I’m not an engineer or scientist. I can’t adequately prove any of this in a scientific arena, and don’t care to try. Most of what I have to say here is from a combination of intuition, experience, logic and reflection. A purely formulaic engineering approach to axe handle design would almost certainly fail. I’m more interested in experience informed by science than the other way around. Feel free to make any efforts, intellectual or physical, to verify or cast doubt on the ideas presented here. To me this is an ongoing conversation with myself and whomever else cares to chime in.
There is much more to be said on axe handle design than what is presented here, some of which I wouldn’t consider myself up to the task of addressing. This article focuses mostly on the problem of breakage near the eye, which is very common for a number of apparent reasons which I’ll outline in this article. Another incidental focus is on side to side stresses and factors that affect breakage in that direction, because that is a way in which the design is especially weak.
An axe head with a wooden handle has some inherent problems. The head and handle are made of very different materials that behave differently. Steel has a very high density compared to wood. Wood is much more flexible than steel and will dent and break more easily. When using an axe, these differences can cause problems, such as the wood being damaged by forceful contact with the hard unyielding metal head, or the relatively high density of the steel head behaving differently than the handle, thereby putting stresses on the weaker wood. Breakage just below the eye is a very common occurrence. This article and video are an attempt to explain some common reasons why axes frequently break near the eye, having to do with design, or perhaps lack of design in some cases. Breakage in the main body of the handle can of course also occur obviously, but I’m not really dealing with that here. I’m pretty sure that the greater percentage of axe handle breakages are initiated right where the handle meets the bottom of the eye, or within the first few inches of handle, especially if the breakage is not due to wonky grain or other defects. Shear stresses seem to be particularly high in this part of the handle.
In this article, I will be assuming that we are dealing mostly with American axe head patterns, which tend to have thinner eyes than European and Scandinavian axes. Even though American axe styles migrated back to Europe (many axe patterns on that side of the pond are actually American or modified American patterns) the axe eye sizes largely remained bigger than American axe eyes. This is an important point when we look at overall handle design, because with any given axe head, the eye shape just is what it is, and the size and shape of the wood where it enters the eye is therefore pre-determined. Some of these problems are obviated by the use of tapered axe eyes, in which the handle feeds in from the top and fits by friction, but that is a separate subject also. The assumption here is that we are dealing with American style patterns that are wedged from the top. For whatever mix of cultural and practical reasons, these axes have pushed the limits of strength and resilience of the wood used in handles, by evolving toward a small eye.
Aside from the size of the axe eye being fixed, there are two other things that are pretty much givens as well.
One is that the section of handle just below the eye, lengthwise (poll to blade) is wider than the rest of the handle. If the whole handle was the same front to back dimension as the eye length, it would be unusable, so the body of the haft has to slim down soon after leaving the eye.
Predetermined factor number two is that we need a slight flair in handle thickness just below the eye at the back of the handle, as well as on both sides, so that the head seats firmly around the bottom as it is driven on. The front edge of the handle can come straight out of the eye if desired, with no rise, but the other three sides need at least some flair, though not very much. In my view, it is always unnecessary, and also a detriment, to come out of the front of the eye and then immediately outward, forming a shoulder. I see no reason to do that, and every reason not to. If the handle isn’t completely straight coming out of the front of the eye, the rise is best made as a gentle transition.
WE CAN CONSCIOUSLY WORK AROUND THESE FIXED PARAMETERS. IN OTHER WORDS, DESIGN AROUND THE THINGS WE CAN’T CHANGE.
Both Dudley Cook, in The Axe Book) and Peter Vido in his article, “The Devolution of Axe Handles” on Axe Connected, have documented the increases in axe handle thickness over decades. Mid last century and earlier, handles were much thinner on American axes than they are now. Presumably the increase in average thickness is an attempt to make axe handles stronger. “Strength”, however, is something of a sloppy concept to superimpose onto this system. RESILIENCY is really what we want, and although strength, as most of us will conceive that word, is a factor. As we will see, excessive rigidity in one area, which may equate to resilience or toughness under some circumstances, may lead to undue stress elsewhere under actual use. In other words, strength in any given area not only fails to equate to overall functional resiliency under stress loads, it may actually worsen resiliency.
