[Nate C.]

Greetings,

I have had this question running around in my mind for a while and thought I'd ask for a little help. My question concerns the sizing of the pommel and guard for proper balance of a sword. How do/did Swordsmiths/Cutlers decide on the size of pommels and guards for a sword? I have been playing with the idea of trying a project sword and I am really at a loss on this point. Especially with cast hardware, how is this done? Is it by trial and error? The "magic formula of the ancients" ? Divine Revelation ?

I'm assuming weight/balance comes in to play here as well as harmonics, etc. But, how? As has been mentioned often, COB isn't everything.

Thanks for the help,

[Peter Johnsson]

Divine Reveation works fine for me;-)

Never mind those pivot points or the pacing of the nodes...

Seriously, I will try to make an attempt to explain how I do it later..

[Craig Johnson]

There are a couple different ways to approach this issue. Often the end goal is the determining factor, as one person's cutting sword may not be anothers ideal. One excellent way to do it is if the blade feels good with no furniture you are on the right track. I am not sure if this is something I have gotten used to as I form the blades but I have a feel of the heft of the blade when gripping the tang that I am looking for and that often carries you a long way through. If this is one of your first swords I usually tell people to pick their hilt components then reduce them by half and you are usually at a good starting point. It is very rare that someone starting to make swords makes their first examples to light.

(side note here, I say above "pick their hilt components" as I dislike the term "designing a sword" as just about everything has been done before and unless its one of those multibladed Gil Hibben extravaganzas no one is really "designing" a new sword they are just repeating what someone with more experience has done before IMHO. Now if you take the examples of the past and use them as a design and guide book one can get well along to crafting a very very good sword :-) but this of course is my personal opinion)

Hope this helps a bit, its a big subject

Craig

[Angus Trim]

[Nate C.]

Thanks for your replies. To expand on my question, do you start out with a weight for the components in mind or is it trial and error again. Is this kind of thing a try until it feels "right" thing? I'd eventually like to make my own swords and I thought starting with cutling (is that a word?) and work my way up.

Craig, Gus - Granting that the blade is properly designed for the intended purpose, what then? I'm trying to get an idea of the process if there is one. I definitely intend to try and replicate some existing designs but would also like to eventually try a few Fantasy-ish pieces still keeping to the historical "feel" (i.e. no Conan/Anime boat anchors ).

Peter - I look forward to hearing from you when you get the chance.

Cheers,

[Peter Johnsson]

This is a question it is difficult to answer in a short post like this. It would take more time and space to do this thoroughly. I´ll try to cover some basic principles and add some to what Craig and Gus already have written.

Part of this needs to be intuitive just as Craig said, but to aquire a good intuition you need to study a lot of swords. There are some basic principles it is good to be aware of. These ideas are your concious tools that help you analyze your work as it proceeds. It is also good to have a critical eye to these ideas. Ideas can very easily become blinds instead of lights as your work.

Number one rule: To get a well balanced sword you *need* to have a good blade. A crowbar will always be a crowbar however much of a pommel you put on in one end.

We could divide most swords into three basic shapes:
-(fullered) blades with little profile taper (X, XIII, XIX and so on)
-blades with medium profile taper (XII, XVI for example)
-blades with acute profile taper (XV, some XVIII, XVII)
All of these can be long swords or single handed wpons.
You can expect different placing of the pivot points with these three types, but there are of course exceptions. There are *Always* exceptions!
Blades with light & narrow points usually have their forward pivot point close to the point (but not always!).
Blade with broad points often (but not always) have their forward pivot point closer to the vibration node in the blade.
Blades with medium profile taper will as a norm (with exceptions) have their forward povot point some where in between the other two types.
Ratio between grip length and blade length can have a great effect here, but as always, there are exceptions.

A longsword with a slim point and a long grip is the school book example of a sword that can be expected to have to forward pivot point very close to the point. *But* I have seen and handeled a very well preserved long bladed single hand sword (type XI, early 13thC ?) with a short grip and a *massive* pommel that had its forward pivot point some 5 mm from the point. It was incredible. A balance like this is the result of a very specific idea of how the sword should handle, function and perform. It is not ypial, but you will obviously find exmples that deviates from the norm.

