below, thus utilizing the structure that is already there and already necessary for the production of power. This eliminates the usual type of Tainter gate and overflow spillways and substitutes for it a very cheap embankment. It is the result of a rather long development through several years; in fact, it was pretty well developed and then almost abandoned because of the terrific energy that is delivered through so small an area. It was found, however, that there was no great difficulty in this if the tail water depth was enough to absorb the energy and that the erosion in concrete for heads up to 30 or 40 feet was not appreciable. Of course if the water is discharged through the conduits before the concrete is entirely set there is likely to be some erosion. This can be easily cured and prevented by the use of cast-iron liners for the conduits, which cost very little. These conduits can be controlled by any type of valve. Those that we already have installed are controlled by gate valves. Those that we contemplate using in the Alcona Development, which will have no other type of spillway than the conduit spillway, will be butterfly valves controlled by a simple mechanism in the turbine room, which is one of the advantages over the Tainter gate spillway, in that the entire operation of the plant is carried out in one room. The operator does not have to leave the generators while he attends the spillway. I might say that the operating mechanism for these valves is not by electric power but by hand. One man can open the valves of any one of these conduits in about three minutes by hand, and only six are required on one of our larger plants.

The conduit spillway avoids the freezing shut of spillway gates as well as other advantages, such as the ability to lower the pond below the elevation of the sill of the ordinary Tainter gate. The head increaser effect, if it is desirable, can be very greatly amplified, apparently, by rifling the conduit, experiments upon which we are now carrying out. We have not used this heretofore in any of our plants, because we are not very much concerned with the head increaser effect at our plants, I mean, in Michigan. The head increaser effect of the conduit spillway as it is now in use is about equal to the rise of the tail water that would be produced by the water spilled. In the first plant in which we used this spillway we also used the Tainter gate spillway as a precautionary measure only. It has been very seldom used since. In the experiments at this plant we found that the tail water would rise nearly two feet, about 1.8 feet for a certain amount of water spilled through the Tainter gate. We then closed the gate, spilled the same quantity through the conduit spillway and found no rise of water, thus avoiding this 1.8 feet of tail water rise and consequent loss of head.

The decrease in the cost varies greatly with the conditions, naturally dependent upon the water to be spilled, the kind of foundation material, the cost of concrete, the location of the plant and the head. Savings in construction cost are evident up to 41 per cent of the entire development. Sometimes the

savings are small, but the advantages in the design have justified its use. We have it in mind to use it on all of our future hydroelectric plants because of decreased first cost and simplified design.

In Michigan we have also worked out a scheme for preventing the loss of head due to the operation of our plants on less than a 24-hour basis. Our plants are rather peculiarly situated in that we have. a series of dams, one backing to the next one upstream on a river. The loss of power due to operating on a 12-hour schedule as against a 24-hour schedule is about 8 per cent, due to the drawing down of the head water and the rise of the tail water caused by abnormal discharge. This is prevented by building the dams somewhat higher than is necessary to just back to the dam upstream. In other words, the downstream dam raises the tail water of the next one upstream enough so that it will be higher than the maximum discharge of the river. In this way if the downstream pond is drawn down the head on the dam next upstream is increased a corresponding amount. We have a series of thirteen dams, and by raising each one of them four feet it is possible to subtract four feet of storage from each one (a total of 52 feet of head) with the loss of only four feet of head on the total thirteen dams. It is not quite evident at first, but if you lay it out on a profile it is very clear. This also has the advantage of impounding a large amount of water for storage without any loss of evaporation over what the evaporative loss would be with a dam that did not lap the next one above. These evaporative losses we have found in Michigan from experiments over a year are almost exactly equivalent to the rainfall on the same area, and so the storage reservoir at the upstream end of the river, due to its evaporation, subtracts very greatly from the total discharge that can be sent down the stream. There is one other advantage of this lapping of heads and that is that in case of trouble this storage is instantly available at every dam on the series without waiting a week or ten days for it to come down from some reservoir upstream.

