Disclosure: My company is a customer of AustriAlpin products.  These comments are mine, not CTOMS’, nor AustriAlpin’s.  The interest in assessing these two buckles is in determining the best product to purchase as a component of harnesses that I design.  It has evolved beyond that into educating the public about my findings so that they may make safe and educated purchase decisions as well.

A while back we pull tested a couple Raptor Buckle samples to destruction to measure the breaking strength and posted the surprising results here.  The intent was only to satisfy our own curiosity and validate our buckle choice in our products.  While it was purely an editorial on our sample test data, it stirred the pot, and I think in a good way.  As new actions have occurred since then, an update on what is new in this realm is warranted.

I noticed that on Raptor Buckle’s website they updated their test images and data.  One of the most serious critiques of the previous post was that Raptor’s test method didn’t appear to pull the buckles in ladder lock configuration, thus not pull testing it in the configuration of function.  The new images now show the buckle in ladder lock mode…well, kinda. If you look closely, the webbing is threaded in ladder lock through the buckle, but instead of the tail being free, they have stitched it down, creating a closed webbing loop around the center bar of the buckle.  Why this is both interesting and concerning is that, once again, it alters true assessment of buckle performance in real world application.  No product is going to be manufactured with that design because it negates the functionality of adjustability in the ladder lock.  The concerning part is that it alters test results.  It pulls on the bar evenly on both sides, instead of unevenly, in a rotational force, as it would in reality.  As such it most likely will alter the data, hypothetically in favor of stronger test results than had the webbing tail been left loose.  Just like the original testing where the ladder lock wasn’t engaged, whether these oversights are due to naivety or an intent to mislead, I’m not going to speculate and will give them the benefit of the doubt.

Three other interesting facts remain.  Once again, their tests seem to only have been conducted in loop configuration.  The direct pull breaking strengths (stamped onto the buckle) being derived from those results instead of actually doing direct pull tests.  The problem with that, just like sewing the running tail in the ladder lock into a closed loop, is that its not a truly accurate conversion – can’t derive direct pull results from loop configuration pull results.  You have to do the actual test, which based on the data available to me, Raptor has yet to conduct or at least publish.  The second is probably also worth mentioning that 1 3/4″ webbing appears to have been used on the 1.5 inch and 2 inch buckle tests, which may or may not alter accurate results.  And finally, the last concern is that their results are so different than ours and other much more thorough and reputable third party testing.

The AustriAlpin camp took the testing a step further.  They sent 5 buckles of each size; Cobra Buckles and Raptor Buckles sourced by 3rd parties from online suppliers of the buckles, to TÜV SÜD for independent testing.  I’ll let you visit their website and determine their credentials for yourself.  TÜV SÜD chose 3 of the 5 samples to test.  I have to say, kudos to AustriAlpin for the initiative.  Obviously they have a bias and a business motive, but to incur the expense for an objective third party to do head-to-head testing shows that the company certainly stands by their product and has the respect for their customers and end users to educate them.

AustriAlpin shared the results of that testing with me a couple weeks ago with permission to post them.  I saturated myself with the test results then incubated them for a couple weeks so as not to make any knee jerk interpretations.  I think I have enough illumination now to post the facts and my editorial comments on them in the hopes of instigating some critical thinking and constructive discussion.

The first thing that struck me about the TÜV SÜD testing was that in the smaller buckles, the webbing broke before the buckles. As the buckles got larger, the buckles would break before the webbing.  I had a hard time accepting this as valid testing if the test is trying to measure break strength of the buckles.  Then during my incubation, I realized that it’s still a good data set because it still measures a system failure point involving the buckles head-to-head.  Whether it’s the webbing or buckle that fails, the system fails.  The test wasn’t necessarily just testing buckle strength, but rather system failure point in general in a direct comparison.  It needs to be fully understood that a webbing failure below 9kN does not constitute a buckle failure because the buckle didn’t break.  The webbing was just not a high enough quality, though it was at least the correct size.  A buckle company can’t anticipate and rate their product to the almost infinite amount of webbing variations that exist.  Having said that, webbing doesn’t break, it has to be cut.  And in this head to head comparison, it shows the force at which a Raptor and a Cobra ‘cuts’ the webbing.  If at some point in the future, head to head testing is conducted to measure break strengths of the buckle, a stronger webbing would have to be used.  In the larger buckle sizes we can definitely see break strengths vs. stamped ratings.

