Ti vs. Al biners heating: limits on IR and Thermocouple Testing

Climbers and Canyon folk often worry about how hot a rap device gets, and whether the device will burn through the rope when they suddenly stop mid-rappel. Hence they worry about “heat dissipation.” Many think aluminum devices should be much better at heat dissipation, because aluminum has such a high thermal conductivity.

Actually, the high thermal conductivity is the reason aluminum devices get so hot so quickly. Heat is generated in the rope in two ways; 1) by the surface friction at the rope-metal contact; 2) and by “internal friction” by bending and deforming the rope as it goes around the friction device and carabiner. Each is roughly half the heat production in the rope. The heat generated by surface friction gets divided between the rope and the metal device; the faster the rope goes past the device, and the lower the thermal conductivity of the metal device, the greater the fraction of the heat that goes into the rope.  This “direct heating” of the rope rarely increases the rope temperature by more than 10C, because the total mass of rope is so great-- e.g. for a two-strand 30m rap, the rope heating might be spread over 3000 grams. For the metal device, conductive-connective cooling by the air is more important than radiative cooling, and all devices face that limit, though it depends somewhat on the metal used to create them. If the rate of heat absorption in the metal device (heat generated by rope-device interaction) is too slow, it is eventually balanced by the heat losses due to conduction-convection.

The above figure (at left) is mainly illustrative, calculated from Eqn 6 in Berry and Barber (1984, Journal of Tribology vol 106, p 408), for a heat source that “ploughs” into a softer, moving substance. As the speed of the rope increases, the greater the fraction of the heat that gets distributed in the rope; and for Titanium Grade 2 and 304 Stainless, a much greater fraction goes into the rope, compared to 7075 aluminum (which is typically used in carabiners and climbing devices).

In the middle (above) is an IR video of a rope going over a munter on a carabiner made of 7075 aluminum. At right is a video of the same rope going over a screwlink made of Grade 2 titanium; both Al and Ti  devices were ~9mm in diameter. The aluminum heats up much more substantially than the titanium device. The ambient was about 40F, as I had to do the test in pre-dawn darkness, backed up by black plastic. The lower end of the scale for titanium gives a false temperature, as explained below.

There are two artifacts; first is that the IR camera is assuming everything has an emissivity of ~0.95, which is true for the fabric, so the best temperature indications likely come from the center of the Munter. The aluminum carabiner is heavily anodized, and likely has an emissivity of ~0.85. However, the Ti is a wild card, as its emissivity may be from 0.4-0.8. (The Ti is certainly oxidized, but the extent is unknown.) The Ti, which was at ~40F initially, is emitting as if it were at a lower temperature than the black plastic that forms the background. Second, the Ti carabiner is actually reflecting the IR emitted by the hot spot, so the seemingly high readings off the edges are not real.  Nonetheless, I can vouch that the Ti felt cool at the end of the test, whereas the Al felt quite warm.

Thus the temperature at the center of the Munter is the hot spot, but is quite a bit colder in the Ti carabiner. Part of the rope heat is from surface friction at the very contact with the biners, but ~half of the rope heat is from internal friction—the bending and deformation as the rope goes over the biners.

N.B.:

The rope was 6mm polyester-covered Dyneema, and force was applied at ~70lbs, by attaching one end of the rope to a linescale 3 load cell, which was then attached to my harness. I then walked backwards hard, using the other line in my “brake hand” to keep the force reading ~70 lbs on the load cell, over ~45’. Obviously a real rap could generate a lot more force and more heat, but I'm looking for relative changes only.

Postscript: Thermocouple Testing

I took the same Ti and Al devices, and affixed calibrated meat thermometer thermocouples (TC) to them, as shown in the picture below (click on the thumbnail to see the larger image). The tips of the thin TCs are placed as shown; Al foil and thermal grease (Arctic MX-6) were used to attempt a good thermal coupling. The TC itself pertutbs the temperature. The mass of the Ti device is 2x that of the Al device, but most of that mass is well above the TC contact. The attempt here was to place the TCs at the same distance from the munter position on the devices.