We have carried out a large number of experiments and records to see if we could develop a relation between rainfall and the discharge of the rivers of Michigan. There are about 93 of these rivers, and we have been unable to establish any real relation between rainfall and the river discharge except over periods as long as a year, which is of little value in predicting the amount of water that we will obtain from a stream. The lag on the AuSable and Manistee rivers between the rainfall and discharge in the river for the water that falls any appreciable distance back from the river on the water shed is from six to seven months, and on some of the southern rivers of Michigan it is only about a week, so that it is very difficult to average any relation. This great lag is due to the absorption of the rainfall by the sand and its retention as a reservoir within the sand.

W. N. RYERSON: Those of us who remember

back to the time when as a concession to the water power people one or two members were tagged onto the membership of the Prime Movers' Committee, I am sure are glad to see we are coming into our own here as we are generally throughout the country. I am not here to criticise this report, which is, as the Chairman said, simply a beginning; but I thought it might be interesting to you to learn some of our experiences in Minnesota with penstocks. We have four penstocks about 4,000 feet long, the first 3,000 feet of which are redwood and the rest steel. The penstocks are seven feet in diameter and carry water to wheels of approximately 15,000 horsepower each, the head being 375 feet. The first three penstocks were built in 1905, the fourth was added in 1914. Recently we have found that the redwood, which we had thought from our friends in the West to be everlasting, is beginning to go to pieces very rapidly. Inspection showed that the ends of the staves were splitting and portions of the staves sticking down into the interior of the penstocks and materially adding to the friction loss. We thought that that might be due to the age of the pipes, but, unfortunately, the one put in in 1914 shows exactly the same effect, though to a lesser degree.

Another thing that troubles us a good deal is the deterioration of the steel. Within the last two years we have had holes corroded right through the center of the plates, and we have had all the steel doctors in the country pretty nearly out there to tell us what the trouble is and none of them agree. Our own conclusion is that the treatment of the steel, whatever it may be, in the form of paint or other protective coating, is of comparatively little value unless the steel, before it is put on, is absolutely clean. In other words, we are not satisfied now to treat any steel that is not sand blasted first.

Another thing that bothers us, both in the wood and steel penstocks, is the collection of slime on the inside. It grows very quickly and can be removed quite easily, but it has a material effect in increasing loss of head. I have been told during the last month that the vacuum treatment of Douglas fir in creosote will effectually prevent the formation of slime on the inside of penstocks, and it would be extremely interesting if any of you gentlemen have had experience with that and could say whether it is true or not.

In connection with the matter of testing it may interest you to know that we had one of our largest and latest units tested last winter, using the Gibson method by rise of pressure. We found it extremely easy to apply, and, so far as we could tell, the results were perfectly satisfactory.

EUGENE VINET: When we heard in Canada that there was going to be a Hydraulic Power Committee we felt that a long-wanted need had been fulfilled. We are particularly interested in hydraulic power. As the Chairman said a moment ago, 86 per cent of our energy is hydraulic power. We have, as most of you know, possibly ten or twelve

million hydraulic horsepower available, of which slightly over two million are developed so far, so you can quite conceive we shall have a few hydraulic problems to face in the next few years. There is no doubt that this Committee will do work which will be exceedingly valuable to us, and we shall certainly do all we can to co-operate in the work of this new committee, which has been in existence now for a year.

We have heard a good deal about various large installations during this Convention. Yesterday we heard about Chicago being a very large power center and also of the superpower scheme around New York and the New England States, and today of the power possibilities in California. We all know, of course, of the Niagara installations, which are the most famous in the world hydraulically; but I would like to call your attention to the fact that in Canada we also have other big hydraulic installations. The Montreal district, for instance, is very favorably located for hydraulic power. Practically all of the power is hydraulic. Not very far from Montreal there is the development at Cedars Rapids with 160,000 hp., and those of the St. Maurice valley, which are a part of the property of the Shawinigan Water and Power Company. Here we develop nearly 400,000 horsepower in two plants. This St. Maurice River is quite remarkable in many ways. A few years ago regulation was undertaken and a very large storage dam has been built some two hundred miles north. The minimum flow of the river, which was something like 6,000 second feet, has now been increased to more than 12,000 second feet. This storage is the second largest in existence; after the Gatun storage of the Panama Canal it is the next largest one. It is quite a remarkable storage. It impounds billions of cubic feet and means a great deal for all the industries in the St. Maurice valley. It was built under very unfavorable conditions on account of being about fifty miles from any railway, away up in the bush, and it was done on schedule time, which was quite a feat.

