Melbourne Cable Tramways – Macmeikan & Pollock Reports

Cables

The important function performed by the cables, which were so inconspicuous in their operation, merits more than general comment. The following, therefore, is a brief record of the life of the cables from the time of their manufacture to their final scrapping.

The cables obtained for the first lines were of 3 5/8" circumference and 11" lay, and consisted of either 7, 15 or 19 wires strands each. The 19 strand rope was made by Roebling [8] in the USA, the others in England by Bullivant & Co of Millwall and George Cradock & Co of Wakefield. Over the years due to increasing traffic, the size of rope was increased until it became standardised at 4" full circumference, 10" lay of 7 wire construction. Just before the St Kilda Rd engine house was closed, ropes of 4¼" circumference and 15 wire construction were being used, but were not in operation long enough to determine whether they were an improvement over the standard.

On arrival at one of Melbourne’s wharves from overseas, the cable was unloaded from the ship’s hold, hand over hand, on to horse drawn lorries. The lorries were arranged side by side in pairs, the cable being coiled on to the first pair until they were loaded to the maximum weight. The cable was then led past the horses and on to the second pair of lorries. Sometimes up to five pairs of lorries were required to carry a single cable in a ‘train’ to the appropriate engine house.

The cable was then unloaded from the lorries and guided by pulley blocks to be wound on to the cable reel at the side of the engine house. The cable reel was slowly turned by its own individual steam engine mounted on the end of the cable reel stand. Steam was supplied to the cable reel engine by the boilers in the engine house.

Interior of South Yarra engine house, showing rope driven reduction gears on right. The immaculate condition of the interior is typical of the attention taken by MTOC employees.

Photograph courtesy State Library of Victoria.

To put a rope in the road, after traffic is over and the engines stopped, the rope tension is eased to a degree just sufficient to keep the rope on the sheaves and pulleys, and then clamps put on the rope. The old rope is then cut and the new rope roughly spliced to it. The other end of the rope is attached to a second cable reel, clamps taken off and the engine house engines are run slowly, pulling the new rope into the tunnels against a brake, and the old rope being reeled up by the cable reel engine. Pulley blocks were used to guide the new cable to the rope race where the rope changeover was carried out.

Interior of engine house showing cable tensioning race with engine in left background.

Photograph courtesy City of Yarra.

When the new rope is entirely paid out into the tunnels, the outgoing clamp was put on, the rope tightened and the incoming clamp put on. The old rope was then removed from the new rope, and the new rope spliced together. The rope tension was reapplied, clamps taken off and the road inspected to ensure that the new rope is on all pulleys and sheaves. During this process the tension carriage, as per Figure 5, was moved as far forward as possible, in order to allow for as much stretch of the new rope as possible. The lower portion of the tension rack was located by large pawls engaged in a rack on each side of the cable race, permitting the tension carriage to be moved back as the cable stretched. The upper part of the tension carriage, known as the tension brackets, was free to run rails fixed on the lower portion, maintaining tension by means of a heavy weight. The weight applied in the tension brackets varies from 2 to 5 tons depending on the conditions of the particular route and the length of the rope. The 13' 2" cable sheave was mounted on the upper part of the tension carriage.

Figure 5. Cable race arrangement diagram, showing driver and idler wheels, and tension wheel mounted on automatically adjusting tension carriage.

From the Macmeikan report.

The speed of ropes, originally set at about 8 mph, was later increased over the years until nearly all were running at speeds of between 12 to 13 mph, with only two cables running at 11 mph.

However, the installation of the first cable on a route required a different process, as illustrated here by an extract from Treatise upon Cable or Rope Traction by J. Bucknell Smith C.E., regarding the threading of the first cable into Collins St.

The operation of threading the cable from the engine house at Brunswick St to Spencer St and back, a distance altogether of 3½ miles, was watched with interest by a large number of spectators. The rope weighed 28 tons, and was manufactured by Messrs. Cradock & Co. This was rolled on an immense drum, and had to be drawn from same through 1¾ miles of covered track and back again. The end of the cable was made fast to the gripper of a dummy car, which was drawn forward by a gang of men, carrying the rope with it. In a short time, however, the resistance became too great to be overcome in this way, and horses were employed. Eight animals were harnessed, but proved too few, and four more were added. These carried the car along for a time until the gripper was suddenly snapped, and operations had then to be suspended until it was welded. A fresh start was made two hours after, and the terminus was ultimately reached without any serious mishaps. The cable was then led around the end pulley by means of a short auxiliary piece which had been previously placed in position, and then the return journey was made with twenty-four horses.

