The Air Brake System

Air Braking

Compressor mounted on tender
Shown before the doors were fitted

During the early part of 1993, approximately half way through the overhaul, the MNLPS Management Committee took the decision to look into the feasibility of fitting air-braking equipment to enable Clan Line to haul air braked stock. Informal discussions were held with BR and SLOA, who were quite positive about the idea. SLOA felt it sensible for Clan Line to act as 'guinea pig’, setting the guidelines for other main line steam locomotive owners who might choose to go down the same path.

It was not practical to air brake the locomotive itself. The locomotive and tender would still be braked by steam and vacuum brake respectively, operated through the air braking equipment. To buy new equipment was prohibitively expensive so used equipment was obtained, overhauled and re-certified by the original supplier. The compressor was obtained from Steam Traction Ltd, who had imported some air braked locomotives from Finland.


Rear of Tender showing cut-out for compressor.

Looking at the system as a whole, the positioning of some of the equipment proved problematical. The compressor for the air braking system was mounted in a recess cut into the back of the tender. The alternative of positioning it on the front of the locomotive would do nothing for 35028's appearance. The picture above shows the recess cut into the tender, and below, the compressor after fitting.


Compressor fitted to tender.

The three new main air reservoirs were placed on the top of the tender tank, at the back. These can be seen in the picture above. The three vacuum cylinders they replaced were removed from the tender and replaced by four smaller cylinders positioned beneath the frames under the water tank.


Air Brake Handle.

The driver's brake handle was mounted on a separate pedestal, welded to the cab under-frame. It is placed on the left hand side of the cab but to the right of the driver's position.

During our seven years running various minor problems showed themselves, also the fact that the original installation was done in a rush at the end of the last overhaul and that the wall thickness of the 15 and 28 mm copper pipe was incorrectly specified by the railways (it was too thin), it was not surprising that we had a number of problems early on, although fortunately very few problems occurred when hauling revenue earning trains - this was partially due to our policy of trying to find trouble before it finds us. After a couple of years running we started to get problems with the 15mm copper pipe fracturing and as a result of this an on going programme of replacement and re-bracketing with resilient plastic clamps was initiated. At the same time changes were made to the location of some of the valves under the footplate to aid accessibility.

Towards the end of our period of running we saw further problems with some of the air brake components. It should be remembered that all of the valves and components used in the currently approved system were salvaged from diesel locomotives and were normally sited in cabinets within the bodywork of the locos. Because they were not exposed to the elements they were able to be made from aluminium and all of the valve bodies and pipe brackets are cast in this material. After years sited under the footplate, the combination of ash, coal dust and steam had caused considerable surface corrosion to most of these components and in the case of the Low Main Reservoir Pressure Valve, which has its control spring and spindle exposed, caused it to fail due to more extensive corrosion. Fortunately, we have a complete set of spare valves for the system so we had no problem with replacements.

Although we had replaced all of the 15mm pipe with thicker walled material, the 28mm main air supply, main reservoir and automatic air brake pipes were as fitted in 1995, and as such had given no problems. It was however decided that we should replace this pipe with heavier duty material, and to change the fixed metal clamps that locate the pipework to the loco framework with resilient plastic clamps as had already been done for the smaller pipework.

During the 2001 - 2005 overhaul and in order to afford better protection to the valves, one of our engineers decided to attempt to design an enclosed cabinet/pedestal unit to contain all but one of these pieces of equipment. (This other valve, the DV2 air/vacuum proportioning valve, is located in an enclosure between the frames above the trailing driving axle and has been immune from any corrosion problems. As it is quite a large valve we decided that if it ain't broke, don't fix it and so we left it where it was).



The component pipework before assembly

A total of 11 valves, plus three stopcocks, would have to be located within the unit, as well as a considerable amount of connecting pipework. Interestingly, the brake system itself would work with only two of these valves - the main reservoir pressure regulating valve, which regulates the compressor air supply down to the required 100psi for the train main reservoir (yellow) pipe, and the M8A driver's brake valve which controls all of the automatic air brake (red) pipe pressures. All of the other equipment is concerned with the fail-safe and automatic (AWS/TPWS) brake application train protection and air/vacuum integration functions.