The problems which of course exist in hydraulic work vary a good deal with the climate. While some of the hydraulic power in the southern countries have certain problems to contend with, we in the north have particularly the cold weather to deal with. Ice troubles were a few years ago very serious with us, but with experience and as developments have taken place we have now mastered that situation to a considerable extent.

I don't wish to take any more time, Mr. Chairman, but I feel that this new Committee has a tremendous field and there is no doubt that the co-operation and experience of the engineers in the various parts of the United States and Canada dealing with these problems of the south and the conditions of the north is unquestionably going to fulfill a very great need indeed. need indeed. As a closing word I might mention. that Mr. Maclachlan here, of Toronto, has just stated to me that if anyone should be interested in visiting the present Chippewa development, which is

now taking place at Niagara Falls on the Canadian side, they will be welcome.

CHAIRMAN CHEEVER: We have just a few minutes left. I would like to have Mr. Caldwell say a few words, if he will. He has been long interested in water power.

O. B. CALDWELL: I would not attempt to discuss this report in detail in any way whatsoever, but I believe that Mr. Cheever and the Hydraulic Committee are to be especially commended for the excellent manner in which the report has been put up. The outline of future work is certainly going to be of great value to the industry.

I come, as the Chairman says, from the northwestern section of the country, and it so happens that in 1913, before the war, the concern that I am with had plenty of excess water power. We were in a position to get along for a number of years without additional developments, and as a matter of fact we would not have been able to make any had we wanted to because we did not have the money to do it with. The last few years we have been particularly interested up there in trying to ascertain in what way we could, if possible, improve the output and capacity of the plants which had been

installed and in existence for a number of years. We have done some very careful testing of the individual wheels with the idea of ascertaining what the conditions are after that period of operation, and I might say in that connection that we unearthed some very interesting data, and we are very much interested now in finding out if we would not be able by the putting in of more modern runners, or by the application of some of the new ideas to draft tubes and flow increasers, etc., to secure additional capacity and more efficiency from the same settings that we now have. The installation of more modern and increased capacity windings on some of the existing generators might also be involved. looking over the field we are quite well satisfied that there are many possibilities upon which we may well realize. I think possibly that there may be other situations where similar things can be done.

In

I have not anything to report in the way of results which have been achieved along those lines, but I hope in another year or so to be able to say something on that subject.

CHAIRMAN MOULTROP: The next item on the program is the report of the Prime Movers Committee. I now present Mr. N. A. Carle, Chairman of the Prime Movers Committee.

No important development in steam turbine design or construction is on record during the past year. However, from the experience gained in the operation of the latest designs of turbines various changes and adjustments have been indicated, and the efforts of the manufacturers have been largely directed toward improvements in a number of important details affecting the reliability and operating efficiency of their units.

These changes in detail of design and construction, the establishment of a higher grade of workmanship and finish, together with a better understanding on the part of the operators in reference to methods of operation and handling of the turbines, have, in general, resulted in more satisfactory service.

No larger units than previously reported have been constructed,-the maximum size of a singlecylinder units being still 45,000 kw. There is no appreciable change in the steam pressure and temperature adopted in this country, although it is felt that higher pressure and temperature are a possibility for future application. The results of operation of European plants where experiments investigating these factors are now being made will, no doubt, be productive of considerable information in the near future.

No departure from the standard speed of turbines, as reported last year, has been made. There is, however, indication of a tendency toward slightly higher speeds for larger sized units as

well as for the smaller sizes. This is particularly in evidence in European design where units of 15,000 and 20,000 kw., of both the reaction and impulse types, are being built for speeds of 2,400 r. p. m. No data as to the overall diameter or peripheral speed of these machines are available.

The study of wheel vibration has been carried forward and a definite knowledge of operating characteristics, under actual load conditions, has been made possible.

Further investigations of the fatigue of metals are being actively carried out, scientific methods of sampling and analysis during the process of manufacture of disc materials are now adopted, insuring a greater degree of accuracy and control, thus safeguarding against imperfections in material hitherto impossible to detect.

Non-corroding materials for the manufacture of turbine buckets are still being experimented with, and while some very interesting developments have been noted, the results to date have not been sufficiently conclusive to warrant any definite statement at this time.

