|
|
|
|
jt512
Jun 9, 2009, 10:12 PM
Post #76 of 130
(5324 views)
Shortcut
Registered: Apr 12, 2001
Posts: 21904
|
cracklover wrote: adatesman wrote: GeneralZon wrote: Aric, How are these holes being drilled? In a press using a jig? free handed? I am having a hard time trying to figure out how any of these "mis-drilled" cams got onto the market. Seems to be a blatant disregard for any QA/QC. I'm 99% certain that the lobes are being made in a single operation on a lathe that has live tooling on the turret (former CCH employee confirmed my suspicion). Basically they load the extruded bar into the bar feeder and it drills holes a single lobe deep in the end, cuts the slot and then cuts that lobe off the extrusion and feeds the bar out a bit for the next lobe. Because of this there's no fixture to screw up when the hole is off center, so it can slip through unnoticed quite easily if you don't do first piece/last piece inspection. I strongly believe that this is not a QC issue in the sense GeneralZon thinks. What I mean is that it's not just someone at CCH who knocks into the jig, getting it off-center. Rather, it looks very likely to me that the programming for the rig at CCH for some of these cam lobe drilling is simply set up wrong, and has been wrong consistently for a very long time. I'll explain both why I think it's a matter of the program being wrong, and why I think it's been like that a very long time. Check out these three separate yellow cams of mine: Yellow #1 [image]http://i26.photobucket.com/albums/c150/gostriker/alien_pics/my_yell.jpg[/image] Yellow #2 [image]http://i26.photobucket.com/albums/c150/gostriker/alien_pics/als_yell.jpg[/image] Yellow #3 [image]http://i26.photobucket.com/albums/c150/gostriker/alien_pics/old_yellow.jpg[/image] And check out this one of Aric's: [image]http://www.shariconglobal.com/misc/pulltesting/axle_center/sample20-error.JPG[/image] All are off the same amount, in the same direction! A highly unlikely coincidence IMHO. More likely the result of the programming at CCH being off. Now - the dates. Aric's cam is straight off the factory floor - just made a couple months ago. Of my three cams, one was made in '06, one in '07, and the third was made in the very first generation of Aliens (identifiable by a different head shape, different sheath connection method, different trigger wires, etc) from the late 80s or early 90s. I want to emphasize again, this cam: (Yellow #3) [image]http://i26.photobucket.com/albums/c150/gostriker/alien_pics/old_yellow.jpg[/image] is a living fossil. It was manufactured before any modern Aliens (actually, one just like it is in the nuts museum) and is misdrilled identically to all the modern misdrilled Yellows! I think that speaks volumes. GO Edited for clarity
I found that results from this program vary according to the camera angle. Anyone else having that problem?
|
|
|
|
|
adatesman
Jun 9, 2009, 10:22 PM
Post #77 of 130
(5316 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
cracklover
Jun 9, 2009, 10:30 PM
Post #78 of 130
(5304 views)
Shortcut
Registered: Nov 14, 2002
Posts: 10162
|
^^^ what he said.
But frankly, when I tried to create variance by getting further or closer, I wound up with essentially identical results. So, basically, if you can shoot straight on (lined up with the axle) your results should be quite consistent.
GO
|
|
|
|
|
boku
Jun 9, 2009, 10:56 PM
Post #79 of 130
(5291 views)
Shortcut
Registered: Jun 11, 2004
Posts: 278
|
adatesman wrote: Average Lobe Hardness: -Lobe 1: HRB55 -Lobe 2: HRB53 -Lobe 3: HRB52 -Lobe 4: HRB23 (not a typo)
[off-topic]
That reminds me, can you test the hardness of the lobes of that KROK cam I tore apart? I can mail them to you next week. It might make for an interesting comparison.
Thanks, Bob K.
|
|
|
|
|
adatesman
Jun 10, 2009, 11:04 PM
Post #80 of 130
(5204 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
qtm
Jun 12, 2009, 9:16 PM
Post #81 of 130
(5091 views)
Shortcut
Registered: Apr 8, 2004
Posts: 548
|
I just went through all my cams. For Aliens, Black, Blue x2, Green x2, Red, Grey, Yellow, and Gold.
The two blue aliens, one from 2004, one so old it doesn't have a date code, are pretty much identical to the two GO posted. Really if I posted them you'd think I just copied GO's images. All the others pretty much match up as well. All but the gold are pre-recall.
