Some of my earliest ever blog posts concerned axe handles and handle geometry; surprisingly on re-reading them I was not cringing as much as I though I would, I still stand by pretty much all I put those posts. However, after starting to sell heads it seems to be time to put up a post detailing how I put a handle on an axe.

I will give you the dimensions that I have used on this head- which is a prototype for the lighter head that I alluded to in my last post on axes. However I would strongly advise that nobody sticks slavishly to them, this handle fits my hand and my swing. Your hand size may well vary as will the optimum swing for you. I have quite a few axes and have tuned the handles so they all swing the same, this makes it easy to swap on for another in the middle of carving and be able to work accurately right away, without having to dial in on each axe. My heads are not cheap and I assume anyone that buys one will have an axe already, as long as you like it I would recommend making a handle for my head that will mimic your original one.

The other thing I always advise people that buy a head from me to do is to make the first one out of green wood- the reasoning being unless you are really experienced at hanging axes you are likely to a) not do a fabulous job b) spend ages doing it, especially if it is seasoned hardwood c) want to keep it on using it, despite a) and because of b) .

So, make a crude handle out of greenwood- you know it will eventually shrink so you can’t get attached to it, it will be easy to shape and it will give you a chance to dispassionately swing your axe head, make some spoons ( or whatever takes your fancy)  and know what you want to change. Either way the processes are the same.

This was my starting point.

Rather than carve out a green handle each time I have accumulated a selection of handles that I can try out on the head- the long wedge secures it. My best guess was fitted, although I knew it would be too long. The one above it was more curved, the one above that considerably less. The two billets for the handle were split from some fairly small diameter ash, – it had been cut a few year ago and had been in my workshop over a year but had been kept in the round albeit with a huge split down it, so was pretty dry but still quite nice to work.  I marked out the blank I wanted to cut from it, this was split out with a froe- and I managed to get a hammer handle out of the waste on each side.

I then used the axe to hew it to a rectangle, it quickly became apparent that the handle I had selected didn’t suit so I fitted another. I like quite a large palm swell on my axes, this does mean that the blank you start from needs to be quite thick.  The squared off blank  was  380 x 75 x 42mm

I then drew out the profile I wanted, and cut it to length, a couple of stop cuts make it easier to shape the end. An axe is not the easiest way to do this,  band saw and linisher is probably the quickest option, but all the time I was learning about the axe I was going to handle ( and admittedly the more I used it the less options I had to change the design of the handle which is why I recommended carving some spoon banks rather than jumping straight into the final axe handle)

You need to allow a bit for shrinkage- only a few mm if the wood is quite dry as this was but if it is green it is surprising how much it will shrink, which is what happened to the temporary handle I am using- it was prefect in the hand when green but ended up too thin when dry, so was consigned to my test handles box.

I then draw on a centreline- and take it down in the other plane. I also do some rough shaping of the corners especially on the butt- I feel it makes it less likely to split if you don’t expose a mass of flat end grain.

At this stage I let if bake under the windscreen of my van for a week. I weighed it and have checked the weight loss periodically, when stabilised I  could carry on. This happened  in 4 days, it being uncharacteristically sunny in for this time of year in Wales.  Also if possible I like to do my woodworking on a Monday, after a few days out of the forge my hands tend to be clean enough not to make a complete mess of the pale ash. I have marked out again where I expect the head to end up, you can see there is quite a lot of excess- this is intentional and makes fitting the head easier.

I carved down the handle to get the head started, until the handle is protruding it is better to knock it on like this. Use a wooden mallet and make sure you have the corners removed on the butt so you don’t chip off any sections of short grain.

Next you want to check the alignment as it goes on, it is in theory pretty obvious where you need to remove material to sort out issues but I find it much easier to mark it out the areas I need to remove material from whilst the head is still on, mistakes can be made and you can’t add wood. In this case the head needed to be rotated a few degrees around the handle, I marked the area, on this side above the lug, on the other side below. The block I am holding is used to remove the head at this point, it is a loose fit in the eye but almost round at the other end. A few gentle taps and you can feel the head moving out, this is pretty safe although I would advise taping up the edge. However even with the head taped up there are dangers.