One last thing to mention before getting to the test results.  There is always the potential for one camp to claim that the other tampered with the test buckles or test method to alter results in their favor.  My personal opinion is this is a desperate defense in absence of tangible evidence of the same.  This was brought into question after posting my test results.  So in commenting on these tests, all I can do is base my comments and evaluation on facts available to me and leave speculation out.  I will give everyone the benefit of the doubt and assume professionalism and common ethics.  This pertains to proof in chain of custody of sample buckles from procurement to testing etc., and the test methodology of TÜV SÜD and their reputation that would ride on that test methodology and ethic.  Test method is derived from images and descriptions, not speculation.  I personally would have liked to compare test data conducted and or funded by the Raptor camp as well.  But the problem is, the only test data available that I am aware of is results from tests where the webbing tail through the ladder lock has been sewn into a loop and the system is pulled in loop configuration.  Therefore, I have yet to see credible data from the Raptor camp of where they get their direct pull data and even credible data to support their loop config ratings.  I welcome such data for review.

Summary of the TÜV SÜD results:

1 inch buckles

Raptor slipped between 5.6 and 6.2kN then webbing broke between 6.3 and 7.1kN.  Buckle bent.
Cobra did not slip and webbing broke between 8 and 8.2kN.  No visible buckle damage.

1.5 inch buckles

Raptor slipped between 5 and 6.6kN then webbing broke between 8.1 and 8.5kN.  Buckle bent.
Cobra slipped between 6.3 and 7.95kN then webbing broke between 7.7 and 10.1kN.  No visible buckle damage.

1.75 inch buckles

Raptor varied from not slipping in one test and slipping at 4.1 and 5.5kN then the buckles broke between 7.85 and 8.3kN.  All 3 below stamped rating of 9kN.
Cobra did not slip.  The buckles broke between 11 and 11.3kN.  All 3 above stamped rating of 9kN.

2 inch buckles

Raptor slipped between 6 and 7.4kN then the buckles broke between 6.6 and 8.1kN.  All 3 under stamped rating of 9kN.
Cobra did not slip.  The buckles broke between 10.9 and 12.5kN.  All 3 above stamped rating 9kN.

Note that the Raptor buckles slipped quite often before breaking. In the configuration of their test method conducted by J. Henry Holland Corp., this would have been impossible to determine because they had sewn the webbing into a loop around the center buckle bar as described at the beginning of this article.

The bottom line is that based on the TÜV SÜD testing (consistent with my own personal testing), the Raptor Buckles bend and brake consistently below the direct pull stamped rating, and the manufacturer has only published their own loop pull testing results, again in irregular configuration.  The Cobra Buckle doesn’t deform and I have yet to see data where is has broken below the stamped rating.  And I’ll say it again – I have no interest, financial or otherwise in these businesses.  This assessment is objective and intended to determine the best product for my purposes and to educate others based on available data.

Some will argue that the Raptor Buckle is still ‘strong enough’, but I think this has become a question of integrity and accountability.  If it was simply a head-to-head, one product is almost always going to be better than the other and, in this case, that appears to be so.  In a free and open market, competition is a good thing and that difference in quality might be offset by price, location of manufacture, and other preferred criteria when making purchase decisions.  The problem though is that pesky stamped rating that is the same between the two buckles/companies, except that the Raptor Buckle seems to have trouble meeting it, except in questionable test configurations.  My question is, if you were a company, why wouldn’t you do exhaustive testing on your product in proper configuration?

The full TÜV SÜD Test Report can be viewed here. 

I welcome everyone’s comments.