Lets take a hickory billet, 3 feet long and 2 x 2 inches thick and suspend it between two stumps like a bridge. If we stand on it, it is not going to break, because hickory is very strong. Now lets remove 1/8 inch of thickness at a time, along the whole length, and keep standing on it. It will begin to bend in the middle, and eventually it will break. The thicker piece then is stronger, and is more tolerant of that particular stress. But it is folly to think that a thick club of wood is the best thing to put on an axe, because we don’t stand on axe handles and more importantly, the behavior and character of the massive, inflexible, heavy handle affects the rest of the system. If we could have that kind of strength in the body of the handle without compromising function or creating new problems, that would be great, but it is not so.
Lets go back to that 2x2 hickory stave and carve one end down enough to mount an axe head on it. Put it back between those two stumps like a bridge, but this time, the axe head alone rests on one stump instead of the wood. Stand on that and where is it going to break if it breaks? If the axe head has an eye of a fixed size, and therefore the size and shape of wood where the handle enters the eye is a given size and shape, what can you do to modify that handle to decrease stress on the thin wood where it enters the eye?
It’s all about load sharing. With a thick, inflexible handle, we not only fail to distribute stress, we actually concentrate it.
When making a bow, an all important step is tillering. Tillering is adjusting the bend of the wood, so it bends just how you want it to. Of great importance in tillering is making the bends gradual or even. Usually the bow is carefully scraped and frequently tested to fine tune it for an even arc. If one spot is much too thin, stress is concentrated in that spot when the bow is pulled and that is likely where it will break unless there is a flaw in the wood that causes it to break elsewhere. The bend of a bow is not always made as an equal, even perfect arc, in fact rarely so, but any changes in the arc are always made gradually, or weak spots will result. An axe handle with a thick body and a thin eye section is somewhat analogous to a bow with a thin spot. When stress is applied to the system, a thick handle can lead to undue stress put on the thin eye area, because the body of the handle fails to flex. A viewer pointed out to me that this phenomenon is known in engineering as a stress raiser, or stress riser. Essentially, it is where an object changes shape and size very quickly. Because of this inherent difference in size and shape, the thick handle behaves very differently than the thin eye area of the handle when energy is applied to the system.
Not to be excessively pedantic, but again, it is really important to note that the weakness of the eye is actually compounded by a very thick handle. The eye section will always be the same size, but it actual vulnerability is somewhat contextual. Understanding this point is a game changer.
So let’s take the perspective that because the wood where it enters the eye will always be the same fixed size, we need to work with the rest of the handle design to accommodate that fixed design feature. Further, lets take the view that the design we choose for the body of the handle can be used either as a means to distribute stress evenly, or as a means of concentrating stress near the eye. Lets next look at how we can easily fail to distribute stress.
The following diagrams with text are more descriptive than I can put into just words. Given a fixed axe head design, the main principals we are working with are as follows:
The axe eye size is fixed, thus the size and shape of the wood where it enters the eye is fixed.
The handle will be wider just below the eye in the bit to poll direction than for the rest of it’s length.
The grain of the wood is cut across where the axe handle enters the eye and in the transition as the wood thins down into the rest of the handle, creating a vulnerability toward splitting along the grain at those places.
These changing shapes and sizes in the various parts of the handle necessarily create concentrations of stress under use, because they don’t bend and behave the same. I think it benefits us, to design in such a way that we minimize these differences, while retaining ergonomics and overall functionality.