You really don’t want to hold it like this. I once made this mistake when knocking out a hammer handle, one huge blow on the tool and the handle went flying out and hit the floor, the handle removing tool was driven deep into the eye and wedged in. Unfortunately it had also caught the web of skin between my thumb and forefinger. I was now trapped, had to go next door with a hammer head dangling from my hand and get them to drive a drift in from the other direction to free me,  luckily only lasting damage done was to my pride.

To start off with you want to make largely straight cuts, go a little at a time and keep checking the eye is going on straight, rationally and also that the edge is following the centreline of the handle. You should reach the point when the end sticks out pretty quickly. Don’t worry about a perfect fit at this point, all you are really aiming to do is get the eye on straight- you will remove this section finally.

You can now knock the eye out much easier- I still tend to hold the head a make a series of progressively heavier hits until it starts moving.  You can then make adjustments and knock it back on like this.  You can see at this point the head has to move up well over a centimetre.  If you hold the head and knock it up the weight of the edge will tend to make the head rotate so the beard is moving closer to the handle. You can steer the eye on to some degree like this. The wedge I use will expand the handles thickness,  so it is OK to make the handle a bit thin, however do be careful of to much off the width as this is not really something that you can correct. It will still stay put but an air gap top or bottom doesn’t inspire confidence.

As you start blending in where the handle meets the lugs I find you need to make a cut that scoops out a slight concave, a draw knife will tend to cut in too deeply, I find the much maligned thumb push prefect for this, as it tends to cut the right shape naturally, admittedly this is a better shape for rounded lugs- pointed lugs would benefit from a more complicated V – cut out, in practice though this feel fine in the hand.

As the head progresses up the handle you can see the points of contact as the head marks the wood, I remove these each time the head goes up the eye as it is tapered in this direction. for greater clarity I rubbed some black ( moly) grease inside the eye before as I approached the final fit. This gave me an idea of the amount of contact I had achieved.

This was pretty good, as the wedge would take care of the other end. I then trimmed the end down and  cut the slot for the wedge. Here you can see the concave section I was talking about, not very easy to do with a drawknife. I have actually shaped the handle at this point, but would recommend cutting the slot before final shaping- you can then put the handle in a vice with no fear of marking the handle. I saw slowly and make sure to go right down the centreline, aiming to cut to between 2/3 and 3/4 of the length of the eye. I make a cut as this stops the handle splitting to accommodate the wedge, these splits become apparent when you take the handle off (which is the point of this method of handling- or at least having the option to)

Shaping the rest of the handle is next, or should have been.  I tend to rough out the four main flats with a drawknife.

I then refine with a knife and even spoonknife for the concave sections.

I like 12 sided, octagonal looks good but can feel a bit sharp, with 12 sides you can keep the facets crisp but still make a comfortable handle.

The optimum circumference of the handle is dictated by the size of  your hand. I used to play a lot of badminton, all quality rackets were sold in different grip sizes but players still would fine tune between this sizes with wraps. If you have gone to small on your axe handle wrap of suede is a great way to build up a grip. It is personal preference but a starting point is your fingers just touching your the ball of your thumb. There is a good case for saying that as you are going to be moving your hand up and down the handle of the axe as you carve that the circumference of the axe should be constant along its length. In practice I think this looks a little odd so do opt for a slight taper. When you are holding an axe short you tend to grip it looser than when you are holding it long and doing full blooded swings anyway.

When I am happy with the handle I will cut the wedge ( although there is no reason why you couldn’t do this after cutting the slot ) I used some dry oak, and roughed with an axe and finished with a knife. I have never measured the slope of a wedge, aiming for about 6mm where it will be cut off. I make sure the width of the wedge corresponds to the width of the eye not the handle, which despite your best efforts may now be slightly undersize.  As a test I drove the wedge in then removed it.

