and its real economy this type of turbine with high specific speed is now generally used for all low-heads. When these units have been built for heads of 15 feet or less their diameter and height of gates are usually so large that the siphon setting must be used to provide the necessary submergence for satisfactory cperation. The routine cperation of any hydro-electric power plant requires that the current be generated to meet the demands for power. In an interconnected system the duty of the taking of the variations in load is usually assigned to the units having the best efficiency under part load cperation. Economy of operation in a steam.generating station is best obtained by a high load factor and where possible load changes are assigned to Page No. 62 the hydro-electric units. In run-of-river plants having no storage however, this plan of operation is impossible and the steam plant has to assume this duty of taking the variation in load. This leaves the hydro turbine to operate at its most efficient capacity. If it were possible to operate in this manner throughout the entire year a maximum.of kilowatt hours would be obtained. Variations in water supply on rivers without storage, however, make it necessary to operate turbines at part- gate for considerable periods. This is especially true during years of drought such as was experienced in 1950 and 1951. The Francis type or mixed flow turbines had reasonably high part-gate efficiency but the new propeller type turbine was not so satisfactory for part-gate cperb ation. This fact was brought cut in the discussion of Mr. Naghar's paper in 1919 when he first described these turbines. He answered this objection by stating that the part-gate efficiency of the high-speed runner was 6 to 8 percent lower for half load mien the mixed-flow 1o! specific speed runner. He added that "the real field for the runner is in such an installation as that at Keokuk or that on the St. lawrence River, where with.a large number of units, the part-gate efficiency would not be a de- ciding factor in the problem." In his paper Mr. Nagler also stated that "a runner may be taken.out and another substituted for capacity variation under flood conditions without removing any other turbine parts except two or three Page No. 63 guide vanes. In two plants this was made use of to install a high-capacity runner for obtaining more power during flood periods." The changing of turbine runners, havever, was not a practical method of providing for variations in flow. Engineers who were concerned with the design of plants usually proportioned the size of units so that there would be as little operation of the turbines at part-gate as possible. In some plants units of differ- end sizes were installed to overcome this difficulty of low efficiency at part-load. On the Flint River in Crisp County, Georgia two 10,000 h.p. units and two 1,400 h.p. units were installed to provide for the variations in flow and load. At the Dixon plant the turbines selected were of 800 h.p. capacity and utilized about 1,000 cubic feet of water per second. The lowest recorded minimum flow was about 800 cubic feet of water per second and it was thought that this size unit would require a minimum of part-gate operation. However, in 1931 even lower flows continued for long periods and considerable operation of the turbine at part-gate was necessary. Plate No. 19 shows an efficiency curve for one of the units at Dixon when operating with a head of 8.5 feet. The physical dimensions of the Dixon units are large for units for such low heads but even larger units might have been used and a plant of four or per- haps even three units to give the same total capacity might have been installed had it not been.for this 1 as Q.) 1.. , a, , v» F/L/L/VCY 23‘ f"; .e'ez'xv‘ r 1 . .l'"" :1," 11;),li131 A. .‘: "'3; H" A”? [9 ,.___..-——-._.—.. c“. -.-.- A {91.21.35 Vixen‘rm‘z‘ - 5.7.7,, Cure v55 5 #0 mm; [ H- 1/ (MM Y of 7:133 w; s f i a .4 T 5 TER L jiNG A IAVD {Dir/L1- ( IN, 171/. I. Knit: '3 747 3P5UF/C 5 P1133 .5 (47.) £93 ’r’fz ll- C) Y v _ 80 PERCENT [FF/C/[NC 60- ‘ I TYPICAL [kF/UFNL v C c .1 3 v1. 5 FOR films/~55 wz .7 H 7-5” 51. / ‘3 U SPEC/Fm Sperm: (N); ' K 1 1 -....- .. i. 1 -7 . _.,,‘-W . _ ,7 f i _ ._ ,- _ . .. - A i a I ; PERCENT .CF Full. LOAD -6‘Q.,..w __. l [/O_ _, _, 7 (SC) J ,_ _ 90 {I‘d/"(3 Page No. 65 limitation of low efficiency. at part load operation. Limitations as to size of units has been removed as far as submergence is concerned by the use of the siphon setting. Following closely on the development of the high specific speed propeller type turbine an even more radical step has been taken.with the propeller type tur- bine itself by a design which permitted the articulation of the propeller blades at the hub to permit a change of blade angle under different conditions of operation, a thing that was impossible to attain with a Francis type turbine. In this country the first adjustable blade turbine was manually adjusted and required the unwater- ing of the wheel pit as the adjustment was made at the hub of the turbine. fihile this was An improvement over changing of the entire turbine far one with different blade angles it was nevertheless not an entirelysstis- factory method of operation. Later improvements per- mitted the blade angle to be changed at the point where the turbine shaft and the generator shaft were coupled together. This was an improvement but still required the shutting down of the unit to make the change. Later improvements permit