Life Safety Incident:
Petzl Ascenders Design Flaw Releases It From Rope
Normal Action of Climbing the Rope Releases Safety Latch
The following report and correspondence documents an actual life safety incident involving the basic design flaw of the Petzl ascender, a climbing gear failure wherein a Petzl model Ascension B17, unexpectedly disconnected itself from the climbing rope while the climber was climbing on a cliff side. The author personally witnessed this incident on Saturday 27 March 2004. Petzl was informed, and chose neither to follow through with a design review nor to notify the public how to avoid possible danger due to its flawed product safety design. Petzl instructions for use of the ascender contain no information about this life threatening safety problem. As of this date, the Ascension B17 ascender has not been recalled. The previous models of Petzl Ascension ascenders did not have this dangerous design flaw. Tough decisions yet remain for Petzl, for their conscience, and for their company's bottom line. When the manufacturer does not make it clear to users what the safe limits of use are for the products they manufacture and sell, it falls to the users to educate and inform ourselves, determine the life safety limits of use, and to gain full knowledge of possible failure modes.  Buyer beware, there are serious product safety problems out there. Do you have one in your pack?
Rope Rescue of Injured Climber
Many climbers, cavers, and rope rescue personnel are injured every year due to gear failure and product safety design flaws. Educate yourself and use good safety procedures. Never use climbing equipment that has been badly designed or is prone to failure.
 
DESCRIPTION OF INCIDENT | INCIDENT INFORMATION REPORT | PHOTO | CORRESPONDENCE  | PDF OF PETZL RESPONSE
Danger: Petzl Ascension B17 Ascender for Rope Climbing SRT
CLICK: ZOOM
CLIMBER, ASCENDER,  ROPE AT BASE OF PITCH
Photo taken just minutes before the incident.


Petzl ascenders manufactured after approx 1999.
Description of Petzl model B17 Ascension Ascender Failure Mode
The design of the nylon (plastic) cam "safety catch" in Petzl ascenders manufactured since 1999 has a curved crescent surface protruding for the thumb to manipulate it. The prior design did not have this. This design "feature" is described in Petzl promotional literature as a new improvement, which is "ergonomic". It is the protruding edge of this curved surface that unfortunately forms an arc wedge with the elongated smoothness and continuous angular position needed to induce lateral displacement of the safety catch when it is moved upward against a rough rock surface. That movement unlocks or disengages the safety while at the same time pulling the cam down and away from the rope capture enclosure. Once the rope is no longer captured, the ascender falls loosely off and away from the rope thereby disconnecting.
After this dangerous life safety incident, and notification of the ascender design flaw was reported to Petzl, they decided to do nothing.
"We think Petzl ascenders are very well designed in this regard, have no plans for a recall or design review in light of the reported incident."
---Hank Moon, Technical Information Manager - Petzl America

