Blade Material and Forging

[Guerre]

I am currently trying to put together a small project studying the manufacture and durability of blades (focusing on foil blades). The intent is to create a spreadsheet or database detailing stiffness (in a numerical value), force needed to cause a buckling break in the blade and general longevity of the blade (via a derivation of required forces for permenant deformation).
I am assuming that few forges would produce the information i'm looking for because it would give away too much about their processes. The most important pieces of information are the Young's Modulus of Elasticity (E), the yield strength (Sy) and ultimate strength (Su). For those not familiar with the terminology these three factors all contribute to how your blade bends, how much force it takes to cause a permenant bend and how much force it takes to cause deformations in the blade leading to breakage in the material. Over the next year i will have access to some machinery which will allow me to perform tensile tests on some blades, in an attempt to find this information. If anyone else has done similar research or has information (even if its dated) on forging processes (which may not actually involve forging), or can direct me to a site or an individual that may have some of this information i would appreciate it.

[CvilleFencer]

My only suggestion would be to contact Alex or Barry Paul at LP. They produce most of their own blades in house. They do a great deal of testing of their own and competitors blades and I am sure they have a great deal of data not just on their brands but others. Also they are very knowledgeable about testing and the engineering aspects of fencing gear in general, even having helped some other forges establish their procedures IIRC. I am not sure how much of their actual "recipes" or processes they can divulge, but I am sure they would be a very valuable resource for you.

[Alex_Paul]

You beat me to it Cville!

We test a lot of blades but on blades we buy in we only test durability in terms of bending cycles as per the FIE test.

We may be able to help / send samples for testing. e mail alex (at) leonpaul.com I am off to the JO's now so I may take a week to get back to you.

Alex

[SJCFU#2]

Have you tried looking at Appendix A to the Material Rules for mechanical characteristics?

This sounds like an interesting study. The big problem I see will be difficulty achieving any consistent results if data is collected based solely on blades that are used in actual fencing. It seems like the varying number of beats, misapplied bends, and other flaw-inducing events any individual blade might experience would sufficiently alter the results to prevent drawing anything more than the most general of conclusions. Of course if your study is strictly limited to testing of brand new blades under laboratory conditions then you will be able to get more consistent results, but then the question will how well do these results compare with blades that are actually being used.

[jjefferies]

Of course if your study is strictly limited to testing of brand new blades under laboratory conditions then you will be able to get more consistent results, but then the question will how well do these results compare with blades that are actually being used.

Such a study would almost like be statistical in nature. So how many samples can you obtain of new weapons? Old i.e. broken weapons, you can have as many as you want. But then how to perform meaningful tests.

Of course if you can do it we would all be interested in the results particularly any statistics determined.

[Guerre]

The thing about it is that there realy isnt going to be alot of difference in the data. All three values are determined by the process and the material make up. Changing the recipe for the steel one way or another just a tiny bit wont realy make that much of a difference in terms of those values. The current testing is bending and how many cycles the blades can withstand, im trying to look at it from the viewpoint of re-setting yield strength (with every permenant bend that section requires more force to cause another permenant bend). This has little to do with the way someone actually hits or with what kind of hits, the whole point is to find out just how much force or repeated force it takes to deform the blade to the point where it is no longer functioning or breaks. In any case thanks for the replies.

There should be almost no variability between blades in the same batch in these terms. I plan to test the forte of the blade since minor imperfections in the steel will have a smaller effect with a larger cross-sectional area. Those imperfections are the reason that two blades from the same batch could have a massive difference in lifespan.

[Pescados666]

You beat me to it Cville!

We test a lot of blades but on blades we buy in we only test durability in terms of bending cycles as per the FIE test.

We may be able to help / send samples for testing. e mail alex (at) leonpaul.com I am off to the JO's now so I may take a week to get back to you.

Alex

Free samples? If so I could do a few of my own tests...