The wheels used for driving the cable were of 12, 13 or 14 foot diameter and were known as the drivers, being situated at the front end of the cable race, as at Figure 4. The drivers were fitted with segments, which were renewed or reconditioned when they became worn or when a new cable was placed into operation. The segments were cast steel with hardwood blocks the arrangement of which is shown in Figure 6.

Figure 6. Renewable driver wheel segment. Machined wooden blocks inserted into each segment applied friction drive to cable.

From the Macmeikan report.

The driver rim into which the segments (A) were fitted, had loose segments (B), which were bolted to the main wheel (C) by the bolts (D), and bolted through the segments at (E) as per Figure 7.

Figure 7. Cross-section of renewable driver wheel segment.

From the Macmeikan report.

The new or reconditioned segments were always employed for driving a new cable to prevent undue wear. From the rolling stock side, new dies were fitted to all grips operating on a line where a new cable was running. This minimised wear on the rope because the old dies, being worn, tended to damage the cable by forcing it into the deep grooves in the worn dies, giving rise to severe and dangerous operation of grips.

The rope driver shaft was fitted with a revolution counter, and by this means the splice and any faults in the cable could be tracked after being recorded on the blackboard. The blackboard was mounted in a convenient position for noting records of cable faults. The permanent marking on the board was a line representing the cable and figures representing the length from the splice. Separate blackboards were used for each rope driven from the engine house.

If a fault, such as a loose strand, kink, or any other damage to the rope was observed, it was noted on the blackboard. The appropriate section of the rope was then examined after service hours, unless urgent attention had to be given immediately. When repairs had to be carried out during service hours, serious delays occurred. This was often avoided by slowing the speed of the rope down for inspection when the fault came through the engine house, to ensure that additional deterioration did not occur. Sometimes a stoppage of a few minutes sufficed to execute temporary repairs, such as cutting off or tucking a strand, which enabled the cable to continue running until after service at night.

Stranded cable at the Fitzroy engine house, 1918.

Photograph from the Pollock report.

Originally, each rope was taken around two drivers which were geared together. This proved unsatisfactory as well as noisy, so back driver was moved further back and the connecting gear wheels removed, so that the back driver now operated as an idler. As larger drivers were installed to increase the rope speed, the former back drivers were discarded and replaced with smaller and lighter idler wheels.

An alarm bell was installed in the cable race, which rang if the incoming cable had a protruding strand. If the alarm rang, the engine driver slowed down the engines, allowing the ropeman to examine the fault, determine the required action, and record the fault on the blackboard.

The cable when leaving or entering the engine house is guided around sheaves known as ‘horimontals’ as at Figure 8. These sheaves revolved in a horizontal or slightly angled orientation, depending on the direction of the rope to or from the cable tunnel, and were of 12' diameter. These sheaves were housed in large pits, approximately 6' deep under the track, and could be reached from the engine house by means of passages or tunnels.

Figure 8. Vertical cross-section of engine house cable sheaves or ‘horimontals’, showing both horizontal and angled versions. These guided the cable from the engine house into the tunnels in the street.

From the Macmeikan report.

Auxiliary ropes were used in short lengths to pull cars around some junctions and curves. Where there was rapid wear, these were cut quickly, but the life of the main rope was considerably lengthened by not having to do this duty.

The rope went around curves by means of a series of pulleys and drums (known as curve pulleys) set at intervals around the curve, the grip being prevented from striking these by means of a rubbing bar. This bar was an angle iron set round the curve under the slot at such a height that the back guard or top die holder, as the case may be, of the grid bore on this angle. The curve the pulleys were set at a distance from the road equal to the distance of rope when carried in grip.

