additional evidence of the great scope of their work and contains specimans of the greatest value. Mr. Ellwood Hendrick gave a humorous exhibit on electron furnaces in which he transformed a mixture of saw dust and like miscellaneous materials into fresh hot roasted peanuts. The remainder of the evening was devoted to fun, refreshments and song. The new buildings of the Massachusetts Institute of Technology are marvelously compact. The right wing of the inner court, over which the illustrious names of Pasteur and Lavoisier are carved, is devoted to the chemical departments. The laboratories are, it is needless to say, the best that could be built. It comparing the chemical engineering equipment with that of some of the other engineering laboratories, the casual visitor might think that there was room for considerable improvement. The commercial size hydraulic, steam and electric power plants make the small Swenson evaporators look small indeed and they are the giants of the industrial chemical laboratory. However, Prof. Walker has overcome these difficulties and set all the President A. D. Little introduced an amendment to the method of selecting candidates for membership whereby the admission committee could place those candidates up for election who do not quite meet all the specifications but have real meritorious qualifications. It is hoped to make the Institute the real representative body of the American chemical technologists and engineers. Dr. David Wesson invited the society to be his guest at Savannah, Ga., at its winter meeting. Great cotton oil and naval store plants are located there, and not least of all, Savannah is the most typical of all southern towns. PAPERS PRESENTED The major part of the program was assigned to the symposium on electric furnaces. Extracts from three of these papers are included below and selections from the others will be given in a future issue. The Institute semi-annual report will be published shortly giving the texts in full. The Growth and Development of the Manufacturing Plant of the Providence Gas Company DIVISION OF CHEMISTRY, HARVARD UNIVERSITY other engineering divisions a new example. Technology instructors are now placed in several of the chemical plants around Boston and the students are given more than the birdeye view of the manufacturing business. The Wolcott Gibbs Memorial Laboratory' is the forerunner of what might be called the Harvard College of Chemistry. A view of the group of buildings as now planned by the architect, Mr. A. W. Longfellow, is given. Dr. Morris Loeb, Mr. James Loeb, T. Jefferson Coolidge and several members of the alumni are the donators of of the first two buildings. These buildings will be instrumental in the development of pure and applied chemistry for generations to come and it is certainly a rare privilege that wealth has bestowed upon the founders which will enable them to contribute so much to the future welfare of humanity. Professor T. W. Richards personally welcomed the Institute. Though he was able to show the prize points of the laboratory such as 6-ft. drawers and other extraordinary facilities for research, the guests knew that their real opportunity was in visiting America's greatest chemist in the building entirely devoted to research under his guidance. For full description see Harvard Alumni Bulletin, Vol. XV, No. 26, P. 424-429. by Mr. W. H. Russell will be given in three parts. Dr. Chas. L. Palmer gave a splendid review of the gaps left in the chemical study of the wool grease. The most amusing debate was precipitated by Dr. Edward Gudeman's argument that chemists should be licensed. Dr. James R. Withrow, who was scheduled to support the issue, sidestepped the matter and Drs. A. W. Smith and David Wesson went at the argument so vehemently that the old western proverb which said, "Never stop to argue with a Smith & Wesson" certainly proved to be as wise a maxim for oratorical combats as those fought with lead. Symposium on Electric Furnaces The program for the symposium was arranged by Mr. C. T. Bragg of the Michigan Smelting & Refining Co. The first paper was read by H. W. Gillett, of which the following are extracts: UTILIZATION OF ELECTRIC BRASS FURNACES Four years ago there was no electric brass furnace in commercial operation. Today some forty firms are using or now installing approximately 100 electric brass furnaces. One rolling mill has about 30, though most of these are small 300-lb. furnaces. One smelting and refining company is using four one-ton and four threequarter-ton furnaces. Another firm has four one-ton furnaces and the various offices of the U. S. Mint will soon have five furnaces. Batteries of two or three furnaces are quite common. Five types of furnaces have advanced far enough so that the makers can cite truly commercial performances. Various other types are in the experimental or semicommercial stage but are not yet on the market, and a few are on the market