If you can get the clock running without encountering, or creating, any sort of an unsafe situation, well and good. If not, the best procedure is to correct the unsafe situation first, whether you do it yourself, or whether the clock owner has someone come in and make repairs. It's a real thrill to bring one of these old clocks back to life, but, don't endanger the life of anything else--a building, the clock itself, or of any person in the vicinity of the clock. John Stutsman mentioned in his "Clockmaker's Corner" in a recent issue of the (NAWCC) BULLETIN, that state governments had given up on trying to set up and impose licensing requirements for clock repairers. Therefore, before you become a 'fool rushing in where professionals fear to tread', stop and think and plan and ask questions--in other words, license yourself. I would rather not have a bureaucrat looking over my shoulder every time I restore or re-oil a tower clock.
I've seen and repaired the handiwork of too many incompetents on too many tower clocks, much less household clocks. The ability to work on a tallcase clock does not qualify you to do similar work on a tower clock. Someone's life may literally be in your hands; please be careful.
Modern nomenclature of various accessories and mechanical contrivances used in hoisting and wire rope systems will be defined in appropriate sections in this article, and in the appended Glossary. Nomenclature with which we are already familiar will be used as much as possible to avoid the confusion of multiple terms which refer to the same object. New terminology will be proposed, and its use justified. Mathematical formulas and tables will be found in the appropriate appendices.
Weight Line Definitions and Nomenclature:
The literature is filled with references to systems with two lines at the weight referred to as doubly compounded, and systems with three lines at the weight referred to as triply compounded--on the other hand, within the pages of a single reference, I found identical weight systems referred to as "compounded once"; later, double compound: another referred to as "compounded twice"; later, triple compound. This lack of any kind of specific definition is confusing, and confusion isn't needed when you're going to be working in mid-air, so to speak, with 800 pounds of strike weight. Since a simple compound system already has two lines at the weight, double compound suggests four lines at the weight, and triple compound suggests anywhere from six to nine. The clockmaker working with tower clocks also needs to be able to 'speak the language' of the rigger and the crane operator, because these professionals daily deal with all of the components that make up the tower clock weight drive system. Their 'language' is no more difficult to learn and use than is the vocabulary you already use as a clockmaker--a deadbeat means something entirely different to you than it does to the 'man on the street.'
Those parts of a weight line directly supporting the weight, that run between the winding barrel (or a guide pulley3 at the top of the weight channel), and the weight, (or the pulley(s) attached to the weight), have no specific names or definitions in current horological dictionaries, merely those catch-as-catch-can terms, such as double-compound, etc., mentioned already. These specific parts of the weight line are used in calculating and defining the compounded vertical weight drop and the mechanical advantage of the system, and should be easily identifiable in terms easily understood by everyone who works with either hoisting or clock drive systems.
A fall, or a part of line, is defined as that section of the weight line between the weight, or the pulley(s) attached to the weight, and the winding barrel, or the guide pulley(s) at the top of the weight channel. The following nomenclature is proposed to define the degree of compounding, as well as to bring the horologist's terminology on weight lines into logical agreement with applicable modern standards and methods as much as possible:
a. with one line at the weight: single fall, or one part of line;
b. with two lines at the weight: double fall, or two parts of line;
c. with three lines at the weight: three falls, or three parts of line;
d. with four lines at the weight: four falls, or four parts of line;
and so on, up through however many falls, lines, or parts of line are found in the system. (Lord Grimthorpe wrote that a system with more than three falls has excessive inherent friction losses, and that the decrease in vertical weight drop is not sufficient to justify the increased friction, heavier weight required, and the much longer weight line of a system with more than three falls4.)
3 The common, or layman's, term for a sheave or block. The accepted technical terms for
the various parts of weight systems will be defined in the Glossary, but familiar terms, unless they are not deemed appropriate, will be used as much as possible throught this article. A moving pulley is referred to as a 'running block'.