Later, in many cases these in turn, the pulleys were removed and the right taken around a 12' sheave housed in a pit under the road set so that the rope left each end of curve at the tangent, being carried in a subway to the sheave. The trams in such cases went around the curve by momentum, ropemen having to throw the rope before entering the curve and automatically picking it up again after exiting the curve. This practice greatly increased the life of the rope.

On up grades, or when traffic was heavy, if for any reason trams could not go round a curve by momentum, curve pulleys were retained. In some locations where this was the practice, short auxiliary ropes were used in order to minimise wear on the long (and expensive) main cables.

Places where auxiliary ropes were used included:

• Junction of Brunswick Street and Victoria Parade;
• Junction of Nicholson and Gertrude Streets; and
• Lonsdale Street between Russell and Swanston Streets.

The latter was interesting as this auxiliary rope was powered from a driver attached to the Carlton terminal sheave, and there was a tension weight attached to the auxiliary terminal sheave working in a pit in Swanston St, unlike the other two cases, which were driven directly from an adjacent engine house. Trams coasted on the up track and only used the auxiliary cable on the down track.

In places where a grip had to cross the rope entering or leaving a curve after having thrown the rope to allow the tram to proceed around the curve by momentum, and there was any likelihood of the rope rising above the grip, a conical drum was placed in the tunnel which ensured that the rope was kept below the bottom of the grip.

There were 32 different designs of drums and pulleys, 17 of which were in general use. The balance of the designs were rarely used as they were built for specific and unusual locations or purposes.

In straight sections of track, line pulleys were placed in the tunnel every 33 feet, supporting the cable. These line pulleys were carried by cast iron yokes, attached to the main yokes carrying the slot beam. All pulleys were of cast steel, the harder the better, roughly ground on the throat. In places where the rope was thrown, line drums replaced the pulleys. The line drums were carried by cast iron brackets bolted to the main yokes. These drums were bored for a 1¼" spindle with 1" ends and secured with set bolts so that they could be set to the most suitable place on the spindle to catch the rope when thrown out of the grip.

The bearings for these drums and pulleys were merely a cast iron block slipped into a light cast iron grease box. No machining was performed at all, except to bore a hole through the side of the grease box to take the spindle.

At the top of a hill, a crown pulley was placed to take the additional strain applied by the rope to pulleys in this position. To take this weight, these pulleys were 2'6" in diameter on a 2" spindle set.

At termini the rope was went around a 12' horizontal sheave to change direction, located in a terminal pit constructed to allow simple operation and maintenance. To bring it into alignment with the new slot, it was deflected into position by a 6' diameter sheave.

At rectangular crossings with other routes such as occurred at several locations, trams on one route carried the rope through the crossings, while on the other route it was necessary for trams to throw the rope and cross by momentum, and to pick up the rope on the other side. This was required as otherwise the grip of the crossing tram would foul the rope on the other route.

Therefore, the rope on this route was forced under the other rope by a depression pulley, and pass on to an elevating wheel which would guide the rope back into the jaws of the open grip, allowing the gripman to close the grip and proceed. There would be a ‘stop bar’ placed in the tunnel at some distance from the throw mark to protect the depression gear from damage inflicted by the grip. The rope passed on the opposite side of the stop bar to the grip. Should the gripman have failed to throw the rope, the rope would pull the grip into the stop bar, breaking the hold the grip had on the rope, and more often than not, breaking the grip as well.

An example of this was in Bourke St where trams had to throw the rope when crossing the Elizabeth St routes. Here the depression gear performed an additional function, preventing the rope from rising in the tunnel due to the sudden sharp climb up Bourke St towards Queen St.

In places where the rope had to be picked up, usually it was arranged so that it could be done automatically, but in some cases this was not possible. In these instances, the rope had to be raised manually. This was done by the conductor, who pulled a wire mounted at the side of the track, which raised a conical pulley, lifting the rope into the jaws of the open grip.

The rope could also be guided into the jaws of the grip by means of a hook carried on the dummy, which was inserted into the slot and used to manually lift the moving rope. This task required both strength and knack on the part of the gripman.

The longest period of a cable in service was experienced on the Fitzroy line, when a cable was removed in 1912 after 4 years and 7 weeks service, covering a total distance of 286,185 miles.