with some at least of the experimental troubles left for the purchaser. While recogniz ing that there is yet no perfect electric brass furnace, and that future development may produce a type superior to those now in commercial use, prospective users may properly pay first attention only to the types which have been commercially proven in users' hands. These types are (1) Direct arc type, with the Snyder as the only make as yet really used on copper alloys out of the dozen of that type that are used for steel, (2) the vertical ring induction furnace type (Ajax Wyatt), (3) the granular resistor, reflected heat type Baily), (4) the stationary indirect arc type (Rennerfelt), and (5) the indirect ore type with stirring of the melt (rocking furnace). Clamer,' Baily,' vom Baur,' Miller, the writer,' and especially St. John' have previously discussed the possibilities and limitations of these various types and a mere summary is sufficient here. The Ajax-Wyatt is the most efficient in use of power, uses up no electrodes, gives thorough mixing of the charge, perfect temperature control, and has the steadiest electrical load. It must be "primed" with previously melted metal after a shut-down and hence is not well fitted for 10-hour operation, though it can be used on 10-hour operation by keeping some current on and holding some metal molten overnight. The need for "priming" makes it difficult to change from one alloy to another. It has been mainly used on yellow brass, though it is applicable to red. Its greatest drawback is that so far no refractory lining has been found that will satisfactorily withstand the action of alloys containing over 3 per cent lead. It is best fitted for 24hour operation on the same alloy. It could be mechanically charged. The Baily is the least efficient in use of power, does not mix the charge, but has a steady electrical load 'Clamer, G. H., Melting Brass in the Induction Furnace, Journ. Am. Inst. Metals, Vol. 11, 1917, p. 381. Baily, T. F., Resistance Type Furnace for Melting Brass, Trans. Am. Electrochem. Soc., Vol. 32, 1917, p. 155. Vom Baur, C. H., The Rennerfelt Electric Arc Furnace, Trans. Am. Electrochem. Soc., Vol. 29, 1916, p. 497. 'Miller, D. D., The Electric Furnace as a Medium for Heating Non-Ferrous Metals, Journ. Am. Inst. Metals, Vol. 11, 1917, p. 257; Met. & Chem. Eng., Vol. XVII, No. 9, p. 537. Gillett, H. W., and Rhoads, A. E., A Rocking Electric Brass Furnace, Journ. Ind. Eng. Chem., Vol. 10, 1918, p. 459; Met. & Chem. Eng., Vol. 18, No. 11, June 1, 1918. Melting Brass in a Rocking Electric Furnace, Bull. 171, U. S. Bur. Mines, 1918. St. John, H. M., The Present Status of Electric Brass Melting, Chem. & Met. Eng., Vol. 19, 1918, p. 321. and uses up no electrodes. This type cannot be powered as high as others without grave danger to the résistor troughs, and is hence at a disadvantage as to efficient use of power, especially on alloys of high melting point. It is best fitted for 24-hour operation. When used for 10-hour operation it is usually necessary to put power on part of the night to keep the empty furnace hot. It is not readily charged mechanically. It has poor temperature control, due to its heat storage and consequent sluggishness. It can be used on any alloy but is better fitted for yellow brass than for the higher melting alloys. Its greatest advantage is its simplicity of operation. The Snyder or any other direct arc furnace is applicable only to true bronzes or other alloys low in zinc, 5 per cent zinc being the probable limit. It is entirely impractical for use on alloys high in zinc. The only installation used for copper alloys is used on leaded bearing bronze, the new lead being added in the ladle. Much lead is volatilized from the scrap in the charge, and much fume results, giving bad working conditions. The installation of Snyder furnaces has a very poor power factor, though this is not necessary in a direct arc furnace. The greatest advantage of the Snyder furnace in its limited field is its adaptability to mechanical charging, which aids in securing large output and hence in reasonable power consumption. One drawback is the singlephase arc load. The Rennerfelt or any other stationary indirect arc furnace is most applicable to alloys low in zinc. It has been used on alloys up to 22 per cent zinc with fair results, but as the zinc increases above 10 per cent the metal losses increase, and 10 per cent zinc is probably the economical limit of its application. It has met with decided success in melting cupro-nickel, bronze and silver at the Mint. It is not readily charged by mechanical means. The power consumption is fairly low. The rocking type of indirect arc furnace is applicable to alloys of any zinc content, gives a low power consumption, is apparently efficient in small as well as large sizes, can readily be mechanically charged, has good temperature control and mixes the charge thor oughly. Its drawbacks are the