the articulation of the turbine blades while the unit is operating. This is done by means of an electric motor controlled at the switchboard. The new turbine at the Sterling plant, referred to above, has adjustable blades that are adjusted manually at the Page No. 66 coupling when the unit is at rest. In Europe the pro- gress of the adjustable blade turbine design has been much more rapid and for several years they have had in successful operation units in which the blade angle is changed by oil pressure from the governor. For every change in load which requires a change in gate opening there is a corresponding change in the setting of the blade angle so as to provide efficient operation over a wide range of load change. Plate No. 19 shows a family of efficiency curves for these several types of units. At the bottom of Plate No. 19 are shown efficiency curves for a turbine of the Francis type with a specific speed of 75 and a curve for a propeller type turbine with a specific speed of 150. It is readily seen that the part-load efficiency of the high specific speed turbine is considerably below that of the Francis type runner. The upper curve is for a Kaplan turbine with a specific speed of 150 and the high efficiency of this turbine at pcrt-load is readily apparent. These curves are taken from an article published by George A. Jessop, Chief Engineer of'the S. Morgan Smith Company, in Electrical Engineering for February 1931. These curves indicate clearly that additional kilowatt hours can be obtained from.the adjustable blade turbine when part-gate operation is necessary. Just how this works out in practice is shown by the curves at the top of the curve sheet. The Francis type units at Sterling, Illinois .__—' Page No. 67 have not been tested but from test data on this runner, (Holyoke test No. 2182), the efficiency curve has been constructed. This unit is one of the early turbines built by the Allis Chalmers Manufacturing Company to meet the demands for higher specific speed turbines and has a specific speed of 93.4. While the efficiency is not as high there is a reasonable resemblance to the curve for a turbine with a specific speed of 75. An index test of one of the Dixon units provided the data for the curve for the unit with a specific speed of 170. This curve is for a head of8.5 feet. Typical of mirbinss of this class the part-load efficiency is below that of the Sterling Francis type units. While the usual opera- tion at the Dixon plant is with turbines Operating at the best efficiency there are considerable periods when part-gate operation must be resorted to. Recognizing this loss as being considerable, the new unit selected for the Sterling plant in 1930 was designed for manual adjustment of the turbine blades. After installation this plant was tested by the index method. A current meter was installed in the flume about 8 feet upstream from the entrance lip and at mid-depth. Simultaneous readings of power output and flume velocities were taken and from later tests to calibrate the index meter suffi- cient data was provided to construct complete efficiency curves of reasonable accuracy for each setting of the turbine blades. The curve shown on the curve sheet for this turbine which has a specific speed of 175 is an Page No. 68 envelop of several efficiency curves obtained from these tests. The automatically adjustable blade turbine cper- ates at efficiencies as represented by this type of curve if the prOper cans have been.worked out either by tests or from theoretical calculations but with the man- ually adjustable blade turbine it is necessary to set first the turbine blades and then adjust the turbine gates for best efficiency. Once the information Has been worked out this is not a long or difficult process. Turbines of either type will give increased kilowatt hour output and the frequency of load changes determines the necessity for the automatic feature. Where changes are seasonal the manually cperated turbine blades will give satisfactory results at a lower investment cost. In this country the European type of turbine was first manufactured under license by the S. Morgan smith Company of York, Pennsylvania. In paper entitled "Greater Efficiency for Low-Head Hydro" published in "Electrical Engineering" for February 1931 by George A. JeSSOp, Chief Engineer for the S. Morgan Smith Company, he states that "Economy in first cost has stimulated 20 years of effort to secure a higher speed in hydro-electric units; principally because generator costs go down as the speed in- creases. Until reeently an increase in speed has resulted in a reduction in effici- ency - particularly at part loads - to a large extent offsetting the advantages of reduced investment. Improvement in generator efficiency with increased speed has only partially offset these reduced turbine Page No. 69 efficiencies. Turbine designers now have succeeded in maintaining the maximum or peak efficiency at a given figure over a wide range oi‘specific speed and by means of the Kaplan or automatically—adjustable bhide runner have improved the pcrt-load efficiency far above that formerly obtained even in.the slow-speed Francis turbines." This develOpment in hirbine design has made available to engineers turbines with high efficiency over a wide range of load. Low-head plants, where pro- peller type units are applicable are usually on streams where little or no storage is possible and where the development, if made, must be a run-of-river insralla- tion designed to use a large