[quote="Jean Henri Chandler"]

Jonathan Dean wrote:

Quote: Speaking of myths, the "Welsh" longbow is a 19th century myth. I think that deserves another thread all it's own.

If you want to make one and argue that late 19th/early 20th century scholars were right and the last 50 years of scholarship are wrong, be my guest.

Jean Henri Chandler wrote:

Jonathan Dean wrote: Tell me, Jean, where have I used Victorian or 20th century scholarship instead of the archaeology and the primary sources to inform my views? Where have I not made an effort to get into the actual historical record? Where have I been lazy, as you seem to think I am? You're the only one of us to make use of Victorian scholarship so far. You seem to think that nobody on the continent had or could shoot 'real' English longbows. To me that is a myth. Considering that English longbowmen were fighting on the continent as mercenaries for centuries, that England was the direct ruler of parts of what is now France for generations, that longbows were being imported (and the wood to make them exported) via Continental European polities since the 13th Century, and that at least thousands of English longbowmen died or were captured in Continental Europe - not just in France during the 100 Years war but well beyond and in other conflicts, I'd say you are not entirely relying on facts here. That sounds more like a legend. You are aware that the Duke of Burgundy was using English longbowmen as mercenaries and training his own people in the use of the English version of the weapon in the 15th Century, right? English longbowmen may have been considered the best at using that particular weapon in aggregate, or for large armies, but I was referring to small shooting societies, militia units and mercenary companies.

I'm sorry, but what? I used a late 15th century primary source, which Victorian and early 20th century scholars didn't know about, to caution against accepting late 15th/early 16th century written records uncritically and stressed the need to use archaeological evidence to test both the late 15th century source and your assumption.

In what way am I relying on a "legend" or using Victorian scholars? Seriously, what evidence from primary sources or archaeology are you using to accuse me?

Were you or were you not trying to suggest that "real" longbows or longbowmen didn't exist on the continent?

Were you or were you not trying to dismiss the notion of longbow archers guilds in Flanders and elsewhere on that basis?

Were you or were you not trying to suggest that records from such organizations would not be relevant data to understanding period longbow accuracy because they weren't using "real English longbows"?

If I am wrong and that is not what you were suggesting, then I apologize.

If that is what you were and are suggesting then I call that a legend.

Ok then, I am sorry if I misunderstood your post. When I wrote

That said, there were urban guilds (often dedicated to St. Sebastian) which did archery as a competition / sport and by no means just for clout shooting. I know that some of the ones in Flanders did shoot the popinjay and presumably also shot at regular disc type targets.

So I would say even if you can't find records in England for some reason (I don't know anything about that as it is outside my area so to speak) I'd be surprised if you couldn't find something in Flanders or Burgundy or somewhere in Germany where they also had archery guilds. I know that the St. Sebastian guild in Bruges is still there and they do still have some 14th-15th Century records, you can do a tour with an appointment.


and then you replied:

Quote:

The question is whether the archery guilds were using bows as powerful as the English. Dominic Mancini considered English bows and arrows to be significantly thicker than those used elsewhere in Europe, and at least in the painting you linked the bows are significantly different from English bows. Which is not to say that it wouldn't be handy to have the records (or to be able to read the original manuscript text, a skill I don't have), but just that some caution would be needed in interpreting them. At the very least we'd need some archaeology of what are unambiguously arrowheads that match typically socket diameters for English examples of the period.

I don't know the quote by Mancini but neither his remark (presumably a generality) nor the tiny details of one painting are any kind of evidence that (and I am quoting you directly) "The question is whether the archery guilds were using bows as powerful as the English" is answered in the negative.

If you didn't mean what I summarized upthread, then I really don't grasp what you meant, but one thing is for sure - I have definitely wasted a lot of time reading and posting to this thread. This started when I was addressing the issue of whether or not any specific data on accuracy existed for longbow shooting. For anyone really interested in finding that out, those St. Sebastian (etc.) guilds are probably a good place to look. This was my attempt to help out with sources, in other words.

So I will close the book on this specific non-debate then with this assertion: They did have longbows in Flanders and many other places in Continental Europe which were every bit as powerful as any in England. They may not have had entire armies of longbowmen, but that is a completely different matter.

