Counter Pressure Brakes
Aim - To describe the use and operation of counter pressure brakes.
If you wish to provide any feedback on this page, please use the contact page. It would be great to have some feedback as this helps to ensure the accuracy of the information and models.
Index
Early History of Dynamic Brakes of Steam Locomotives
Riggenbach Brake System - Test Report
Riggenbach Brake System - Operating Instructions
Introduction
To assist braking of steam locomotives on very steep gradients a braking system capable of producing a constant retarding force was developed. This system was developed by reversing flows in the steam cylinder effectively converting it into an air compressor or a "reversed" steam pump. This produced a steady retarding force that supplemented the other braking systems on the train.
Two principle systems were devloped:
- Air Counter Pressure System - sucked air into the steam cylinder so that in compressing the air a retarding force was created. Counter-pressure brake was developed Niklaus Riggenbach, and reduced wear and tear on the friction braking systems.
- Countersteam Brake - reversed the operation of the steam cylinder so that when steam was injected into the cylinder it opposed the movement of the locomotive. Countersteam Brake
Typically the counter pressure brakes were used on rack mountain railways, and were used in some Central European countries such as Germany, as well as providing a known load when testing locomotives with dynamometer cars. With a counter-pressure brake, it is not only possible to maintain a set braking force depending on the speed; the braking performance even increases at higher speeds. This is where it differs from friction brakes.
Most of the information on this page is based upon a partial extraction from an engineering paper published by Journal Institution of Locomotive Engineers. It was titled, Dynamic braking of steam, diesel and gas turbine locomotives, and written by Koffman, J.L. (Volume 41 (1951) - Paper 505.)
Early History of Dynamic Brakes of Steam Locomotives
The use of dynamic brakes with steam locomotives dates back to reversing the motion in emergency so that the pistons aspirated combustion gases and forced these into the boiler. Already on 10th June 1836 a 2-4-0 Norris-built locomotive, "George Washington" of the Philadelphia & Columbia Railroad, descended with an 8.7ton train, a 850 yards long incline of 1 in 14, braking being by reversing the motion before starting downhill. The introduction of Walschaert valve gear in 1844 made reversing possible whilst in motion.
The use of gascs having an intake temperature of about 750oF as the working media led to troubles with the early hemp and tallow stuffing boxes. Forcing the gases into the boiler resulted in a rapid rise of pressure whilst the injector did not take kindly to a jet consisting of steam with a substantial addition of combustion gases which did not condense when coming in contact with water. Last, but not least, the introduction of gases containing ash and cinders into the cylinders, caused considerable wear of pistons, cylinders and valve gear. In fact some railways found that this method of retardation could render the locomotive due for overhaul after 3 to 4 minutes of continuous application. On early French engines the reversing lever pointed backward when engine was moving forward, so that when applying reverse the forward pointing lever could become very dangerous to the driver if not securely locked in position. Reading between the lines of early Papers this had on occasions even lethal consequences.
To make the counter-pressure brake suitable for use when descending long grades it was necessary to prevent the aspiration of combustion gases; provide sufficient cooling during compression; prevent the forcing of compressed gases into the boiler and provide means of controlling the pressure of compressed gases. Early tests with a controllable steam counter-pressure brake were carried out in 1858 by Stuymecher, the locomotive foreman of the Madrid-Saragosa-Alicante Railway at Albacete. The arrangement was similar to that known as "Rouch's Steam Retarder" and used on the Stockton and Darlington section of the N.E.R. In 1860 the Austrian Western Railway adopted throttle valves (due to Zeh of the Sigl Co.) in the exhaust pipes close to the cylinders. With valves and throttles shut, only a small volume of gases was aspirated and 100ton trains were brought down the 1 in 40 Simmering incline. During 1864 Bergue introduced a modification to the original scheme whereby the motion was reversed, but instead of being forced into the boiler the combustion gases were led into a tank provided with a safety valve and an adjustable blow-off valve. In July, 1865 Le Chatelier, then Ingenieur en Chef des Comite de Paris, of the North Railway of Spain, requested the local manager and the C.M.E. (Ricour) to carry out certain tests which in the following year resulted in the introduction of the counterpressure brake utilising steam as the working fluid, whilst water was injected to reduce the operating temperature, a solution of which Le Chatelier stated in May 1866, "Pour moi, l'eau est toute la solution; cette solution est neccssaire et generale" (For me, water is the whole solution; this solution is necessary and universal). In operation it was necessary to shut the throttle, reverse the motion, then open the throttle and a valve feeding steam and boiler water to the cylinder exhausts, thus preventing the aspiration of exhaust gases. This method of braking became very popular on the originating railway which in some sections has 5 mile grades of 1 in 66. Within three years it was also adopted with some 1,400 locomotives in France, whilst another 1,220 were in course of conversion. On the other hand the Austrian railways experienced a spate of connecting rod, crankshaft and crank fractures attributed to the enginemen admitting too much water into the cylinders.