Cable Maintenance

Ropes were made from what was known as ‘special acid steel’ of highest quality picture available, having a large admixture of Swedish material. The construction of the rope was six strands of seven wires over a medium hard laid up manilla hempen core. The strands of the riots were of 1 3/8" circumference, 4½" lay, 6 wires of 0.142" diameter over a soft iron core 0.144" diameter rope core of manilla hemp, usually about 2" circumference, and 2" lay-weight of rope approximately 2.6 lbs per foot.

The lengths of ropes varied from 17,000 to 30,000 feet, the latter weighing approximately 35 tons, and breaking strain of 66-68 tons psi. It was found that ropes with a breaking strain of 69-70 tons did not wear as well, and it was apparently difficult for makers to keep down the tensile strength and still maintain the quality of the wire.

All wires in the rope were either brazed or welded, so that the wire was in one continuous piece for the entire length of the rope.

When a new rope was run into the road, it was filled with a composition known as ‘rope filling’ to keep the water out. The rope filling consisted of a mixture of Stockholm tar, sludge oil and plumbag in the ratio 6:4:1. The mixture was boiled in a copper by engine house staff and applied to the rope warm. A container at floor level fitted with cock and pipe leading down to the out-going rope held the rope filling, which was permitted to trickle on to the outgoing rope 3 or 4 times per day until the rope was filled, generally taking about 3 weeks to completion. Rope oil was then applied when required, usually once or twice per day.

Rope filling must be liquid enough to run into the innermost spaces between the strands, it must set but remain plastic enough that it does not crack and fall out when the rope is being bent around curves or sheaves. It must be waterproof and of a lubricating nature. Keeping the water out of a rope is an important factor on its working life.

It was the practice to use new ropes in city traffic and then take them out before they were worn and put them in suburban sections; this resulted in a longer average life and less risk of stoppages due to rope trouble.

Two factors operate against the life of the rope: wear by grips, and wear by curves. Grips wear the wires and make them thin in the crown, and curves by alternate bending and straightening break the wires. By far the more potent factor is the curves. The majority of ropes were discarded through broken wires and not because they were worn in the sense that the grip wears the rope down.

Situated in the engine house pit, on the incoming rope, was a device known as a ‘strand alarm’. Should there be a protruding wire or broken strand on the rope, it would strike this device and close an electrical contact, ringing an alarm in the engine house giving the engine driver time to slow down the engines and the ropeman time to examine the rope. Quite often, this alarm was rung by pieces of rag which small boys used to delight in dangling down the slot until they were caught by the rope, often with an empty can attached.

The stop bar was a mild steel bar of some 3" diameter placed in the cable tunnel at various places in the network where it was necessary for the gripman to throw the rope. The objective of the stop bar was to break the grip and release the cable in the event that a gripman forgot to release the grip manually. This protected the rope from damage, as well as protecting various types of underground gear such as sheaves and depression gear. It was always more cost effective to break the grip rather than to repair the rope or the underground gear.

When a rope was overcarried by the gripman, it was always kinked to a greater or lesser extent. In a severe case, it must be cut and re-spliced, which required the service to be stopped for about an hour. If only slightly kinked, the rope would only be slowed in order for the ropeman to view the damage. If a strand is cut out and bunched, stoppages may be up to 4 hours duration.

When a gripman failed to or was late in throwing the rope, he was supposed to notify the engine house by using the alarm signal box, when all would be on alert in the engine house for the potentially damaged section of rope to come into the engine house. Unfortunately, this was not always done, and the first indication of a problem to the engine house was the ringing of the strand alarm.

A bunched strand, the most severe type of damage, if of any magnitude always broke some of the yokes carrying the line pulleys and always displaced the line pulley drum. Substantial wooden frames were kept at the engine house to drop into the tunnel to carry temporary pulleys until such time as the yokes could be replaced. These yokes differed slightly from the original ones in that they were made from two pieces bolted together, and were not bolted to the main tunnel yokes which dropped over the head of the yoke. Ends of varying length were kept in stock to meet all circumstances.