Beyond that, I'll leave you to your own opinions since I apparently can't parse the meaning from your sentences, I'm not going to try to change your mind.

J

Maybe we are talking past each other, I don't know. Crossbows can be a bit contentious, but anything to do with longbows on the internet seems to make everybody crazy for reasons I can't fathom. Why don't we al least keep the longbow discussion to the other thread, so that this one can more or less stick to the original topic.

Hi everyone

The FEM work is progressing, but not there yet.

Bartek hit the nail on the head - my goal with this is to try to find out what improvements are possible, because although I know there are limits, I don't know what they are. So its pointless for me to make any claims without trying to at least get a better idea what the limits are. If there are any surprisingly well performing possibilities, we can check for matches any extant historical crossbows. And also, when I have the geometric information on a given historical crossbow (which some forumites have kindly PMed me to help with - thanks!), I can try to model its performance and see how well I think it will do, given a range of possible input parameters. As I said, forwards and backwards, and hope they meet in the middle...

So, since the question of metallurgy is going to come up again with the FEA, I'm going to give one last potted description of what I know about the metallurgy of iron-carbon alloys.
I think most people here already know something about the different phases of iron-carbon compounds present at different temperatures and at different temperatures and carbon concentrations. This is why we can quench-harden swords and why we need to temper them. There are actually only a small number of phases, and fewer still that are present at room temperature: ferrite and cementite (martensite is only a quasi-stable phase, and in swords, the tempering process actually reduces it to a combination of ferrite and cementite with a microstructure that has better mechanical properties for blades). Ferrite is relatively soft, while cementite is hard and brittle. Pearlite, tempered martensite, bainite and even wootz all owe their properties primarily to the way the smaller quantity of hard cementite is distributed through the ferrite (the microstructure).
Now the material properties of pure ferrite and pure cementite (Fe3C chemically speaking) can both be determined very accurately based on the interatomic bonds in their respective crystal structures. The property of any of the various combinations and microstructures of these two phases is therefore some weighted average of the two. But the modulus of elasticity can't exceed that of the cementite. In the case of wootz, the very hard cementite takes the form of thin rods on a nano-scale, which occur in large numbers because (in part) there is an unusually high percentage carbon than is usually the case. This makes it strong for loosely similar reasons that fiberglass-resin or carbon fiber-resin composites are strong, as far as I understand (Bainite is notionally similar but the microstructure is on a much larger scale and its properties less dramatic, though still good). I found a value of 400GPa for cementite in one publication, but would like to look further to check it. Pure cementite is also far too brittle to use for a bow that will flex, as far as I know. It is actually probably more appropriate to refer to it as a ceramic in its pure form, namely iron carbide. Also bear in mind that the amount of cementite present is limited by the ratio of carbon to iron (even in wootz or cast iron it is only of the order of 2-3%) so there will always be much less cementite than ferrite, therefore the material properties will tend more or less towards that. Of course, other alloying elements (vanadium, chomium, nickle, nitrogen, and many others) will "nudge" the properties of steel this way or that. But the dominant feature is always the strength of the bonds between the iron atoms.That is the long version of the reason why I don't think we are going to see vastly different elasticic properties in medieval steel. Ductility is not relevant because that implies that the bow lath would stay permanently deformed, rendering it useless. For those who know metallurgy better than me, please correct me as necessary. For those who want to know more or want to check me, my primary source is "Callister's Materials Science and Engineering", the handbook that I used for my udergrad material science course, supplemented by various bits of literature such as the wootz paper I mentioned before, and this rather interesting paper made available by NIST
https://www.nist.gov/publications/elastic-constants-and-internal-friction-martensitic-steel-ferritic-pearlitic-steel-and
And for those who can't get the textbook, there's a nice phase diagram for the iron-carbon system on the website of the Technical University of Kiel, which is the core of the thing:
[url] https://www.tf.uni-kiel.de/matwis/amat/iss/kap_6/illustr/s6_1_2.html [/url

