A version of this article appeared in the July/August 2020 issue of Good Old Boat Magazine.
It might have been an omen or a couple of them that drove me to inspect the 26-year-old Harken Traveler Car on Second Star, a Sabre 362. A young sailor documented the repair of a similar traveler on YouTube and a week later a fellow Sabre sailor experienced the destruction of his car during an accidental gybe. While not normally superstitious, I try to be prudent and inspected my traveler car for wear.
The traveler car is an item that can get taken for granted. The common complaint is difficulty moving the car under load; so long as it rolls easily on the track, all is good. However, lurking beneath the lower mainsheet block are two shackles that silently grind away at each other eroding the stainless steel. Eventually the shackles reach a breaking point during a jibe or tack in heavy weather, resulting in a wildly flailing boom.
The Harken traveler car is a simple design, consisting of a 5/16” shackle attached to the traveler with an interlocking ¼” shackle attached to the lower mainsheet block. Nominally 5/16”, the larger shackle measures .315 inches and the smaller shackle .250 inches. Inspecting the shackles revealed significant wear with a reduction in diameter and strength. At the narrowest weakest point, the larger shackle measured .279 inches, an 11% decrease in size and strength. The smaller shackle did not fare as well measuring in .180 inches, a decrease of 28% in size and strength. Closer inspection revealed another weak point, the pin holes in the larger shackle; the holes showed elongation with a thinning of the sides. This is particularly troublesome as the holes are not visible without disassembling the traveler car.
How serious is this wear? Quite serious. Let’s do the math. The small shackle has a working load limit (WLL) of 320 kg or about 700 pounds. A diameter reduction of 25% reduces the WLL by 175 pounds to 525 pounds. If the design load for traveler is 350 pounds or 50% of the WLL that same load is now 67% of the WLL of the worn shackle. Put another way, the safety margin has been reduced by half, instead of a 350-pound margin, the margin is now 175 pounds. Some might say this is an accident waiting to happen.
The cause of the wear is pretty obvious, movement of the shackles while sailing and at the dock or mooring. When the boom swings back and forth the shackles bear on each other and wear. Salt, dirt, and grit all compound the problem by adding an abrasive element. Completely eliminating wear is unlikely, however it can be reduced by stabilizing the boom at the dock. Securing the boom in two places will reduce swinging and subsequent wear. Likewise, when sailing in sloppy conditions, reducing boom swing will decrease wear. Adding a sacrificial cover to the shackles might also work, however, this repair is simple, the cost minimal (about $30), and the life span pretty long, in my case 26 years.
Accessing and removing the traveler car shackle is not immediately obvious. The first inclination might be to remove the traveler car from the track. This would be ill advised. The old-style Harken traveler cars have loose ball bearings and require a special tool or a piece of track to remove the car without scattering ball bearings across the deck. While these instructions are specific to Harken cars, other manufacturers may use similar methods, explore those options before removing the car from the track.
The car shackle is held in place by a stainless-steel rod captured inside the traveler car. Accessing the rod is relatively simple. Remove the turning blocks to reveal two set screws. These screws prevent the rod from sliding out of the car. Remove the set screws and insert a screwdriver into a hole on the end of the car and push the rod out. This frees the shackle.
Before installing the new shackle, a small modification is necessary. The threaded side of the shackle has to be drilled out to remove the threads. The unthreaded side of the shackle is 5/16 in diameter; however, the screw threads reduce the hole’s diameter. A drill press makes quick work of this.
Changing out the mainsheet block shackle is straightforward, unscrew the shackle pin and attach the new one. Note the attachment hole on block will not accommodate captive pin shackles.
Reassemble the traveler by inserting the shackle and pin, reinstall the set screws, and reattach the turning blocks. Applying an anti-seize lubricant to the screw threads will make future disassembly easier.
Removing six screws and 10 minutes is all it takes, and traveler is good to go for another 25 years.
Sidebar: By the Numbers
After submitting the accompanying article, the editor asked, “Are you sure about your numbers?” I confidently replied, “Of course I am.” Then I looked for a sure-fire way of confirming my numbers.
The wear on the shackle effectively reduced the diameter of the shackle. The small shackle was nominally a ¼” shackle which had worn from .2450 inches to .1750 inches, which is a little smaller than a 3/16” (.1875) shackle. How do the mean breaking loads (MBL) compare? Would the comparison yield similar results to my back of the napkin calculations? Using information available on the Hayn website, a standard 3/16” D shackle has a MBL of 3300 pounds and a ¼” shackle has a MBL of 4290 pounds. Stated another way, the smaller shackle has 77% of the strength of the larger shackle, or in the case of wear, a 23% decrease in strength. This is very close to the calculations made in the article.
Having piqued my curiosity, I looked the MBL for larger shackles; would the same relationship hold, a 25% reduction in size yield a 25% reduction strength? The short answer is no, there is a greater reduction in strength for a similar percentage size reduction. For example, a ½” shackle has a MBL of 16500 pounds, a 3/8” shackle (25% smaller) weighs in with a MBL of 10560, a 36% strength reduction. The larger the shackle the more pronounced the difference, a 5/8 inch compared to 7/16 (30% size reduction) showed a 57% decline.
If I was a materials engineer, I could no doubt give a lengthy explanation, the simple answer is the strength of the shackle is due in part to the cross-sectional area which is, of course, related to its diameter in a nonlinear way. One way to see the relationships between diameter, cross sectional area, and MBL is compute correlations. The correlation between diameter and MBL is strong at .967, however, the correlation between MBL and cross-sectional area is stronger at a near perfect .991. These figures suggest calculating loss of strength is more accurate using cross-section area, however, changes in the diameter are a good proxy and yield results within a few percentage points. Not to mention, they are easier to calculate.
Other factors can contribute to strength reduction besides metal loss due to wear. Stressing metal with heat, excessive loads, bending, and impact can contribute to metal fatigue and failure. The discussion here is relevant to all metal to metal bearing surfaces, such as sheave axles, clevis pins, halyard shackles, and anchor chain to name a few. Even an apparently small reduction in diameter due to wear can lead to significant reductions in strength.
So, do not ignore metal to metal bearing surfaces, inspect them regularly.