4 The decrease in required weight fall varies as the reciprocal (1 divided by the number of falls) of the number of falls; in other words, with a required vertical weight drop of 72 feet for 8 days, a three fall (reciprocal of 3 = .3333) requires 24 feet of vertical space (72 x .3333 = 24). The difference between the two falls (36 foot drop) and 5 falls (14.4 foot vertical drop) is only 21.6 feet. The increases in friction, length of rope, and weight required do not justify the use of more than 3 falls excepting in unusual cases.
Figure 2. Common weight line suspension systems.
Note that only supporting weight lines, those running directly to and from the weight or the moving pulley(s) attached to the weight, are counted in specifying the number of falls or parts of line. A non-supporting weight line section, e.g., one which leads from the guide pulley at the top of the weight channel to the winding barrel, even though it transmits the effective force of the weight to the winding barrel, should not be counted as a fall. In short, the number of falls defines two different parameters. First, the total vertical weight drop (Tvd) required, multiplied by the reciprocal (1 divided by the number of falls) of the number of falls yields the compounded vertical drop (Cvd) of the weight line system. Second, the number of falls determines the mechanical advantage of the compounded weight line, which mechanical advantage is used by the crane operator to lift a heavier weight load with comparatively little effort. Unfortunately, the mechanical advantage acts against the clockmaker, becoming a mechanical disadvantage, so to speak, requiring him to use a heavier weight to achieve the required effective weight for the clock drive system5. It should be noted here that there is a definite difference in the application of clock and crane rope and weight line systems: the crane operator is interested in the short-term dynamics of safely lifting and moving widely varying weights through widely varying distances; the clockmaker is interested in the long-term dynamics of safely supporting a defined weight, and in safely controlling its rate of fall through a limited vertical space.
5 If the total required drop is 72 feet with a 50 pound weight, a two fall system requires half the vertical weight fall; (1/number of falls) = .5 x 72 = 36: but requires a 100 pound weight; 2 (falls) x 50 = 100. For more accurate calculation of the total weight necessary to drive the clock, see Appendix II/9.
The weight drives for household clocks can probably follow current convention and use the terms 'weight cord', and 'weight cable' as well as 'weight line', and 'pulley' rather than 'sheave' or 'block'. The latter two terms are used by the crane and derrick operator, and are considered the 'proper' technical terminology, but for the sake of convention and that we are already familiar with and use the term 'pulley' rather widely, pulley will be used throughout this article to avoid the possible confusion of three different words which refer to the same item.
As an introduction to tower clock weight systems, consider first some of the weight drives with which we are all familiar. The common O.G. has a weight cord which runs from the winding barrel over a guide pulley in the top of the case with a weight attached to the end of the weight line. The system in this case has a single fall, or single part of line supporting the weight, so the entire force of the weight is felt at the pulley and (less the friction losses) at the winding barrel.
Figure 3. Single fall weight line system used in a weight-driven shelf clock.
The total vertical weight drop (Tvd) and the compound vertical drop (Cvd) are equal. The mechanical advantage is 1. The O. G. clock, with its single fall, is designed to run (usually) for thirty hours before the weight bottoms out in the case and the clock stops. A compound, or multiple fall, system is used in the typical tall-case clock which is designed to run (usually) for 8 days.
For all that it might sometimes look like a tangled web of ropes and pulleys, a compound weight system is really a simple affair. For instance, the weight line in the tall-case just referred to is attached to the winding barrel at one end, and (usually) to the seat-board at the other, with the driving weight suspended between these two points on a pulley. The degree of compounding is 2 fall: the driving weight has two supporting parts of weight line. In a 2 fall weight system, the effective driving force for the clock is one half of the actual weight and the total drop of the driving weight is one half of what it would be if it were suspended as is the O.G. weight.
Figure 4. Typical tallcase clock weight line compounding.
