A take-down bow is a bow that comes in pieces, usually two or three. I have been experimenting with building such a thing and it turned out to be much more difficult than I thought.
The project began when I cut a hazel bow blank in half with what I hoped would be a self-locking finger joint but the thickness of the saw kerf made the joint too loose so I packed it with leather (hard and grippy) but the bending forces are extraordinary and the joint just levered open no matter what I tried – I even tried through bolting it with a single bolt but to no avail. What I probably should have done is to use a scarf joint and two bolts, one at each end of the scarf – may have worked but there would be very little timber surrounding the bolts, so probably not……
Anyhow, now I didn’t have any wood left in the handle to try the scarf joint so I had to try something else. So I made a handle (riser) the same size as the one I cut away and notched the limbs to fit it. This was a big mistake on many levels.
First off the notches act as pivots which pull the limb outward instead of anchoring it to a bearing surface (this interface between the limb and the riser is called the ‘bed’). Also, as was pointed out by several of the guys on Primitive Archer forum, the bed needs to be, ideally, 4 inches long – still don’t know why but I think you need a surface this size to provide enough room for the bolt and for the locating pin/lug that stops the limb slipping sideways – certainly less than 3.5 inches is not enough space. This, coincidentally, puts the through bolt several inches away from the end of the riser, which is important as we shall see.
Secondly, the timber used for the riser is critical – I used Sapele as I thought the interlocking grain would be useful (it is) but the wood was too soft and prone to splitting. In actual fact the bolt was pulled out through the end of the riser so huge was the sideways force imparted by the notch . Next I lashed up a pine riser so as not to waste more expensive wood prototyping and found that soft wood compresses under the huge force pulling upwards on the bolt when the limb starts to bend – a 2.5 foot long hard-wood lever pulls really hard. The washer and bolt pulled more than a centimeter through the pine on the first test but at least I was able to test the bed-size and effect of hooking the limb until I could see why a smooth flat bed above 3.5 inches length is important.
Having established all this I carved a very pretty handle out of cherry and glued in some brass receivers for the bolts, as I really strongly dislike the look of through-bolts, at the very least I could have counter-sunk and plugged the receivers in from the back but I have an old commercial riser with the receivers just sunk and glued in place into the face of the beds, so I know this works well. More lessons – the receivers just pulled straight out! This would not have happened if I had set them in from the back… I had not toothed the receivers and had used resin (I love Cascamite!) not epoxy, so I heavily toothed each receiver, cutting scores of slots into their surfaces, then mixed a liberal dollop of Araldite and glued them in. Brilliant! Nope, the cherry pulled apart – I had the grain running across the handle and beds and the stress on the receiver literally pulled the grain apart – there was possibly a tiny shake/weakness at that point that I hadn’t seen but I had closely inspected the wood. More likely is that cherry is easily split and it was just not suitable.
OK, so I pulled out a lump of the hardest, most homogonous timber I have – hawthorn – and made the final version: chunky and strong, not delicate and pretty, grain running the right way, very dense hardwood, homogonous grain known to resist splitting, with toothed receivers epoxied in place into beds more than 3.5 inches long – total success, got there in the end, only took 3 months!! Hope that if you ever want to build your own 3-piece take down my experiences will be useful!