You can see how far the wedge went in as the oak has been grooved by the pressure from the ash. You can also see how the head has now been expanded to make nearly full contact along the eye. When forging these heads I also reverse drift to put a slight taper in that the wedged section can expand into, in this way the head can come loose but still not fly off. however to be really safe I also leave about 5mm of handle exposed, you can see how the head has formed a step here- this is a measure of how high wedging forces are. Don’t oil a wedge with BLO, this just lubricates it to the extent it pops out like an apple pip between your fingers as you try and drive it in, comical the first time, but ultimately a little frustrating. I then lined up the wedge with the slot and drew a pencil line at the point that the wedge would bottom out. I drove the wedge in again then trimmed it and bevelled the edges hopefully to stop it splitting. at this point you can drive it in a bit further if you wish as it is more rigid being shorter.

I find this style of wedging really useful, having it long allows me to take up any slackness, but it is simple to put the wedge in the vice to remove it ( and yes you can just see that I cracked this in two, a first for me, I will probably replace it when I take do finally take it out) When the pencil line reaches the handle you know it is time for a new wedge rather than trying to drive it into the bottom of the saw cut.

Earlier on I mentioned making an axe swing the same as another one you have, one way of looking at this is comparing the orientation of the handle to the eye. On my axes I like the eye pointing up slightly as this tends to make them slice as opposed to chop.

If you come to fit a head it is useful to know the orientation of the eye to the edge- this will effect the handle shape needed to give the swing that you want to achieve. I try and forge my heads so the eye is pointing straight backwards, but they do all vary a bit. If you buy a head from me I will send a tracing showing the eye orientation of your head, here is how I check- with the drift it was made on.

A lot of this is down to personal preference, I  simple formula I have found that works aesthetically  when marking out a handle is to divide the section behind the lugs into thirds and aim to put the point of inflection ( where the bends change direction) at these points. I have also found that having flare on the top of the butt 30mm further back than the one at the top feels better.

I tend to use ash, but most hardwoods are suitable, carving axe handles are pretty under stressed. As to oiling the head- some thinned Boiled linseed oil is my choice, then when that has dried a top up of neat BLO whenever the handle looks like it needs it.  Apologies for the  length and depth of this post- I was going to provide these instructions to all my customers that have ordered axe heads but it seems easier to put it up here.

I have now processed the results of my sets of experiments and even found time between grinding to conduct a new one. To recap though, I wanted to see if there was anything you could soak a loose axe handle in to swell it sufficiently to fix it.  I tested Veritas chair doctor glue, Water, PEG, Boiled linseed Oil and thinned BLO

I will start off with the Veritas Glue- This worked to a degree but running any glue that is thin enough to penetrate between the handle and axe would also work. I could find no evidence of the wood actually swelling, but then I was not using the glue as intended. You can see from the picture that the glue did not penetrate very far and rusted the socket. Gluing is not a method I can see myself using as I like to be able to disassemble my axes so leave a protruding wooden wedge.

Water-An average of 4.1g soaked into the wood. This swelled the wood massively, but in drying the wood had been crushed by the swelling and were so loose that they could be rocked in the socket. You can see the constriction here. So not a long term solution but in an emergency it will temporally secure a loose head. It was interesting to measure the differential expansion coefficient- this came out at 2, the Ash moves twice as much tangentially as radially.

PEG-  This didn’t work as hoped. The wood swelled, but as it dried it shrunk back, it was slightly looser than before. Ideally PEG should be applied to wood that is green and if using dry wood it is recommended the wood is soaked for a few days to rehydrate it, this allows the PEG to penetrate the wood better, it was very viscous and the amount absorbed by the wood was less than for the thinned Linseed oil at 1.8g.  I don’t see that soaking a loose head in water for three days and then soaking it in PEG a realistic idea. I wouldn’t like to leave my axes submerged and rusting for the best part of a week.

