I'd guess they tried to get their act together post-recall, too, but the cam Aric pulled to failure and cross-sectioned the joint was 2002. As in, a few years *before* the outsourcing and the recall.

Aliens are da bomb. Nothing beats them in versatility of placement. I have climbed on them and trusted them for years and will continue to do so. If past problems have led to better QC, all the better.

Proper placement of cams will continue to be the most critical and common aspect of avoiding cam failure.

Snipped from RC since its late and I don't see the point of retyping it...

It'll be a couple days until I get this all written up, as rather than testing the 5-8 cams I had hoped we'd be able to get we did 13 new Aliens, 9 used Aliens and one each BD pre-C4 and Metolius TCU. Took ~5 hours to go through it all and I'm trying not to think how long it's going to take to edit all the pics, tidy up the data logs and run all the pre-test pics through that spiral center software (got pics of a bunch of other brands they had as well for comparison).

In the meantime here's a bit more info:
Of the 13 new Aliens (including 2 offsets, dates ranging from 408 to 509), only 5 failed above their rated strength. The percentage of the rating held ranged from 63.3% to 116.7%, with an overall average of 94.3%.

Of the 9 used Aliens (dates ranging from early/mid 1990's to 1204), only 2 failed above their rated strength. The percentage of the rating held ranged from 54.7% to 105.0%, with an overall average of 81.9%.

Its been a long day, so I'm turning in. I'll work on the documentation some over the next day or two, but since I'm probably going to cross section most if not all of the heads it may even be sometime next week before I get it done.

Thanks again to everyone who chipped in for this, and a special thanks to Rich @ Rock&Snow for helping us out by discounting the samples considerably.

Actually, there're some missing details in the Mountain Project recap that is present in the Rockclimbing.com post.

From "gunkiemike":

"The black and blue Aliens failed at low loads (like 4-5 kN) due to their tracking out of the fixture. The larger sizes were close to their rated strengths, but many were below it, like 100-200 lb lower than spec. Some tracked out (with MUCH deformation/smearing of the lobes), other suffered cable failures. But bad brazes were not found in the new units. Bad brazes WERE found in a couple older Aliens. But none were "scary low" as in the 5 kN range. The folks watching concluded that the lobe metal is very soft (yes, some folks love Aliens for this reason) and this limits their strength, esp. in the smallest sizes. Also that the ratings by CCH might best be taken with a 1-2 kN grain of salt."

From "adatesman":

"That's how the majority of them failed... A combination of the lobes smearing off and the axles bending. Neither the BD nor Metolius had problems like this, so I don't think it was an issue with the fixture."

I'm certainly not buying any more CCH cams, but I think this result isn't actually as bad as it seems at first blush. Certainly the cams are failing below spec using a test very similar if not identical to the UIAA test, but the failure mode doesn't appear to be a manufacturing defect. It's impossible for me to know how the duration of time that the force is exerted over impacts lobe performance, but it seems plausible that the long pull is more likely to scrape off and deform the lobe material. The soft lobe issue has been known for a long time, and the soft lobes may in fact have certain benefits when used in the wild especially on hard rock. I tend to place my cams by sliding them--properly cammed--behind constrictions whenever possible. I find it hard to believe that enough metal could shear off to cause these placements to fail. Obviously, there's no protecting against a bad braze where the head simply pops off of the cable.

I'm curious to hear what the rest of you think.

Good point about the data actually representing different failure modes involving several different parts. It sounds educational, but it also makes the data rather useless. If you really want to study the braze, then it sounds like the cam would need to be disassembled and the braze section put into a special fixture.

Yeah, some more information leaked out on RC since there were ~a dozen or so people present for the testing. The full report will be coming in a couple days. There's a bunch more posts over on RC now discussing the implications of the softness and slippage if you're interested. My only beef with it is that it usually happened well below the rating and would indicate that the rating needs to be moved. There were no braze failures in any of the new cams and I will be disassembling some of them and testing just the braze joint shortly.

There was one braze failure in a 2002 though (which makes it 2 for 2 for ones from 2002 I've tested), which brings this all back around to my earlier point that if you are climbing on old, untested Aliens you really should get them tested before climbing on them again.

