Dealing with tapering of sword blades

[Alveprins 213]

I have arrived at the conclusion that I wish to make a multibar viking sword...

However - I am curious as how to deal with the tapering of the sword from the guard to to the tip.

 

Getting the blade symetrial, and grinding equal measure of steel on each side is one thing - but the tapering? ...

[steven smith 97]

if the edge bevels are the same angle then the thickness will be proportional to the blades width, so if you make the blade narrow towards the tip then it will also thin towards the tip, i imagine it would be harder to make a sword with no distal taper because as you shape the blade you would have to constantly change the angle of your tool unless the blade was the same width all the way. some swords have a complex distal taper, knives can be pretty complex as well, im not going to get into that though. 

 

hope that makes sense, its something i started doing with my miniature swords because it can be hard to see a ~1/4mm different over 5 inches of shiny steel, i scribe in my grind lines and when everything is ground i (hopefully) have a distally tapered blade without having to think about what the heck im doing the whole time. 

[peter johnsson 147]

The distal taper, or tapering in thickness is rarely straight in most blades. As a starting point it is useful to divide the blade into 5 equal parts. Let the first 1/5 have a drastic distal taper, the next 2/5 have a moderate or minimal distal taper and give the last 2/5 a distal taper that is less severe than the first but more than the mid part of the blade. 
This lay out of the distal taper will give you a lively blade that is stiff for its weight.

Needless to say, there are many variations to this and you have to tweak the proportions of the parts and the severity of the distal tapers according to blade type, cross section and dimension of the blade. As a starting point, or rule of thumb, it works well however :-) 

Quote

 

 

[Al Massey 62]

From what I've seen many of the composite bar examples had very little tapering in profile anyway.

[steven smith 97]

Thank you, Peter, that's some good info. I don't seem to recall seeing much about complex distal tapers here or on bladeforums, I'll have to read up on that. 

[peter johnsson 147]

Migration and Viking period blades tend to have subtle distal tapers, but it is still rare to find blades without this.
The fact that it is mild should not be taken that it is unimportant. As a rule these blades taper both in width and thickness and it is worthwhile to spend some time to note just how they do this. The effect may be surprising.

There is very little published on distal tapers. The info you find on the net is mostly inferred from ideas of what swords *should* be rather than facts.
Thickness, cross section and distal taper of sword blades are the key features for the character and performance of the weapon. This holds true for most any type of sword. 

One example of a pattern welded blade:
Total length of blade: 802 mm
Width at base: 55.4 mm
Thickness (in mm) at 50 mm intervals from base to point: 6.8, 5.9, 5.7, 5.7, 5.7, 4.7, 4.6, 4.9, 4.9, 4.6, 5.0, 4.8, 4.3, 3.9, 4.2, 4.1, 2.5

If you plot this out as a graph you will see the trend of the distal taper conforming fairly well to the basic principle I outlined above. The distal taper is not very even and part of this is from rust damage (=pits in the surface). The blade was not much thicker originally. Most measurements were taken in spots that were close to original surface.
Total weight of this weapon is 1353 gram and a total length of 945 mm.

I might add that the base thickness on 6.8 is unusual thick. This blade also has a fairly deep fuller. Many blades start out 4.5 - 5.5 mm.

Hope this helps.
 

[Brian Dougherty 1,167]

Just in case anyone else was trying to visualize the data Peter listed, here is a plot of it:

 

[Wes Detrick 487]

Peter, if you don't mind me asking...

Why do you think the distal taper dropped off like in some places by a comparably large amount, say 6.8 to 5.9 and 5.7 to maybe around 5.0 (it drops more, but this is the rust I think you were talking about as the blade gets thicker a little later down the blade, unless that was intentional as well) rather than a smooth progression?

Edited March 22, 2017 by Wes Detrick

[peter johnsson 147]

Wes, it is a combination of things.
First of, the blade was not perfectly even in distal taper to begin with. This is normally the case (some ±0.1 mm variation is typical).
Secondly, a sword may suffer damage during use that affects also thickness, but this is rare.

Finally and most importantly, the degree of corrosion plays a role. Corrosion naturally eats away thickness but it might also swell thickness in places by build up or delimitation.

In this case the sword also had some iron inlay in the fuller. It is rather common to see the area of the iron inlay being overground in some places. The fuller can be made wider and the ridges on either side slightly thinner.

However, the drastic drop in thickness at the base is a design feature that this blade shares with many other blades, not a result of corrosion, or mistakes in the making. It is a feature that on Japanese blades is known as "Fumbari" if I am not mistaken. It is an effective strategy in dealing with blade stiffness in impact and bending stresses. It also helps making the main body slightly more lean without loosing rigidity or resiliency and thus allows you to make a lighter blade. 
A blade with composite distal taper will be more agile and handle smoother. It will feel lighter in the hand than a blade with a more constant distal taper.

[Wes Detrick 487]

Thank you so much for the great answer Peter; I really appreciate you taking the time to answer in such a clear way.  The drawing is also immensely informative!

If you don't mind me asking yet another question...

Is this drastic drop off of thickness near the base a very common design with in European swords from both the Viking (maybe Migration too) and the Medieval periods?

[peter johnsson 147]

Yes, it is a common strategy and it becomes more common and more pronounced in later periods.