A rigid handle, because it fails to adapt and flex when stressed, stacks up stress on the weaker thinner eye. A handle with some flexibility to it is like the evenly tillered bow, it distributes that stress by flexing along it’s length. Over time, I have come to think of the handle as working and flexing all the way up into the eye. How much it actually bends up in the eye is unknown, but it must certainly flex a little bit. Regardless of whether we can quantify that, it’s a useful view to take because it highlights the importance of aiming for a more even stress load along the whole handle length, and not just thinking of the handle as something that exists solely from the bottom of the eye down. We may then conceive of a theoretical axe handle that will bend in an arc all the way through the eye from the top of the eye to the butt when pressure is applied to the middle of the middle, thereby making for an even distribution of stress, just like a bow. While that concept is good for gaining some insight, there is much more at play.
The forces at work on an axe handle are not that even or predictable. For one, we have the juncture of two very unlike materials. Steel is very hard and very dense. A cubic millimeter of steel weighs a lot more than a cubic millimeter of hardwood. The wood is soft, more flexible, but decidedly weaker and lacking in toughness. So, one problem is that the wood is continually working against this very hard material at a defined and narrow juncture. Usually, that juncture at the bottom of the axe head is straight, which is even worse. It is also often sharp, compounding the problem yet further. The ears protruding from some axe head designs probably help with this problem, because the juncture is effectively longer and the stress is not applied at even right angles across all the wood fibers all at once. (thanks to Rooster from Axe Junkies for that insight).
Being that the wood and metal are very different densities, they behave differently when energy is applied. The wood and metal are joined, but they won’t act completely as one unit because of their material differences. The axe head of greater density will have more inertia than the handle, either when moving, or when still. The lighter handle, having less inertia, may move easily to follow the head when it’s moving under most circumstances, but there must also be some violent forces working against the wood as well when the heavy, hard head changes directions suddenly. Remember that inertia is the idea that a body that is moving wants to keep moving and a body that is still wants to stay still. As the heavy axe head is either moving and wants to keep moving, or is still and wants to stay still, there must be some conflict between the two materials.
The mass of the handle may also have an effect on breakage. Impacts to the side of the head, are not uncommon during bucking, felling and limbing, even though that’s not how we use an axe on purpose. For argument’s sake, imagine a handle that is just a 4x4 with an axe head on the end. Not only is there no stress distribution along such a thick handle, because it will not bend at all, but the large mass of the handle has a great deal of inertia. If we slam the side of that axe head against a tree, the 4x4 in it’s great momentum, is going to want to keep traveling past the tree and the head will probably snap right off. Imagine now the same axe head hafted onto a thin light handle, and lets slam that against a tree. The thin handle will flex of course, but the lower mass of the light handle is also not going to have nearly the momentum and forward inertia of the heavy 4x4 handle. When the heavy mass of the axe head rebounds off of the tree, it’s going to be easier for the light handle to reverse direction and follow it. You could think that the thicker handle is less likely to snap from being jerked about by the suddenly shifting heavy mass of the axe head reversing it’s direction, but remember that the eye portion of the handle is exactly the same size and shape in both handles, and that is the area likely to break in a side impact, not the main handle body. Another way to think of it is that when you have a heavy compact head and a thick, heavy and inflexible handle body, the unfortunate thin, weak eye portion of the handle is caught in between two masses that are potentially undergoing violent actions and reactions. An analogy that might be useful is to imagine that your brain is made of steel, so your head weighs 80 pounds and your spine is fused to you can’t bend or twist your body. Unless your neck is as thick as your head, your poor thin neck is going to be in danger of snapping! If we can make the spine flexible so that it adapts under stress, it should reduce some of the stress on the the neck.
I’ve talked about side impact a lot, but what about front to back impact? Front to back impact, from either chopping, or striking with the butt of a poll axe, is after all the most common purposefully applied stress. The wood of an axe handle, especially at the eye, is longer in this front to back direction, and therefore stronger in terms of shear strength. By making the eye much longer than the rest of the handle, the design creates a situation similar to having a thin eye and a thick handle, where the body of the handle is thinner this time and not the eye. But, it is probably a pretty safe assumption that this design evolved for legitimate reasons.