Nonetheless, I do intend to test a wide range of possible moduli of elasticity, which will exceed beyond what I think is the "known range" of properties. Because I could be wrong, and there might be some hitherto unknown quirk of metallurgy. I certainly don't claim to know how absolutely element will affect steel when introduced into the alloy. Frankly I was surprised it was measured as high as 250GPa in a wootz sample. But I stick by my claim that there is some definite physical limit based on the atomic structure of metal alloys composed mostly of iron. We are almost certainly not easily going to see steel with a modulus of elasticity of double or half the generally accepted value of about 210 GPa. This is a falsifiable claim, and as such, if a counter-example is available, I will be happy to correct or recant it in the light of the new evidence. It then becomes a matter of checking whether the alloy involved may be found in any extant steel crossbow laths. We can never prove definitely that the hypothetical alloy was never used as we cannot test every crossbow ever made, but we can disprove the claim by finding even a single counter-example. And if we don't find any, we can at least say that to the best of our knowledge, it was not used in any extant examples. Then if we find an example, we can thoroughly investigate whether it will have a positive or negative effect on a crossbow lath - we can claim there is no crossbow design which will benefit from it, then try to prove this statement wrong. And so on...

As for my ballistics calculations being crude: Well, they are simple, certainly, but not crude in the sense that they are not fit for purpose. My results match the measured behaviour of an arrow shot from Tod's Lockdown Longbow and 960lb crossbow to within a few percent. They also match the large body of available data for olympic javelin-throwers and javelin guns and at least one paper where javelin drag force was measured directly to a good degree of accuracy. Late last night, as a final check, I applied data for a shell from 16-inch british naval gun such as was used on the HMS Nelson or Rodney in WWII, and I was able to predict the maximum range to within 15% (a larger margin of error, but certainly not terrible considering the much greater uncertainties involved). These sorts of ballistics calculations are inherently simple; they were pioneered and largely solved in 1537 by the brilliant self-taught mathematician Niccolò Fontana Tartaglia https://en.wikipedia.org/wiki/Niccol%C3%B2_Fontana_Tartaglia
He sounds like a fascinating character, and deserves to be far more famous than he is. His work was vital to Gallileo's study of falling objects, and was apparently accurate enough to be used by artillerymen 200 years later. It was not quite as accurate as Gallileo's work, of course, and still less than Newton's (whom I used) but it was very close (one of the great things about this exercise is how much I am learning along the way). Tartaglia's model and its long subsequent practical use demonstrates that even an inaccurate model is still very useful if the degree of error is small and quantified. By analogy, a gun or crossbow does not have to be perfectly accurate to hit a target, it just has to have some ammo(data), a small enough spread (an acceptably small degree of inaccuracy) and a shooter who knows his weapon...

Of course, as I said in the initial post, I can in no way claim that my hypothetical scenario with a tailwind was the truth behind the Payne-Gallwey's famous shot. It is purely speculation, but it is based on solid physics and demonstrates one alternative possibility (and also illustrates that we can't neglect the effects of air resistance, wind and arrow streamlining when discussing crossbow ranges). To claim more than that would be unscientific, though, so I don't and didn't. Its also not based on Tod's or Adreas Bicher's measurements or work; I only used those to check that my calculations were reliable, ie. that i hadn't made a programming error. It is based on the dynamics of a projectile from first principles.

Jean, please do write that paper. I may disagree with the way in which you reach some of your conclusions as far as I understand that now, but I am also painfully aware that I don't have all the data that you do on which to base your ideas. I feel that if I read a step-by-step logical account of how found your information and reached your conclusions, I'd quite possibly be more able to see what you see. And even if I still disagree with some of your conclusions afterwards, the historical data you have access to is fascinating and very valuable to all researchers of this topic, and that alone would already be a valuable contribution.

Andrew

I will do a paper on it one day, though it is at the end of a long queue.

And I say to you, by all means do your calculations, but I'd be wary of leaving out what seem like 'outlier' data points like your wootz sword. May want to check to see if it's actually an outlier first ;)

If there is one thing I have learned in studying history, it's don't dismiss the outlier data points because they fall outside of a pre-concieved pattern.