The total vertical weight drop, [Tvd = 8 ft] required, multiplied by the inverse (1/# of falls; in this case, 1/2 = .5) of the number of falls yields the compounded vertical drop, [Cvd, = 4 ft] of the two-fall weight line system. Note that while the weight itself drops only four feet, eight feet of cable actually unwind from the winding barrel. The mechanical disadvantage is 2; each fall supports half of the total weight, therefore the weight has to be twice as heavy as is the weight used with a single fall. That's why a tall-case runs for 8 days with a four-foot weight drop, (but with a heavier weight) without a lot of extra wheels (as in a month or year clock) between the main (or great) wheel and the minute (or center) wheel. The O. G. weight line setup in an otherwise normal eight day clock would need a weight drop of 8 feet, with a weight half as heavy. Lighter weights driving tall-case clocks might save a lot of bottom boards in clock bases when the weight lines break, but we would have to have either very tall clocks (and very high ceilings) or more expensive clocks with more wheels--which would probably require heavier weights--which would...why not just compound the weight line system?
O.G. and tall-case weight systems are simple and easy to understand, as are the weight systems for most tower clocks, roughly 75% of which are 2 fall systems. But how does the clockmaker, faced with an unfamiliar tangle of weight lines, ponderous weights of half a ton or more, and more pulleys than Cap'n Ahab6 had on board the Pequod, deal with this seeming complexity? Common sense and some elementary math will untangle all your problems on weight drive systems. A small calculator will speed up your math a bit, but clockmakers had it all figured out
6 Cap'n Ahab would probably break out the cat o' nine tails if one of his sailors were to use the word 'pulley' on board ship. Every profession has its own vocabulary: on board ship a specialized vocabulary avoids confusion when confusion is the last thing needed. Since the term 'pulley' is familiar to the reader, the technical terms will be defined and explained, but will not necessarily be used in the body of the text.
several centuries ago. But, before you even lay a hand on the clock, do remember that tower clock weights and their weight lines (on which more later) will be handled, at all times, with respect, great caution, forethought, and, if necessary, advice or help from the professionals in a local rigging company. The weight lines, if they are under working tension, will always be handled in the same manner. Gravity might play tunes on a musical clock, but it does not play favorites.
The logic in using a compound weight system is to allow the clock a reasonable and pre-determined running time (with a reduced vertical weight drop) with a reasonable amount of weight. Assume that the clockmaker designed his tower clock to run for a week with a 50 pound weight dropping 72 feet, but the church in which the clock will be installed has only 25 feet of weight drop available. Several options will be considered. It would be possible to copy the O. G. setup, and suspend the weight on a single line from a guide pulley 70-odd feet up in the steeple, but the chances are that a weight thus suspended will invade otherwise inhabited areas in its descent, whether that invasion is in a controlled manner, or in free fall, which latter tends to put the inhabitants off their feed. With a single fall 25 foot weight drop, the custodian would have to wind the clock every 2 .88 days, but that would wear out ratchets double-time, and he would probably complain. A two fall rope system will double the running time to 5. 76 days, but will now need at least 50 feet of weight line, another pulley, at least a 100 pound weight, and the clock winder will probably still complain that he has too much work to do too often. Once a week seems to be enough for most folks, whether they're going to church, visiting the in-laws, or winding a clock.
7 A single line in a tower clock weight system will have two faults: the weight line will tend to unlay, unless a special (more expensive) type of line is used, and any small disturbance will tend to make the weight oscillate like a pendulum.
Figure 5. Typical three fall reeving diagram.
The clock winder will be much happier if a three fall system, with two compounding pulleys, and three falls at the weight, is used. Multiply 72 by .3333 (1/3 = .3333), and note that the clock now needs only 24 feet of fall to run a full week, with a bit left over.
