Bow and fletching, arrow speed: Lepers 2020 study

Two experimental archaeologists at the University of Liège used two chronographs, 1 and 6 metres from the bow, to measure how much speed and energy wooden arrows lose when only the bow, or only the fletching, is changed. The first comparison matched a bow made from a single stave of elm against a modern fibreglass-laminated bow of the same straight shape, both drawing 34 pounds at 23.5 inches. From the wooden bow the arrows left 13.90 to 14.25% slower and with about 26% less energy: for the lighter arrow, 36.31 m/s against 42.17 and 16.91 joules against 22.82. With both bows the heavier arrow (30.35 grams against 25.66) came off about 2 m/s slower but carrying more energy. Between the two wooden bows, at 29.5 inches, the 55-pound yew longbow launched the same arrow at 48.99 m/s against 43.69 from the 48-pound flatbow, but over the 5 metres between the chronographs it lost 9.42% of its energy against 4.65%; it still reached the second chronograph faster.

The second experiment always used the flatbow at 29.5 inches, with 17 arrows that differed only in their fletching. The bare shaft was the fastest off the bow (44.93 m/s) but lost 12.70% of its energy in 5 metres, and that figure rests on a single valid shot: another bare shaft broke on its first impact, and the flight was so erratic that testing was stopped to protect the chronographs. Among straight fletchings of three natural feathers, the smallest loss came with 4-inch feathers (1.93%), then 5-inch (4.65%); the shortest, at 2.5 inches, lost 7.51%, almost as much as the longest at 8 inches (7.09%). Plastic vanes lost more than feathers of similar size: the 4¾-inch vanes lost 4.88%, more than the 5-inch feathers with 20% more surface, although their speeds were more consistent from shot to shot. Flu-flu fletchings, built to slow the arrow down, lost between 12.14 and 22.72%. Mounting the feathers offset or helical reduced the loss with 2.5-inch and 5-inch feathers, but increased it with 4-inch ones.

The arrows were shot not by an archer but by a mechanical device: a tripod staked into the ground, a rail on which the bow is mounted horizontally, a trolley carrying a mechanical release, and an adjustable stop that sets the draw between 10 and 39 inches. There were three bows, all straight: an elm flatbow built by the first author on the pattern of the Holmegaard bow, a yew longbow by the same maker that had been in use for ten years, and a laminated bow with a wooden core between two layers of fibreglass, made by the bowyer George Merlin. The two wooden bows, which like prehistoric bows have no cut-out shelf, had linen strings, the modern one a Dacron string. The arrows were commercial Port Orford cedar shafts, picked as the closest matches for weight and spine measured before fletching, all with a 125-grain iron point and with the back of the fletching 3 cm from the nock. The 17 configurations in the second experiment combined natural parabolic feathers of 2.5, 4 and 5 inches, plastic vanes of 2⅞ and 4¾ inches, and flu-flu feathers used whole or cut down, mounted straight, offset, helical or wound in a spiral.

Speed was measured with Chrony chronographs using two light sensors, rated at 99.5% accuracy, placed 1 metre from the bow and, when a second one was used, at 6 metres. Each arrow was shot until there were at least ten consistent readings, meaning within 2.5% of the mean. Kinetic energy was calculated from speed and arrow mass, and the loss is the difference between the two chronographs, so it covers 5 metres of flight.

Each configuration is a single arrow, so differences between fletchings are also differences between individual shafts, which weighed from about 31.4 to 38.6 grams. The authors themselves put some anomalies down to mass (the arrow with 4-inch feathers was heavier than the one with 5-inch feathers), and mass changes how much energy the arrow takes from the bow. There are no statistical tests: many of the differences in loss are one or two percentage points, on speeds measured to a stated accuracy of 0.5%, so the order of similar configurations could change if the test were repeated. The losses cover only 5 metres, the stretch in which the arrow is still straightening out, and the authors themselves call for a third chronograph further downrange. Nothing was measured on accuracy or grouping.

The whole second experiment uses one bow, a wooden flatbow without a shelf, which makes the arrow bend round the handle more than a modern bow does; the authors attribute the erratic flight of bare shafts and small feathers to exactly this. Their conclusions on fletching therefore hold for that pairing of bow and arrow, and the stated aim of the paper is to help archaeologists who use bows in experiments on prehistoric weapons. The comparison between the modern and the wooden bow rests on two arrows, with six and eight consistent readings for the modern bow, which was hard to secure on the device. In the text one offset and helical pair is labelled N7 and N14, while the table gives N17 and N14: this record follows the table.

For traditional archers, especially those shooting wooden bows without a shelf and wooden arrows, the study puts numbers on something many know from experience: speed off the bow says little about how much energy the arrow still has a few metres on. The bare shaft was the fastest at the first chronograph and among the worst at the second; the flu-flus left almost as fast as the rest and lost up to more than a fifth of their energy in 5 metres. On this bow, 4-inch natural feathers mounted straight gave the best compromise, while feathers that were too short could not settle the arrow quickly enough. The authors think that with a hand-held shot, less regular than the machine, 5-inch feathers may cope better with small errors; that is an opinion, not a measurement.

For an Olympic recurve with a rest, pressure button and small vanes, the numbers do not carry over: the shelf reduces how far the arrow has to bend round the bow, and the authors themselves link this to the smaller fletching used on modern bows. The study does not say which fletching groups best, because accuracy was not measured. The bow comparison does point to two more general things, although measured with few bows and arrows: at the same draw weight a fibreglass bow returns more energy to the arrow than a bow made from a single piece of wood, and a faster arrow loses proportionally more speed over the first few metres, as expected from air resistance that grows with the square of the speed.