You’re trying to teach strangers on the internet how to safely rappel yet you don’t know how to do so yourself.

Others have pointed out the many things wrong with the way you have set this up yet still it’s here for others to ‘learn’ from.

A better configuration again would have been isolated strands from your tree to each bolt, but even then you would have to take care not to create cross loaded caribiners.

Since you know the issues with the “American Tirangle” also called the Death Triangle or Yosemite Death Triangle, a better approach to this anchor would be if you created a load sharing anchor out of your webbing. By creating two isolated anchor strands you would be able to share the load between the two points. As you set it up, you have created a load amplifying anchor. Luckily in this configuration (equilateral tirangle) each point of the triangle only experiences 1 times the load. but if you lengthen the top side of the triangle (shorten the loop of the figure 8) the load is actually amplified at the weekest point, the knot, this amplification becomes extreme as the triangle flattens out and failure is more than possible as the rappeller transitions over the edge (the highest force experienced in the rappel.)

In your set up, a force strong enough to fail one of the anchor points would generate an even greater shock load on the other bolt and likely fail it too. While you are correct that a failure in your rope would be the end of this excersize a failure in one of the achor bolts need not be catastrophic, unless you set it up so that one failure will cause the other to fail as you have here.

On another note, it also appears you have split a figure 8. Most knots weaken a rope by about 1/3. A split figure 8 is more like 1/2 to 2/3 of the strength of the rope compromised. Add to that an easy error to create, amplification of the load from a flattened triangel, and it is a recipe for failure. This would be an example where multiple cascading failures causing a catastophic failure.
A boline, with a backup would be a much better choice in this situation as it would not be weakend by pulling it in two directions as the figure 8 is. Several variations of the boline will create good multi-point anchor systems as well.

When we design rope systems we should always do our best to isolate possible points of failure so that they will not cause catastrophic falure. Ingraining this philosophy from the begining of training is the best way to encourage safe technique. As in all disciplines coming back to eliminate bad habits is always harder than starting with good habits.

This setup probably worked just fine when you set it up for the article and it will probably work fine 99 times out of 100. It is that 100th time where the unexpected happens. Like you tie the figure 8 too short and create more amplification of the load and one of the bolts isnt as rock solid as you thought… and POP! we have a problem. If we start by eliminating the possiblities of those failures we set ourselves up for success instead of failure.

On the webbing portion that backs up the system, a tape knot was used http://www.itstactical.com/2009/05/11/knot-of-the-week-tape-knot/ and it does have overhand safeties on both sides.

Thanks for the comment!

Thank you for the link and information. We’re aware of the american triangle, and have spoke about it in this article. We’re looking to do an update of these rappelling articles to offer other ways of setting up a rappel.