One way to make the handle stronger might seem to make the handle the same dimension as the eye along the entire length of the handle. But, that would make a very uncomfortable long oval, to the point that it would be nearly unusable. Also, it would not just make the the body of the handle stronger, it would also increase overall rigidity, thereby decreasing load sharing by the handle body in a more general sense. We can conceptualize these simple directions of stress for theoretical purposes, but the reality is that an axe undergoes complex, dynamic stresses under real life use. I favor the idea that the resiliency of the handle body as a whole, and not just in one direction or another, results in a more complex stress distribution.
I think that the reason axe eyes are much greater in the front to back dimension than the rest of the handle is that they need to be because the shear stress at the point where the handle works against the hard axe eye is so high. While side impacts are not uncommon, they rarely are equivalent in intensity to the repeated forces that we typically use in the normal tasks of chopping and splitting wood. This high bit to poll force, creates stresses that are probably too high for a much smaller dimension of wood to endure. Note though, that we can slim the handle down pretty quickly just below the eye, yet we do not easily break it under normal, reasonable use, even if the thickness of the handle where we hold it is not much greater than the side to side axe eye dimension. That probably speaks to the difference in the degree of stress each of those areas is actually under in real life use. Like most traditional tools, these dimensions probably evolved gradually out of a combination of tradition and repeated exposure in the crucible of experience. As axe eyes became narrower for thinner handles, they also had to become longer. A longer eye increases the strength of the wood just below the eye when stressed in the narrower side to side direction, not just the poll to bit direction. As all builders and many others know, increasing the thickness of a board contributes to it’s rigidity much more than adding to it’s width, but adding to width does still add appreciable strength, especially when the length added is over twice the width as it is in an axe eye.
Someone brought up the point somewhere that a sledge hammer handle is not long the way an axe eye is, but they still work. True, but a sledge hammer also does not get stuck in wood and have to be wrenched free, which is one of the greater stresses an axe handle undergoes. The amount of stress that I regularly apply to axe handles to un-stick them while splitting wood is very high. Sizable maul and sledge eyes are also usually oval and at least one inch wide. A quick measure of some splitting maul and sledge heads laying around the place compared to a couple of full sized axes:
Sledge and splitting maul heads varied from 1” to 1-1/8” wide, all at least a little bit oval
Full sized axes, Craftsman (American) about 3/4” x 2-1/4”, Hults Bruks (Swedish) 7/8” x 2-1/4”
This brings us to another very interesting point, which is that stress in never going to be consistently evenly distributed. Take a ruler or piece of wood that is of a uniform width and thickness over it’s whole length. If you affix or hold one end of this and bend it from the other end by simply pushing on it, it will not form an even arc. You might expect it to, because it is a uniform thickness, but it doesn’t. The evenness or unevenness of the bend will depend on how the ends are held and whether one is held in place while the other is bent. In real life use, an axe handle will bend unevenly along it’s length depending on the stress and the position or status of the head. If you drive an axe head into a stump deeply so that it can’t move and push on the end of the handle, it will bend more near the eye than at the butt end, assuming the handle is of a fairly uniform thickness and width along it’s length. This demonstrates stress stacking at the eye, like I’ve been talking about. How many different things is that now that cause stress and vulnerability near the eye? 5 I think? It’s a lot. The possible usefulness of that information is that we might be able to relieve more stress by tapering the handle gradually toward the butt end. I have a basque axe handle here that drops in from the top which is tapered pretty much from end to end. Since the butt end of a handle seems to be under the least stress, it can probably afford to be very thin.
So, the long eye may compensate somewhat for the greater pressure that seems to be experienced by the head end of the handle. Since the handle at the eye is always working against the hard, heavy metal, sizing the eye up at that point makes sense.
Between two theoretical extremes of handle thickness, there will be functional grey areas which may favor certain types of breaks, or stresses. Sometimes the eye will just shear off almost flush, and other times the handle will split lengthwise or diagonally in different ways. If the handle is made too thin, it may eventually end up breaking more in the body of the handle instead of at the eye, which will also depend on the stresses applied, which will in turn vary with both the work and the user’s work style.