Climbing Rope Using Ascenders - Gear Failure - Petzl Ascenders



Petzl Ascension Ascender Failure Model B17 Unintentional Rope Disconnection

DESCRIPTION OF LIFE SAFETY INCIDENT
Climber starts ascending the rope at ground level using typical frog-style motion. Witness is clipped in to safety line at top of pitch while viewing climber. Climber is using right hand in handle of upper ascender to push the ascender up. Climber continues slowly ascending rope with two boots in the foot loop until reaching mid-pitch, whereupon climber removes left boot from foot loop. Climber pauses while sitting in harness, holding upper ascender handle with right hand. Climber continues slow ascent of the rope against the face of the rock while balancing away from the rock using the free left hand, left leg, left knee, and left boot. Climber pauses upward ascent to rest while sitting in harness and holding the handle of the upper ascender with the right hand. Climber continues slow ascent of the rope against the face of the rock, approaching the point near the top of the pitch where the angle is approximately 30 degrees from vertical. Climber pauses and sits back in harness, while putting both hands against the rock, and not holding or touching upper ascender. Upper ascender is about climber's head level on rope. The climber's right leg exerts some weight on the footloop sling and thus pulls downwards and about 15degrees angularly to the right from vertical on the upper ascender. At the position of the upper ascender, the rope is taut against rock surface. As climber rests, ascender is laying against the rock on its natural point of equilibrium, with the side which has the cam and safety catch against the rock. Climber starts to climb again, reaching to the upper ascender. Climber grasps bottom lower part of the handle of the ascender and starts to push upwards slightly on it while sliding fingers through lower part of handle opening and getting a better grip in the handle. Ascender instantaneously and unintentionally disconnects from the rope. Climber is immediately surprised to be holding the loose ascender in hand, within a few centimeters of climber's face. Climber utters exlamatory remark. Witness at top of pitch is viewing the incident at a distance of 2meters and utters exclamatory remark. Climber is now supported on rope solely by the singular lower ascender. Witness urges climber to re-connect the upper ascender to the rope. Climber re-connects upper ascender to rope. Climber continues climbing rope and reaches the top of the pitch without physical injury. Climber clips in safety lanyard to safety line at top of pitch, and sits down on top of the cliff. Witness notes climber's elevated breathing rate and climber's symptoms of anxiety. Discussion between climber and witness. After resting and calming of nerves, climber rappels down the pitch. Witness rappels down the pitch. On the ground, in a safe situation at the base of the pitch, witness is able to re-create the ascender's unintentional rope disconnection with the rope placed against a large boulder. Witness and climber discuss incident with other personnel present at the site.

INCIDENT INFORMATION REPORT


TITLE: PETZL ASCENSION B17-L ASCENDER UNINTENTIONAL ROPE DISCONNECTION

TYPE OF INCIDENT: Gear Failure, non-injury.

REPORTED BY: Bonnie Crystal

INITIAL REPORT DATE: 18 APRIL 2004

DATE OF INCIDENT: Saturday 27 March 2004, approx 1530hrs.

INCIDENT LOCATION: Cliffside above the entrance to "R" Cave, California.

INCIDENT LOCATION WEATHER: Clear, sunny, temperature 17C degrees, no precipitation.

LOCATION DETAIL: Dry limestone cliff, with rough uneven face.

ROPE: PMI, 9mm Static in good condition.

SRT METHOD: American style, no rebelays.

RIGGING: Top anchor of rope to trees and rock outcroppings above cliff face. A rope pad with separate rigging is positioned between the rope and rock approximately 1meter from the top, where the rock is rough and the top of the pitch becomes more horizontal.

PITCH DISCRIPTION: 10meters approximate total pitch height. Excess rope is flaked into a bag at ground level at the takeoff position. Slight overhang at mid pitch, rope hangs free to the ground for the lower half of the pitch. Above mid-pitch, rope lays against rock with gradual curvature slope of about 10degrees to 30degrees. 2meters from the top, the rope lays in a wide angle V-notch in the rock (approx 130 degrees).

OBJECTIVE: Practice ascending the rope from the bottom to the top of the pitch.

WITNESSED BY: Bonnie Crystal

WITNESS/REPORTER EXPERIENCE LEVEL: 33years+ Comm towers; aid-climbing; vertical-caving

CLIMBER: "Mike" (surname withheld)

CLIMBER HANDEDNESS: Right handed, right footed.

CLIMBER EXPERIENCE LEVEL: Novice

CLIMBER SYSTEM TYPE: Frog ascending system.

CLIMBING SYSTEM UPPER ASCENDER DESCRIPTION: PETZL ACENSION B17-L LEFT HANDED ASCENDER, recent model, excellent condition, clean, functioning freely as designed, very little indication of wear marks or scratches, no visible indication of damage. Attached to a 5/16" webbing foot sling through a bottom hole. The other bottom hole is attached via a lanyard to the seat harness D screwlink.

CLIMBING SYSTEM LOWER ASCENDER DESCRIPTION: PETZL CROLL ASCENDER, recent model, excellent condition, functioning freely as designed, very little indication of wear marks or scratches, no visible indication of damage. Ascender attached direct to seat harness D screwlink. Figure-8 style GGG Frog chest harness attached to upper hole of ascender.

PHOTO OF CLIMBER, ASCENDER, AND ROPE AT BASE OF PITCH
This photo was taken just minutes before the incident.