However, at least a 150 pound weight is now needed to drive the clock, because there are three falls, and each one of those falls supports one third of the total weight--the mechanical advantage is 3--the clock is still being driven by an effective weight of 50 pounds. The addition of a few extra pounds to the driving weight will compensate for the frictional losses (see Appendix II/9) due to the two compounding pulleys. The clock, amazingly enough, is still nearly as easy to wind as it had been with a fifty pound weight (without any pulleys), because of the mechanical advantage of 3 provided by the
compounding of the weight line system. The clock winder will have to reel in 72 feet of line each week, and will also have to be fairly punctual about winding the clock, because that one foot of weight drop available after 7 days will only run the clock for another 8 hours8. If the custodian still complains that once-a-week-winding is too much work, we could install a 12 fall system, just to keep him quiet. The clock will now run a week with a vertical weight drop of only 6 feet (just over a month with 25 feet of vertical weight drop), but now needs a 600-plus pound weight, 300-plus feet of rope, and at least 11 pulleys. The cost and complexity would skyrocket out of all proportion to any advantage gained--even the winding barrel would have to be redesigned--and Lord Grimthorpe's ghost is always there to remind us that three falls are just right, the shortest weight fall with the least friction. The time required to wind the clock with a 12-fall system would be ridiculous--assume that the winder can spin the winding crank at 6 turns per minute, and that the winding barrel circumference is 2 feet. He will wind the clock for a solid 25 minutes once a month, as opposed to a mere 6 minutes each week with a three fall system.
Figure 6. Reeving diagram for a 12 fall weight line system.
A 12 fall weight line system can easily be installed, but somehow it seems to be a bit less than practical.9 The design of a maintaining power system which would run the clock for the 25 minutes needed to wind the clock would be a clockmaker's nightmare. Might as well put in an electric clock.
8 I recently worked on a tower clock that had a three-fall strike weight system and a minimal amount of weight fall space available. Winding the clock just after it struck 12 on Sunday meant that it had to be wound after striking 12 exactly 7 days later--there just wasn't enough fall left for the clock to strike 1, one hour later.
9 Ball bearing pulleys each contribute about 2% of the total friction losses in a weight system. A 12-fall system will require at a minimum, 12 pulleys. Friction losses would be on the order of 24%. A clock designed to run with a 50 pound weight rigged with a 12-fall system would now require a 600 pound (12 x 50) weight, plus 24% (144 pounds) for a total weight of 744 pounds. Bronze/brass bushed pulleys (5 to 5.5 % friction) would require a 930 pound weight in a 12-fall system. No allowance has been made in either of these examples to compensate for the loading of the outside dial hands due to weather, or for the losses due to bending losses in the weight line. Determining the correct weight to drive a tower clock is a totally empirical process, otherwise known as a SWAG (Scientific Wild Ass Guess). For a fuller discussion of determining frictional losses see Appendix II.
An odd method of compounding a weight line recently appeared in Tower Talk, the newsletter of Tower Clock Chapter 134, which was referred to as a 'winding aid'. To understand its operation, we have to make a couple of assumptions for the sake of clarity, as far as various diameters, circumferences, etc., are concerned. The clock winding barrel is two feet in circumference, the large barrel is 6 feet in circumference, and the small barrel is again 2 feet in circumference. Referring to the diagram on the right below, 114. 5 feet of cable will be required, but the clock will be able to run longer before the weight bottoms out. Total cable length depends on the physical separations of the clock winding barrel, the 'winding aid', and the pulley at the top of the weight channel.
Figure 7. A comparison of conventional compounding and an ingenious compounding method for a weight system.
The differential in cable lengths is due to the 'winding aids' ' requiring two separate lengths of cable, while normal compounding can use a single length of cable. The 'winding aid' was probably used because it would provide for a slightly longer running time, and would avoid the friction and cost associated with the three (extra) pulleys needed for conventional 3-fall compounding. Disregarding friction, and the half-diameters of the wire rope, this assumed 3:1 ratio will reduce the force felt at the clock winding barrel to a third of the actual weight, which will require that the weight be increased by a factor of 3. In this case, to reduce the weight drop required, 2 fall compounding would require a doubling of the weight, and would yield a total weight drop of about 22 .4 feet. Three fall compounding would yield a total weight drop of almost 15 feet and the driving weight would have to be three times as heavy.
Whoever designed this system had his head on his shoulders. He realized that three fall compounding was out of the question because of the limited weight drop available, and came up with an ingenious solution that used a little more rope, and achieved the same end, with less friction and expense for the extra pulleys. The need for a very short weight drop is, I suspect, the reason this odd system was used. Very simply, it reduces the overall complexity of a three-fall weight system.
Multiple fall, or compound, tower clock weight systems are simple in concept--and used the world over--but must always be treated with caution. Compound systems provide the clock with a mechanical advantage which allows it to run and strike for an extended time with a reduced weight drop, yet require a much heavier weight. The conveniences of the shorter weight drop and an extended running time only come at the expense of the heavier weight, a longer, more expensive weight line, more pulleys, increased friction, and a tired, irritated, clock winder--there ain't no such thing as a free lunch.
Pulleys are used in a weight drive system to make the rope load uniform throughout the system. A pulley is also known as a sheave, a block (a pulley or multiple pulleys housed in a common frame), or a grooved wheel, and is critically essential to the proper and safe operation of a weight-driven clock. Unfortunately, pulleys are usually ignored. The general tendency of most clock owners is to re-use what the clockmaker put in the steeple--sometimes as much as 150 years ago. Chances are that the original cable was a natural fiber rope, and the clock owner ends up putting in a "longer-lived and safer" wire rope weight line system with cheap, inadequate, and unsafe pulleys. Whatever the case, be very careful if you (re)use an old wooden pulley for a wire rope--and caution the owner about the possible problems. Wire rope on wood may work--for a while--but the wire will very soon begin to cut into the wood, the pulley will wear badly out of round (because of the grain), the thin flanges of the groove may break off, the line may then slip off the pulley, and will either be seriously damaged (probably kinked), or may do some other damage. A kinked wire rope is an unsafe rope, and has to be replaced. A wire rope is safest running on a metal pulley, either of cast steel or of cast-iron.
Figure 8. The localized abrasions caused by running a kinked wire rope. The kink is due to mishandling or incorrect installation.
Illustration courtesy of Williamsport Wirerope Works, Inc.
A pulley is defined by its shaft diameter, the rope diameter for which it is grooved, the outside diameter of the flanges, tread diameter (the diameter to the base of the rope groove) and its pitch diameter. Pitch diameter is the diameter to the center of the rope on the pulley, i.e., one half the rope diameter plus tread diameter. Tread diameter is the most critical parameter to be considered in the selection of a pulley for a wire rope (steel cable). The ratio of tread diameter to rope diameter is referred to as a D/d ratio. Optimal pitch diameter of a pulley is determined by the ratio D = X x d, (usually written D/d) where D is the diameter of the pulley, X is a Federal specification/rope manufacturer10 recommended constant, and d is the diameter of the rope. For instance, a given D/d ratio of 45, for a 1/4 , 6 x 18, fiber core rope, requires that an 11 .25 inch diameter pulley must be used for optimal rope life. The service life of a 1/4 " 6 x 7 (much stiffer) rope would drop by about 55% if it were used on a pulley this small. Wire rope bent 180° around a pulley is stressed or fatigued each time it passes around the pulley. The smaller the pulley diameter, the higher the stress or fatigue factor, which has a direct bearing on rope life, rope strength, and on the safety of the system. (See Appendix II/4) For this reason, American codes specify fixed D/d ratios which are directly linked to rope size, rope construction, and apply to both pulley and winding barrel diameters. The recommended size for the groove in a pulley is 1/64 of an inch wider than the nominal diameter of the rope. Too narrow a pulley groove will prevent proper seating of the rope in the bottom of the groove, and the uneven load distribution will damage the rope.