Given the inherent limitations of the materials used, and the fact that an axe is subject to varying types and degrees of stress, and also that it must be usable for any given unique person, no handle can be perfect. Within the extremes of a 2x4 with a head stuck on the end, and a handle that is exactly the same width as the eye from top to bottom, or continuously tapered from top to bottom, there is surely much leeway in designing a handle with adequate resiliency. What those limitations and compromises are exactly is not a debate that I care to engage in much. To turn these ideas into a very dogmatic approach to handle design is no doubt a mistake. There are too many variables involved. A person could really dig deep into this subject with slow motion studies, analyzing different stresses an axe undergoes and how various handle configurations behave, complete with rigid engineering formulas applied to an inherently variable material. Those pursuits are expensive in materials and time/thought. I do think they could be fruitful in coming to a better understanding if approached with intelligence, due caution and humility, but they don’t interest me enough to pursue at this point. And really field testing is where it’s at in the end. What’s going to happen if you used the axe regularly through a whole season for lots of different work. Or what is going to happen if you hand that axe to a novice. Not only might they be more likely to break it, they might break it in a completely different way. It’s enough for me for now to understand that I need an axe that feels comfortable and works for me, and which does not violate these principals too far; then it’s back to actually chopping wood to see what happens.
As many know already, I’m not a big fan of dummy rules. But some guiding principals could prove useful when engaging in handle design and modifications. To simplify things, I would make the following bullet points:
Design and modification should be informed by ergonomics and actual use, not just theory
Design around the things you can’t change
Think of the flex, or lack of flex, of the handle under stress as a means to either distribute or concentrate stress
Think of the handle existing and working all the way from the tip through to the top of the eye
Where changes in shape and size of the handle are necessary, favor gradual transitions over sudden ones
Make the handle body flexible enough to absorb at least some of the stress by bending
The ideas presented here could seed some potentially interesting innovation and experimentation. As things stand now though, I would just like for people to see the ways in which more radically sloppy design and execution, or overbuilding of certain parts, out of context of the rest of the system, can result in increased vulnerability to breakage. That understanding could inform the manufacturing side of things. A good start would be making handles at a reasonable thickness to avoid unnecessary stress on the eye area, and avoiding radical transitions and unnecessary protrusions of wood which result in unnecessary stress risers. And, honestly, that is probably good enough. We are dealing with a tried system here that does work pretty well if we stay within certain parameters. But I also think that for a motivated person, it’s worth asking questions, innovating and testing any assumptions and traditions. Just do it for reals and not just from the arm chair. Which brings me to the point of theory v.s. application.
It is important to approach this problem with a functional attitude. There is no perfect solution and compromises will be made. There are many factors involved, Like the type of work we are doing, axe head weight, handle length, head design, wood species, wood quality etc. In the end, the tool has to pass muster in the crucible of real life application. If you commandeered a non axe using engineer to design a handle, based around the existing eye size to theoretically compromise in all the right areas for maximum strength and resiliency under what are perceived to be the most commonly encountered stresses, that handle would probably not be that fun to use. To take the resulting concepts and fit them to our needs might be very useful though. Don’t make it about just ergonomics and usability, or just about maximum strength and resiliency, but rather about whatever meeting of those two actually seems to work. Again, that is not a sweet spot, it’s probably more like a sweet grey area.
Way back when I first picked up an axe and actually started using it quite a bit, I did not come at the idea of handle shape and flexibility from a theoretical direction at all. I probably read some stuff that suggested handles should not be too thick, and I began to tune them by feel until I felt they were “right”. It wasn’t until later that I started thinking some of the theory out. That internal reflection, coupled with continuing observation, experience and continuing cycles of experience and reflection has steadily evolved my insight into the subject. I think it is okay to approach the problem from either direction, but if I had to choose one, it would certainly be experience and intuitive feel, over a more analytical approach.
