subject rather than the title given by the author. I consider an explanation is due from Mr. Hobart for this inconsistency.

2. Since the object of the paper is to discuss electric operation of railroads, why does Mr. Hobart stop at the substation? Note the table of efficiencies he has presented for the two systems. On the three-phase side the chain stops at the substation. Is there no line loss on the d-c. distribution system? Wireless transmissions have made great progress of late, but I doubt their practical application in such a case as this. Again, why throw in the step-up and step-down transformers on the singlephase side? Just because they are on the three-phase? Or because it balances the table, thus making four items for each table? As a practical example, they are not used in the case of the single-phase electrification of the New Haven road. Why charge up 8 per cent on line losses for the single-phase side, when, for example, our commercial experience allows a loss less than half that amount as the total average loss between generators and locomotives.

3. In the history of engineering, can any one point to the method Mr. Hobart has used to show the relation between the efficiencies of two systems-starting at the power house and working toward the railroad? I am glad he stopped at the substation, for had he arrived at the driving wheels of the train, which by the way would have something to do with the schedule, his explanations would indeed have been impossible.

Mr. Hobart starts with 160,000,000 kw-hr. per annum as the outputs of the two stations. Now it is manifestly clear that unless the efficiencies of the two systems are identical from the generators to the driving wheels of the train, if one system provides the tractive effort necessary, the other will generate either too much or not enough power, depending on whether the efficiency of the second system is respectively greater or less than the first system. A graphical example of this is as follows:

In the above figure, A in both systems 1 and 2 represents the output of the station. B represents the amount of power that must be delivered to the drivers of the trains, and thus system 1 provides the exact amount needed. System 2, having higher efficiency than system 1, not only provides enough power C = B for the railroad, but has left over an amount D (not needed). I believe this simple diagram will make clear the error Mr. Hobart has made in starting toward the driving wheels from the power house, rather than vice-versa.

Let us reverse the direction now and work from the tractive

effort requirements to the generating station, still adhering to our simple diagram:

B = C again represents the power requirements of the railroad. Both systems supply the exact amount required, but now note the difference in output between the two stations; the A of system 2 no longer equals the A of system 1.

If, to produce an equal tractive effort, a system employing a chain of higher transmission efficiencies requires less output than another, is it fair to compare transmission efficiencies starting with equal outputs from each station? I think it is only necessary for me to point out this erroneous assumption to make clear the fact that the deductions must be equally

erroneous.

Before leaving the question of transmission efficiencies, let us look at a table that has associated with it the facts of actual practise, and this time, following the plan as outlined in Fig. 2, we will trace the power requirements in the two systems from the driving wheels to the steam turbines, starting with-say 100,000,000 kw-hr. as the necessary driving-wheel power.

or 16 per cent more power for the three-phase d-c. system.

Thus it is apparent that for the same tractive effort developed in the driving wheels of the trains the number of kilowatt-hours to be generated in the single-phase power station is 86.6 per cent of the three-phase station, or, stated in another way, the

three-phase station will have to generate 16 per cent more power than the single-phase station to accomplish the same result at the driving wheels of the train equipment.

Having now drawn attention to the attempt to disqualify the single-phase system by this fallacious treatment of the transmission question, I wish to point out a few isolated errors in assumption and design that Mr. Hobart has made.

Let us take a look first at Mr. Hobart's 8000-kw. machine. Now just why did he select this size? It is a rather odd one. I do not recall ever having seen this size of generator in print before, and certainly not on the floor of a power station.

By a selection of this size Mr. Hobart has admittedly and erroneously assumed it will carry him out of the environment of 1500 rev. per min. for the single-phase generator, as after he has allowed for the larger capacity made necessary by the star winding using only two legs and increased the copper for his 75 per cent power factor conditions, he has built the unit up until it represents in three-phase capacity a generator of 16,000 kw. Of course, had Mr. Hobart allowed his generators to retain the modest capacity of 5000 kw., then his double generator single-phase unit would not have been safe.

Notwithstanding, however, Mr. Hobart has calculated without his host, for there are other manufacturers who have built 1500-rev. per min. generators, whose three-phase capacity has been 14,000 kw., and have offered tenders on machines of 20,000 kw. capacity. Thus Mr. Hobart's double generator proposition fails. Two hundred and forty tons is indeed a weighty generator to charge up for a 16,000-kw. machine, when one of 14,000 kw. has already actually been built which weighs but 95 tons.