All the others; 8 WC Tech Friends, 2 WC Zeros, 5 Camlot (old), 2 Camalot C4s, 2 Metolius Master Cams... all there rest are pretty much spot.
Aric, if you want the images, I'll zip 'em and email them to you. But given the results, I don't think it's worth posting them to the thread.
|
|
|
|
|
jfield
Jun 13, 2009, 10:41 PM
Post #82 of 130
(5029 views)
Shortcut
Registered: Apr 26, 2005
Posts: 5
|
Excellent that somewhat is using the program at last. It has been most frustrating to me that this very important safety issue has lingered on and on for years, and I am glad for everybody's is attention to it now, especially adatesman. 
Anyway, I just wanted to mention a few things for people who are interested in the cam fitting.
1) I'd love to hear from you. Please drop me a line. I am at 'jfield' on my home address on dorringtoninstruments.com
2) Yeah, the automatic mode is very picky. Sorry. I never really finished the image processing. Maybe if the interest continues, I'll go back to it.
3)Care must be used when fitting Metolius Supercam angles. These cams use a specially designed non-constant cam angle[ log(a*theta +b*theta^3) ] that flares out so as to improve safety near the tip out point - e.g. towards the outer end of the cam range, there is a point where you become more worried about tipping out than about slipping, and therefore it makes sense to trade off the forces applied to the rock near the maximum extension to reduce axle bending when near the tip out limit. Therefore, exact matching is not possible in the program for supercam axles. The induced error is pretty small though.
4)For the geeks, go check out the CT scan X-ray of the supercam at dorringtonclimbing.com/CT
It's fun to play with and see the insides of the thing.
cheers
John Field
|
|
|
|
|
bradley3297
Jun 14, 2009, 1:49 AM
Post #84 of 130
(4987 views)
Shortcut
Registered: May 5, 2008
Posts: 83
|
all three of my aliens had at least 1 mis-centered lobe on all of them. there are very new... cch pissing me off. maybe ill whip on one with a bomber c4 backing it up to see how well it does.
|
|
|
|
|
jt512
Jun 14, 2009, 1:55 AM
Post #85 of 130
(4986 views)
Shortcut
Registered: Apr 12, 2001
Posts: 21904
|
bradley3297 wrote: all three of my aliens had at least 1 mis-centered lobe on all of them. there are very new... cch pissing me off. maybe ill whip on one with a bomber c4 backing it up to see how well it does.
You need a control group in order to determine whether you are aligning the camera lens properly with the cam axle. I got alarming results with my Aliens, but then found the same alarming results with my Camalots. A little experimentation showed that the results are sensitive to the camera angle. I still can't figure out how to align the camera reliably, however.
Jay
|
|
|
|
|
holdplease2
Jun 14, 2009, 3:41 AM
Post #86 of 130
(4961 views)
Shortcut
Registered: Dec 18, 2002
Posts: 1733
|
Search the forum on "Yellow alien pulled" and you will find a 2006 thread discussing the mystery of yellow aliens failing in 'perfectly good placements.'
-Kate.
|
|
|
|
|
cracklover
Jun 14, 2009, 3:55 AM
Post #87 of 130
(4953 views)
Shortcut
Registered: Nov 14, 2002
Posts: 10162
|
Strange, Jay. You had me wondering about my results, so I just took a whole round of six photos of the same cam, from slightly different angles, in different light. They all show the same thing, except one very blurry one where I guess the software didn't pick up the edge properly.
Unfortunately, the only other small cam I have to run through the software is a single TCU, and that one came out looking perfect.
GO
|
|
|
|
|
jt512
Jun 14, 2009, 4:09 AM
Post #88 of 130
(4945 views)
Shortcut
Registered: Apr 12, 2001
Posts: 21904
|
cracklover wrote: Strange, Jay. You had me wondering about my results, so I just took a whole round of six photos of the same cam, from slightly different angles, in different light. They all show the same thing, except one very blurry one where I guess the software didn't pick up the edge properly. Unfortunately, the only other small cam I have to run through the software is a single TCU, and that one came out looking perfect. GO
I'll post the results of my varying the camera angles when I have a chance.
Can you give me some information about how you're shooting: type of camera, lens, distance from the cam, etc? And how are you positioning the cam? If I lay the cam on a surface, I find that it tilts to one side or the other? Are you doing anything to keep the axis of the axle vertical?