percentage of the flow of th water which is available for at least a part of the year. i fur- As pointed out above the high peak of the efficiency curve of fixed blade turbines requires that several units be installed and then the units either operated at the most efficient gate opening or not at all. On the other hand the adjustable blade turbine has an efficiency curve of such shape that it is almost ideal for a run-of-river plant. Variations in load can thus be provided for without a serious departure from peak efficiency. Just what bearing this development in turbine design has on the economic value of the siphon setting is best shown by the fact that it is now possibka to build hydro-electric plants with still larger units than.have been used heretofore for the low heads and for this reason fewer units are required for the same Page Uo. 70 total capacity. The high efficiency at part load makes it possibha to obtain an even greater number of kilowatt hours from an installation than would be possible with fixed blade prOpeller type units of equal capacity. If siphon settings were necessary for the successful utili- nation of the propeller type turbines for low-head plants they are even more essential for the settings for adjustable blade turbines whether manually or automatical- ly controllzd. Kr. L. F. Harza, Consulting Engineer of Chicago, was responsible for the installation of one of the first Kaplan turbines in this country. This plant was located on the Devels River in Texas. In writing about these turbines in an article entitled "Water Turbines of the Propeller Type" published in Electric Light and Power for April 1931, Mr. Harza states that "Although the Kaplan turbine costs more than either the Francis or fixed blade propeller types, for fairness the comparison must be made on the same total capacity, but a smaller number of units, thus tending toward a reduction in cost. The alternative procedure is, of course, to put in part fixed-blade propellers and part Kaplan units, the former to be oper- ated at best efficiency or not at all, and the latter to handle the variations in load." In the spring of 1930 a plant was placed in Operation on the Rock River at Rockton, Illinois in which this latter procedure was followed. The plant contains one fixed blade propeller type turbine of 725 h.p. capacity and one Kaplan turbine of 994 h.p. capacity. Page no. 71 The head is 11 feet and both turbines are in siphon settings. Had it been necessary to install only fixed blade turbines the installation would no doubt have been made with three turbines and even turbines of this size would not have been passible for such a low head were it not for siphon settings. Plate No. 20 shows a cross section Of the Rockton plant along the centerline of the : adjustable blade unit. So far as the author has been able to learn this is the first automaticallywadjustable blade turbine installation in this country in a siphon i setting. Several manually adjustable blade turbines, D however, had previously been installed in siphon settings. The high efficiencies obtained with adjustable ; blade turbines over the wide range in power output and the increase in the size of these turbines built in 1931 indicates that even larger units will be built in the future for low-head plants than have been considered practical in the past. As the size increases it will become increasingly necessary to apply the principle of the siphon setting to even higher heads than heretofore has been considered necessary. it Ottumwa, Iowa three units were installed in 1931. The head at this plant varies from 12 feet to 14 feet. The units are 1,400 h.p. capacity under a 12.25 foot head. The turbine pit is sealed to form a siphon setting but under certain condi- tions of high head-water the turbines may be operated under pressure. It can readily be seen that increases t ; ' ’ Plate No. 1 I, ? ' 9 ((44 'mwl’ Ms! ,1: ”-2! I" f! r; 1 (and? Iwmqw 57 .5 #15 a ”arm 40% Hy 7”“ (a 3 ¥ . ”(01' ; '5' ..... in size and C and 30 feet '4 econoziwl it the 3001‘: I31 BOXEEW to 0 42,000 11°?" etion on the under a heat Where a' cate the ’51” adjustable creased 03? in an inst?- 1nstallati< uphon set head inste practical . power with constructi the alpha: 14 feet. largest e1 three inc?" Page No. 73 in size and cap:city of turbines for heads between 15 and 20 feet will require siphon settings for their economical installation. The 21,000 h.p. turbines for the Rock Island Development of the Washington Electric Company to operate under a head of 32 feet. and the 42,000 h.p. units for the Safe Harbor Water Power Corpor- ation on the Susquehanna River in Pennsylsania to operate under a head of 55 feet which were installed in 1931 and whichhare automatically adjustable blade runners, indi- cate the trend in turbine design toward larger and larger adjustable blade turbines. The economic value of in- creased capacity per unit and decreased number of units in an installation has been clearly demonstrated by the installations of the past few years. Combined with the siphon setting this economy can be extended to those low head installations which lie on the borderline of practical development. In Sweden where there exists abundant water power with high heads there is at the present time under construction a plant that will utilize the principle of the siphon setting to develop power with a head of only 14 feet. The turbines to be installed will be the largest ever installed in a siphon setting being 26 feet three inches