Before leaving the generator end of things, I am forced to draw Mr. Hobart's attention to the fact that his statements concerning steam economy at high speeds is at variance with the statements his company has advanced with reference to economies at low speeds. We will grant that two generators, each of half the capacity of one, weigh more than the one generator (and we reserve the privilege of using one), but his statements as to steam economy for the type of turbine used are contradicted by other advocates of the same type of turbine. However, it is all summed up when it can be stated as a fact that the manufacturers who have been in the business of real singlephase machinery can build one generator in the form of a single unit, large enough and with the same speed, to match up with Mr. Hobart's single unit three-phase machine. Let us not forget, however, that Mr. Hobart will owe us either a refutation or explanation of the low economy he applies to high economy, four-pole, 750-rev. per min. Curtis turbines. Throughout his paper in text and foot-notes we are reminded constantly that he is being liberal with the single-phase. A too great liberality might do a great injustice to the three-phase. Why not be just? Say what it is, not less or more than it is. I might say Mr. Hobart

showed, for instance, great liberality in choosing that 8000-kw. generator, as 240 tons of single-phase generator was a far too liberal supply.

Now having shown the error in proposing the double generators, let us have a look at some of Mr. Hobart's thoughts on generator design. It is evident that Mr. Hobart has had little to do with dampers that is, field dampers. He tells us the loss in the dampers is just equal to the armature loss, and backs it up by quoting 13 kw. for the field damper of the single-phase generator and 13 kw. for the armature loss. Evidently Mr. Hobart has assumed that the damper has to neutralize the complete magnetic flux due to the armature turns. If it did, even then its loss would not be what he has stated. He evidently does not appreciate that the armature field is made up of two components, each revolving in opposite directions and of equal value. One of them is in synchronism with the rotor and is therefore not acted upon by the damper. The other component revolving in the opposite direction to the rotor flux is compensated for by the damper. Here, therefore, Mr. Hobart's figures are reduced to one half. Now let us go a step further: On account of the long armature connectors necessary to a two-pole winding, and the short connectors inherent to the squirrel cage wound damper, the proportion again is cut in half due to the resistance of the damper being in that proportion to the armature, and so Mr. Hobart's figures sink to 25 per cent of their value.

Still in the domain of machine design, let us discuss for a moment Mr. Hobart's views on regulation. He says “Incidentally the three-phase unity power factor installations will have some 6 to 8 per cent inherent regulation, whereas the inherent regulation of the single-phase 75 per cent power factor installation will be of the order of 15 per cent or worse, and will be thus so inferior as to require that some type of automatic regulators be provided." Can it be possible at this late stage of generator design Mr. Hobart is not awake to the fact that an inherently poor regulation in a machine, which compared kilowatt for kilowatt with another of equal weight and better regulation, gives the best account of itself when measured on the scales of efficiency. 6 per cent machine has no regulator, less kilowatt capacity and poor regulation. The 15 per cent machine has a regulator, more kilowatt capacity and perfect regulation. Thus the regulator pays for itself many times over. Besides this its inherent regulation is in marked economical contrast to the revolving mass of substation apparatus which the author recommends to buffet the short circuits of the line.

The

In conclusion, let us compare Mr. Hobart's hypothetical single-phase system with the actual single-phase system in operation on the New Haven road. He gives the single phase aggregate annual efficiency as 81 per cent thus involving a loss of 19 per cent. The aggregate loss between the steam turbines and the electric locomotives on the largest single-phase road in operation

Mr.

is 7 per cent; thus giving an efficiency of 93 per cent. Hobart's assumed losses are, therefore, 2 times as great as the actual losses. Hence, there is nothing left but a choice between the theory propounded by the author in his paper and practise as we find it; which?

Edgar Knowlton (by letter): In the first part of Mr. Hobart's paper is a comparison of the weights of the three-phase 8000-kw. generator at 1500 rev. per min. and a single-phase 8000-kw. generator at 750 rev. per min. I believe that the weights of these two generators will be more nearly represented by 100 and 200 tons since this is the ratio of the three-phase ratings of the two machines and the reduction in speed will not greatly increase the weight of the single-phase machine.

A single-phase 8000-kw. 1500-rev. per min. generator is a practicable machine but, as explained above, its weight, cost, and efficiency, would vary but little from that of the same capacity machine at 750 rev. per min.

For the losses and efficiencies given a little further on I would be inclined to substitute the following:

THREE-PHASE 8000-KW., 100 PER CENT POWER FACTOR., 1500

THREE-PHASE 6000-KW., 80 PER CENT POWER FACTOR

SINGLE-PHASE 4000-KW., 75 PER CENT POWER FACTOR, 1500 REV.

A comparison of the estimated losses is given in the following

table: