Introduction

Prusiking, the ascending of a single static rope, has reached a very high standard of sophistication particularly in America. There, prusiking is often treated, quite justifiably I think, as a sport in its own right, independent of caving. Prusik races seem to form a major pastime at caving conventions, and many people have devoted a lot of time to the development of fast racing systems or 'rigs'. These incorporate some incredibly subtle innovations. Unfortunately, with the races being held on quite short continuous loops of rope run over pulleys, and with the various individual prusik rigs having the inevitable 'trade off' of safety for speed, it is sometimes difficult to see the link between prusik racing and prusiking underground. Obviously it is better if you can prusik GGMS safely in ten minutes rather than thirty, but speed should not be the ultimate aim of any caver ascending a rope. Because of the need for the widest possible safety margins, along with the need for relatively simple systems, many of the innovations used in prusik racing have little relevance to caving. An exception to this, however, is the floating ascender.

A few years ago prusiking speed records were held by people using the standard ropewalking rig - fixed ankle ropewalkers, fixed knee ropewalker, and a chest mounted ascender. Kirk Macgregor then developed the floating cam. This was a length of elastic from the shoulder to the knee ropewalker, as shown in fig. 1. The knee strap could then be discarded and the ropewalker lifted by contraction of the elastic. The application of this method to racing was immediately appreciated and Macgregor stormed to both the world one hundred and four hundred foot prusiking records. At the time of writing, his time over four hundred feet (and I understand that it is somewhere around the four and a half minute mark) still holds the world record. His one hundred foot time however, has since been beaten by Bill Stone of Pittsburgh, Pennsylvania, also using a floating Gibbs system, who recorded the incredible time of 29.8 seconds (1, 2).

Along with this undeniable suitability for speed, the floating cam has other subtle advantages that now make it distinctly preferable to the fixed knee ropewalker, for both racing and caving. I shall outline these advantages later. Because of the floating cam's historical context the discussion so far has involved only ropewalkers. However for caving rigs , the floating of sprung cam ascenders too, has advantages over its hand-held counterpart. As far as I can see, floating cams have an application to most short-step prusik methods (3). Despite this and the floating cam's general acceptance in America, employers of the method in Britain seem sparse. This is possibly due to the fact that, while the idea of the floating ascender is essentially simple in theory, in practice there are several subtleties to the method which, if unappreciated, can cause problems. The solution to these problems is unlikely to be found in any of the literature commonly available in Britain with the possible exception of Montgomery's new book (2). Even this contains a few ideas about floating ascenders that differ from Macgregor's original arrangements. I shall now explain the setting up of a floating cam, under some appropriate headings.

Position of Upper Attachment of Elastic

The upper attachment can be to the shoulder as shown in fig. 1, or alternatively to the waist, either in front, or at the back (having been passed over the shoulder) (2, 4). Consequently various lengths of elastic are possible. The attachment to the front of the waist is simplest and possibly the most convenient, but for many systems, the resulting length of elastic will not be strong enough to raise the ascender. The shoulder attachment involves a longer length of elastic. It is still convenient and is probably the most common attachment point used. When a chest harness is used a choice of shoulders is possible. With a ropewalker rig, Halliday shows an attachment to the shoulder that is on the opposite side of the body to the foot using the floating Gibbs (5). Conversely, Macgregor attaches his elastic to the shoulder corresponding to the foot using the floating Gibbs. With chest harness based ropewalking systems the choice of shoulders seems theoretically arbitrary, and the user should experiment to find his 'best side'. With other rigs such as the Mitchell method - chest harness with standing rope run under it short sling to one foot, longer sling run through the harness to the other foot (fig. 2) - and its variations (for example, APS after Cowlishaw (6)), the lower ascender can be floated, and an attachment to the corresponding shoulder is preferable. An attachment to the opposite shoulder would mean the elastic crossing the upper ascender sling.

Passing the elastic over the shoulder and down to the back of the sit harness is cumbersome, but with some elastic materials it is the only way to obtain sufficient tension. If a back support strap is used (as in fig. 1) and not a chest harness, this method of attachment may be the only practical alternative to the upper ascender shoulder attachment shown.

Lower Attachment of Elastic

Sprung cam ascenders are much simpler to float than ropewalkers. Of the makes commonly available - Jumar, Petzl, and Clog - all have krab holes conveniently placed at the top of them. This allows for simple attachment methods (see below) . The only important point is that with the Petzl ascenders, only one of the two top holes should be used. If both are used then the ascender will have to be 'defloated' in order to remove it from the rope - an unnecessary procedure.

When deciding whether or not to float a sprung cam ascender, one has to decide whether the freeing of a hand, the main advantage, merits the increase in complexity caused by the use of the elastic, and the fact that the lower sling may have to be shortened in order to give a sufficient length of elastic. With the Mitchell slings that I usually use, I have got the elastic tension such that there is no need to shorten the lower sling.

Ropewalkers have no single obvious point of attachment and several methods have been proposed. All have their various pros and cons, but a short discussion of the ultimate aims behind floating the ropewalker will soon elucidate the better methods. The main point behind floating a ropewalker for caving, I think, is the spring loading of the cam. Obviously, at least to prusikers, the convenience of the lifting mechanism is a motivation for floating a ropewalker, but I believe the elimination of the 'flicking' or other unfortunate motions normally required to make the ropewalker bite, is more important. Another good point about a well-floated ropewalker is that the cam is held nicely positioned and angled in space even when the ropewalker is in pieces. This facilitates a quicker operation at re-belays or other circumstances where the ropewalker must be removed from the standing rope and then replaced. The only other important consideration when floating ropewalkers is the possible increase in cam wear caused by some elastic attachment positions (4). This arises if the elastic lifts the body of the ascender and not the cam directly. If the body is lifted, then the drag supplied by the weight of the foot on the rising ascender pulls the cam into the rope. With fixed-knee ropewalkers, the cam falls away from the rope during lifting, and so cam wear is not generally a problem. With alloy cams, wear could occur quite quickly, and so well floated cams should avoid this. This can be done by having the cam critically pivoted by the elastic, such that the cam only contacts with the rope when pressure is applied by the foot, and not during the lifting. The intricacies needed to find this ideal pivot point in practice are the main disadvantages of some of the attachment positions outlined below.

Fig. 3a shows an attachment illustrated by both Halliday (5) and Montgomery (2). It would be quite easy to set up and consists essentially of a metal plate with two holes drilled in it. The elastic is tied to one hole, and the other hole is placed over the pin of the assembled ropewalker, and secured there by the wire that retains the pin. With quick release Gibbs the plate would have to be placed on the other side of the ropewalker ie. the side where the pin enters. I have not tried this method. It would certainly spring the ascender, but would also almost certainly increase the cam wear, perhaps critically. I think that from a theoretical viewpoint the method is unsound, because of the need to defloat the ascender in order to dismantle it. As mentioned earlier, this is an unnecessary procedure, and should be avoided if possible.

Fig. 3b shows the elastic connected to the body of the ropewalker (4). Some Gibbs have a spare hole near the top of the body suitable for this type of attachment. Alternatively, the elastic can be fixed to the wire fastening. The method again guarantees to spring the ropewalker, but since the lifting is done via the body and the pin, an increase in cam wear is also likely. It is not necessary to defloat on dismantling the ropewalker though, and this is a strong point in its favour.

Fig. 3c shows the elastic tied through the hole in the cam. This is unlikely to increase cam wear, and should nicely hold the cam poised while the ropewalker is dismantled. Unfortunately, the spring loading of the ascender cannot be guaranteed. Indeed, when I tried this method, I sometimes had to give more of a flick to make the ropewalker bite than is normally required with a fixed-knee cam. This is an overwhelmingly serious disadvantage.

Fig. 3d shows another direct attachment to the cam (2, 4). The elastic acts around the middle of the cam near its centre of gravity. If the attachment is chosen carefully, the cam is sprung-loaded and yet held away from the rope during the raising of the ascender. There would consequently be little or no increase in cam wear. Also, on dismantling the ropewalker, the cam is held nicely poised to accept the other parts. This attachment is awkward to attain correctly, and I had to hook the elastic onto a short length of thin cord that I put encircling the cam. Alternatively, a hook placed through a loop of cord that is attached to the strap that joins the cam to the body, may act in the right place (see also 'Mechanisms of Attachment below). Once achieved, I found the arangement shown in fig. 3d to be the best ropewalker attachment position out of those that I tried.