CORRESPONDENCE:


TO: Mr. Hank Moon (Petzl America)
FROM: Ms. Bonnie Crystal
DATE: 18 APRIL 2004

Dear Hank,

I originally sent an informal email report on an incident involving Petzl Ascension B17-L Ascender Unintentional Rope Disconnection to [xyz Petzl dealer] on 13 April 2004. I sent the informal email report to [xyz] because I own and use a lot of Petzl gear, which I have purchased from the good [xyz] company for many many years. He replied to me by email, indicating that he was forwarding my message to Petzl. On the following day, I posted an informal message on the San Francisco Bay Chapter of the National Speleological Society grotto's egroup describing the incident, because the incident happened at a cave during the SFBC grotto's regular vertical practice session. I was the only close eyewitness to the incident which involved a caver who was practicing vertical rope ascent on the cliff face above a cave entrance.

My informal message about the incident on the SFBC grotto egroup received a quick reply from another a Petzl dealer Mr.[abc]. He said that he had known about Unintentional Rope Disconnection with the updated B17 ascender since shortly after the new version of the ascender came out (1999). I asked Mr. [abc] a few questions in response to his comments. One of the questions I asked him was if he was speaking as a Petzl representative. At the time I am writing this, Mr. [abc] has not yet answered my questions.

I have not received a reply from Petzl to my initial informal report forwarded through [xyz]. I expected it would take some time for the report to be digested and a response or course of action to be formulated. In this email, I am providing Petzl, and other individuals directly involved with analysis of the incident with a more detailed report of the incident (attached below). Please be aware that correspondence regarding this may also be forwarded to concerned individuals, and thus may show up on the internet.

After the initial incident, back on the ground in a controlled environment, I re-created the failure mode of the ascender to try to understand it better. The failure mode is repeatable and requires only gentle motion for it to disconnect unintentionally from the rope. In other words: the ascender can come off the rope without the climber intentionally manipulating or disengaging the "safety catch", and in doing so, the ascender does not need to be torqued or pryed from the rope by force.

Upon close inspection, I observed that the design of the nylon cam "safety catch" has a curved crescent surface protruding for the thumb to manipulate it. This is described in Petzl promotional literature as a new improvement, which is "ergonomic". It is the protruding edge of this curved surface that forms an arc wedge with just the elongated smoothness and continuous angular position needed to induce lateral displacement of the safety catch when it is moved upward against a rough rock surface. That movement unlocks or disengages the safety while at the same time pulling the cam down and away from the rope capture enclosure. Once the rope is no longer captured, the ascender falls loosely off and away from the rope thereby disconnecting.

I have not encountered the above-described failure mode with such ease on the previous metal "safety catch" version of the Petzl Ascension Ascender, pre-1999. The previous version has a metal post safety catch manipulator protrusion instead of the nylon thumb crescent safety catch of the present-day version.

I look forward to Petzl's response to this report.

Best Regards,

Bonnie Crystal
18 APRIL 2004


Subject: RE: Petzl Ascender Unintentional Rope Disconnection
Date: 4/20/2004 4:46:13 AM GMT Standard Time

In a message dated 4/20/2004 4:46:13 AM GMT Standard Time, [Hank Moon at Petzl.com] writes:

Bonnie

Thank you for your very detailed description of the incident - much
appreciated. I have copied Steve on this post because of the following
statements he made in his post to the SFBC group:

"If any ascender opens under any circumstances, without the operator
deliberately opening the safety gate, this IS LIFE THREATENING."

"Other brands do not have such a problem."

Bonnie, you seem to share both of these opinions with Steve, so I guess
this question is for you: would you briefly describe your view of how
Mike's life was endangered by the incident? Your report states that
after the ascender was re-attached to the rope, "Climber continues
climbing rope and reaches the top of the pitch without physical injury."
This would seem to imply that there was an expectation of physical
injury, presumably from the lower ascender failing while the upper was
briefly detached from the rope? If this is correct, would you also give
your ideas of the possible mechanisms of such a failure? I am wondering
in all this if old fears about cast-aluminum jumars spontaneously
breaking are haunting this discusssion? Or am I reading too much into
the report?