On a side note, a couple of them didn't just have material abrade from the lobes but had the lobes buckle as well.

help: http://mountainproject.com/v/climbing_gear_discussion/zero_f>>>>>

I climb with the understanding that climbing is inherently dangerous and it doesn't matter what the rating on the gear is after a certain point. Even if the gear holds you could still be looking at a serious pendulum into a very solid wall at a high rate of speed. Placing undue faith in your gear to save you from sub-optimal decisions is no different than believing that 4WD will bring your car to an immediate stop on a greasy, snow covered road.

Question for all the people who have the itch to break climbing gear. How come the pull testing rigs I've seen pictures of don't use some sort of stone face for the device under test to reside in? Granite and machined metal don't exactly have the similar properties, at least from a layman's view.

A comment on the original plots Aric posted on RC also. It seems like 1 second is an awful long time to be applying the force. Doing a thought experiment I would expect the force from a fall factor of 2 to be more like an impulse to the rated breaking strength of the device under test, with a damped spring resonance after the impulse leveling out to some much smaller fixed force around the weight of the average climber assuming one is climbing with a dynamic system.

Repeatability. Different labs can order jig faces with a spec texture in steel. Natural stone has too much variation.

It's difficult/expensive to build testing machines that can give repeatable results with high loads and short duration. In the real world, a cam that pulls out of a jig at 10kN with a slow pull quite likely would hold much more for a fraction of a second (depending on the rock, etc). It's another reason not to be too concerned about "breaking strengths" of cams--it's among the least important factors in selecting gear for your rack.

@Clyde- You may be right and the data may not be all that meaningful. The main thing I was looking for was braze failures in new manufacture gear, of which I didn't get any. I did get one more braze failure in a 2003 vintage cam though, which is quite troubling given that its the second pre-recall failure of that nature I've happened to run across. My whole thing with this is to try and get people to get their old gear tested so that these bad ones get weeded out, and the easiest way to do that is (should be...) to get CCH to expand the braze recall to cover earlier units. The lobe softness/axle bending are secondary issues IMO, although almost as concerning since they result in the cams not meeting spec when tested similarly to the UIAA spec. I've not had any trouble with cams from other manufacturers when doing this, so I have reservations about the ratings CCH gives as being accurate. Heck, Yellow through Red SL's spec a rating 100 pounds higher than the rating typically given the cable used to make their stems. This makes no sense to me, but it is what it is. Oh, and BTW, the minimum UIAA/CE spec called out on the copy I of it I have is 5kN, not 2kN, calls for 2 samples of each size and doesn't say anything about averaging tests or using 2 different friction surfaces (only a max roughness of 500um). Perhaps one of us has an outdated copy? Mine's the EN12276:1998 version.

@Nathan- I'm afraid I can't comment on the slow pull v. impact testing... The UIAA test is what's used for this, so that's how I modeled my testing. The copy of the spec I have states 20-50mm/min, or 50-200mm/min if there's a load bearing textile component. There's a good thread on Supertopo re: use of real stone for the test fixture and Mal@Trango chimed in to say that it works but wears out fast. I can only assume using steel with a specified surface finish makes for more repeatable and more durable fixturing.

How does one know if when their Aliens were manufactured? I know that I should have looked into this a long time ago when there was a recall but I was to buys climbing then and not sitting at a computer every day.

To another point if this is all a quality control issue then theoretically many CCH cams are good and many could be bad, but If I have individually tested all of mine by falling on them multiple times then theoretically they should be good as long as I don't apply a force greater than what I have already applied.

What do ya'll think about this?

Manufacturing date is stamped onto the trigger bar.

...And typical falls don't put anywhere near the force on the gear as you think they would. That piece that held the 50 footer when you were 100 feet out may well break if you accidentally slip while clipped to it with a runner since the static material will generate a much higher load than a fall on that much rope out. I've personally hit 5kN+ bouncing around on a nylon runner while bored and playing with the test gear.

BTW, I think I was told yesterday that they didn't start marking the date stamp until ~1994. I've got 3 used ones in the tested gear pile with no date stamp that are known to be from around then.

I dunno Clyde... The 2002 and 2003 failures I had failed by way of the stem pulling from the head with absolutely no braze on the end of the cable, both when pulled in the fixture. Definitely looks like a manufacturing defect to me and worthy of a recall. The braze on the 2003 looked perfect and we were all quite surprised when it failed so low and the inner bundle of wire in the cable was clean (no braze on it whatsoever, 7x7 construction). The 2002 (Red) came apart at 4.5kN and the 2003 (Yellow) came apart at 9.3kN, neither of which is very good for a cam that should hold 12kN.