in diameter and will develep 16000 h.p. maximum. The runners will be of the Kaplan type and I111 weigh 150 tons each. It is expected that this plant will be placed in operation sometime in 1934 and Page No. 74 is known as the Vargon plant. It is located on the Gotha River at the outlet of Lake Vanern and is being built by the Royal Board of Water Falls. For low-head hydro—electrical developments there is now available equipment which can be operated over a wide range of load without a serious reduction in efficiency. This type of turbine permits the in- stallation of larger units than have heretofore been considered practical because of this feature of high part load efficiency. The increase in turbine size with fewer units in an installation tends toward reduced cost per h.p. installed and the wide range of operation at high efficiency provides a means of increasing the kilo- watt hour output when part gate operation is necessary. The higher speeds aVailable with the fixed as well as the adjustable blade turbines has reduced generator sizes and the cost per kilowatt and at the same time in- creased the generator efficiency. These advantages which were pre icted for this equipment by Mr. Nagler in his paper before the American Society of mechanical Engineers in December 1919 have been proven in practice and are not only available for heads above 20 feet but also for lesser heads when the turbines are installed in siphon settings. - Q—' ‘4‘. m fwd»: -400 75 Reduction in Investment by a deduction in the '1 V Size 0: the To er Ilant. In the preceeding pages we have considered the economies nade possible by ihprovements in the hydro- electric eiuipnent. “Increases in efficiency are desir- able when they can be obtained at little or no increase in cost. he have seen that while {aplan turbines are more expensive than fixed blade turbines of equal size their use can be considered economical when the size is increased and the number of units reduced. Bifteen years ago the open flume was considered the practical and logical sett'ng for low-head turbines. I.groxwments in the hydro-electric plant equipment have dam nded changes in plant design and as the cost of .h plant for low- head developments may exceed the cost of equipment by a considerable armunt it till be well to give some consid- eration to this item of investment. For the mall Francis type turbines of low specific speed the open flu e setting offered the most economical type of installation. Some attempts were made to improve on this type of turbine setting in the early days and for the smaller units steel jackets were provided instead of the wooden flumes or flumes of ashlar masonary. At Sterling, Illinois there can still be seen almost all these types of turbine settings. Tie earltest plants had timber flumes, later plants had ashlar masonary walls and one such plant is still Operating after not kiss [full-III. .rjw‘... . H a. Page No. 76 than 40 years of service. This plant represents an economical use of the natarials available at the time of its construction but is hardly to be recommended as a modern installation. he the use of cement concrete came into more common use it was applied more and more to power plant construction. In 1904 the plant shown on Plate No. 1 was built at Sterling, Illinois. It was de- signed by Professor Mead and the three turbines were placed in separate wheel pits with concrete walls between them and an attempt was made to improve the hydraulic conditions by making the downstream wall of plate steel and semicircular in shape. As turbine speeds increased engineers began to improve flume designs. The improved turbines would take almost twice the water through the same throat diameter and one of the mistakes of the early designs for high specific speed turbines was failure to take into account the flume velocities. The units at the Government dam at Sterling were placed in open flumes and the average water velocity at the intake is about four feet per second. At the Dixon plant the design is such that the flume velocity is about two feet per second and in addi- tion scroll type flumes were employed. It cannot be denied that improved flumes with scroll type walls cost :more but as the art of concrete forming has improved the building of these curved surfaces does not present the problem.that it formerly did. he with the turbine costs Page No. 77 the fact that fewer units are required to make up a given capacity the total plant cost may even be less for the improved flumes than for a greater number of box flumes. The redeveIOpments of the past few years in which propeller type turbines have been installed in scroll type cases with siphon settings emphasize the economy of fewer units for a given capacity development. In most cases even greater capacity has been achieved, within the same space limitations by fewer units of larger capacity. In Plate No. 2 there was illustrated the saving possible with a single large unit to replace a quadruplex horizontal runner. In the building of the first vertical type unit in a siphon setting seven units were replaced by three larger units. In the rebuilding of the plant on the Rock River at Rockford, Illinois two units replaced four and the old substructure was used by removing the concrete in the intermediate walls. Illus- trations might be cited without number to show the ten- dency during the past few years to decrease the number of the units in an installation and increase the plant capacity. The result of this change in plant