single-phase arc load and the possibility of electrode breakage in rocking too early if unskillfully operated. The performance of the various furnaces properly operated may be expected to lie in the ranges set forth in Table I. The figures in Table I do not in all cases agree with the catalog claims of the makers of the furnaces, as they are based on data obtained both from makers and users. The output and power consumption depend not only on the analysis of the charge, but also on the condition of it, i.e., whether all ingot, heavy scrap, light scrap, borings, or mixtures, as well as on the way the furnace is operated, just as gasoline consumption varies with the roads and the way the car is run. It will be noted that there is a wide variation in power consumption among the various types of furnaces, and that for the same type, the efficiency increases with the size. If the types of furnaces unfit for use on alloys high in volatile metals are confined to use on alloys free from or sufficiently low in such metals, and if the furnaces are kept tightly closed after the charge is melted, all types of electric furnaces will give very low metal losses indeed. No fuel-fired furnaces can compete with properly chosen electric furnaces on this score. Of course, the lost metal is mainly zinc and lead, so that the value of the loss in fuel-fired furnaces is not as great as it would at first sight appear to be when the percentage loss is considered. This loss, however, usually amounts to about as much as the fuel or labor cost, and, even with cheap zinc and lead, is worth eliminating. In making a choice of furnaces the user must first 9 F. M. DE BEERS 10 HENRY HOWARD 11 PAUL R. CROLL 12 B. R. TUNISON 41 MRS. B. R. TIMISON 44 JOHN C. OLSEN, Secretary 13 W. C. BAINBRIDGE 14 GEO. C. PUCKHABER 15 WALTER B. MURPHY 16 A. H. KRAUSE 45 MRS. JAMES MUNN 46 MRS. WALLACE SAVAGE 47 MRS. G. A. PROCHAZKA eliminate those types which will not operate satisfactorily on alloys he must melt in them, for example, the rolling mills making yellow brass would thus eliminate the direct arc and un-rocked indirect arc furnaces, and the makers of alloys high in lead and the people who must change from one alloy to another often would eliminate the Ajax-Wyatt. The user should then select from the remainder the type and size of furnace which will best fit his particular work. If he has very cheap power available, the granular resistor type may be chosen over a more efficient arc furnace because of the avoidance of electrode consumption. Where power is high, the balance would be against the granular resistor type. One rolling mill may prefer to continue to pour very small billets and may find the Ajax-Wyatt exactly suited to its needs. Another may wish to operate on a larger scale and may find a one-ton rocking furnace more suitable. A maker of bronze who must take his power from a very small power plant may find that a single-phase arc load would be unacceptable to the central station and hence might choose the two-phase Rennerfelt instead of the single-phase. rocking furnace. The size of furnace chosen will depend on the output desired and on whether 10- or 24-hour operation is called for. While the largest size furnace that can be kept busy should be used, it is poor economy to purchase a large furnace and then operate it with charges much under its capacity or to allow it to lie idle a good share of the time. An electric furnace is expensive in first cost. The one or two furnaces required to melt 5 tons per day on ten-hour operation will cost from $10,000 to $20,000 completely installed. For purposes of calculation assume the cost to be $15,000. At 300 days per year the output is 1500 tons. Taking interest at 6 per cent, the interest is $900 per year or $3.00 per working day, this amount being lost every day the furnace is idle. At full output the interest charge is 60c. per ton. On the other hand, if one furnace costing $10,000 is operated for 300 days per year 24 hours a day with an output of 16 tons, the total output is 4800 tons, and daily interest charge $2.00 per day, or 12 c. per ton. On account of the initial cost and the interest charge, one should in general not have electric furnaces standing idle to handle small peaks of production in excess of the normal, but should utilize for this fuel-fired furnaces of lower initial cost. On the other hand, the greater the number of electric furnaces which can be kept busy the lower the cost of power per ton. BASIS OF POWER RATES Industrial power contracts usually have two factors, the demand charge and the energy charge. The first pays the power company for the equipment it must maintain to supply the maximum power needed, while the second depends on the total amount of power used. Suppose we have a maximum demand of 300 kw. and that the average power consumption per ton of metal is 335 kw.