It has been my understanding that the primary factor which causes self-drive anchors -- Spits, M8s, Petzl Cheville Autoforeuse and probably several other names -- to wear out is a chemical reaction between the aluminum alloy hanger and the hardened steel bolt/sleeve. Petzl's own page for this product points to a dead link about corrosion in climbing hardware that would probably be helpful here if it still existed. I have two questions:
(1) Does anyone have a reference which discusses this corrosive process?
(2) Can it be entirely prevented by using the Raumer stainless steel M8 bolts and hangers, which work with the Cheville Autoforeuse sleeves, or is there still a reaction between hardened steel and stainless steel? Please note: these hangers are not available in this country, though they can be purchased from international speleo-vendors.
This is the only reference for this issue I can presently find, from Alpine Caving Techniques, which simply acknowledges the problem and makes a recommendation for avoiding the aluminum hangers (which, of course, are in caves everywhere all over the globe and are sold directly by Petzl for use with the sleeves).
Alpine Caving Techniques wrote:Most hanger plates are manufactured from a light aluminum alloy (zicral), for obvious considerations of weight. These hangers should never be left for long in a moist environment. They will corrode as a result of an electrochemical reaction between two metals (with water acting as the electrolyte): the steel in the anchor, and the aluminum in the hanger. The aluminum oxidizes into alumina, a white powder that forms a translucent gel in the presence of water (this cannot occur under water due to the lack of oxygen). After a long period of disuse the hanger can corrode to the point of passing right over the head of the screw (and extracting) the moment It is loaded. So in particularly humid caves. always use stainless steel hangers.
What is this process? Does the use of stainless M8 hangers (again, not available in this country, at least not the Raumer version) prevent the problem, or simply mitigate it?