design has been to reduce the plant cost by a decrease in materials required per kilowatt of plant capacity. There are certain necessary parts of a power plant installation which contribute to the cost of the installation but not in proportion to the capacity. This auxilliary apparatus can be reduced in quantity and cost Page No. 78 by a decrease in the number of units for a given capacity. If we first consider the switchboard and electrical wiring for a small hydro plant it is evident that one panel must be required for each unit installed and conduits for wir- ing placed in the concrete must be run to each unit. If the number of units be reduced one-half by making their capacity twice as large it is evident that there will be a considerable saving in this item of equipment alone. The wire and conduit for the larger unit mighttae in- creased slightly but the labor and materials for the switchboard would be cut in two. In a large installa- tion the switchboard cost is not a large percentage of the total cost of a hydro development but for the small low-head plant such as is usually built with siphon settings the cost of the switchboard in percent of the total cost is much larger. A saving such as has been suggested above I111 materially reduce the investment in any low-head hydro-electric development. The cost of the governor for the turbine is another item of plant cost that is reduced by the re- duction in the number of units. A larger governor'may be required for the larger turbine but the cost of the turbine governor does not increase in proportion to the capacity in h.p. of the unit to be governed. The cost of this equipment is sufficiently large to affect mater- ially the investment in a plant and in some plants in order to keep the investment down to a value low enough so that I38 ;_:e 1‘30 . 79 a fair return might be earned, the governors have been omitted entire y. The unit installed at Oregon in 1921 is small and has no governor. The run—away speed is not high and usually the operators who are on duty are able to close the gates before the full run-away speed is attained. This unit has cperated successfully for 10 years without a governor. Considerable study has been given to this subject of reducing plant costs by omitt- ing the governors and several small units and some larger ones have been so installed without depreciating the value of the plant as an operating unit where it is Operated on base load with no load changes to provide for and where operators are on duty to shut the unit down in case of trouble. At the plant located at the Government dam in Sterling this item of cost was given considerable study and as a means of reducing the in- vestment the units were so designed that one governor controlled two turbines. At the Dixon plant smaller capacity governors are controlling units with capacities nearly equal to two units at the Sterling plant. The necessity for a governor to control the hydro unit is not determined by whether the unit is in a siphon setting or not. That is a problem.of engin- eering and is determined largely by the size of the system.with which the unit is connected and the method of operation. If governors are required in the hydro- electric plant it is evident that:when larger capacity Eage No. 80 units are installed in siphon settings the cost of gover- nors per h.p. installed will be less than with a greater number of units installed in open flumes. The exciter is another plant auxilliary that can be reduced in cost by the increased speed and larger size of units now possible for low-head installations. Several illustrations of typical power plants included in this thesis show exciters direct connected to the main shaft and mounted on top of the generator. In larger power plants separate water wheel driven exciter units I are sometimes provided but in smaller plants it is a f more common.practice to make the exciter an integral part of the machine which it serves. For slow speed 3 machines the cost of this type of exciter is high and as with generators the cost reduces and efficiency in- creases as the speed increases. Where the speed is too low for economical exciter units direct connected on the machines, motor generator sets are employed. As this involves generating the current and then passing it through a motor to operate the exciter the losses make it less efficient than the direct connected type of ex- citer. With speeds of 100 r.p.m. or over satisfactory direct connected exciters can be built at reasonable cost. At the Sterling plant excitation was first supplied by exciters driven by belts from.a jack Shaft geared to the vertical water wheel shafts. This arrangement was not very flexible and later-a motor.genen:tor set was instdlled. The speed of 60 r.p.m.'at which the Page E0. 