I have no hard data on whether other brands might have "such a problem"
- or not. I have heard stories of various non-Petzl ascenders
inadvertently coming off rope, but again, nothing I have personally
witnessed. I have, however, seen other types of failure. I was caving in
TAG with my friend John about 15 years ago; can't remember the exact
cave, but it was multi-drop. John had just bought a pair of CMI
ascenders, having swallowed the misleading claim of "strongest on the
market". When I first saw them, I was impressed: their beefy frames sure
*looked* stronger than my Petzls. Plus they were black (always cool).
Having bottomed the cave, we were headed out when John got a funny look
on his face while getting on rope. "It broke," he said, holding up one
of his brand-new CMI's. "What"?!!I moved closer to take a look. Somehow
during our descent, the safety catch on one of the CMI's had broken. It
wasn't completely broken off, but it would no longer function as a
safety catch. So much for "strongest on the market", I thought. Being a
mild paranoid type, John had 3 other ascenders with him, so the ascent
out of the cave was uneventful.

I will reply in more detail to your report later this week. In the
meantime if you can provide your perception of how the situation was
life-threatening, it would help me to prepare a more meaningful
response. I hope that you do not perceive this request as disingenuous
or evasive; if we were speaking in person, I am sure that you would not.
However, the limitations of electronic communication can easily foster
misunderstandings. I hope that despite this difficulty, we can arrive at
a clearer picture of the causes and importance of this incident.

Best regards

Hank


Subject: Re: Petzl Ascender Unintentional Rope Disconnection 20APR2004
Date: 4/20/2004 8:12:47 AM GMT Standard Time

TO: Mr. Hank Moon (Petzl America)
FROM: Ms. Bonnie Crystal
DATE: 20 APRIL 2004

Hi Hank,

Thank you very much for your courteous and friendly reply to my report.

Perhaps it is the responsibility of the manufacturer to make it clear to users what the safe limits of use are for the products manufactured and sold. Perhaps it is the responsibility of the users to be informed, determine the safe limits of use, and to gain full knowledge of possible failure modes. Perhaps it is good for independent organizations to determine safety guidelines or point out suggested limits of safe use for products. Or perhaps it is a combination of the above.

I'm not personally in a position to determine what the safe limits of use should be for Petzl products. As you have noticed, I have not admonished Petzl to go forth and tell the world about this problem or to recall or redesign the ascender. I leave those tough decisions to you, your conscience, and your company's bottom line. That's probably something you and your company will have to wrestle with in the days ahead. With the liability situation as it is in USA these days, I certainly don't envy your position, Hank!

As for your question of hanging on a cliffside by a rope with only one solitary ascender... I tend to go with the conventional wisdom that it isn't safe. The fact that the solitary ascender in this particular incident was a Petzl Croll may have made it somewhat more scary to me at that moment. In fact, thinking about it now, it seems a little scary even with a single prussik knot or a Rock Exotica or a Gibbs, but yes, the fact that it was just a Petzl Croll is a bit more scary than some other ascenders I can think of if I had the choice...

As to whether or not the incident was "life threatening";  that phrase was injected into the discussion by one of your Petzl dealers [Mr. abc]. But when asked about it, I responded that indeed, when the incident occurred, the climber and I both reacted with the feelings and physiological symptoms of a threat to the climber's life. Adrenaline flowed.

In my report to you, dated 18 APRIL 2004, you should note that I've made no value judgment on the incident. I have simply stated the facts as best I know them and remembered them from the incident. I suspect that Petzl will want to investigate the ascender failure mode further, and I suspect that a design review might be initiated. The failure mode is so easily recreated by anyone with the simple description of it, that I don't think there is a need for you to be supplied with diagrams or photos or a video of it being recreated. But if your staff has any problem recreating it, I can at some point make an mpeg video or other photos of it being recreated here.

At this point, I want to give you a couple of corrections and clarifications to the 18 APRIL 2004 report, a result of [B.F.'s] research today ([B.F.] is the SFBC grotto vertical chair). B.F. was present at the site of the incident, but was not in a position to directly view the incident in action easily. He has discovered a photo that he had taken as "Mike" was attaching his ascenders while starting to climb the rope. This photo was taken just minutes before the incident. Please note the photo image shows the gold anodized color of the Petzl Ascension Ascender model B17-L, which is a left-handed ascender.  I have attached the photo to this email. [...]
 