There are three on/off cocks in the system. The AWS shutdown cock is a combined air and electrical on/off switch and was originally situated on the return bulkhead behind the drivers seat - this valve must be switched on before the braking system will work. The AWS isolating cock was under the cab floor near to the drivers left foot and was accessed through a trap in the floor. The effect of this valve is to override the AWS electro pneumatic air control valve by isolating the effect of the electrical input from the AWS control system in order to enable the brake to be created in the event of an AWS electrical failure. It is normally wired in the closed position. Finally, the Assist Failed Train cock was under the cab floor to the right of the brake pedestal, this cock is wired in the open position but must be closed to allow a rescuing locomotive to supply air to the brake on our loco in the event of a compressor or AWS failure. Its effect is to isolate the automatic air brake pipe from the AWS unit, and it is a requirement that the handle of this cock projects through the floor as a reminder that it has been operated. These cocks have all been re-sited in the new layout, with the AWS shutdown cock on the front face of the brake pedestal and the other two cocks behind a removable panel in the side of the pedestal. This panel cannot be replaced if the Assist Failed Train cock has been closed.



The first trial assembly.

In order to save time on a project that was going to be quite time consuming, our engineer took the valves and the existing brake pedestal to his home, where all of the initial design work, construction and pipe manufacture was carried out in his garage.

The first stage of the cabinet design was to establish the physical size that would fit under the cab floor in the required position and to make a plywood box to these sizes - actually about 20 inches wide, 22 inches long and 16 inches deep. The air control valves use a system of pipe brackets which are permanently fixed to the infrastructure of the locomotive and are linked by the appropriate pipework, with the valves being bolted to these brackets with 'O' rings to seal the pressurized air joints. This is an excellent system which makes it much easier to replace individual valves if required, but it does increase the amount of space taken up by each valve.

Having made a plywood box it took quite some time to work out how to fit everything into it, manufacture dummy fixing brackets and bolt all of the required valves to the box. Many hours were spent bending and silver soldering the individual pieces of pipe to connect the valves to each other - although it was not practicable to complete all of the pipework, about three quarters of it was done offsite. In order to save money, unions, connectors and fittings were salvaged from the old pipework by heating them up and unsoldering them. This in itself took quite a while to do because the fittings have to be brought up to near red heat to unsolder the joints and then boiled off in dilute sulphuric acid to clean up all of the oxidisation that occurs when they are unsoldered. With literally dozens of fittings involved at a cost of £5 - £10 each, the saving made the effort worth while and is an example of the thrifty approach taken by the engineering team towards expenditure on the overhaul.

On completion of the fitting out of the dummy box, the unit was stripped down and all of the valves (except the M8A Driver's Brake Valve) were taken apart, cleaned and inspected. Apart from a film of oil on the internal surfaces and components of these valves they were all in very good condition and suitable for further use. All components were checked against the manufacturers' specifications, overhauled and re-assembled before everything was returned to Stewarts Lane - itself no mean feat as the whole unit now weighed almost 2 cwt. The M8A driver's brake valve is very complicated internally and was sent away for professional overhaul.



Assembled Box.

Back in our workshops at Stewarts Lane a system to join the box and pedestal, and to fix the complete unit to the support rails under the footplate floor was devised. We then consulted with our design engineer to ensure that the proposed unit and fixings had sufficient structural integrity and strength for the job in hand. Agreement reached, we cut and welded up a cabinet from 6mm steel plate, and drilled the required fixing and pipework connection holes into the box. All of this took a lot longer to do than it takes to read this!

Finally, with the box complete it was possible to manufacture the remaining pipework, including the feed to the three pressure gauges on the pedestal desk, also to fit the conduit and wire up the required electrical connections to the AWS shutdown switch in the pedestal and the electro pneumatic valve inside the cabinet. With the setup physically complete the whole unit was then stripped down and, using the workshop compressed air supply, all of the valves were tested and set up to work at the required pressures. This was only a rough setting to within a couple of psi, as the whole system was to be fine tuned as part of the engineering acceptance certification before our return to traffic. All pipework was then softened and the silver soldered joints tested for leaks (two were found and repaired), before the whole lot was reassembled with all unions and couplings finally tightened up. Our spare M8A driver's brake valve was fitted, blanking pipes were fitted to all necessary outlets and an air feed supplied to the unit connections.

Then the moment of truth - apply air and see if it worked. Immediately the first problem was obvious. The brake pipe pressure gauge, which should read between 0psi and 78psi depending on the drivers brake valve handle position, read a steady 100psi while the main reservoir pipe gauge, which should read a steady 100psi - yes, you've guessed it! - fluctuated with the brake valve setting. The two gauges were cross connected. The problem was soon rectified and the whole system again tested for leaks - three leaking coupling joints were soon dealt with. A battery connection to the electro pneumatic control valve in the cabinet was used to simulate AWS/system failure mode operation. An immediate problem that appeared was that when connected to the AWS equipment, the system kept creating and dropping the brake pressures and would not stabilise with the system pressurised. This was traced to an inadequate test air supply and our expectation that this would not be a problem when supplied with the loco's much greater air supply capacity proved correct. All pressures were within the expected limits and the simulated AWS application worked correctly. Success!



The Final Assembly

Article produced by John Adams