81 generators cperated was too slow for economical exciter design. Even the 80 r.p.m. generators at the Dinah plant do not have direct connected exciters and moter generator'sets supply the excitation in this plant. The new unit at Sterling operating as it does at 128% r.p.m. is supplied with exciting current from a direct connected exciter meunted on top of the unit as shown in Plate No. 18. The advent of the high specific speed turbine which has made possible higher Operating speeds also made it possibka to design direct connected exciters for generators connected to them. As with the governors the increase in unit capacities reduced the number of units required and so the plant investment can be reduced by considerable savings on.the excitation equipment. It is true that as tha units increase in size the capacity of exciters connected to them.must also increase but the cost does not increase in proportion to the capacity and even reduces with an increase of speed. One of the major items of cost of a power plant is the building which houses the generators. Be- low the generator floor all of the space is utilized fer water passages and very little of the area can be con- sidered as useless or waste space. Above the generator floor however, questions of crane clearance and size of the equipment determines the size of the building neces- sary. In some of the larger plants for higher heads _‘- A are remove ‘: in; and {18' tith 51:10: in 1925 (P1 dispense wi sary will 1 building oi tendency tc dimension :3 erator diar Page No. 82 the building has been dispensed with and protection only provided over the generators. hese generator covers are removable and gantry cranes are provided for install- ing and maintaining the equipment. A plant of this type with siphon settings was installed at Otsego, hicaigan in 1925 (Plate No. 21). It is not always practical to dispense with the building for low-head hydro plants but any savings made possible by reduction of the space neces- sary will tend to reduce the plant investment. While the building of larger and higher speed turbines has had a I tendency to expand the size of the plant somewhat in its di ension from intake to discharge, the reduction in gen- erator diameters due to increased speed has reduced the width of power house necessary to house the generators. ‘ In the past some plants have been built to in- clude inside the building not only the generator room but also the rack structure and raking platform. This is not a common practice however. In the more modern plants the generating room.is designed to cover only the generators and auxilliaries. The switchboard, trans- formers, and the other equipment being placed in.a separ- ate room built either at one end or as a wing to the main plant. Crane clearances usually control the height of the power house although other considerations such as appearance or materials available may enter into the de- Sign. It is evident that generators of large diameter with large rotors will require wide power house super- I! _ «\J‘JI AI ni.JI.-V.\.. HJ¢‘1.\J‘ .H 1, fl WiJJ‘ ‘ <2AWJ-Hv \u.J< .u‘l‘J fig J‘N 22286 stated a $5 ...oeem cause»: sass: oweo .26 825$ Team 9:395 Juan— xooficom amino 2:. Page NO. 83 21 Plate No. 8258300 52. But, 638m 3323: u_.a80.:< 0380 mZOHH/VHM UZ~Hf the runner a suction action equivalent to at least a EHDrtion of the velocity head available in the discharged Water. This later purpose is accomplished in the draft 1lube by a gradual increase in the cross-sectianal area Page No. 88 and a consequent decrease in water velocity as the water passes from the top to the bottom of the tube. The early tubes used were short conical sections of doubtful value as velocity head regainers but serving very well as suction tubes to p rmit the use of the entire head.of water. Prasil, an experimenter of Switzerland, worked on this problem of the iraft tube and as a result of his investigations he developed a tube with a flare at the bottom in which the change in the velocity of the water was constant for each unit of length or in other words the veolocity variation was a straight line function. The form.of the curve for the sides of the tube was that of a hyperbole which is asympototic to the plane of the bottom.of the draft pit. For reasons of simplicity and cheapness however, the hyperbolic draft tube was not often used and the truncated cone made from cast iron or steel plate was generally furnished as the standard draft tube. Various relations of diameter to length of tube Shave been used. Experiments by Venturi showed that for (iiverging tubes a maximum.discharge was obtained when the Ilength was 9 times the smallest diameter and the diver- Eence or flare was equal to 5 degrees. The National Ifllectric Light Association has suggested that for the EHerose of testing model draft tubes a standard reference tube be used, the length of which is 5 times the diameter 31nd the flare 5 degrees. Some tests have been.made where IVeference tubes have been used and in these tests publish- Page Ko. 89 ed by the N. E. L. A. the straight conical tube gives the best results when compared to other forms of draft tubes. ‘ It can readily be seen that for the larger sized turbines such a draft tube would have considerable length. The logical reason for not using the long straight conical tube is found of course in the excessive excavation that would be required in vertical plants. To overcome this objection engineers a few years ago attempt- ed to make use of the conical tube by bending or turning it through a 90 degree angle and discharging the water downstream instead of against the bottom of the draft pit. The fallacy in this design lay in the fact that the water would not fallow around this bend in parallel elements. Where the bends were short the draft tubes were not very effective. Where longer sections, which continued to flare after the bend was made, were used a reasonable draft tube action resulted. Elbow draft tubes have been considerably improved and are used today with very satis- factory results when properly designed. In Hay 1921 Mr. W. M. White presented before a Ineeting of the American Society of'Mechanical Engineers 61 paper entitled "The Hydraucone Regainer, Its Develop- Iment