Best Regards,
Bonnie

Bonnie Crystal
20 APRIL 2004



Hello Bonnie
 
In my haste to throw off the keyboard shackles, I omitted a phrase from the first line under "Redundancy" in the pdf I sent yesterday: "normally due to misuse." Please find attached the correct version (BC_UDAR_reply_050204.pdf).
 
Other stuff:
 
My reply to your report was originally much longer than the final version, but the word count got a bit out of control with delving into tangents on cause and effect. One such tangent addressed the "misuse" phrase. I know many people who have had ascenders pop off unexpectedly, most of them big-wall climbers. It nearly happened to me once (on a "small wall" desert tower), but instead the anchor failed and I took a big, scary swing :-O   Misuse can sometimes be a slippery word. For example, is it misuse when one is cleaning an aid traverse with radically loaded ascenders and one pops off? Sure, but sometimes certain situations demand one takes the risk of this happening. Bad weather, rising water, rockfall and other factors may induce a necessary haste that could lead to misuse and something unexpected happening. These types of situations arise often in sporting activities and some people seem to routinely get away with misuse without serious consequence. This can be due to something as simple as the person's weight: the weight of the pack or rack might make all the difference between staying on or popping off. So, if we can expect "routine misuse" to happen, what (if anything) can be done? Good question! I don't have any simple answers at the moment, but awareness, good judgment and technique, etc. are always the best ways to fend off accidents. One advantage climbers have over cavers is the rope they use. They can just tie into the (dynamic) rope every now and then and have a super-redundant system for climbing up the rope. When climbers hear that cavers rarely tie in this way, they are typically horrified until realizing the fall arrest forces produced by static rope don't provide a reassuring backup...
 
One section that I began to write, but ultimately left out was one on modifying the safety catch on ascenders. Following your report, I devised an informal "office UDAR test" that is probably similar to your "gentle motion" test, only mine initially used only the grabby office carpet under my desk - probably a much more demanding surface than most any rock! Laying an ascender open/cam side down against the carpet and moving it about with one hand, I was able to get the Ascension w/nylon safety catch to open. I then tried it on concrete and asphalt. Initially, the catch skittered over the concrete without opening, but would open on the bumpy asphalt. I tried some real pressure on the ascender (2 hands) on the concrete and noticed after a few tries that the catch had ground down a bit. As the point on it flattened out, it began to grab the concrete better, presumably due to the increased surface area and friction. I kept grinding until there was a large flat surface - it still opened. <many other boring details omitted>
 
From my informal fooling around, it seems as if the greatest contributing factor to opening are the abrupt edges and ridges on the safety catch. Of course, these are what makes the thing so nice to use when wearing gloves! The smoothest catch I made (using a Dremel tool) can be hard to open with insulated gloves, but seems to resist opening quite well, at least in my cursory testing. Another one I modified by cutting the point off with a heated razor blade and this seems to do a good job of smoothing off the sharper edges, making it harder to open. Ultimately, I have reached no conclusions on this topic and thus thought it premature to include anything definitive about it in my reply to you. I would like to continue to explore this issue (UDAR, safety catch modification, etc.) with you and other other cavers. To that end, would you mind if I send you the three units I've modified? The idea would be that you subject them to your "gentle motion" test and send me your informal results and opinion. Some cavers I've spoken with about this issue think the UDAR issue is a non-problem, others think it is. Balanced testing standards can emerge from such issues and differences of opinion, provided all parties are generally interested in achieving a common goal. I'll have to ask my counterparts in France why there is no test for UDAR in the EN ascender standard. Sometimes it's because nobody brought it up, but more commonly because no consensus could be reached. On vera...
 
hank
 
P.S. some of the folks on the UDAR group have not seen the detailed report you sent to Petzl - would you mind if I posted it to the group?

PDF of PETZL USA FORMAL RESPONSE

This website  (C) 2004, 2006 Bonnie Crystal
Petzl and the Petzl logo is a trademark of Petzl. Drawings and images of Petzl equipment are provided only on an educational basis.