Front View	Front Close-up	Rear Close-up

Left Side Close-up	Right Side Close-up

Technical Details

I made this roller rack in 2022.

The Storrick Roller Rack is 771 mm. tall, 52 mm. wide, 54 mm. thick, and weighs 3476 g.

The frame is made from 6061-T6 aluminum. Twenty-one 8x22.5x14.5 mm. stainless steel deep U-groove track roller bearings ride on 8 mm. stainless steel steel bolts. The bolts pass from rear to front of the frame, through the bearing, and into a milled 6061-T6 support bracket held to the frame with two 1/4-20 UNC stainless steel bolts. All nuts are stainless steel lock nuts.

I stamped my logo on the roller rack.

Comments

Everyone knows that a rope running through a device generates friction, and that the rubbing friction is reasonably approximated by the snubbing formula T1/T2=e‑μθ with everything having the obvious meanings. Less well appreciated is the internal friction generated by bending the rope. I am not aware of anyone actually measuring this friction. Bob Thrun and I discussed this many times, and I designed a roller rack based on those discussions, but I never finished the construction. We debated the “how many rollers do we need?” question. Our guess was 20 to 24, depending on the rope, but the Aldo Moschino descender provided some empirial evidence that intuition may not be complete and accurate.

I had bought some ball bearing rollers for making chest boxes, and decided to design and build a roller rack with these. This would be for test and measurement only, not for practical rappelling (due to lack of expendable volunteers). This is the result. This rack used 21 rollers because, well, that’s how many I had.

Ideally, the rollers should be placed inline and spaced to give a 180° bend at each roller, but to allow for simplified rigging, I offset the rollers and widened the spacing. This will affect the results. Neglecting the unneglectable rope compression, this rack has about 172° contact for 11 mm. rope and 164° for 9 mm., and effectively linear in between (the second-order term is negligible).

Testing this has been enlightening. So, what is the result? First, the roller friction is not a large contributor, being in the 1 to 2 lbf. range for using all 21 rollers. Second, “Obviously, it depends on the rope” is an accurate intuition. So are “Stiffer and larger ropes give more friction.” Qualitatively, the effect of different rope characteristics aligned with my expectations. Quantitatively, there were some real surprises; for example, I was somewhat surprised how much friction 11 mm. PMI Pit rope gives. Rappelling on 9 mm. ice climbing rope with only a few rollers would not be habit forming.

My testing is not complete and it will be quite some time before I finish. My test setup, like all test setups, is not beyond criticism. I will not be making my results public. There are too many haters in the world and I see no need to start another online flame war just because someone else would have done this differently. Anyone who is interested is free to build a device matching their own ideas of perfection and test it at their own risk. Just don't do it in an unsafe manner - keep it in the lab.