I'll throw a couple on the puller once I get it set up again to test just the head/stem joint (maybe tomorrow or the next day) and cross section a bunch more on the mill and see if anything is amiss...

I've been following this thread just out of an unbiased interest. I've never used an alien, but I'm starting to think I never will. To me, the bottom line is CCH won't stand behind their work. The company seems a little shady on the customer service side and to me that is bad business. Safe or not, my choice is to not support them as a business. Just my thoughts.

Derek

Clyde,
I think maybe you should go back and read the rest of the thread as well as the links to RC.com. Some of your Alien defenses have already been covered extensively i.e.- (testing) the brazing.

Now personally, if I place a piece of gear, and even for a millisecond question the integrity of the piece, I will not use it. You entrust life and limb to your gear when you climb. So why, WHY would you ever expose yourself to an unnecessary, completely avoidable risk like that? These are the reasons why the only Alien that has ever been on my rack was a bootied gray alien. It was in good shape, with no visible defects, but I didn't completely trust it. Hence it was taken to the Sports Re-cycler and replaced with a Mastercam.

Some other CCH issues that tweak my noodle are;

A: Their complete failure (IMHO) to acknowledge the LEGITIMATE concerns and issues brought to their attention by Aric. This reflects a company that does not stand behind their work 100%. (Would you hire a contractor that doesn't stand behind his work? Now think, that is your home, not your personal safety and well-being.)
A: Using steel which is of sub standard strength TO WHAT THEY SAY THEY ARE USING! Be it due to improper treatment of the metal or just not using what they say they are using.
B: Using steel cable that is rated to 100lbs. lower than their "CERTIFIED" GOD DAMNED PULLING STRENGTH! Holy crapola! This is what I would consider to be an outright lie. And you would still trust your life to this gear because CCH says it is safe?

= No effin way will I ever use an Alien in such a place/manner that I am entrusting my life unto it. Not a chance. To all of those who will not part with their beloved Aliens, I wish you all well and hope that you never have to find out first hand why I will never trust one. Be safe, and if this thread makes you second guess them, drop 'em.
-Z

Sorry Phoenix, I won't waste time reading the silliness at rc.com.

Don't get me wrong, I am not defending the quality control or customer service of CCH! But I do think they are getting a bit of a bum rap for a product design that has proved excellent for around 20 years. Personally, I consider the original U-stem Camalots that many climbers still use a far more dangerous product than Aliens. Who cares if their quality control was better when the design itself is defective? Kong cams were even worse but they still turn up on racks.

At this point, Aric's testing merely raises more questions than it answers about Aliens. More data is needed but he doesn't have the resources to do it right and the AAC (who should be doing this stuff) doesn't care. The one cam that tested seriously low was compromised by a damaged cable before the pull so it's a moot point. The other cams are still in the ballpark for reliable placements even if they aren't to 12 kN spec. Even an Alien that merely holds 9 kN with a long, slow pull is pretty damn good in the real world--and still superior to many others that hold more in lab tests.

Bottomline nobody has made a cam that replaces Aliens, yet. BD, Metolius, and WC (among others) have all tried and failed. Not that their microcams are bad but they sure aren't equal. As soon as another company delivers a better Alien, CCH is history. Until then, I see no reason, so far, to doubt new ones.

The killer issue for me was the lack of braze in the pre-recall units: you just don't know what they'll hold. Somewhere between 0 and full strength just isn't good enough for me (all of mine are pre-recall). In the ballpark isn't even enough. I want to know, for sure, that the strength ratings of the pro I place are well *above* the ballpark.

For this failure mode, it's not all-or-nothing; failure depends on how well it was brazed, and there's just no way of knowing what's inside that joint. You can't definitively test them, unless you kill them - or, if you want to keep them, you can only test up to the strength where you start getting flat spots, which is well below rating.

Clyde, if the bootied Alien had failed due to the (remaining) wires breaking, you might have a point about it being compromised already. But it didn't - it failed by cables pulling out of the joint. The proof in the pudding is the cross-sectioned joint: it's clear as day that it was empty.