When I started out making swords I aimed to keep the weight of the pommel around one third of blade weight.
This worked ok in the beginning, but it gave less satisfactory results when I made more and more swords based on documentation of actual originals. As I got to handle and documnet more originals I saw that they did not conform to this rule of thumb.
The variation is great. A single hand sword can have a pommel that weigh around half the weight of the blade or a third of total weight. There are also many single hand swords that have a pommel weigtth of less than 25% of the blade weight, a fifth of total weigth. A lot of variation.
A long sword could perhaps have a pommel that is around a third of blade weight, but this depends very much on the type of blade and ratio between grip length and blade length.
It is also important to remember that different blades are going to be more or less sensitive for the weight of the pommel. Some blades can take quite a bit of variation in pommel weight and perform very well. Other blades are going to be sensitive about the size of the pomel or they will begin to be temepramental...
If you can adjust the length of the grip even just slightly, this can also be an effctive effective way in changing the balance and properties of the sword.

When I design a sword (and Craig, here is where I am of a different opinion than you : I think you can say that you design sword today even if you closely follow the example of historical styles and principles basically without "inventing" anything new. It is just like a typographer can design a book without having to reinvent the alphabet or the traditional proportiions of the page) and get to the setting of the pommel weight, I always start with the almost finished blade.
I have a pommel with adjustable weight that locks down on the tang so I can try the effect of different weights.
I do not look at where the point of balance happens to end up. That is just an end result of the other two important factors: first the placing of the pivot points, and second in importance, the placing of the vibration nodes.
With higher weight of the pommel the more the forward pivot point (the one that corresponds to the place in the grip just behind the guard) will be pushed towards the point of the sword.
At the same time the vibration nodes will wander back towars the hilt and pommel. By adjusting the weight of the pommel carefully you can in some blades get very obvious higher orders of vibration nodes in the grip. This can be of help in long slim blades to keep them more stiff during the cut.
The point of balance will move less and less as a result of an increase in pommel weight and will really just be of secondary importance for the handling of a sword. To my mind you cannot gauge the balance of a sword by observing how the point of balance moves with different pommel weights. What counts is the correlation of pivot points in the blade and different parts of the grip. This is what you intuitively feel when you move a sword about. With a careful balance the blade will index your arm and give you an intuitive feel for where the point is and where you have the centre of percussion (and this is contrary to popular belief *not* always the vibration node in the blade. It can be in some cases, but it will vary depending on how and what you cut)
How much you can move the nodes and pivot points are determined by the type and character of the blade. Some can carry more, other will start to behave badly sooner. Some blades cannot take much of pommel increase or they will start to have problems with vibrations or become just strange in handling, other blades are much more tolerant for this. It is impossible to list all the aspects that comes into play in determining this here in this short post.
I hope you still get an idea of what is involved.

Where you end up with these three aspects Pivot Points, Vibration Nodes and Balance Point) is a result of how you want the sword to perform and what type of dynamic balance you want it to have (dynamic balance is the way the sword behaves in motion). To sum this up as "harmonics" is too simplistic as it is much more than harmonic correlations of vibrations in the blade and hilt. Dynamic balance is as important and that is something quite different than placing of nodes.
If you want to learn this you will have to spend some years doing experiments. There is now way you can apply some principles and hope they will take you all the way. They form the basis for the work, though.
It is also a good idea to try to get access to originals. They tend to surprice you. Whenever I thought I had the general picture clear in my mind there usually came along that odd sword that made me throw my fancy theories out the window. It is both refreshing and frustrating! :-)

[Greyson Brown]

Peter,

Is there any way we could prevail upon you to show us a picture, or even a drawing, of your adjustable-weight pommel? I would really like to see that!

All,

I am enjoying this thread quite a bit (I started blacksmithing because I wanted to make a sword, and have still yet to do it), but I'm not sure I understand all of it. Thanks for the info anyway; I'll sit down one day and read it all through while I draw myself little pictures, that should help quite a bit.

Thanks again,

-Grey

[Nate C.]

Peter -

Thanks for your reply. It helped clarify things a great deal. The mention of an adjustable pommel intrigues me . Not a bad idea at all. Your post did give me another question though. What are rotation points? I understand the others from playing with my Mercenary and my physics classes but I'm drawing a blank on rotation points.

Thanks again,

[Alexi Goranov]

[Peter Johnsson]

I mean the same with rotation point and pivot point, sorry for sloppy nomenclatur.

You do not need milimeter exactness. That is beyond reason. You need to check the general tendency. If this is within a cm or two you know enough. A sword is not a space shuttle...

