Tuesday, 1 May 2012

Comparing Two Longbows

I've got two bows finished, 55# and 60#.
Well is the 60# any faster? With some trepidation I shot it through the chronometer and averaged the 6 results getting a respectable 173.7 feet per second (fps) average.
I then strung the 55#, would it be just as fast? Again 6 shots and average, 168.9 fps.
I'm happy with those results and it's a bit of a relief to find the 60# is faster (I shot the same 3 arrows with each bow, 5/16" shaft 100gn point). I took a pic once I'd finished as I happened to get a nice group with the last 3 arrows from the 55#, you can see the reading on the chrono'.

I take the arrow speed divided by the draw weight as a sort of "efficiency" indicator. The 55# bow gives 3.1 whereas the 60# gives 2.9 implying the 60# bow is somehow less "efficient". I think the reason for this is light arrow favouring the lighter bow. If I repeated the test using my 11/32" arrows with the 100 grain points I expect the 55# bow would loose more speed than the 60 and the "efficiency" figures would even out or swing the other way. The arrows used for the test feel fine with the 55#, but the 60# definitely feels happier with the slightly heavier arrows.
Out of interest my best speed/draw weight figures tend to be the primitive style flat bows with 'Twister' my Yew primitive (45#) giving a figure of 3.7 Of course the longbow is designed for delivering energy rather than velocity as is better matched to a heavier arrow.
Figures need treating with caution of course and reducing the weight of the arrow will always give increased speed, but it's about matching the arrow to the bow rather than pure speed, unless you are into flight shooting where you are after maximum range with little interest in smoothness or longevity of the bow.
Explain more? Not sure what to explain really, I put "efficiency" in quotes as it's not really a measure of efficiency as it doesn't take into account the arrow weight, but it gives some sort of comparison.
These 2 posts explain more about velocity/energy/ speed and power.
The chronometer has two photo cells which detect as the arrow passes over it. I've stuck thin pieces of cane in to show the active area which I have to shoot over. (The chronometer comes with metal rods which get severely bent when hit by an arrow so the thin cane is better!) I have two100w incandescent lights which hang down above it to give suitable illumination (fluorescent lamps flicker and can confuse it) Sometimes it's tricky to get reliable readings as the arrows can be flexing at short range as they leave the bow, I generally shoot at about 5 paces and take several reading to make sure I'm getting reproducible results. A chronometer is the only way to really tell how a bow is performing, as they can feel quite different and with the 2 bows I tested here it would be hard to tell which was shooting faster with out the chronometer.


  1. I was the one ticking the 'explain more' box.

    What are the dimensions of the two bows? Are they the same length/width? I'm just thinking of difference in limb masses here, if any, and how big it might be.

    What were the 6 individual chrono readings for each bow? I'm used to look at RSD and searching for groupings in the data so whenever the raw data is missing I feel cheated =)

    Great post, no doubt your evening was better than mine (spent doing my tax returns).

  2. I was the one who ticked the 'explain more' box, but for some reason Blogger ate my comment.

    I was curious about the dimensions of the bows, primarily thinking of difference in limb mass and how large it might be, if any.

    Also, what were the 6 individual chronograph readings for each bow? It's a force of habit to look at RSD and possible grouping, so whenever the raw data is missing I feel a bit cheated.

    Great post, looks like your evening was better than mine (spent doing my tax returns).

  3. Hi, your comment took a while to come through.
    The bows are similar, being from the same batch of Yew. The 60# is about 1/2" shorter amd a bit thicker (I can post some figures if you really want).
    Chrono readings:-
    55# bow 163.6, 170.6, 170.6, 168.3, 170.3, 170.3
    60# bow 171.1, 174.1, 175.2, 173.4, 173.8, 174.4
    The only slighty iffy data is the first from the 55# bow. I could have discarded the best and worst from each data set but I don't think it would have made much difference to the average.
    I might make up an arrow with a 125grain point for testing, it will help me recommend the best arrow for each bow.

  4. No need to post the data for the bow limbs, sounds like the difference would be small.
    Thanks for the chrono data! Discarding 2 values from such a small population of 6 seems a bit heavy handed, but using Dixon's Q-test* for outliers the 163,6 reading can be rejected at 95%, but kept at 99%.

    Rejecting the value would change the average from 168,95 to 170,02 and the RSD from 1,63% to 0,57%
    For the 60# bow the average is 173,67; RSD = 0,80%

    (*) Q=|(suspect value-nearest value)|/(largest value-lowest value) which is compared to table values for a certain level of confidence and # of observations.
    If Qcalc > Qtable then the value can be rejected.
    (wikipedia entry at http://en.wikipedia.org/wiki/Dixon%27s_Q_test with table values for 3 to 10 observations and 90, 95 and 99% confidence level)


  5. I've not come across Dixon's test, ta for the link, interesting. I have a smattering of stat's knowledge which comes in handy occasionally.