-hr. per ton on 9-hour operation, 275 on 18hour, and 250 on 24-hour. The total power used per day is then about 2000, 3565, or 5250 for the three cases. In a 25-day month this makes 50,000, 91,000 and 130,000 kw.-hr. per month. Assume that the plant, before it installed its electric 25 P. E. KRAUSE 26 W. S. PUTNAM 27 C. D. CARPENTER 28 J. M. WEISS 56 P. S. BARNES 57 CHESTER G. GILBERT 58 R. L. DAVIDSON 59 H. S. DAVIDSON 29 H. L. SHERMAN 30 JAMES MUNN 31 A. M. BRECKLER 32 A. L. GARDNER 60 ROBERT P. CALVERT 61 JAMES BROWN 62 STEPHEN L. TYLER 63 GEORGE P. HAHLWEG furnace equipment, had a maximum demand in lights and motors, of 200 kw., and used 20,000 kw.-hr. per month for those purposes. Taking a concrete case where the power contract calls for a demand charge of $1.80 per kw. per month for the first 50 kw. and $1.00 per kw. per month for all over 50 kw., and the energy charge schedule is 2500 kw.-hr. per month at 2.Oc. per kw.-hr. for the next 35,000 kw.-hr. per month at 0.8c. per kw.-hr. for the next 310,000 kw.-hr. per month at 0.5c. per kw.-hr. all over 347,500 kw.-hr. per month at 0.4c. per kw.-hr. The 200 kw.-20,000 kw.-hr. lighting and motor power cost 50 X $1.50 $75.00 2,500 X 2.Oc. $50.00 150 X 1.00 = 150.00 17,500 X 0.8c. = 140.00 = $225.00 demand = $190.00 energy charge Total $415.00 or 2.07 c. per kw.-hr. used. The plant now has 500 kw.-hr. maximum demand, and 70,000, 111,000, 150,000 kw.-hr. used per month on the three assumptions. 50 X $1.50 $75.00 450 X 1.00 450.00 = = This table shows the advantage of continuous operation of electric furnaces, as well as that of large installations. The exact figures will vary in each particular case, but the ratios will remain approximately the same. The elimination of delays will give a 25 per cent increase in output in an hour's less time, and decreases the cost of power per ton by about 40c. The cost of power is in most cases the largest single item in melting costs in electric-furnace practice. Every effort must then be put forth to keep this cost down. The way to do this is to keep the furnace at I work at its job, which is melting metal. This is important on 24-hour operation, but even more so on 8to 10-hour operation, for an extra heat per day on single-shift operation means that it is obtained at the end of the day when the furnace has recovered from cooling off through the night. Any furnace will illustrate this fact. Take an actual day's run of a small Rennerfelt on red brass for example. Idle Time Charging Kw.-Hr. Charge, Total Arc and Kw. per Heat Lb. Time On Pouring Hr. 100 Lb. 123456 525 2:10 With the exception of the granular resistor type in which it is difficult to hold metal after it is once ready to pour because the response to changes in power input is so slow, it is possible to hold metal ready to pour as long as one likes in any of the electric furnaces. This is a convenience in an emergency, but it is very poor practice from the point of view of furnace efficiency. In addition to using efficient discharging and charging devices, another way that power can be saved in electric melting is to operate on a definite schedule of power used. The Ajax-Wyatt takes power at so steady a rate that a time schedule works just as well on that type, but the granular resistor furnace and the arc furnaces may vary considerably in rate of power input. After a few test runs, one can, for any particular alloy, particular proportion of ingot, scrap and borings, and for any particular weight of charge, make out a perfectly definite schedule of kw.-hr. needed on each heat. In 10hour operation the kw.-hr. per heat will be higher in the morning and approach or reach a lower constant figure at the end of the day. By adhering to such a schedule and running on kw.-hr. input instead of on a time schedule at a supposedly constant but actually variable kw. input, the heat can be brought out with astonishing regularity as to temperature. Each furnace should have its individual kw.-hr. meter, and one readable to certainly not less than 5 kw.-hr., and better 1 kw.-hr. With individual kw.-hr. meters the performance of each furnace in a battery, and of each furnace tender can be watched. Inasmuch as the heat losses through the walls and the electrodes are approximately constant even though the rate of power input may change, it is obvious that the higher the rate of power input, the more of it is usefully employed in melting. Suppose & furnace takes 100 kw. and loses 35 kw. through shell radiation and electrode losses, 65 kw.-hr. then do useful work. If the same furnace is given 125 kw. it may lose 37 kw. in shell and electrode losses, but 87 kw. do useful work. The furnace will do one-third more work in a given time at the higher rate. The upper limit of rate of input is that at which the local temperature is so high that refractories fail or that local overheating of the metal and consequent loss of volatile metals ocur. It is quite probable that on the larger furnaces automatic control of power input would be desirable. |