A variable pommel is any contraption that allows you to add mass by some locking device and graded weights.
Anything along those lines will work.
The exact design is not that important.
Try to keep close to the final shape of the pommel.
You might have to devise and build a few different variable pommels to serve you in different situatiions.
This is not difficult engineering, so I leave it up to you to find a good design ;-)
If constructing a varable pommel is too difficult you should rethink your ambition to make a sword

[Alexi Goranov]

[Craig Johnson]

Mornin Nate

The dynamic balance that Peter describes is the crucial hurdle to cross when trying to understand swords. The rotational/pivot points are those spots where the item will want to rotate around as it is moved in the hand. This is so for any long object. If you take a dowel that is the same dimensions along its length it will have two evenly spaced rotation points. If you put more mass at one end of a stick i.e. a tapered stick the pivot points will adjust to compensate. Thus when you pick up a branch in the woods and it feels really good in the hand for smacking something you are actually admiring the sticks dynamic balance.

A quick demonstration to see the affects of pivot points and such is take two long thin objects and put an equal weight on each one in the middle and one 1/4 of the length in from an end. The two items are the same weight but will have different dynamic balances.

Greyson

If you are interested in doing blades and swords the best way to start is just do it. Look at the best examples you can find and handle and try to produce something you like that is similar. There is absolutely no better teacher than experience and repetition. An old blacksmith once told me the reason most guys never produce very much is they are afraid of goofing up and that the essence of learning to smith was making mistakes.

Alex

Not sure if there is a simple test for that kind of accuracy. WHen I have talked to Physics types it involves some calculating on the pendular motion of a piece, if I understood correctly.


Craig

[Craig Johnson]

Hello Peter

Well when you lay design out like that I would have to agree with you The term "Sword Designer" has been a little to freely used of late IMHO and thats what I was a bit discouraged about. Even had some one send me a design that had been done by a "professional in the field" who I had never heard of and it was obvious they were unable to wrap their design around the constructs of physics and nature as the specs and the design were not attainable with any of the known elements. So of course I sent them your way na just kidding.

Another interesting trend I have seen is the "erganomicly perfect sword". I suppose much of what we are discussing could be classified as a form of ergonomics but when the sword or rapier is "designed" from that standpoint only it is a very strange thing. Not to mention a real challenge to make if one so chose.

Best my friend

Craig

[Angus Trim]

[Blaz Berlec]

Great posts!

Just wondering loudly… I think computer mechanical analysis program should be able to calculate all the needed parameters. Not that it would be any easier than doing it by hand – you would still need to design a sword very accurately in a CAD (Computer Aided Design) program, and there are many programs available that can calculate lots of stuff. Weight, centre of mass, pivot points and vibration nodes are quite easy to calculate for such a “simple” object (compared to, let’s say, Space Shuttle) even in today’s home computers. Well, it could take a while, but it can be done.

The one thing that could be calculated and can be measured on real swords is their rotational momentum around certain spots – around beginning of grip close to the cross (where the sword rotates while striking) and around centre of gravity (that’s where you mostly rotate the sword in transitions from one guard to another and when parrying). It would really be interesting to see the numbers for some known swords – I wonder if this can explain the difference in feel for some swords that have similar weight, point of balance and general geometry, but handle quite differently. I guess you would feel the difference in momentum much sooner than difference in pivot points or vibration nodes – it would be as apparent as difference in weight, as soon as you try to rotate the sword.

Anyway, back to computer stuff…

Such analysis could also show you the potential weaknesses (for instance any weak spots due to sharp corners) and lots of other stuff. But this would be only useful for “fine tuning” – you’d still have to start with a decent sword design (and that’s pretty hard to make for us non-initiated).

Not that I’m an expert in the field, I have only seen such programs from afar, and they were running on workstation computers.

I mean, if they can precisely calculate as complex system as tennis racket or golf club, the simple symmetrical steel sword with very little parts should be no problem… Right?


Oh, I don’t know if this article has been referenced in this thread yet, so here it is:

http://www.thearma.org/spotlight/GTA/motions_and_impacts.htm

Sword Motions and Impacts
An Investigation and Analysis

By George Turner

I know the article has had some opposition, but I think it is still very useful to get you started. It describes all the important physical stuff that happens to a sword – it explains centre of gravity, vibration nodes, rotational nodes…


Carry on, don't mind my ramblings...

