wheel, or rod, and at the top end of the live brake-beam lever. To find the power at the brake-beam, multiply the power at the top end of its live lever (at D) by the distance from D to F, and divide by the distance from F to E. This would mean, multiply 4,180 pounds by 321⁄2 inches and divide the product by 62, but we know that the long end of the lever is to the short end as 4 is to 1the actual measurements are reduced proportionately and the result is the same if we add 4 to 1, call the whole length 5, and dividing by 1 being unnecessary, as it will not affect the result, we simply multiply the power of the top end of lever by the proportion of brake-beam lever. 4,180 multiplied by 5 equals 20,900. 20,900 pounds is the force applied at the center of each brake-beam and as there are four beams, we multiply 20,000 by 4, equaling 83,600, which we figure as the total brake power. To go to the end, we may divide the power of the brake-beam by 2, and have the force of each shoe against the wheel, multiplying by 8 to find the total brake power.

We were required to supply a total brake power of 82,800 pounds; our calculations prove that we did it as nearly exact as necessary, 83,600 being only 800 pounds heavy. The figures would have balanced even, if we had not eliminated fractions in the measurements of the cylinder levers; the required lengths were 16.57 inches; and 14.93 inches, and think how impractical to send levers to be drilled to such fine measurements; so we

called them 16% and 15, and the difference is no more than the weight of three or four heavy people in the car.

The power at the outer end of the dead cylinder lever is the same as at the outer end of the live one; to figure it out, we must find the power of the cylinder rod that supplies its force. Multiply the force at A by the distance from A to C; divide the product by the distance from B to C, and the result is the power at the central connection of the two cylinder levers-17,563 pounds. Now to get the power of the dead lever; multiply the force at S (17,563) by the distance from S to R, and divide the product by whole length of lever-P to R. This will give 8,360, same as the other lever. The rest of the rigging at that end is a duplicate of the other end and not necessary to go over.

We figure the power of the dead brakebeam levers the same way as for the live one turned upside down. Power is applied to the bottom end of the dead lever; to find that power, get the power of the lower end of the live lever by multiplying the long end, which we will call 4, by the power applied to it (4,180) which equals 16,720 pounds; then divide by the short end which we call 1, not affecting the result, and we have the force at lower end of both brake-beam levers at 16,720 pounds. Multiply it by whole length of dead lever, and divide by the upper end and the result will be the power at the brake-beam-20,900 pounds-same as the other beams.

able circumstances, such as a break-intwo, result in their being pulled apart.

Being an engineman, or so it is assumed, you have not had the opportunity of seeing one not very unusual damage that follows where the practice you speak of is followed, but which the switchman knows of very well; this is, hose being torn in two, and, in some instances, the angle cock broken off from the pipe.

But what is in the aggregate probably the most serious damage is one that even the men following the practice do not see, the weakening of the hose that leads to its early removal by reason of sponginess or bursting. The following bearing

on this is quoted from the proceedings of the Northwest Railway Club:

"Leakage at hose coupling packing ring does not necessarily imply other than a faultily made connection, this being likely to result in permanent leakage until the packing ring thus distorted is replaced. However, the dummy coupling recently put on the market by the Westinghouse Air Brake Co. will, in addition to making other improvements, surely effect a betterment in this particular on account of keeping the ring pressed back into position.

"Again, rings are distorted by bent couplings, the lack of proper clearance rendering it almost impossible to connect the latter properly, this having been explained in a previous paper before the club.

"While pulling couplings apart is always liable to induce leakage in them and hose, the bent coupling almost insures these results; being, also, a common cause for torn off hose.

"The following is a record of pulling apart tests made at the St. P. & D. R. R. shops. Two coupled hose, with pressure connection, were attached to an air hoist, the lower hose having a weight connected. This weight was, in each test, gradually increased until sufficient to pull couplings apart.

With new packing rings and coup

lings rotating as pulled...... 80 lbs. With new packing rings and couplings not rotating as pulled....105 lbs. With worn packing rings and couplings rotating as pulled.. 27 lbs. With air pressure of 70 lbs. in hose: New rings (coupling would not rotate) Worn rings (coupling would not

...

rotate).....

.....267 lbs.

..230 lbs.