J

What level of detail do you need on geometric information, Andrew? I'd be curious as to the performance of some timber crossbow lathes, but I only have a couple of reference points for one, although I think some extrapolation should be possible.

Hi Jonathan

What I need besides the basic measurements (length, brace height, center thickness and height is the section measurements at a few different points along the lath, as well as how far along the lath this occurs. The more sections along the lath I have, the better, but in practical terms, if I have 3 or 4, I can get a decent approximation of the way the bow tapers. Luckily nearly all the steel bows I've seen in photos and my rare visits to European museums are more or less rectangular in cross-section, which is nice and easy to define - only 2 parameters. If there are significant bevels on the edges of the belly or back, as someone here mentioned, that can make quite a big difference (from what I remember of structural mechanics and my own attempts to make wooden selfbows), and I'd ideally want to know their dimensions as well. Oh, the other thing which would be very useful would be the string diameter. If data on the bow unstrung is available, that would be great too. I've read that steel bows tended to have some deflex built into them?

One question that I've had answered already by Mr Da Vinci is that they were not necessarily using a straight linear taper on their bows. Its difficult to judge this from photos of extant bows or even if they are behind glass in a museum case, but it is pretty easy to see in Leonardo's sketch that he at least knew a non-linear taper was needed, and therefore it is likely that this was done by at least some crossbow makers of the time.) I know that technical workers of the time - the people building cathedrals and artillerymen in particular, were using some quite good geometric and numerical methods to do their calculations, so it would not surprise me if the same were true for crossbow makers, though I have no definite evidence either way. Apparently Leonardo did some design calculations for his bow, though not reading medieval italian beyond a little of Fiore de Liberi's fencing terminology, I don't know how, and I feel some caution is needed with them in any case (one data-point doesn't mean much, and the fact that he didn't always design things that worked).

Doing a simulation of a solid wooden lath will be slightly more challenging for a few reasons, but I'll certainly have a go (probably after I can get verifiable numbers for a steel bow) The challenge with wood is that it is not a truly homogeneous material and can vary quite a bit in density and elastic modulus, as well as the fact that I think some wooden laths have a more complex d-section that needs more measurements to accurately define? But even so, it should give some interesting insight into the way the design works, and a rough estimate at least of what sort of performance to expect (I hope).
If you have data for a wooden bow, please send it on!

Andrew

I think Da Vinci would be a really bad starting point for any kind of data. There are hundtreds of antique crossbows which can be looked at and measured directly, there are reams of records from the crossbow crafts, and books of regulations on craft guilds like the Balthasar Behem Codex. I know there also are some papers that have been published on craft guild regulations for crossbowmakers in Venice as well.

If for some reason you felt like you needed to look at artistic sketches of engineering designs similar to Da Vicni, I'd recommend finding a more plausible military engineer. For example Taccola created many designs which were actually built and used (and I think Taccola was an influence on many of Da Vinci's sketches).

There are also the various Kriegsbücher and many more specialized manuscripts like the Codex Löffelholz which depict and describe specific crossbow designs, a couple of which have been reproduced by modern artisans. I know Andreas Bichler built a copy of the famous self-spanning latchett crossbow from that MS (which you can see on page 27 of the scanned manuscript I linked). It has a steel prod. He also shows the tools used to make and assemble these devices.

[ Linked Image ]

Here is Bichlers version on youtube

It is obviously a lightweight weapon of course compared to the others we've been discussing.

Yes I actually do agree with you about the taper.

But I also think rather than go by my hunch or your hunch, you should check the real numbers. Someone has (several people have) definitely done measurements of late medieval and early modern crossbows. It would be worth a little effort to look for some academic articles, if you have JSTOR access or even on Academia.edu, but I think that would be worth the effort.. It might take some digging and some work to find the right search terms etc., but I guarantee that data has been published somewhere with measurements for multiple antiques. There are so many surviving antique crossbows all over the place even in the US there have to be thousands of them. Partly because they were still making late medieval style crossbows up to the 19th Century as prestige hunting weapons.