[Alexi Goranov]

So I did a little test with the NG Baron and Duke trying to determine the lower ( corresponding to the point behind the guard) pivot point. What I observed is that the pivot point of the Duke is fairly close to the tip, where as, the pivot point of the Baron is somewhere midway between the start of the fuller and the tip of the sword. Are these observations accurate, or had I completely screwed up the "wiggle" experiment. the reason I am asking is because the location of these pivot points seems to be the exceptions to the rules you mentioned in your long post (The rule of the thumb being that swords with little taper have points closer to the vibrational node, and tapering swords have pivot points closer to the tip of the sword).

So first off, are my observations accurate, and if they are, why this choice. What were the properties you were shooting for with these two swords, so that you chose the respective pivot points.

Now that I have these swords I really want to appreciate them. Any comments are welcome.

Alexi

[Angus Trim]

"I mean, if they can precisely calculate as complex system as tennis racket or golf club, the simple symmetrical steel sword with very little parts should be no problem… Right?"

What's simple about a sword blade? *g*

Its a three dimensional object, that has some fairly complicated geometry. Some historical pieces have the distal taper change direction several times down the length. Whether copying something like that as exactly as possible, or merely copying the parameters of something like this to see what it does to performance parameters, this is not a simple symmetrical piece. Each change of direction of the distal taper, changes the relationship of the profile taper to the distal taper, and has both harmonic and dynamic implications. Then this changes the angle of the main bevel... you can measure the main bevel included angles all the way down the blade, and get different readings.... Simple? *g*

Now throw some curved surfaces in there. What happens to the geometry of a blade, if it begins a gentle curve to the point some 11 inches from the point? Have we now induced a curve all along the plane of the bevels? Or is it just the edge itself that's curved. How does this effect the main bevel angles {from spine to edge}?

Lot of different things to consider. Better yet, at times certain software wants to generate a toolpath which will miss certain characteristics of the blade, even if you've carefully plotted in the numbers.... This can be very subtle, but still there. Some times the only right way to do it is to edit the cutting media, the tool paths if you will, to custom generate what you need. In other words, bring back the human factor.........

To answer your question, yeah, you probably could do all you ask on a computer. But it likely would be more expensive than would ever pay out {have to make a lot of swords, a lot more than most of us do to pay the cost of software etc} for most of the small operators to do it. And all of the right info would have to be input, otherwise you have gigo, garbage in, garbage out...

Someday, it probably will all be done on computer by someone. I don't think the day has come yet though, though computers are used by many of us to some extent now.

[Patrick Kelly]

The problem with using a computer is that you already have to have this knowledge in order to program the computer to do the work. IMNSHO using a computer for this kind of work is a self-defeating exercise.

Not to mention the fact that I don't like computers.

[Craig Johnson]

Here are some more fun musings I like to add to these kind of discussions. Not trying to side track the thread but I think they are serious components to the use of the sword.

As Gus says above they are not a simple shape. I have talked to a golf club designer about some of these things and he thought the golf club was a much simpler issue. A tapering circular shaft in most cases with all the weight in one end and the need for a certain combination needed in the amount of flex and stiffness to impart. The designs are computer assisted but as he said if the computer could make a perfect club there would be only one design. I am sure swords would fall in the same category.

Each sword, as we have described above, is a slightly different taper and configuration. They were also hand made items. The result is that often you will see asymmetries in a piece. Do you factor these in? What if the guard is 1/2 inch wider on one side or the pommel is asymmetrical or one side of the blade has an extra 1/8 of width. What happens when the distal taper is on the majority of one face of the blade, does that change the dynamic balance? Almost certainly these issues have an affect.

But and this is a big "BUT" can you tell????

When the gross dynamics are being adjusted yes we can feel a difference but at a certain point and I am afraid the vast majority of us are not experienced enough to just feel the difference in the hand when the finer elements are being tweaked. As Peter said a few cms close on the pivot points is a pretty decent accuracy for the average sword user. This also does not take into account one of the most important factor for a swords effectiveness, the hand on the grip.

One wants to manage the vibration and other elements in a sword to make it as good a weapon as possible, but in all honesty the vast majority of people using swords today are unable to detect the difference on some of the fine adjustments. The hand on the tool has far more to do with how well it works than the tool itself, as an old carpenter I used to work with would say. To use the vibration while cutting issue to illustrate. If the person cutting is not meeting the target with a nice clean arc in the plan of the edges and smoothly using their strength to advantage the cut will be producing more vibration then the sword will induce working in a free environment. i.e. ideal conditions with no influence by the cutter. Even a slightly rotated cut will make the sword start vibrating more than the actual flex induced by a "perfect cut" would generate in the blade.