[Kayla92]

My name is Kai, and i'm a student studying physics at a college in england, and i was interested in your study. I was wondering if you've found any findings so far? I would love to be able to use your work so far as an example to my study of steel use in foil blades.
Thanks.

[the ancient one]

There should be almost no variability between blades in the same batch in these terms. I plan to test the forte of the blade since minor imperfections in the steel will have a smaller effect with a larger cross-sectional area. Those imperfections are the reason that two blades from the same batch could have a massive difference in lifespan.

Do eddy current testing periodically with the stress testing and include that data. I'd love to see what you come up with.

[brtech]

Eddy current testing was once believed useful to diagnose impending failures in blades. There were some testers constructed that could be used by armory staff at World Cups and Gran Prix events. Dan DeChaine has one made in Russia. That idea didn't gain much traction, and is not part of any test regime.

[Superscribe]

There should absolutely be some sort of cyclic loading. Can you do some of that, and do you have access to the necassary machinery? Verification of fatigue strength for published numbers would be helpful, and you could account for stress concentrators (wire groove). I'm curious as to what you mean by buckling break. Specifically, what your boundary conditions are going to be for the specimen.

[Mergs]

The term 'buckling break' has me wondering, too. I think what was meant was buckling failure. Which is a totally different failure mode. Or maybe not, and the OP is thinking that if a blade fails when it is bent, as in a hit, or during the flex test, that constitutes a 'buckling break'. Sorry, a break is a break, and in the case of metals it is caused either by tension or torsion. As for stress concentrations and cyclic loading, the issue here is a non-uniform cross-section along the length of the blade, as well as whether or not you are using a new blade that has not had the edges nicked up (fracture initiation points, which are really only smaller stress concentration points).

The things that the OP are looking for are best found in samples of the material in an as worked condition, not the finished product (although, the samples can be obtained from the finished product). There are too many variables associated with the forging process, in my mind, to effective find the information being sought, from finished products, let alone draw any definitve conclustions regarding the durability of blades.

[Superscribe]

It would be important to try and get something that wasn't nicked. That's MTS operator 101. I don't know that cross section is something that would be important if you used a small section of something closer to the foible. Then again you could extrapolate based on varying cross sections. There are formulas.

This smells like sophmore mechanical engineering or material sciences. Except for the buckling break part.

[the ancient one]

Eddy current testing was once believed useful to diagnose impending failures in blades. There were some testers constructed that could be used by armory staff at World Cups and Gran Prix events. Dan DeChaine has one made in Russia. That idea didn't gain much traction, and is not part of any test regime.

For poor cost vs benefit or lack of meaningful indicators?

[Guerre]

I've had this sitting on the backburner for quite a while now and have not had the time to revisit it. A colleague and I have looked at several ways to test the blades (he had more of an interest in sabre blades which are considerably harder to model). As some of you have mentioned the geometry is a very important component of lifespan (stress concentrations blah blah). The only way to really look at one of these blades completely is to put the exact geometry into a finite element analysis package like CATIA. The only thing i was asking about was the tensile strength, because a simple model can be constructed from that information to find out a stress cycle number by basing the theoretical model on the weakest component of the blade.

I have recently resigned to giving up on tensile testing, because i can get a decent approximation of the Sy and Su with a Brinnell Hardness tester. Once i have the time to finish the tests i will post the data (current samples are Absolute FIE, Vniti FIE, BF FIE)

[Guerre]

buckling break: to specify that i was not just looking at compressive or tensile stress breaks (since that would be nonsense), which is what is normaly derived from Sy and Su values

Cyclical loading: cyclical loading is what is currently used by the FIE (as far as i've been informed) and does not account for the variance in the rate of loading.

FEA based on numbers from tensile tests or brinnell/rockwell tests is the most accurate way to test this without building a cyclical loading rig with variable loading rates.

[brtech]

For poor cost vs benefit or lack of meaningful indicators?

Believe it's the latter