caprotti

CAM Modification

As many will be aware, 71000 was fitted with new exhaust cams during the recent overhaul.

For those who are not intimately familiar with the design of A L E Caprotti camboxes, I will first of all explain the functions of the cams. The camshaft in each box is fitted with two inlet and two exhaust cams. The shaft rotates at the same speed as, and in synchronisation with, the driving wheels.

The two inlet cams, as their name infers, control the operation of the two inlet valves admitting steam in turn to each end of the cylinder. Two cams are required to control the operation of a valve. An inlet valve is operated by means of a tappet, which in turn is worked by means of levers and a pair of rollers mounted on a swing beam which follow each cam profile. For an inlet valve to open, both raised cam profiles must overlap to some degree and the cams will then have their rollers in the lifted position concurrently, enabling the swing beam to move the inlet lever and operate the tappet. As the camshaft rotates and the leading edge of the raised profile of the first cam meets its roller, the roller is lifted. The leading edge of the raised profile of the second cam then meets its roller and the inlet valve begins to open. The second cam here is controlling the point of opening (admission). The inlet valve actually starts to open just before the piston reaches the end of the cylinder to begin its stroke, providing a small amount of "lead" steam. The trailing edge of the first cam controls the point at which the valve closes (cut-off) as the roller is allowed to fall off the raised profile. By rotating this first cam in relation to the camshaft, the point of cut-off can be varied.

Each of the two inlet valves is operated by its own rollers, levers and tappet working off the same pair of cams with events being 180 degrees apart. When travelling tender first, the roles of the two inlet cams are reversed.

The two exhaust cams control the operation of the two exhaust valves releasing exhaust steam in turn from each end of the cylinder. As with the inlet valves, two cams are required to control the operation of a valve but the principle is slightly different. An exhaust valve is operated by means of a similar tappet, worked by means of a lever, which this time has a single roller follower directly mounted on the end. The roller is wider and follows both exhaust cams, being lifted to open the valve when either cam profile is raised. The two raised profiles again overlap to some degree.As the camshaft rotates, so the leading edge of the profile on the first exhaust cam controls the point of opening of the exhaust valve (release) by lifting the roller.The trailing edge of the profile on the second cam controls the point of closing of the valve (compression) by allowing the roller to fall. The exhaust valve begins to open before the piston has reached the end of its stroke to allow a free passage to exhaust during the return stoke.

It is necessary for the smooth and efficient operation of the engine that the residual exhaust steam in the cylinder should be compressed up to the inlet steam pressure during the final stage of the exhaust stroke. The exhaust back-pressure in the cylinder varies with cut-off being lower at short cut-offs. In order to achieve the correct degree of compression the point of exhaust valve closing is varied with cut-off.As the engine is 'notched up' the exhaust valve is closed earlier. This is effected by rotation of the exhaust cam controlling the point of compression in relation to the camshaft. In practice it works as follows. The two cams have identical profiles. In full gear, the overlap is minimised. The cams are described as being at full extension, and give the longest period of exhaust valve opening. As cut-off percentage is reduced the second cam as mentioned above moves, advancing the point of compression, increasing the overlap and reducing the period of opening. The first cam remains stationary in relation to the shaft and hence the point of release remains constant. As the cut-off approaches about 30 %, the two profiles come into line. The second cam continues to move and its leading edge advances ahead of that of the first cam. At shorter cut-offs this cam thereby takes over the control of the exhaust valve opening and release takes place earlier. The trailing edge of the first cam now lags that of the second cam and this cam takes over the role of controlling the point of compression, which now remains fixed at 60% as cut-off is further reduced until as mid gear is approached when the first cam starts to move and compression is reduced to 50%.

Each of the two exhaust valves is operated by its own roller, lever and tappet working off the same pair of cams with events being 180 degrees apart. When travelling tender first, the roles of the two exhaust cams are reversed.

Design of the events of release, compression and degree of lead provided by the point of admission is a matter not only of theoretical consideration but also of experienced judgement and evaluation of testing. The valve events provided by A.L.E. for 71000 were B.R.’s choice. The changes currently being made follow from recommendations made by L.T. Daniels after initial main line testing in 1990.

The transformation of performance resulting from fitting of the Kylchap blastpipe as advocated by the Caprotti engineer during restoration was the pivotal episode in the 71000 legend. After the initial main line test run Daniels recommended increasing the blast pipe diameter, and modifying the exhaust event to give a slightly later release. The former change was made but the latter has waited until now to be trialed.

The design of one of the exhaust cams, that controlling the release event at long cut-offs in forward gear, was already available, having been put forward to B.R. originally for consideration. The design of the other, to control the forward gear release event at short cut-offs was made by the Trust and approved by Daniels. The new designs both have the forward running leading edge profiles retarded to open the exhaust valve 15 degrees later. The profile design, which controls the rate of opening has been maintained. The effects of the changes will be to increase expansive working of the steam and thereby obtain maximum power out of the stroke. As with the blastpipe modification the effect should also be to make the blast a little smoother. Some reduction in exhaust back pressure results from the retarding of the release event, and further work will be needed to optimise blast pipe nozzle design during main line operation to ensure correct draughting.

These changes are effective only for forward running. Making the changes for tender first running would compromise the forward running compression event and in any case is not considered to be worthwhile. The changes will advance compression in tender first running but as no serious tender first running is contemplated, this is considered an acceptable compromise.

Provision of equipment for obtaining indicator diagrams electronically will be progressed and hopefully will enable full evaluation of cam changes to be made.

It is interesting to note from research that comments were made to B.R. about the valve events of 71000 by André Chapelon to the effect that the release events were inappropriate and should be modified to give later and slower opening at longer cut-offs. (The French had better success with Caprotti valvegear on Alsace Lorraine pacifics particularly in the area of boiler efficiency!).

Dr Gary Shannon
Email: engineering@71000trust.com