Spider gear mounted the winch from a US military truck on his Unimog. A little bit of digging can turn one up for a couple of hundred bucks. A winch is a substantial challenge to engineer from scratch.
I had been noodling on the engineering of an F64 winch, per the OP, and this is now the invitation to flail away on the subject. I re-read the previously posted spec sheets for the F64, and found there was more to work with than I originally thought. Mostly as a substantiation of the above, from Speedwoble, but also hopefully illuminating for the OP ( who seems not to want to calculate), here is how I see the numbers. For my convenience, I converted all into inches/ foot-lbs, etc.
Per Werner specs: drum diameter is 5.0" and rope is 0.56" diameter. To be especially nuanced/ persnickety, I used the diameter at the rope centerline of any given wrap to be the winding diameter. This assumes equal rope compression and tension below and above the centerline of the rope cross section.
radius to centerline of: first wrap 2.78", second wrap 3.34", third wrap 3.90"
circumference of: first wrap 17.47", second wrap 20.99", third wrap 24.50"
drum width is 318mm divided by rope diameter of 14.3 mm = 22 turns across the drum, per wrap
total rope wound on: first wrap 32 feet, first two wraps 70 feet, three wraps 115 feet (close to the stated 32m = 105', so far so good)
line pull 16000# first wrap x 3.90" divided by 2.78" = line pull on third wrap = 11,400#
11400 divided by 16000 = .712 from Werner 7000 daN x .712 = 4984 daN (again, close to published values)
Werner first wrap line speed, 540 rpm PTO is 10m / min = 394 inches/ minute
394 in/ min divided by 17.47 in/ rev of drum = 22 revolutions/ minute required
540 rev/ min divided by 22 rev/ min = 24 which is the necessary reduction ratio
torque on the drum or driven/ worm gear 16000# x (2.78" / 12 in/ foot) =
3706 foot-lbs
This is the point where anyone contemplating this build should step back and go "W
hoa,
Dude, that is a LOT of TORQUE". The forces on winch components are potentially HUGE. Reducing the drum diameter will reduce the torque load for a given line pull, but the big trade-off is that more wraps will be needed for a given rope length, and the diameter is increasing at a faster rate. This is results in a relatively greater loss of line pull from first to last on small(er) diameter drums. This is the bane of the small electrics, and a strong argument for the use of small diameter (for a given WLL) synthetic rope, as there is less diameter increase on the drum for a given length of rope (vs steel).
To continue, 3706 ft-lbs output divided by 24 (reduction ratio) = 155 ft-lbs torque supplied at input
Note: I have not factored for friction losses, theoretical is good enough for me, here.
I don't know the engine speed at which the rated 540 rpm output is obtained, someone can chime in.
Where we are coming to is some quick research at McMaster-Carr. The 24:1 listed worm reduction gearboxes are in the vicinity of 70 ft-lbs max, and cost circa $700. The only worm and worm gear pair shown at 24:1 reduction cost $96 and $156 respectively, but no telling if they can handle the loads and speeds. The 4 inch pitch diameter of the worm gear means a force of 22191 lbs on the tooth face, which will also be the force on the thrust bearing of the worm/ input shaft. Kind of substantial. I suspect lube is also important.
If one procures a ready made gearbox, it must attach to the winch frame sufficiently to resist that 3706 ft-lbs, or it under load it will simply wind itself around the winch frame until it rips the PTO drive shaft apart.
Checking with Fastenal's bolt engineering guidelines, I found an easy example that was relatively applicable. If the worm gear is bolted directly to the drum, and assuming a 3" diameter 4-bolt circle, there is 35625# shear load on the bolts. This would decapitate a single 1/2" GR 8 bolt, even if applied on the shank and not the threaded portion; the four 1/2" bolts per example would give about a 3X safety factor.
One can see why the electrics and the hydraulics use multi-stage planetary gears. They can be big, and speed decrease/ torque increase occurs at the very last, and acts directly on the drum, without a lot of other components to get loaded up. And, one can see why winches are expensive. For cost alone, Speedwoble has it pegged; look for used, and do minimal mods to fit.
EDIT: I have built a bunch of things in my life, and I can say with certainty, if someone is mass-producing an item, you cannot duplicate it as a one-off for even close to equal cost. Disregarding one's own time contribution is side-stepping the true cost issue. Not to say doing cool stuff can't have non-monetary rewards, but I prefer to not practice self-delusion.
Disclaimer: I am not an engineer, do not certify any of the forgoing, and publish strictly for the purpose of acquainting interested parties with the level of intensity this endeavor requires, for safety of all. And I definitely do not volunteer my carcass to stand by and observe the first home-brew winch created.
FWIW, I have attached my crib sheet:
Lee
Per an old ME I knew long ago:
Figures don't lie, but liars can figure