and Applications in Hydro-Electric Plants". This Paper brought out considerable discussion as the re was Eit that time, and still exists, some conflict of opinion as to the relative merits of this and a similar type of Page Ho. 90 draft tube known as the Kocdy Spreading Tube or Whirl Re- gainer. Both of these tubes however, were developed to meet the conditions of the high specific speed turbine which discharges the water from the runner at a much -higher velocity than the lower specific speed turbines. These draft tubes were of especial interest to designers of low-head plants as the depth of draft pit below the runner was considerably reduced by these designs. Their use permitted the placing of the turbine runners well above tailwater level and when used with plants having siphon settings the necessary excavation was reduced by a considerable amount. An examination of the illustrations of some of the typical plants which are included in this thesis will show just how this saving has been effected in these low head plants. At the Edison Sault Electric plant (Plate No. 4), which.was the first plant to have a single runner direct connected unit, one of the elbow type draft tubes was used. The runner diameter was 71 inches and the lowest excavation was 20 feet below the centerline of the runner. The ratio of depthtb the diameter of the runner in this plant is nearly 5.4. At 'the Sterling plant (Plate No. 5) the lowest point of the (excavation is 25% feet below the centerline of the runner éand.the ratio of the depth to the diameter of the runner 1:3 2.5. Two plants built almost at the same time wifii Page No. 91 siphon settings and with units of about equal capacities are the plants at Dixon, Illinois on the Rock River and at Appleton, Wisconsin on the Lower Fox River. The former has a Koody Spreading type of draft tube (see Plate No. 14) while the later has a hydraucone type of draft tube (see Plate No. 22). The Dixon units are 125 inches in diameter and the Appleton units are 124 inches. The Operating heads are about the same. The units with Moody sprfading draft tubes have the lowest excavation 21 feet below the centerline of the units and the ratio of depth to the runner diameter is 2.02. The runners with hydraucone draft tubes are set 2 feet higher than the ones at Dixon and the lovest excavation is 21% feet below the centerline of the unit. The ratio of depth to the runner diameter is 2.08. Because of the lower setting of the units at the Dixon plant the excavations are about la-feet deeper than at Appleton. Looking at the above types of'drafts tubes from the construction angle any reduction in depth of excavation such as is possible with either of these tstpes of draft tubes is certainly a valuable considera- tion. Deep excavations are not onlyvery expensive but tine element of risk is always that much greater. The innproved elbow type tubes also effect savings in excava— tion and do not require as wide a draft tube discharge as <10 the concentric type of tubes. At Rockton, Illinois (Plate No. 20) the unit is about 110 inches in diameter -- plant at 22 O N e t a 1 P Appleton, Wienonsin. Samoa ll eflam widow - xbu ; “4 .- ~\ ,‘\ . ' . I ‘ f nh~ \. e . ‘ 1' .1 « . V 5‘ l n 1.! ... X , n.“ -.n. -..-m. » ~ .7 .. “...-— v r- t: r . l x- , . ‘ f .1 / M- as a VHS URHTED STATES 3P DRWMHDAL HYDRO ELANTS MEANS. S? Dt—iOTM F ./ SHQVV 1 N6. ELECTRlC up». I if" (1‘! ‘t»:" I I ‘ . x . I ‘ . > 4 . > . \ IN . '_ I 1 Y' “ _ " . ‘ ’ I, ‘ ’ A ' 1‘ ‘ ' a ” L, a} 4: ~‘ ‘ f / < ..." '\ K , ‘ — ‘I . . ( 1’ - ‘. .73.. - a . ’- lm. v 4" . , 4“ ‘ 4 \ , a? . ', 22.. . ‘.- ,‘l‘. "r \ . 4 1: 7., , '_ . 4‘ - . r.~ , 4, ‘ Li’s ‘ . - . _ \ . 5 p 7‘ _ 1."; / A r‘ (h ~ - , [ 1" 31-4“, _ ‘_ ~ .. A, IV ‘ V l .1 F; -.: _ _ 3:; “ . , .' F \| I" . h. F ‘ ‘ _. 4 . t V. - \ D \~ ‘I \y}. _ '5 ' . k, r". ' \,‘. 1 _ .'\ - ‘ \ ‘1 \ Y» I‘;' I ,' . 34‘) ,’ 5 ‘ x» ‘ - -~ - p I. ' “ 'K ‘ l ’ ‘ a / . . . . ‘a . _ - . s I r. :L , . ' .1 . .L4 1: _ . . u _ ', \‘ r _‘ o » p . > ' ‘ , ' - 4"} u" f . _ x '_ a u.’ "-Hn ~b , -{Iw‘ ‘11-. .._. . » ‘75! c“A"~:' v _, r . 1 . . w . _. ’ " *1. I 4‘1- - y‘ ‘ l , ‘ u. K1 25.-w . -- ; M '. ' 4“ l ,/ ~ ~ . Mg. . ’ ~ I :— (I A ..r :x‘k . .. ' :‘ t 4);: Vi: . {fl ’ — . ’ I l a - 351‘ r ..L‘ { z”! ‘ ‘ ‘ ‘ 7,.» IN . \j - ' ‘ . _- .4 . t—\ }— l \ 1“ 2 _.. ' ' K . ’ -' . {,1 y x 1: A v r. ,' x . . f , ‘- ' /< }- _ I \ _ 1" . ' ‘ (__. \‘ \\ ,' y.‘ _J “ ‘ .v‘ r" \ '7” a v ‘ I . X ’ --. fl ' >\ " ’~. I ‘1 x, “i. r’ \I .‘r '. ‘~‘ & a / t ' ""‘ — ' E «”1‘ / l _ 7“ ‘ - ”‘H ' - PL r3 1, . , ’ v. f. -- 1 -\ . / 1‘ ‘ ‘ p- X > (I _ . { . / '\ ,- ‘ , ‘4 ' x ‘\ I. 1 __ \:\ “L , ‘ _ _ {5" —;->\ . _ _ r ' \ , Q ~ “‘4 / ‘ - I I ' ~ / .... ' J-f’“ . ._, l ‘\ . <>_ , k z x. ‘ u/ \ ‘ '. _ - ~ ' M" 'I .1. I \k ‘\1 ’ ’ y -’ . t -\ z v \ / ... ..- _ '_ ‘5“ r \. a . . , .r m / Q ”r . I '1'" V» .4 ‘ 7 / \ A, “~. ; - '1 \_ ‘.L A: / ~; _ V f“ I | z' _ ‘ 1 ? Fl 3’ "' ‘“-'—-~ ‘4 ‘ x - . ($1 ~. 0 I K the t , ‘ l.,_l : ‘ ‘ " Y— / ‘7’ \‘ ..: {'__ ‘ _— ,4 ‘ ~ ~ \ - --v _'_ l a (1' ‘7 "f‘ ‘. 5‘ i ~a. Y.-- r I / J '0 7. \ I -. I ’I/ 3; \ > ‘ ‘ I ‘ ‘ J. >-_ n "I ..... '3 ‘ c‘ ’W) | > < \x !d* \‘ "' \. L.- —‘ " ‘ -~_/ \ _TFK. \.. V 1. r .1- x \ I ~ \\ 1 W. - T A ‘ . K a \ _ _ /' £1: ‘ 1“ ‘L \' §\ \ _ a *1 fl" \. , I" l \ '- >,, ‘ 1 # ,- _1 \\ . ' A. l .\ \ / 1 .- - ... P \\ .j R. J ‘ . , ‘r p} '. r”! f" ,- _.- - \ ff. [I t f" \ . p. -— —~ - — MA by 4...... p , f ,. I ; T. a». | . ' ' \ "" 4' I ~ ‘ .- \ ‘ A a l ‘O . n a} . ' ‘\ I " _-__ ‘ ' ‘l ‘ V‘ l ’ I U. up ' I j ’ \ . < ' . ‘ ’ “ I ‘ .... .. -—~O \ _. -\ . r 2 ; \ 0 1 I , . “I _ .. .. 