I must also stress that you will find many examples that do not have this type of taper to a noticeable degree.

Below you will see four examples of swords of the same basic type that show variations in distal taper: composite, convex and concave (but no simple straight taper).

Another thing to take away from this is that if you only have two or three measurements of thickness of the blade, you will completely miss this important element of design.

[Alan Longmire 2,717]

Okay, I'm gonna have to pin this one.  Thanks, Peter!

[Al Massey 62]

Peter, I believe that in Nihonto fumbari actually refers to profile taper, a marked widening of the blade as it approaches the guard, rather than thickness taper.

It coincides with an increase in thickness as well.

[Wes Detrick 487]

Again, thank you Peter.  Eventually, I would to try my hand at a sword, and I do not have in live examples to document, so my only exposure to what I see on the web.  Your information and experience are invaluable.  

[peter johnsson 147]

4 hours ago, Al Massey said:

Peter, I believe that in Nihonto fumbari actually refers to profile taper, a marked widening of the blade as it approaches the guard, rather than thickness taper.

It coincides with an increase in thickness as well.

Hi Al,

This increase in width coincides with an increase thickness as well, as far as I can tell and have seen. It is slight but it is there.
This increase in thickness at the base of blades is a subtle thing and easily overlooked. Sometimes it does not become evident until you take callipers to the blade and measure it. To my experience, it is still important and worth attention.

[Brian Madigan 17]

I've always thought of fumbari as 'tree trunk' or mountain shaped. When forging I start with a lot of material where the hilt will be. I will upset that portion of a bar if its not wide/thick enough. Good question/thread!

[Alveprins 213]

It seems this topic has evolved quite a bit on it's own.  I thank you all for the feedback - it will be most valuable for me when I eventually start working on my sword.

EDIT: Some incredibly detailed explanations from several of you guys... You are quite the assets for the community on this board! I am glad to have the priviledge of being able to post here... I sincerely mean that. Great stuff!

Edited April 11, 2017 by Alveprins

[Viktor Johansson 19]

This stuff is gold! Thanks Peter.

[Robert D. 105]

I second this, I would like to start a sword this winter and I will for sure be referring back to this when I do. 

Thank you good sirs. 

[Al Massey 62]

Here's some stats from a Napoleonic heavy cavalry sword, 1796 pattern, I acquired last year. Note the extreme degree of thickness taper:

Weight w/scabbard 2.035 kg or 4.49 lbs. Weight sword alone 1.093 kg or 2.41 lbs (2 lbs 6.5 oz.) Blade length (spear point mod)- 34.5 ". (There is a small chip at the tip- without the chip I estimate it would've been 34.625") Overall (unsheathed)- 39.5". Inner grip length about 4.25". Thickness at guard- 8.35 mm or 0.329" "" PoB- 5.1mm or 0.200" "" midpoint- 4.25 mm or 0.167" "" end of fuller (about 6.5" from tip) - 2.1 mm or 0.083" Width at guard- 39.5 mm or 1.55" "" at midpoint of blade- 35 mm or 1.38" "" at end of fuller- 35 mm.   Not a lot of profile taper but the thickness 6" from the tip is barely a quarter of what the blade at the guard is.

[Florian F Fortner 127]

Peter is spot on with his analysis!

In my research with original blades, which ranges from 15th to 18th century european blades, a constant bevel angle is a rare thing. To get good handling in fencing, the cross section curve will be necessarily nonlinear, this implies a constantly changing bevel angle along the blade. This applies to all swords - two-handers, sideswords, rapiers, smallswords and others that are intended for proper fencing as described in the various treatises! (Meat cleavers like Pallasches and some cavalry sabers are exempt from this, there the momentum in swinging cuts is the focus of design)

Usually the cross section curve has a double S-shape. The cross section at the hilt is big and drops significantly within the first 1/5th of the blade, then stays level for some time before dropping again before about a 1/3rd from the tip, then tapering out towards the point (see picture, a very nice Juan Martinez rapier blade from around 1600)

With a linear taper, handling will always be less than optimal and many people try to correct things through the weight of the pommel which makes matters worse.

A good blade will handle well without tiring the arm too much if you try it on its own without hilt, grip and pommel. If it doesn't, it never will. If you close your eyes, the subjectively felt weight will be in the 1/3rd closest to the hilt. 

Here are some cross section diagrams of original blades:

Various one handed swords

 

Some big two-handers:

[Al Massey 62]

Ok, the termnology- is the bottom scale thickness in cm and the y axis distance from guard? Which scale?

[Jerrod Miller 443]

The y-axis is labeled as "cross-sectional area", so it doesn't really tell us much since there is distal taper and profile taper.  The x-axis is labeled "standardized distance".  I assume that means it is essentially unitless.  

[Florian F Fortner 127]

Yes, the y-axis is the cross section area in mm^2, the x-axis is the standardized distance from the hilt (sorry for the german labels, I took those directly out of a paper). A curve of the distal taper would look similar but will be incomplete information to judge the blade performance, only the cross section says it all. If you want I could give you the distal and width taper curves as well...  I just wanted to get across that none of all those curves will be linear and grinding a good blade will be more work 

[peter fontenla 98]

Amazing Post !!!! I do not have words to thank everyone for such accurate information