Figure 9. A. Illustrates a worn rope in a worn groove. B. Illustrates a new rope in a worn groove.
C. Illustrates a new rope in a new sheave.
Illustration courtesy of Williamsport Wirerope Works, Inc.
10 Federal Specification RR - R - 571a specifies minimum drum and pulley diameters in relation to rope diameters. Safe and proper operation of cranes and derricks and other systems using wire rope systems are governed by Federal, American Natonal Standards Institute (ANSI) amd manufacturer standards, from which the OSHA standards are derived, and which are the basis for this paper.
Too wide a groove will not provide adequate side support, and the rope will tend to flatten. Specifications for crane pulleys require that the grooves be smooth and free from surface defects, that a retainer (guard pin) be fitted if the rope system might be momentarily unloaded (as with the weights all the way down), and that pulleys be equipped with a means to lubricate the bearing. Alignment of the pulleys in the system must be checked, to avoid excessive wear of the rope and the pulley flanges.
Friction losses (See Appendix II/9) due to the pulleys can seriously affect a tower clock weight line system--losses with common bronze-bushed pulleys are on the order of 4 .5 to 5 .5 per cent., and those from precision ball or roller bearing pulleys about 1 .5 to 2 per cent. Total friction losses are affected by the construction of the rope, pulley bearing types, and the ratio of pulley-to-rope diameter. Imagine how much drag is created when the rope installer, trying to save a few dollars, reuses the old unbushed, too-small, wooden pulleys installed in the steeple and over the weight chutes 150 years ago. A great deal of weight has to be added to overcome the extra drag. Incidentally, the total lack of any safety considerations in the reuse of some wooden pulleys with wire ropes is appalling. I have removed badly cut, worn and wobbly wooden pulleys with the groove flanges broken away, still at work 150 years after the clock was installed, with several hundred pounds of stone looking very much like the Sword of Damocles. However, it is a shame to remove the original wood pulleys from a two-hundred year old clock--if the pulley has an adequate D/d ratio. If the groove is in good condition, and the pulley is rebushed correctly--the pulley can probably be safely used--for a while. If this is done, make sure that the custodian (or you) checks these pulleys periodically to ensure that they remain in good condition. It is possible (indeed, recommended, if the pulley is to be retained) to fit the groove of some of these wood pulleys with a suitably formed sheet metal hoop, to avoid the cutting of the wood by the wire. The wooden pulley should only be reused if it is in good condition, and its radial loading (See Appendix II/6) will not exceed 200 pounds. There is no specific data available on radial loading limits for wood pulleys; the above is an empirical value derived from experience with several weight systems using wooden pulleys.
Radial loading of a pulley should be limited to 500 psi for cast-iron, and 900 psi for cast steel11. Pulleys may have a bushing of bronze (high friction) or ball/roller bearings (low friction) and should have some easily accessible means of lubricating the bushing or bearing. Lord Grimthorpe (and, incidentally, the Howard instruction sheet posted in many clock rooms) recommend lubricating the pulleys annually. Personal experience suggests that once every six months is better. Pulleys and blocks should have a guard installed to keep the rope in the groove, in the form of a pin or a bolt through the cheek-pieces which will just clear the rope when it is seated firmly in the groove. The purpose of the guard pin is to keep the rope in place if the rope is completely unloaded.
Under no circumstances should you ever use a cheap pulley made up of two separate stamped pieces of steel spot-welded together, or one with a formed sheet metal rim (sometimes called a gin pulley). An inadvertent shock load can split the former, and can seriously deform the latter. Do not use the typical hardware store pulley--they are too small, and they will break. It is not an afternoon's light entertainment to wrestle with a 300 pound weight jammed three quarters of the way up the weight chute because the rope slipped and broke a pulley while you were working on the weight system. If you find yourself in a situation such as this, unless you have the tools and the expertise, it is best to call in the riggers. It's also a safe bet that a mistake like this will only happen to you once!