Minor update 1-Jan-15

Automatic Electric Bow

Many thanks to Peter Broad who provided aluminium bar for the barrel, plate for the gearbox and an old pair of recurve limbs. He also rode to the rescue with some good old solvent based varnish when I realised that the water based stuff is CRAP and was ruining everything.


The specification I devised called for the following.


1) A full auto-firing rate of 1 bolt in 1.25 seconds; 6” of draw length and a minimum of 50 lbs draw weight.   (Firing rate of 1.5 secs achieved, see below). (Draw length  and draw weight achieved).

2) The magazine was to house 10 bolts 9” long.   (Achieved)

3) The bow was to be capable of discharging the full magazine with imperceptible slowing down and have an effective range of 20 yards.   (With the batteries fully charged, I was able to discharge 100 rounds. However it was struggling with the last 10, and the last 5 in particular).


Battery update

I have obtained a set of Sanyo NiMh batteries which are 1900 mAh and don't loose their charge like normal NiMh batteries. These provide the current needed to to power the bow at full speed for at least 300 shots (I stopped the test at 300 with no perceptible slowdown of operation). At last NiMh has come of age and I can dump the old tired NiCd batteries.


The draw length is not limited by the throw of the crank as in the prototype design but is restricted by the relationship  between the diameter of the timing belt pulleys and the time to complete a full discharge cycle commensurate with the near stall torque that can be pulled out of the motor / gearbox assembly. The need to balance these relationships must be offset against the voltage / number of batteries / weight.


The above was achieved using 12 AA rechargeable NiCd batteries giving 14v. However the cycle rate is 1.5 seconds. This increase of 0.25 seconds on the desired 1.25 seconds is mostly due to the need to slow down the speed of the carriage as it goes to pick up the string. When traveling to pick up the string there is no load to slow it down. It operates the reversing switch too roughly at 100% available voltage. Tests showed that 58% voltage seemed ideal and that 66% was a little over the limit. With construction as it was at that point, 50% voltage was a reluctant choice. This means the no load travel is slower than drawing the bow under load.


The bow was made using broken limbs from the clubs practice bow (scavenged by Peter). I cut off most of the recurve and riser fillet. It draws about 50 lbs at 6”. I think increasing the draw weight beyond 50 lbs will over stress the limbs.


The bolts are fed upwards against gravity by 4 tension springs looped around 4 pulleys in the magazine. The bolts are held underneath the barrel assembly, the opposite way round to a conventional bow.

To load the magazine, a catch is operated which allows the top half of the bow to rotate upwards away from the bottom half containing the magazine. Bolts are dropped into the open top of the magazine after pulling the sprung loaded feeder down and locking it.

Accuracy is good with all bolts easily placed on a 12” boss at 30 yards. At 40 yards, bolts group well on an 18” boss.


Tests showed that bolts without fletchings flew OK over 30 yd; however as I solved the fletching clearance problem with the previous design and simplified it in this design: I retained fletched bolts.


If doing the job again I would increase the voltage to 18v using another 3 of the Sanyo NiMh batteries mentioned above. This would give an increase in the speed and torque enabling the mechanical advantage to be adjusted by reducing the diameter of the timing belt sprockets, thus allowing the rate of fire to be brought back down to around 1 bolt per second while providing more torque. The additional torque would enable the draw weight to be increased.  The motor is rated at 12 – 15v so the overload to 18v would be OK as there is plenty of time for it to cool down between reloads assuming the insulation holds up. I would also replace the 5 mm pitch belts with 2 capstans as per the ‘Self Cocking Bow’.

I would make the magazine able to accept Bolts from a caddy (like a speed loader).


I would also try to get heavier duty switches whilst retaining the small size of the current 5 amp switches and attempt making the top half rotate about one side of the longitudinal axis. I rejected this idea as the trigger assembly would become more complicated.


I have thought up a design that will allow for the doubling of the rate of fire without prejudicing the battery life or adding weight. I doubt I will get around to making it now.

<< Back to Cross Bow Page

Automatic Bow Photos

Click for Photos

Click for Video link

Cross Bow Videos