No problem Dynosore. That stuff is all elsewhere here on RC but I left it out of the writeup since its already huge. I built my rig around what's called out in the EN12276/UIAA-125 spec document, which I have a copy of. The rig is a converted press frame with a hydraulic ram and power unit that's good up to ~10,000 pounds. The ram is bolted to a 10,000 pound load cell (the one in question is from Transcell Technologies, although I have others from different manufacturers) that feeds into a Daytronic 4077 Strain Gage Indicator. That in turn drives a National Instruments 6008 DAQ module, which then feeds into the NI SignalExpress software you see in the force charts. The Daytronic will scan at 1000 samples/sec digital, but it also has an analog peak capture function to do away with issues surrounding scan rates and impact testing (so its still scanning at 1000 s/s, but getting the true peaks between cycles). I have it set to output this data out the device's analog output port, which feeds into the DAQ which samples at 250,000 s/s, so way faster than the data coming out the 4077 (the reason I did this was so I have enough bandwidth to hook up five 4077s for doing drop testing on anchors). The load cells were all purchased new and came with certs, the 4077 was used from Ebay and the NI gear is new.

Anyway, the CE/UIAA spec calls for a ram speed of 20-50mm/min if the unit has textile components and 50-200mm/min if the unit has a load-bearing textile component. I've not measured the speed I use, but suspect it is closer to 200mm/min than 20mm/min. The crack fixture is a pair of 1" thick 1018 steel plates with threaded rod through the corners for width adjustment. The plate was textured by milling with multiple passes of an out-of-tram flycutter to give a nice stippled finish. The spec calls for a maximum surface roughness on the plates of 500um, which I think I'm in the ballpark of but have not verified as I don't happen to have a surface finish gage. There's more texture than a nailfile, but much less than a metal file and my feeling that I'm in the ballpark comes from working in a machine shop for 15 years.

Sorry if I missed anything; its been a busy day.

EDIT- Did miss something... The plateau was a combination of deformation and slippage; the cam came out the bottom of the fixture as the axle reached the end of its deformation.

Aliencamsbycch.com says every main cable since January 2006 has been tested in this manner. This makes me believe that the issue with poor "braze wicking" would be prevented from reaching the market.

Curious what post-recall (post-January 2006) Alien stem and braze failures make you believe?

Just thought I'd point out that these two statements contradict each other. 0105 is near the beginning of the recall period, but well within it. Doesn't affect the validity of your results in any way, but you might want to correct the original post just to avoid confusion.

I have 3 non-dimpled Aliens from the recall range (1104, 1204, 0105). They are in good condition and I can attest to their history. If enough people wish it, maybe I'll send them to you for destructive testing.

What say you all?

Sorry, I'll edit that post to make my statement clearer. The pull-testing of 100% of units to 7.7+ kN makes me believe that the same issue could not make it to the market. Had they pull-tested to 7.7+ kN back then, there's a good chance (knowing what we know) that this product wouldn't have made it out the door.

Not sure you're getting it, so again, what does post-'7.7+ pull-testing of 100% of units' stem and braze failures make you believe?

No need to play games. If such failures have been documented please link sources.

Don't worry. It didn't turn out the way he intended.

Thanks Patto, I needed that right about now. After almost 3 full days of calling and emailing CCH about this they're finally doing something... questioning my credentials, background, impartiality and test procedures.

Sigh.

What I mess I've created for myself. Its important to get the word out though, and I'd do it all again in a heartbeat.

-a.

<snip>
Units made after the recall began (1/05). Have been tested, and they DO NOt NEEd to be re-tested. <snip>

No need to play games. If such failures have been documented please link sources.

You've been here long enough to be paying attention - you go find them yourself - maybe try searching on 'Tensile Tested'

OK, here's what I dug up based on memory and a quick search. If the threads I linked here aren't the best sources, or my descriptions are innacurate, somebody please post up a correction.

5/15/7 Non-dimpled Blue Alien fails at 900lbs when tested by Russ Walling http://www.rockclimbing.com/...2;page=unread#unread

Souders Crack 11d groundfall (broken cable, non dimpled, post recall)http://www.rockclimbing.com/...=groundfall;#1585733

Faulty Swage (post recall)
http://www.rockclimbing.com/...n%20failure;#1316820

Dimpled Orange Alien Braze Failure at Indian Creek (the cam that started the recall)
http://www.rockclimbing.com/...n%20failure;#1277756

Gray Alien braze failure (2005, pre-recall)
http://www.rockclimbing.com/..._reply;so=ASC;mh=25;

Non dimpled Paradise Forks Orange Alien bodyweight braze failure (post-recall)
http://www.rockclimbing.com/..._reply;so=ASC;mh=25;

Misdrilled Axle Holes (rei recall thread)
http://www.rockclimbing.com/..._reply;so=ASC;mh=25;

Latest Indian Creek Purple Alien braze failure (newer cam, post-recall)
http://www.supertopo.com/...59&tn=0&mr=0

Well hell, one just flashed by me on another search so here you and spike go (just click on the dead, post-'7.7+ pull-testing of 100% of units' Alien):

[image]http://xs216.xs.to/xs216/07242/Blue_TESTED_s.JPG[/image]

i guess what i'm trying to say is maybe its not necessarily CCH's QC that needs immediate review so much as it is CE's standards (if my vague memory is correct). If you make a batch and only 'need' to test a very small percentage, that leaves a lot of room for some big oopsies to slip through the sift.

If your results aren't anomalous and 5kN Aliens are common then I guess that goes to show that the typical forces they are exposed to are actually quite low. A bit of indirect evidence for the Real World kN thread.

I have heard in the past that only a certain percentage of random samples from an entire manufactured batch need testing in order to give an overall product a certified strength rating. the friend that told me about it has been a long-term employee at a climbing gear shop, and i specifically remember the reason he was even telling me about this was because that minimum percentage required to be tested was alarmingly low (we were talking about testing biners). i believe this standard is found buried somewhere in some document outlining the CE standards.

...maybe someone can help me out here or knows more about this...has an actual number, perhaps? surely it is NOT 100% of every piece manufactured. in the same way, i doubt that every single cam is subjected to the 7.7kN test in order to "say" that they are all good by CE's standards. Is this true??

i guess what i'm trying to say is maybe its not necessarily CCH's QC that needs immediate review so much as it is CE's standards (if my vague memory is correct). If you make a batch and only 'need' to test a very small percentage, that leaves a lot of room for some big oopsies to slip through the sift.