Worn rings and one coupling bent down 1-32 inch.. ..356 lbs. Worn rings and one coupling bent down 1-16 inch stuck with....588 lbs. "The latter weight which failed to separate couplings though assisted by blows from a hammer, was the limit of the air hoist.

"While not claiming these tests are absolutely accurate, yet they are sufficiently so to leave no doubt on two points. First, that every time hose . couplings under pressure are pulled apart the resultant strain weakens the hose, for the reason that the necessary angle of hose nipple throws the entire strain on but about 1-3 of the hose section, or

possibly less if angle cock stands vertical. The second deduction is of even greater importance, being that a loss of 1-32 inch of clearance between coupling faces, even though the error is at but one end, increases the separating strain over 50 per cent, while 1-16 inch raises it somewhere above 155 per cent.

"The few roads that gauge all remounted couplings to insure their being standard are finding a large number of bent ones, 1-16 inch not being unusual. Let those interested furnish their inspectors with a small piece of metal 3-32 inch thick, 1-64 inch less than the proper clearance, and have reported the number of instances where with air braked trains it could not be entered between the connected couplings."

It may be of interest to know that there is in use a hose coupling which is made to be pulled apart, that it can be coupled and uncoupled in no other way than automatically when the cars are brought together and separated and that the air signal and steam hose as well as the air brake are coupled and uncoupled at the same time. The only duty of the trainmen is to open and close the several cocks at the proper time. This device, which has been in use on both passenger and freight cars for a considerable time, is fully illustrated and explained by the article commencing on page 704 of the November number of this MAGAZINE.

Driver Brake Slides Wheels. 72. "The road I am on got some new Baldwin four-cylinder compounds about a year ago and they had not been in service long before a couple of them had flat drivers from sliding, and in one case at least, I know the engine was not reversed and they say it was not in the other. Since then they keep long piston travel on the driver brakes on these engines so as to prevent them from sliding.

"These engines have the largest brake cylinders I have ever seen used and I am wondering if they do not give too much power."—W.

Answer.-Cases similar to this, possibly this is one of them, have been carefully investigated. At first it was presumed that by some mistake the brake force was too high for the weight on the drivers, but a careful inquiry showed it was not. By making some tests with one of these engines with main rods disconnected, it being pulled by another, it was found that the wheels would not slide with the piston travel correctly ad

rods up.

The Westinghouse man who began the investigation learned from an engineer that these engines would not drift down hill, but would "lay back" on the train when shut off. The test previously mentioned and the use of a steam engine indicator demonstrated that, as expected a ter this statement by the engineer, the sliding was due to the aid given the driver brakes by the steam cylinder resis

tance.

It is altogether probable that this is the cause in the case you mention.

Percentage of Discharge by Brake Valve.

73. "On making an emergency application of brakes on a four or five car train with ordinary size of cylinders what

percentage of air passes through the engineer's valve with maximum pressure in the train line?"-T. F. H.

Answer. The percentage varies with different conditions, but mainly with the length of train and time the brake valve is left in emergency. With the engine and one car the percentage of the total train pipe pressure which the brake valve would discharge in emergency would be considerable owing to the train pipe reduction having to commence before the triple would make an opening to the brake cylinder, and because of the short distance from the brake valve to the rear end of the train pipe.

With each additional car the discharge at the brake valve would be a lesser percentage of the whole, and this would again be varied, considering the entire train pipe pressure, by how long the brake valve was held in the emergency position. While it is true that in a case of emergency the brake valve handle should be left in emergency position until the stop has been made or the reason for stopping no longer exists, yet this does not always mean emptying the train pipe, particularly with long freight trains. It is hoped this will make clear the impracticability of a more definite answer to the question.

Equalizing Port in Brake Valve. 74. "Is equalizing port g in seat of rotary valve in communication with chamber D when brake valve handle is in full release position?"-T. F. H.

Answer. It is. Answering farther, it may be said that in full release, in run

izing port g is at no time cut off from chamber D, though the amount of opening bewteen the two definite positions named varies somewhat.

Preliminary Exhaust Port.

75. "Why is preliminary exhaust port e in rotary valve seat bushed when all other ports are not?"-T. F. H.

Answer. To facilitate manufacture and repairs and insure greater accuracy. The diameter of this port at its smallest part is five-sixty-fourths (5-64) of an inch and it is very important that this be accurately maintained. Increasing it is liable to result in quick action when making a service application, particularly with a very short train.