You can also see detailed photos of them on many museum websites, many in Europe including the Vatican, and a few in the US (Louvre just put all of their stuff online a day or two ago)

If you find reference to a manuscript (MS number) you can look for it in university websites like this one

And many museum sites have searchable archives where you can find good images. For example here is a composite prod crossbow from the Metropolitan museum of art's website

[ Linked Image ]

... and here is a pretty heavy duty looking 18th Century steel prod hunting crossbow which they claim is halbe rüstung. They list some measurements.

You should also look for auction catalogues (online or physical) which sometimes include detailed measurements. Hermann Historica et al. The private auctions actually tend to have better stuff than the museums when it comes to medieval antiques and they often include a lot of detailed photos and measurements.

Also, look at sites like Vikingsword.com, (ethnographic arms and armor) there are many threads on there with many close up detailed photos of antiques, like this one or this one.

Andrew Gill wrote:

Ah, I see you said Kriegsbucher, not Fechtbucher, though something like Talhoffer's 1459 manuscript seems to try to be both?

Yes the Talhoffer 1459 has elements of both, as do many others like Ludwig von Eybs Kriegsbuch Ms B.26. Those can be good sources of information though a lot of them are not translated yet. Some are and a few are searchable.


Trying to help you here, in the hopes you will follow my recommendation to spend a proper day or two gathering some data so you'll have a good historical baseline, to which you can add more and more as you go along with calculations. And again, please don't dismiss the outlier data even if it doesn't match your expectations. Put an asterix next to those kinds of things if they don't make sense, because obviously some outlier data is noise, but keep track of them in case you need to come back to it. Ultimately the data should create the conclusions not the other way around right?

J

Thanks Jean

Absolutely! Without historical data, to back it up, whatever I find will be somewhat conjectural, though some things might still be usefully established with some degree of certainty - I hope I can determine the ideal efficiency attainable for a steel prod, for instance, and can give a good indication of the likely effect of varying any one design parameter, all others being equal. Some of these we can do just from old-fashioned cantilever beam calculations, but there is room for improvement, as that mostly assumes small deflections of the free end, and crossbow tips move quite a bit, of course.
What I've been doing up until now (and still need to do some more of) is to make myself very familiar with the behavior of a crossbow (the physics of crossbows if you like). I'm certainly not going to do a simulation based on thumb-suck estimates derived from da Vinci's (or anyone else's) sketch and claim that proves anything :lol:

I work for a research organization, so I can get my hands on at least some academic articles through the internal library service; its just a matter of finding the right articles, which takes time, especially as its not my field of research. That is definitely my aim, though, which is why I've been asking about sources of information. So for instance, When I did the review of steel, I linked to the one paper in the interests of brevity, because it seemed most relevant, but I spent the better part of a day in total (spread out over several days, in my spare time, of course), collecting and reading information. Even when I've had to estimate numbers, I always try to braket them within a plausible range of values based on what I find in literature and reputable online sources (and I almost always try to confirm with more than one independent source if I can, otherwise I flag it as a point of higher uncertainty). Much like I'd do when I'm writing a research paper, in other words.

And I do appreciate your help very much, thanks.

Andrew
Edit: Jackpot! Stuart Gorman of Trinity College Dublin did a thesis titled "The Technological Development of the Bow and the Crossbow in the Later Middle Ages"
It has an exhaustive literature survey, including mention of a few works containing detailed cross-sectional geometry of crossbow laths. It also has (for my purposes incomplete but still useful) overall measurements of literally hundreds of crossbows in various museums, with photos in some cases.
It's available here: http://www.tara.tcd.ie/handle/2262/77397

See there you go, good find!

Been reading that article, it's a gold mine for this discussion.

He goes over all the published literature on the crossbow, and there are several detailed analysis of both composite and steel prods. The publication "Crossbows in the Royal Netherlands Army Museum" apparently includes (this is a direct quote) - "metallurgical analysis of a steel crossbow lathe" - apparently a destructive analysis. Sadly it seems to be out of print and printed copies anyway are very expensive. Maybe there is a PDF available.

I'm only on page 30 but apparently none of the published literature makes the assumption that composite prods were more effective than steel, some (including Gallwey) suggested the opposite.