Jay
|
|
|
|
|
adatesman
Jun 14, 2009, 2:03 PM
Post #89 of 130
(4901 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
jt512
Jun 14, 2009, 2:40 PM
Post #90 of 130
(4889 views)
Shortcut
Registered: Apr 12, 2001
Posts: 21904
|
adatesman wrote: Hey Jay, Are you manually selecting the points or letting the software do it automatically? That's the biggest issue I had with it and found it to be much more tolerant of skew when manually selecting them. I got an email from the guy who wrote it last night and he said he didn't finish the image processing as well as he wanted to because there didn't seem to be much interest in the axle location issue so he moved onto other things. My pics were shot on an Olympus Stylus 1020 point&shoot with most being done in super-macro mode (lens ~2" or so from the cam). I simply did it by hand, centered on the axle and then rotated just enough to hide the lobe on the other side and give a nice sharp edge for the top lobe. The other thing I tried (and had success with) was putting a thick washer around the nut on the cam (so the cam lobe would balance on the washer rather than the nut/axle) and then carefully balancing it on a flatbed scanner. You have to make sure to have the other lobes rotated a bit so they don't show up in the scan, but it isn't too hard to do.
Originally I was letting the software fit the curve. Later I tried picking the points myself, which did result in the center being better positioned. However, when the software drew the curve it visually appeared to be a better fit than when I selected the points. So, I am not convinced that the better center found when I picked the points was not due to my error.
I also tried using a scanner, but all I got was a nice sharp scan of the axle; the cam lobe was too blurry to use.
I'll have to fool around with it more. But it's pretty academic for me at this point, given the apparent issue with the brazes.
Jay
|
|
|
|
|
qtm
Jun 14, 2009, 11:03 PM
Post #91 of 130
(4856 views)
Shortcut
Registered: Apr 8, 2004
Posts: 548
|
I'm using a canon a640 in macro mode, holding it about 4 inches away from the cam.
The biggest problem I encountered was with older cams; the outside edges are worn and rounded; it's easy to see in the color photos, but when they are turned to grayscale in the program, the shiny rounded edges can hide the darker, physical middle of the lobe and this throws off the curve. Had to carefully manually pick the darker, middle of the lobe to get an accurate curve.
I think maybe putting a white sheet of paper behind the outside lobe may make it easier to pick out the edges of the lobe but I didn't want to go through it all again when the results of manually picking out the lobes were consistent with the non-aliens.
|
|
|
|
|
cracklover
Jun 15, 2009, 1:39 PM
Post #92 of 130
(4992 views)
Shortcut
Registered: Nov 14, 2002
Posts: 10162
|
Canon powershot sd1000.
I've varied the background, and varied between shooting with a macro at a couple of inches, or shooting from ~ 3 feet away. They both gave me similar results. The trouble with shooting from further away was that it was hard to hide the rear cam lobe and still center on the axle. Of course if you're manually selecting point on the software, it doesn't matter if you hide the rear lobe or not.
I always used the manual selection in the software, as I found it did a better job of matching the edge of the cam.
GO
|
|
|
|
|
TradEddie
Jun 18, 2009, 2:01 AM
Post #93 of 130
(4920 views)
Shortcut
Registered: Mar 29, 2007
Posts: 164
|
Another non-Alien image analysis, Black Metolius, made about 2000.
Also Aric, if you want these lobes for any testing, let me know, and thanks for all the hard work you've done recently.
TradEddie
|
|
Attachments:
|
camresult2.jpg
(86.4 KB)
|
|
|
|
|
adatesman
Jun 18, 2009, 2:09 AM
Post #94 of 130
(4915 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
dj69
Jun 18, 2009, 2:28 AM
Post #95 of 130
(4904 views)
Shortcut
Registered: Oct 18, 2006
Posts: 43
|
mheyman wrote: Someone want to start going through ANAM to see how many (espcially yellow) Aliens pulled?
this accident comes to mind.. i remember everyone laughing.. saying it was impossible that an alien slipped out from the crack. Maybe due mis-centered axel?
e_wire wrote: An experienced climber (a very good friend of mine), suffured injuries to his left wrist after falling 25 feet when climbing in the Montreal, Quebec region this week-end. Apparently, a Green Alien "poped" after the climber felt. The Alien was placed at about knee level when the climber felt. The route was a 5.10b (named: Samourai) located in Val David, North of Montreal. Another climber, a very well know trad teacher in the area, reported seeing and earing the pro pop like a bottle of champaign. The pro was well placed, and didn't walk. The Alien also looks pretty good, after examination. Only explanation from both the climber, belayer and spectator, is that the Alien slipped out of the rock due to some humidity in the crack. It's not the first accident reported on Aliens, please be on the lookout for tricky placements! The climber was operated Sunday and will most likelly suffer for several months. Hopefully he'll get back to 100%, but doctors are not that optmistic. Tks.