Making the port smaller than standard decreases the speed of application when operating less than seven or eight car brakes and is particularly annoying in switching or making very accurate stops, such as at water tanks, etc.

The small diameter of this port extends for a very short distance and then tapers out quite large. The large end is the outlet, and the arrangement just described decreases the tendency to stop

up.

In cleaning it use a sharp pointed piece of wood, but never metal of any kind, as the latter is sure to enlarge the opening.

Feed Valve Defect.

76. "A slide valve feed valve on a freight engine had to be removed because it would maintain practically no pressure slowly to about 60 pounds with the light with a train and would only feed up engine, no matter how much the regulating spring tension was increased. An examination disclosed no other apparent defect than a small crack running part way around the piston on the slide valve side. Kindly advise what the trouble was."-C. P. C.

Answer. It would appear that the flaw so weakened the piston on one side as to allow it to cock and bind in its cylinder, thus preventing the feed port from opening more than slightly if at all. Such little opening as may have taken place and whatever leakage passed the piston raised the train pipe pressure with the light engine to the point where it just balanced the leakage out of the train pipe.

Length of Grain Pipes Affecting Quick Action.

77.-"Will the length of the train pipe affect the operation of the quick action features of the New York triple valves?" -M. L. Q.

Answer. No, not appreciably. Recent tests made with regard to ability to obtain quick action under varying conditions of train pipe and location of cars having brakes cut out, show that if the train pipe be 2,000 feet or more in length, with the usual fittings, there will be a perceptible weakening towards the rear; but at the same time the number of cars having brakes cut out may be increased, the closer they are placed to the rear, and yet quick-action will jump them.

In a test that I made personally on a fifty car train of New York triples, I was surprised to find that quick-action would jump fourteen cut out brakes, immediately in front of the rear brake, while it could be made to jump but four cut out brakes just behind the first brake in the train, this number increasing the farther back in the train the tests were made.

Auxiliary Reservoir and Corresponding Brake Cylinder.

78." About how much larger is the auxiliary reservoir than the brake cylinder, when the piston travels eight inches?"-T. T. Č.

Answer. The size of the auxiliary that should be used with a brake cylinder of given dimensions is as follows:

An eight inch cylinder should have an auxiliary reservoir ten inches in diameter by twenty-four inches in length; a ten inch brake cylinder, an auxiliary reservoir twelve inches in diameter by thirtythree inches in length; a twelve-inch brake cylinder, an auxiliary reservoir fourteen inches in diameter by thirty-three inches in length; and a fourteen-inch brake cylinder, an auxiliary reservoir, sixteen inches in diameter by thirty-three inches in length.

The freight auxiliary will hold about 1,625 cubic inches of air and the eight inch brake cylinder, having a piston travel of eight inches, will hold about 400 cubic inches of air. From this it will be seen that when the piston travel is eight inches, the auxiliaries have a little more than four times the capacity of the brake cylinder, and this is the relation

with respect to volume or capacity that exists throughout the different sizes of auxiliaries and their corresponding brake cylinders, when the brake cylinders have eight inches piston travel.

High Train Pipe Pressure.

79.-"How high a pressure can be used with a New York quick action triple valve on cars equipped with twelve and fourteen inch brake cylinders without the assistance of pressure reducing valves to guard against the danger of flattening wheels in emergency application?”—M. L. Q.

Answer.-On account of not having the data necessary in the way of reliable experiment it is somewhat difficult to give a satisfactory answer to your question.

However, if the cars in a train equipped with air brakes are of such weight as to require brake cylinders of the size you mention, a pressure of sixtyfive, and possibly seventy pounds could be used in the brake cylinders without the use of reducing valves, provided the foundation brake gear is properly designed to withstand the increased braking force and all wheels in the train have brakes acting upon them, all working to their highest efficiency.

Reason for Nipples on Pipe Unions.

80. "Why does the union in air brake

pipes that require fibrous gaskets have a short extension or nipple on them to enter the other part of the pipe?"—W. C. H.

Answer. The short nipple or extension that you refer to is formed in the union to enter the connecting pipe, so as to prevent the gasket from being forced into the pipe where the union nut is screwed up.

Trouble has been experienced on account of the opening in small unions being entirely closed by gaskets that were crushed into them, when this nipple