In this country the Le Chatelier system was applied on some L. & N.W. and Midland engines about 1870, but it was not a success. To quote Ahrons : "The failure in this country of an apparatus, which has been successfully used for many years in France, may be partly ascribed to want of care on the part of the drivers". However, the system became popular, particularly on mountain lines in Central Europe and Russia.
Riggenbach Brake System
The use of air as the operating media, originally proposed by Mars and Aiken, was introduced in 1869-70 by Riggenbach with the locomotives of the Vitznau-Rigi Rack Railway. Because of its obvious advantages this brake superseded that due to Le Chatelier on many locomotives and is dealt with here.
The modus operandi of the Riggenbach brake will be gathered from Fig. 1 above. The regulator (1) and blastpipe (2) are shut and the cylinders connected to the atmosphere via the valve (4). The motion (3) is reversed to ensure full admission. The pistons aspirate and compress air (5). The brake force thus produced is regulated by a valve (6), the compressed air escaping via the silencer (8). To provide cooling, boiler water is injected into the cylinders via the valve (7). The reason for obtaining appreciable brake power and also for the necessity of using water injection are indicated in Fig. 2 below showing the power required to compress dry air at 60oF at various compressor efficiencies and the relevant temperature ratios. Whilst the temperature ratio will remain practically independent of T1, the power required for compression will depend upon density (i.e. T1). The effect of introducing water into the cylinders upon T2 can be determined with help of Mollier charts. The values plotted here were obtained from thermodynamic and the efficiency of compression is defined as isentropic compression work / actual work. Because of the cooling effect of steam the actual work done will be smaller.
The design of the change-over Valve located upstream of the exhaust nozzle requires careful consideration. A typical design is shown in Fig. 3 below. This may however cause undue back pressure losses due to rapid change of direction at entry, and expansion and contraction when passing into and out of the valve, respectively.
The power output curves were plotted as a function of wheel r.p.m. (to eliminate the suggestion of a speed dependence) to illustrate the possible throttling effect of the Riggenbach valve (Fig. 4). The three locomotives thus fitted show a pronounced output peak, the loss of power at higher r.p.m. being possibly due to the presence of the valve. This possibility is suggested by the fact that compound machines have a somewhat similar power-r.p.m. dependence due to throttling effect of the cylinder interconnecting passages.
An aerodynamically more acceptable shape is shown in Fig. 5 although this valve might be more difficult to operate. In either case the valve can be operated by hand or air. The cross-section of the driver's control valve depends upon engine speed and usually amounts to 1 / 160 to 1 / 120 of the cylinder piston area.
Silencing is either by a cylindrical silencer provided with perforated baffle plates or by leading the air to atmosphere via the space between the chimney duct and its jacket or a combination of both. In the latter case the space should be packed with copper shavings and a large number of exhaust holes should be provided as shown in Fig. 6.
With early applications water was injected into the steam exhaust duct whilst with more recent applications injection is directly into the cylinders. Whilst no additional wear is usually encountered under normal operating conditions it might be advisable to arrange for the air intake to be in a relatively clean (high) position when operating under dusty conditions. In extreme cases the use of a suitable air cleaner should be considered.
A Mechanised Version of Riggenbach Brake
The application of the brake is time consuming and cumbersome, requiring up to seven steps, i.e. :-
- Release wheel brake.
- Open exhaust air valve.
- Shut blast pipe.
- Shut by-pass.
- Reverse valve motion.
- Apply water injection to maintain a cylinder temperature not higher than 550oF.
- Control brake pressure with exhaust valve.
To release the brake the procedure will be reversed in the following order:-
- Shut off water injection.
- Fully open exhaust air valve.
- Slowly move valve motion into forward position.
- Open by-pass.