My advice for users at this point is first and foremost to not accept what manufacturers have on offer as is, and don’t take any information offered by anyone at face value. Look at each handle with curiosity, modify nearly every handle you get and then use it with the same curiosity. If it breaks, follow through with said curiosity and try to understand what stress resulted in that breakage, where it initiated, and how that stress might have been avoided. An axe handle is probably best viewed as a consumable item, so as such why not make it an opportunity to experiment? Probably the more common approach is to assume that the manufacturer knows what they are doing and that their design is a good baseline from which to approach modification. Unfortunately, that is often not the case. It is quite clear from the products coming out that axe manufacturers are not axe users for the most part, and are probably not interacting enough with experienced users to influence the design changes that are happening.
Okay, lets tie this thing up with some notes on use. No matter how good the design is, a wooden handle on a relatively heavy and very hard piece of metal, is a configuration that is inherently weak. It is shockingly easy for a complete novice with an aggressive attitude to break an axe handle which might have lasted someone more experienced for a very long time. There is no substitute for skill in using this tool and no way to get that experience except for time served on the ass end of an axe. And this brings us, the users, into the picture as a crucial and influencial part of this system. How we use or misuse the tool has everything to do with the nature and level of stresses applied to the head and handle. Some tips on agile use are as follows:
Do not grip the axe too tightly, or try to push the axe head through the work, use the handle more as a guide and a means to whip or throw the head into the work. As the heavy head jerks around pulling and pushing on the weaker handle, the handle can more easily follow the head, bouncing, flexing and rolling with the punches, if the grip is light. You will also be working less and will absorb less shock if your grip is light.
Avoid hard side impacts to the head when possible, because the handle is weakest at the eye in the side to side direction. Side impacts are common in bucking, felling and limbing especially.
Don’t use the axe as hard as possible. There is generally no need for that. It’s okay to chop with energy, but rarely is it necessary to chop as hard as possible. It may be more necessary and useful to use nearly full force when splitting wood, but side impacts are much less likely when splitting. If a light grip is used and the tool allowed to do it’s work, splitting will rarely result in incidental breakage unless the handle itself is flawed, or is struck against the wood.
It is obvious from this article, and regularly mentioned in my other axe content, that I prefer relatively thin handles. The main arguments I hear against them is that they are hard to grip and that they rob power by flexing too much. They can feel awkward at first, but the better technique becomes, the more natural it should feel. If you don’t like it at first, give it some time. Make sure the axe is not over gripped as that can lead to all sorts of fun stuff like loss of accuracy, breakage and handle shock. I understand that some people have very large hands, but for most types of use, it should not be necessary to grip and control the axe overly much while chopping and splitting. As for robbing power, In normal chopping, it is my opinion that it is best to avoid trying to use the axe handle to drive or push the head through the work v.s. using the handle to whip the head into the work and once it hits the wood, just letting it cut. It’s not that no energy can be delivered through the end of a long handle to the axe head by pushing, but it is very inefficient and hard on hands and wrists. Further, there may even be a mechanical advantage to flexible handles functioning something like a whip. Dudley Cook seems to allude to this phenomenon in The Axe Book and it is demonstrated in the extreme by the video below, where a unique handle material makes a very thin flexible handle possible. This phenomenon is also known in golfing, where the weight of the head trails behind the flexible handle slightly, but is designed to catch up at the end of the stroke giving extra snap. My friend Atlatl Bob also discovered this phenomenon in atlatl (spear thrower) design.
Relatively thin handles used to be the norm out of the box, and experienced users would often tune them down further. If people used thin handles back then, you can probably get used to using them now. Personally, I find a thick handled axe v.s. a thin tuned handle something akin to a beater car v.s. a sports car, but your mileage may vary.
As always, stay safe and happy chopping.
This epic blog post and video amount cumulatively to at least four full work days. This type of deep diving into unprofitable obscure subjects is made possible by my supporters on Patreon Thank you! :)