(thread link http://www.rockclimbing.com/...35838&highlight=)
D
(This post was edited by dj69 on Jun 18, 2009, 2:30 AM)
|
|
|
|
|
mheyman
Jun 18, 2009, 3:06 AM
Post #96 of 130
(4888 views)
Shortcut
Registered: Jul 25, 2002
Posts: 607
|
Although I probably would have agreed with many of the replies at th time, look at tthe bashing this guy took.
Perhaps he'll get it checked out.
|
|
|
|
|
jt512
Jun 18, 2009, 4:09 AM
Post #97 of 130
(4875 views)
Shortcut
Registered: Apr 12, 2001
Posts: 21904
|
adatesman wrote: Posted this in one of the other Alien threads since it was in response to a question there, but probably should have been put here... Here's the scoop on Healyje's Offsets... I'm sure everyone's tired of scrolling past pics, so here's a link to them. Same deal as before, with Red being at mostly closed, Yellow at ~50% and Blue at almost tipped out. All of them are stamped 12/04 except for the Blue/Black, which is stamped 03/05. Blue/Black Lobe 1: -Red: 16 - 5.8 = 10.2 Degrees -Yellow: 16 + 0 = 16 Degrees -Blue: 16 + 2.4 = 18.4 Degrees Blue/Black Lobe 2: -Red: 16 - 9.2 = 6.8 Degrees -Yellow: 16 + 2.7 = 18.7 Degrees -Blue: 16 + 7.5 = 23.5 Degrees Blue/Green Lobe 1: -Red: 16 - 6.7 = 9.3 Degrees -Yellow: 16 + 1.5 = 17.5 Degrees -Blue: 16 + 4.3 = 20.3 Degrees Blue/Green Lobe 2: -Red: 16 - 1.7 = 14.3 Degrees -Yellow: 16 + 1.5 = 17.5 Degrees -Blue: 16 + 2.6 = 18.6 Degrees Yellow/Green Lobe 1: -Red: 16 - 1.8 = 14.2 Degrees -Yellow: 16 - 1.7 = 14.3 Degrees -Blue: 16 + 0.7 = 15.3 Degrees Yellow/Green Lobe 2: -Red: 16 + 7.3 = 23.3 Degrees -Yellow: 16 + 2.0 = 18.0 Degrees -Blue: 16 - 2.4 = 13.6 Degrees Yellow/Red Lobe 1: -Red: 16 + 5.9 = 21.9 Degrees -Yellow: 16 + 2.0 = 18.0 Degrees -Blue: 16 - 1.5 = 14.5 Degrees Yellow/Red Lobe 2: -Red: 16 + 4.2 = 11.8 Degrees -Yellow: 16 + 0.6 = 16.6 Degrees -Blue: 16 - 1.5 = 14.5 Degrees
Aric, every time I look at results like this, the first question that pops into my head is, "How do these compare with other brands of cams?" I honestly think that it is imperative that each time you present results for Aliens, that you include results for other brands, as a control—at a minimum, include a link to such results. Otherwise, you are essentially presenting the results of an uncontrolled experiment. The reader has no way of knowing whether these results are unusual, or not.
Jay
|
|
|
|
|
blondgecko
Moderator
Jun 18, 2009, 4:17 AM
Post #98 of 130
(4871 views)
Shortcut
Registered: Jul 2, 2004
Posts: 7666
|
I'll post this here, too - a quick plot showing the relationship between coefficient of friction and the critical angle at which a cam will slip under any force. Blue boxes = approximate range of frictional coefficients between rock and aluminium. Green = sliding friction coefficient between aluminium and mild steel; purple = static coefficient between same.
|
|
Attachments:
|
critical_angles.gif
(9.19 KB)
|
|
|
|
|
adatesman
Jun 18, 2009, 2:21 PM
Post #99 of 130
(4835 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
|