There was a ‘throw rope’ marker in Gisborne St opposite the Eastern Hill Fire Brigade HQ where the Collins St rope had to be thrown and an auxiliary rope picked up. As this ‘throw rope’ marker was away from any junction crossing, it was often missed by gripmen. Therefore, a warning gong was placed in the tunnel as a reminder to gripmen to throw the rope. It was operated by the grip moving a ‘kicker’ across the slot which operated a striker, hitting the gong. This provided a very audible reminder to gripmen to throw the rope, and avoid damage to the grip, rope and underground gear.

Signal boxes were placed all around the road at selected places and inside a dial with such notations as:

• Stop rope;
• Start rope;
• Run slow;
• Track blocked;
• Fire hose across track;
• Grips broken and wedged; and
• Send help.

A member of the tram crew went to the signal box, set the pointer and depressed a trigger. This sent a signal to the engine house, which was recorded on a telegraphic tape and indicated the message and station from which it was sent. For a stop signal, it was necessary only to depress the trigger without setting the dial. By means of a telephone from any of these signal boxes it was possible to talk to the engine house.

Varieties of kinks and strands were endless and it is safe to say that no two were exactly alike, nor was the method of dealing any two exactly the same – hence the necessity of someone of experience and authority being quickly on the spot. The worst case is when a strand is bunched and the rope parts. When this occurs, the facing end flies back in the tunnel, and as the trams are moving forward, loops form around the first few grips.

All these loops have to be freed from round the grip and slack got up in the roadway before anything can be done towards pulling the ends together preparatory to splicing. The difficulty will be appreciated when it is understood that there is only a manhole every ½ chain and the tunnel is not big enough for an average man to crawl through.

There were four basic methods of dealing with bunched strands, depending on the particular circumstances of the case:

• The bunched strand may be got up on the road and cut off;
• The bunched strand may be taken to a pit and cut off;
• The rope may be cut, removing the bunch, and the rope spliced; or
• The rope may be cut at each end of the bunch and a new section of rope spliced into the rope.

Ropeman splicing a cable.

Photograph courtesy City of Yarra.

A splice put in a new rope is approximately 80' long, and in a new rope it is most difficult even for experienced men to find the splice until the rope is filled and some wear showing on the wires. It usually took seven men about 2 hours to splice a rope.

The record time for a splice was 23 minutes, from stopping of engines to restarting, including putting on clamps, slackening tension, making the splice, taking up tension and taking off clamps. This occurred when it was found necessary to shorten a rope during traffic hours. Plenty of men were available for this record task, which was supervised by James Turnbull and William R. Pollock.

To splice a cable, as per Figure 9, each end is unlaid for approximately 40 feet and the hempen core cut off leaving about a foot. Every alternate strand, three in each end, is cut off leaving about one foot. The ends of the rope are then married, or brought together so that a long strand from one rope is alongside a short strand from the other end, and so on for all the strands. The ends of the rope are then pulled together until the lay of each end of the rope matches. A short strand from one end is then unlaid and the corresponding long end laid into the vacant space. This strand is run out to the end and the two strands crossed, leaving about 4'6" ends. While this is being done, another ropeman would be running out a strand at the other end of the splice.

Figure 9. Rope splice.

From the Macmeikan report.

While these ends were being run out, they would be hammered into place with copper mallets. The other four strands (two each end) would be run out. Similarly, the second at each end would be stopped and crossed 9 feet in from the first one out. The third strand would probably be just run out 5 feet at each end. This would leave space between the 2nd and 3rd tucks and the 4th and 5th tucks. This space is left for running repairing strand into, if necessary, during the life of the rope.

The ends referred to are then parceled up with hessian to make them larger in diameter, but still less than the diameter of the hempen core. The twist is taken out of the strand at point of crossing, the strand flattened and crossed, and the end run into the centre of the rope, the rope meantime having been opened and hempen core cut, and pulled out between the strands of rope. As the end of strand is run into the centre of the rope, original hempen core is taken out with the result the rope at this place has a steel core. This is called a tuck and is done at each strand. The finished splice should be indistinguishable from the rest of the rope.

Footnote

[8] Builder of the Brooklyn Bridge in New York, United States.