1 .. v” \ ‘ 1' ‘ K \ b . \ f- > _ ‘ \I ' CL‘ 1‘" ."J \ C . ‘: \ \ I ‘7. l‘ ' _ A . > / \ ’ ‘~ » v f ‘ ’ ’ , - I \ r‘x . \ a. ‘ ".' .'\ " ' > a -. , \_ ‘ . , . VA I \‘- , 1.3. a . r J p) k ‘ . , k? ’ ‘ . / ‘*~. ‘ - ’ '. , «. ~_ . ~ ,q I \ J. I \ "./ J" .7 . A -. \ - 4 \ \ ~ ‘ ’ ‘ (é ‘ / ‘ x \ 4 . I . .) k K _ V I \ / ’V I 3 I , > . A ‘ \ .‘ \‘ . \\ ’f‘"? ‘. \ —_ ' . _ f" ‘ , . . I , . I / ‘ . P - F IV‘! E. A" ( (”a \\.’-.. ) ' 7- . \p p r ‘ J | f -. M _-__ - 1- “ ‘ '" ”W W” _._.-“ _ ";.':’:::::T:-~“-~ - ——~— 7— ~ _- - --.. _.-“; LO C A. T‘ l. C) N POWER SE.T"T\ NQS rla {3 e E: o 1. 2. 3. 4. 5. 6. 7. 8. 9. Page 1.00 AEIBHDIX TC. III BIBLIOGRAPHY Water Power Engine ring — Textbook Daniel I. Need - Sec. Edition 1915 Hydraulic Engineering - Textbook H. K. Barrows Hydro Electric Power Stations - Testbook Eric A. Lof Forty Years of Research into Water Resources. F.}L Newell Engineering News Record - Jan. 23, 1950. Greater Efficiency for Low Head Hydro Elants. Geo. n. Jessop and C. A. Powell Electrical World - Feb. 1951. Automatic Hydro Station at nockton, Illinois. 3. J. Kallevang Electric Light and Power - April 1931. Water Turbines of the Propeller Type. L. F. Harza. Electtic Light & Power - April 1931. Hydro-Electric Power and Achievement of the Past half Century. J. D. Galloway Engineering News Record - April 17, 1984. Efficient Hydro Operation. Forrest Haghar Electrical World - Nov. 2, 1929. 124 10. 11. 13. 14. 15. 16. 17. Jf‘. g). Page No. 125 Some Recent Tests of High-Zone Xian-Speed Eater Turbines. S. J. Zowski Engineer Record - Nov. 28 and Dec. 26, 1914. Hign—Speed hheels for Eow Head Plants. 1. Streif Electrical World - July 17, 1920. DevelOping the Low Head Plant. Iouis 5. Ayres Electrical world Dec. 25, 1920. Largest Seven Foot Head Hydro-Electric Plant. G. E. tharman and ng Holland Power - July 7, 19£5. Grist Kill Transformed to Profitable Hydro Plant. Lawrence H. Jacobson Power - Oct. 7, 1930. The Hydraulic Development of the Sterling Hydraulic Company. Daniel 5. Head Engineering Record - Dec. 16, 1905. Ilant at Rock Island Arsenal nook Island, Illinois. Hest Electric - Nov. 28, 1901. Low Head Hydro-Electric Developments in Kichigan. Engineering Record - Oct. 19, 1907. Economics or Hydro-Electric Developnents. A. H. Gibson engineer - Vol. 151 - April 8 and 15, 1921. 19. 20. 22. 25. 24. To. 12 9—) 0) e C“. Iater Power in Great Britain. In B.L; Kemit Electrical Times - London - Nov. 12, 1951. Typical Low Head Hydro-Electric Power Plants. Transactions Engineering Society University of Toronto, Canada - Kev. 20, 1908. The Hydraucone Regainer, ts Development and Applica- tion in Hydro-Electric Plants. Wm. H. White f a pa er grosented at the meeting;o:‘imerican NJ Q Society of Kecnanical Lngineors in Chicago, Illinois - Kay 22 to Ray 26, 1921. A New Type of Hydraulic - Turbine Runner. Forrest Kagler A paper presented at the innual.Meoting of The American Society of Kechanical Engineers - December 1919. Published in Vol. 42 of Mechanical Engineering. Changing Reqiirenents in Hydraulic Turbine Speed Regulation. Forrest Nagler A paper pres nted at the Annual Meeting of The snerican.Society of Mechanical Engineers - December 1929. Published in Vol. 52 of Mechanical Engineering. InCJeased Kilowatt Output of Adjustable - Blade PrOpeller Turbines. .6. R. Martin 25. 27. 28. t0 (.0 O CO. 51. 52. 1.3.58 KO. 127 A gage‘ :resented at the Annual heating of The American Society of Mechanical Jngineers - December 1929. Published in Vol. 52 of Mechanical Engineering. Water Wheel Types Combined to Overcome Vsriations in Head and Flow. Engineering News Record - Nov. 15, 1950. Kanufacturers Statements for 1951. Bulletin Published by the National Electric Light Association. Hydro Electric Handbook. Creager anl Justin - 1927 Edition. Control Crofips Simplify Operation. R. K. Stanley and B. D. Wood *1 Electrical Engineprimg ~ reb. 1951. Electric Power Survey, Great Lakes Division, National Electric Light Association - 1924. l-‘ir st Llejor Low-Read Power Plant in the West—Rock Isle nd. J. D. Shannon - Stone and Webster Engineering Copp. Engineering News Record - February 18, 1952. A Report a? the Hater Power of the Rock River at Sterling, Illinois. Daniel H. Head - 1904. Water Power Resources of Hisconsin. George P. Steinmetz - Water Power Engineer Nisconsin Railread Commission. Page Do. 128 Paper in the tenth Annual report of the 4 3 T n- ring Society of Wisconsin - 1928. H. 4 n5 1 \J 55. Swedish Plant Will Use Largest — Diameter Hydro Turbines above Read - Water Level. Power - September 8, 1951. 54. European Low-Head Hydro - Electric Develonments A. V. Karpov Power-—£gufi1 l, 1950. 3:12.311 1:0, IV Acknowledgements The author Wienes to acknowledge hi ness to the folloning men who have aided in t tion of this thesis by contributions oi’data, printed material as well as suggestions which helped in the preparation of the text. C. R. Martin, Engineer, Hydraulic D Allis helmers hanufacturing Co. Milwaukee, Wisconsin. iii. 3 1‘63 iff Fargo Engineering Co. Jackson, Kichigan. George E. Lewis Ayrs;s, Lewis, Norris, 8:. May Ann Arbor, Michigan. J. Robert Groff James leffel 8;. 00. Springfield, Ohio. G. E. Ackerman Holland, Ackerman, and Holland Chicago, Illinois. 0. V. Kruse I. P. Morris Division Page No. 129 s indebted- he prepara- plans, and apt. Baldwin ~ Southwerk Corporation Philadelphia, Pennsylvania. George A. Jessop S. Morgan Smith Company YbIk, Pennsylvania. M'TlTflTflUTQfljMiflflifllflfljflflflflflfflfimfl'ES