I am sure all of this is pretty apparent to all but I think it good to bear in mind that the best made sword is always only as good as the hand on the hilt. Helps keep me humble

Best
Craig

[Doug Gardner]

The discussion of computerized analysis is interesting to me, because as a newbie and a lurker, I haven't really had much to add to these discussions. Until now. ;^) I'm not an engineer, but I do know a fair amount about the software you would use to conduct such an analysis. In the process, I learned a bit about the analysis itself.

First, a few terms which will help with writing my response. The type of analysis you are talking about is called a finite element analysis (FEA), and is conducted with Computer Aided Engineering (CAE) software. This software is frequently part of a suite of software which will include a design portion (CAD), a manufacturing portion (CAM, which I understand Gus is intimately familiar with), and the simulation or CAE portion. There are many different types of FEA, including harmonic analysis.

As Gus pointed out, there isn't anything simple about a blade, and I'm not sure how accurate your model of a given sword would need to be, not being an expert in blade dynamics. However, in practice the 3D model which is used for design or machining is usually GREATLY simplified for CAE. Before you can conduct a finite element analysis (FEA, remember?) you must create a "mesh" to represent your geometry. A mesh consists of points (called nodes) connected by straight lines. Curves are represented by many straight lines representing the curvature. (Remember the string or wire art that was common in the '70s? A mesh is kind of like that.) It seems to me that a blade is typically composed of a lot of curves and non-planar surfaces, and it will be difficult to produce a good quality mesh to conduct a really really useful finite element model (FEM). (sick of the TLAs yet?) Part of the reason I say this is because CAE has been attempted to model certain types of musical instruments, and fails utterly on things like steel drums. In addition to having problems creating adequate meshes to represent the geometry, a major contributer to this failure is the fact that the FEM's require detailed information about the properties of the material the object is constructed from. In the case of steel drums, the resonant area is worked so much (and by working it, the material is differentially tempered) that nobody can create an adequate model of the material. Without a sophisticated model of the material, the analysis is simply wrong.

OK, so much for why it can't be done. Now for the ray of hope. Maybe. If all you want to do is predict the center of mass and the primary and secondary nodes of the object, your mesh may not need to be very sophisticated at all. In fact, I'd predict that the best mesh to conduct that type of an analysis would be a 2d outline of the blade, without a lot of detail. Maybe include a couple of ribs to represent the fullers, because they should greatly effect the stiffness of the blade. You'd model the taper of the blade by specifying the material thickness. An earlier note about the hand on the sword reminded me, however, that your hand WILL change the dynamics of the sword, because your hand will act as a mass damper. The impact of this damper will be less if it is very close to a node, which it is, but there will still be an effect of your hand.

Of course, all of this assumes that a few of the material properties are known pretty well. The modulus of elasticity, (or Young's modulus for those engineers out there) should be particularly important. Without this knowledge, you might as well hang it up.

Now, what would this buy you? You might be able to predict the characteristics of the sword almost as accurately as people like Gus or Peter probably could by looking at the drawing. Someone already mentioned that the software is expensive, as is the expertise to use it. For high-end CAE software, you should expect to pay $10-15 K. Yes. That's right. Put 000.00 after that. Generally, trained engineers expect to take about 6 months with new software of this type before they become really proficient. On the lighter side, though, it can run on PCs now! This software is no longer restricted to high-end workstations. Although, for a reasonably complex FEM (say 100K nodes or so) you ought to have a gig or more of RAM. When you want to load that model into memory to start working on it, you'll want to get it started and then go get a cup of coffee. Or maybe go out to lunch.

So, how would you get it done? Simple. Go to an engineering school and find the guys who are learning to use CAE software, or a prof who needs a class project or 3. Give 'em a sword and see if they can figure out how to model it for whatever analyses you'd like. They could then test that model against the known characteristics of the sword. Hey, good class project so far, right? Then have them modify the model to match another sword, and see if the model works as modified. Rinse and repeat until the process works or they figure out it can't be done. My guess is that it would be easy for Gus to get this done, because he already has a 3D model of the thing (at least in the rough) in his milling machine.

Just my 2 cents.

Oh, by the way.... I'm Doug. I just ordered a Viceroy (my first "real" sword). I have 4 wallhangers, including the UC limited edition Anduril, plus a CHEAP sabre from India I've had since 1982, and a fencing sabre (maybe it is an epee') I've had just about as long.

--Doug