Steel has replaced cast iron for all parts subjected to high temperatures, such as diaphragms and casing. Radial packing on impulse machines of the General Electric design has thus far proved satisfactory, and no cases of serious distortion of shaft due to rubbing have been reported. Also, the adoption of radial pin bushings on the above type of turbines, particularly on the high-pressure

initiate the work. A number of studies and experiments are now being carried out as the result of the agreements reached at the above-mentioned meeting.

The severity of the conditions imposed by steam turbine service and the complicated requirements to be met have made the solution of the problem difficult for the designer and manufacturer, as well as the oil refiner. While both seem now to have this matter well in hand and have produced a system of lubrication and a grade of lubricant apparently fulfilling the requirements and conditions of the problem in general, the results in practical operation and the numerous instances of trouble experienced by a great number of operating companies, indicate that considerable work remains to be done in order to secure the highest degree of reliability in service.

In the past, this Committee has pointed out the difficulties experienced with the lubricating system of the units and has called attention to important changes in the design and construction of such lubricating systems. The turbine manufacturers have, in general, followed the suggestions offered, with the result that operating conditions have been appreciably improved.

The tendency in turbine design indicates that, in the future, higher speeds will be adopted and we may expect that present unit bearing pressures will be increased rather than reduced. With the advent of new designs, therefore, higher bearing temperatures must be anticipated and met. To satisfy the operating requirements of the future, it will be necessary to use oils having the greatest thermal capacities to more effectively carry away the heat generated and maintain the maximum lubricating efficiency. To meet these conditions will probably require the use of lighter oils, a reduced range of temperature in the oil between inlet and outlet of bearings, and possibly greater cooling area of oil coolers.

The wide variation in the grades of lubricants offered by the oil refiners to the central station operator, disagreement of the lubricating experts regarding the quality or grade of oils-their blending and other properties-and, finally, the bargaining spirit of the purchasing agent, are all factors indicating the necessity for preparing a specification setting forth the chemical and physical characteristics of oils to be supplied for the lubrication of steam turbines. The preparation of such a specification will require the combined efforts of the turbine manufacturers, the oil refiners. and the turbine operators. To attain the best results and to prepare a specification which will insure successful operation, the close cooperation of each one of the parties interested is absolutely

recognized, the industry is still without standard oil specifications.

In view of the early developments relating to the question of lubricating oils and other petroleum products, the treatment of this problem is the more arduous, and your Committee in undertaking to deal with it is guided solely by the imperative necessities arising from turbine operation which call for the formulation of at least the most essential requirements to be fulfilled by satisfactory turbine lubricants. This initial and, naturally, incomplete specification will be brought up gradually to its full scope by the aid and joint action of the oil refiners, engineers, chemists and consumers. Subsequent amendments will also be made following the advancement of the art and science of oil refining and lubrication.

The scope of the problem is such that a review of the details of lubrication is necessary in order to properly treat this important question, and, as noted above, the discussion which follows has been prepared with a view of presenting in detail the several factors which must be recognized in considering the subject.

Friction

Friction is one of the most important factors to be dealt with in the problem of lubrication. If it is decided to adopt one standard grade of oil, it will be necessary to standardize bearing pressures, speeds and clearances since the proper viscosity depends upon these factors. Either too low or too high a viscosity will cause an excess of friction, and the viscosity must be chosen so as to give a sufficient factor of safety against seizure, in case of an unexpected increase in temperature of the bearings, without too high. a friction at normal temperatures. Due consideration should also be given to the work which the oil is to perform in other accessory mechanism, such as the governor, couplings, etc.

The cycle of operation, that is to say, the volume of the oil in service and the rapidity of its circulation, its cooling, its method of purification, its circulation and distribution, its contamination with air, steam, dust and water; are also extremely important factors which must receive careful consideration in order to make these conditions as nearly uniform as consistently possible with the different types and designs of machines.

Briefly reviewing the theory of lubrication, we may say that its purpose is to change dry or solid friction—which results from the actual contact of the moving surfaces-into fluid friction by the interposition of a film of lubricant between them, so that the moving part is practically floating and friction is reduced to a minimum.

With dry friction, the resistance varies according to the nature and condition of the surfaces. in contact, as well as with the "oiliness" of the lubricant, but it is always much greater than with fluid friction. With fluid friction the resistance depends only upon the viscosity, pres