As an aside, an ordinary fixed ropewalker can be sprung by hooking a rubber band over one end of the protruding pin of the assembled ropewalker, passing it up and over the cam, and securing its other end to the other end of the pin (4). I would not recommend this for all ascents underground as the method probably results in a serious increase in cam wear. However for the sake of carrying a couple of rubber bands underground, a ropewalking caver could just use them when he is feeling tired and is missing a lot of bites.

Mechanisms of Attachment

The actual mechanism of the attachment of the elastic (at both its upper and lower ends) is not critical. Natural ingenuity should enable a caver to put together some kind of hook and eye arrangement, hinted at earlier. A more permanent attachment such as actually tying the elastic to the ascender is possible, but is less suitable if the elastic has to be replaced by a spare when underground. For one of my efforts I managed to adapt an old belt fastening. More secure is a ring and snaplink-type of arrangement. Whatever method is used, it should be such that it is unlikely to unfasten accidently, but it should not be too difficult to couple up and unfasten under adverse conditions. Some of the many possibilities of elastic attachment mechanisms are illustrated as a secondary feature of fig. 3. Some of the features of the ropewalkers have been omitted for clarity. (Only 3a is exclusively for a lower attachment - see earlier for a description of this.)

The Elastic

The elastic material itself should be strong and durable. A tension of about eight to ten pounds force when the elastic is at its full working extension is all that is required (4). I have found this quite easy to measure using a balance of the sort used by fishermen. It may not be necessary to go to this trouble, however, as the required tension is not difficult to judge. The elastic should be of such a length that it comes just slack when the leg is raised well above that involved in normal prusiking. This ensures more uniform lifting of the ascender. The position of the upper attachment may have to be varied in order to obtain the right tension and slack threshold.

For materials, ordinary rubber bands may be tied in parallel and end to end (4). Although this allows for tension and length variations to be made quite easily, the bands are not very durable. Some sorts of rubber lubricants may preserve the bands, and prevent them from breaking so easily. Both Halliday (5) and Montgomery (2) mention thick surgical tubing - despite its vulnerability to cuts, and I have even seen inner tube rubber used with some success. The best material that I have come across is shock cord, 5-7mm thickness being the most apropriate (2, 4, 5). Some varieties do expose the rubber through the sheath when extended, however, and this type would be of little use in caving where abrasion of the rubber may occur.

If you have been inspired to try rigging a floating cam system, then a few words of caution. Obviously practise on the surface first, and have some provision for the elastic failing. If it does fail, it could result in a black eye, which is inconvenient. With ropewalkers carry a spare elastic or, if the sling length permits, a knee strap. With sprung cam ascenders a spare elastic is only necessary if the ascender is effectively out of arm's reach. When it is necessary to defloat the ascender, unhook the elastic from the upper end first. The other way could be dangerous if the elastic slips out of the hand.

In conclusion therefore, the decision of whether or not to float an ascender depends on whether the advantages outweigh the complexities of the device. With the sprung cam ascenders of the short step prusik methods, I think that it is worthwhile, provided that the elastic can be rigged without having to change any of the existing sling lengths. As for ropewalkers, I do not personally use them underground. However the growing concensus of opinion amongst those that have been using the modification, seems to be that floating ropewalkers are far better than fixed ones in every respect.

Nick Thorne

References:
(1) Descent 33 p 36 - 'Gibbs SRT record'
(2) Montgomery, N. R. - 'Single Rope Techniques - A Guide for Vertical Cavers'. A complete work on SRT of which pp 86-88 are relevant here.
(3) Thrun, R. - 'Prusiking' - pp 38-43 contains descriptions of some short step prusik methods.
(4) Macgregor, K. - Personal communication.
(5) Halliday, W. R. - 'American Caves and Caving' - a floating Gibbs system is described on pp 217-219.
(6) Descent 36 pp 39-42. Cowlishaw, M. - 'Another Prusiking System'