BLO- The immediate swelling I expected to see did not occur, however as the oil cured it should gain weight by oxidation- in theory a weight gain of 40% is possible in raw linseed oil, somewhat less in BLO due to the added driers. A gain in weight suggests an increasing in volume that I thought may give rise to the swelling that I was hoping for, so I have been recording weights for the last few weeks.  This was not found, despite using scales that weighed to one hundredth of a gram, no consistent weight gains were recorded. Changes due to humidity were recorded in all samples and it was found that these changes were less in the BLO samples- the oil reducing the flow of water vapour from the air to the wood.

Thinned BLO- Broadly the results were the same as BLO, I found it interesting though that significantly more oil was absorbed even taking into account that some of the initial recorded increase in weight was just the white spirit used to thin the oil 2.1g of thinned oil went in equating to 1.7g of BLO as opposed to 1.2g of BLO if it is not thinned ( It should be pointed out that it is likely that the BLO I was using had some thinners in to start off with, but I was amazed at how much thinner the oil was with 20% whites spirit, I would guess that there was 5% or less thinners in the ‘neat’ BLO).  This seemed to evaporate off quite quickly. It does seem to be worth thinning BLO not only will it penetrate deeper but more goes in as well. However in splitting the samples open at the end of the tests it was apparent that not all the white spirit had evaporated off- it could be smelt easily. It can be seen how the Thinned BLO penetrated the ring porous sections of the Ash.

So not the set of results I was hoping for, but I remembered reading that wood treated with BLO resisted compression better as the fibres were filled and less likely to collapse. Perhaps this means that if the wood gets wet and swells it will resist being crushed and then shrinking on drying as seen in the water samples?  Also the Axe will transmit force to the eye- these concussive blows may dent the wood allowing the head to come loose.  However in the axes I have used it tends to be periods of storage that cause heads to loosen rather than use, but then I forge my heads with a larger contact area than many reducing the pressures on the wood.

I didn’t have time to run a series of tests to see how the samples fared against compression over time, but I did set up a test imitating the concussive forces that an axe or hammer might impart to a handle.

Not the clearest picture I am afraid but we used a bar of steel that was held a set distance above the samples ( by the box section that also guided it down onto the target peg)  We dropped this three times onto the samples radially and tangentially then measured the degree of crushing.   We tested BLO samples against some control samples I had kept for such eventualities.

In these samples we found that they were crushed much more radially- the growth rings more readily crushed into one another, not really a surprise if you have every tried pounding ash splints.  Tangentially  however the depth of crushing was roughly half.

In BLO loaded samples this effect was not seen, the tangential crushing was reduced slightly, meaning it was slightly stronger. However the real surprise was the radial strength this was now equivalent to the tangential strength.It seems that the BLO filled the pores of the ring porous ash and reinforced it at this weak spot. No statistical difference was seen between the thinned and uncut BLO despite the thinned seeming to have penetrated more readily.

So what did I learn? Well..

Soaking in water is as bad as thought.

PEG and Chair Doctor glue don’t work.

Boiled linseed Oil didn’t work as expected- there is no point soaking a loose head and expecting it to swell and tighten up; however the wood does seem to be slightly more resistant to moisture after oiling so will move less after treating. Ash will also see an increase in compressive strength after treatment that may help it to stay tight in a head. Thinning BLO allows more to go in, reducing the soaking time needed. Probably better to use it neat for a final coat though. BLO is not the cure all that I hoped it to be, but out of all the samples tested is was the best.

As many of you will know Don now does the IT support on my site, and is very knowledgable on the subject of Linseed oil, we have been communicating over the last month or so, it was he who confirmed the idea the the oil should gain weight as it cures. I initially found great comfort in this, I had made an axe stand from some oak years ago and as some of the sapwood had rotted, so I poured litres of BLO into it, the idea being that it would soak into the punky wood, cure and harden it. It didn’t work like that, 15 years later bits still flake off and leave oily smears as it still hasn’t fully cured. I had increasingly found that this stump wasn’t so easy to lug about and had been putting it down to encroaching old age, so the idea that the stump was still getting heavier as the oil cured was encouraging. With these results though I am afraid I have had to revert to the previous theory.