- Shut exhaust air valve.
- Open blast pipe.
- Apply wheel brake.
Obviously whilst this can be tolerated in case of prolonged applications required down long grades, these manipulations make dynamic braking unattractive at high speeds, unless the application of the various valves, etc., could be made automatic and interlocked under the control of a single lever. Whilst it will not be difficult to bring 1, 2, 4 and 6 under the control of a single lever, the overriding control required at 5 and 3 may present difficulties. To overcome these a novel scheme was proposed by Ewald (Fig. 8 below). Usually the position of lever (2) corresponds to a certain position of the reversing lever (7), but for the application considered here it is essential for (2) to move-if required-into full reverse, quite irrespective of the direction and degree of cut-off determined by (1). To achieve this it will be necessary to alter the length of the reversing gear reach rod (2-3). This can be done by using floating cylinders (4) and (5). Under normal circumstances the rod is held at fixed length by the admission of compressed air (or any other suitable fluid) to the cylinders, so as to force the pistons against the cylinder covers. With a forward moving engine, lever (1) is moved forward and thus actuates valve (6) which connects pipes ( 7 ) and (6). If now the wheel brakes are applied compressed air will also be admitted from (7) via (6) to (8), and then to valve (9), thus admitting air from the reservoir (10) to cylinder (1). At the same time air from (6) closes the valve (11) which now connects line (12) to the atmosphere. The piston of cylinder (4) can now move to the right taking with it cylinder (5), the piston of which remains at rest relative to the cylinder because of the pressure of the already admitted air. The rod (2-3) thus moves into full reverse although lever (1) is still in the forward position.
If the wheel brakes are released air will escape from the brake cylinder ( 13 ) as well as from pipe (8), thus permitting valves (9) and (11) to return to their original positions and in turn connect pipe (24) to the atmosphere, whilst valve (11) admits air from container (10) to cylinder (4) moving the piston to the left and returning point (2) to the position corresponding to that of lever ( 1 ). Provision of a three-way valve between the main brake reservoir and the control valves will enable the driver to use either or both types of brakes simultaneously. In addition, the use of a speed-operated governor would cut out the dynamic brake once the speed falls below a predetermined value.
Riggenbach Brake System - Test Report
During his original tests Le Chatelier found that with the then prevailing locomotive and train weights it was possible to hold a 25 per cent. heavier train down a grade at the same speed as the locomotive would be capable of pulling uphill. The mean indicated pressure obtained when braking with "full admission" was about 55 per cent. of that obtained under steam. Tests carried out in Germany in the mid-twenties with a heavy tank (T20) and a tender engine (G10) incorporating the Riggenbach-brake have produced some interesting results.
For operation on heavy grades a rolling resistance of 10 lb./ton may be assumed for the driving axles under normal conditions, whilst with heavy sanding this value may be increased to 20 lb./ton. Descending a 1 in 20 grade the mean effective brake pressure need be only (112-10)/(112+10)=0.835 and 0.695 respectively, of that required to move the train uphill. In actual service it was found that the mean pressure maintained during compression was in fact much lower than that obtained when driving (Fig. 9) because with the large admission maintained when ascending, the exhaust was opened long before the steam pressure was reduced to that at the blast pipe, whilst compression (when braking) starts at a value similar to that of low blast-pipe pressure. In addition slight vacuum is encountered during the suction stroke. The resultant loss in diagram area is not compensated by the fact that the compression adiabate of air is steeper than the steam expansion line or to put it more correctly, the polytropic exponent of air compression is greater than that of steam expansion. During tests with the T20, the ratio of (140-10)/(140+10)=0.865 was confirmed on the 1 in 16.7 sections of the Harz Railway. The main limitations imposed upon the intensity of dynamic braking are those due to high compression temperatures. During the above tests temperatures of over 750oF. were attained, this on one occasion leading to the melting of the glands. It was possible to keep the temperature below 750oF. only by a liberal use of hot (boiler) water. Cold water proved less effective since water at temperatures near that of evaporation, evaporates and disperses more readily and thoroughly, absorbing heat corresponding to the latent heat of evaporation, whilst cold water droplets with relatively small area of contact are somewhat sluggish so far as heat absorption is concerned (Leidenfrost's Phenomenon). With new well fitting piston and valve rings it should be possible to maintain the same constant train speed when descending as pulling uphill. Lighter trains can be retarded as well but they can be stopped only if their weight is substantially smaller than that of trains brought downhill at constant speed. Heavy trains cannot he stopped by dynamic braking alone due to the compression deviating from adiabatic towards isothermal, whilst too much air also escapes past the pistons at low speeds. Generally, it can be stated that at the same speed the weight of trains downhill must be slightly below that pulled uphill. When running down grades of up to 1 in 16 at speeds not exceeding 15 m.p.h. the trains can be controlled by the use of dynamic braking with a slight help from air brakes.
The brake effort developed by a G10 locomotive at a constant speed of 15.6 m.p.h. is plotted as a function of pressure in the steamchest in Fig. 10. The values were ascertained by pulling the locomotive behind a dynamometer car. It will be noted that whilst up to about 100 lb./sq. in. the brake effort is directly proportional to the pressure, this is not so at high pressures. This is due to leakage losses and also due to the throttling effect (wire drawing) encountered by the aspirated air entering the cylinders at higher speeds.
The ability of this engine to hold different train weights down grades at a constant speed of 15.6 m.p.h. is shown in Fig. 11 as a function of pressure. The values were calculated on the basis of a rolling resistance of 10 lb./ ton for the driving and 5.5 lb./ton for all trailing axles. For example, with a steam-chest pressure of 85 lb./sq. in. it is possible to hold a 200 ton train down a 1 in 40 grade. The dependence of compression pressure upon speed will be gathered from Fig. 12. During the tests the temperature did reach 660oF. at 130 lb./sq. in. With water injection similar temperatures were attained at 155 lb./sq. in. The tests have shown that it will be advisable to determine train weights on the basis of the pressure in the steam-chest not exceeding about 85 to 90 lb./sq. in. Since this precludes the possibility of forcing back the (Wagner type) regulator valve, safeguards the glands and leaves a reserve for retardation. On a 1 in 40 grade the resultant train weight amounts to about 2.7 to 3 times the adhesive weight of the locomotive. Pressure of up to 185 Ib./sq. in. was used at temperatures of up to 750oF, but even with oils as used with superheated engines deposits were left in the cylinders. Apart from this, high pressures did not improve the effectiveness because of throttling losses encountered by the air at higher road speeds and-the resultant low volumetric efficiency.
Generally the use of Riggenbach brakes results in a diagram as shown in Fig. 13. The air is compressed between (1) and (2) and admitted from the valve chest between (2) and (3). From (3) to (1) the air is forced into the valve chest whilst the pressure drops between (1) and (5) because of admission to the opposite cylinder. Expansion finally takes place between (5) and (6).
To ensure lower operating pressures and temperatures a number of locomotives were fitted with pre-admission valves, Fig. 14. These valves (1) are provided between the steam-chest (2) and the cylinder ports (3). By moving lever (2) into the brake position rocker (5) is made to actuate the valve levers (6).
The valves (Fig. 15) are operated if the cut-off exceeds 60 per cent. and in the case of the machines concerned increase the admission from 75 per cent. to 92 per cent. A comparison of Figs. 13(a) and (b) shows that although the maximum pressure has been reduced the mean pressure is appreciably increased. The constant pressure maintained between points (a) and (b) (Fig. 13b) is caused by the cutting in of the other cylinder. The effect of the preadmission valve on brake torque F will be gathered from Fig. 16, whilst the resultant forces at the crankpin are shown in Fig. 17.
Other designs may be evolved to improve the effectiveness of dynamic braking but they must always conform with Mr. Cox's dictum that the steam locomotive "as a simple, cheap, rugged tractor can still have a part to play, a part which it cannot sustain should it leave its vantage ground and once more descend to the complexities . . .". Even in its simple form the Riggenbach brake is capable of contributing to the reduction of operating costs particularly down the long grades of overseas railways and possibly in shunting yards as well. It can also play a welcome part by providing assistance when braking from high speeds, but here more effective means of control will be required.
Riggenbach Brake System - Operating Instructions
The following are some operational "rules" that were used by different railway companies.
For locomotives of the German Reichsbahn equipped with counter-pressure brakes, the requirement was therefore that the valve chest pressure must not exceed 6 bar (87 psi).
Useful Links
Railway Economy - Use of Counter Pressure Steam in the Locomotive Engine as a Brake by L. Le Chatelier