Chockstone Forum

      Sponsored By
      ROCK
   HARDWARE

Chockstone Photography

Australian Landscape Prints

Chockstone Forum - Gear Lust / Lost & Found

Rave About Your Rack Please do not post retail SPAM.

Author

anchor testing rig

One Day Hero

15-Oct-2010
10:47:02 PM

On 15/10/2010 GPage wrote:
>
>Manufacturers specs would have a pressure surface- 2.21829 inch2 - 1000psi=9.9428549kn
>3mm wall -ps- 1.857723 inch2 - 1000psi=8.3267157kn
>
>Faders steel screwgate 34kn-(OG 12kn)-( standard issue toprope biner)-
> significant stretch after 4000psi- failure at approx 7100psi- failure
>due to gate key pin extraction. Held 2400psi after fail.
>
>Kong hms alloy screw(keylock) 22kn-(OG 8kn) Significant stretch after
>2000psi- failure 6100 psi-failure due to keylock sidewall blowout- held
>2000psi after failure.
>
>Petzl spirit keylock 23kn-(OG 9.5kn)- significant stretch after 2400psi-
>failure at 5500psi- failure due to keylock fail causing terminal spine
>failure.
>
>Faders pearshape alloy screwgate 20kn(OG 6kn)- significant stretch after
>2400psi-failure at 5600psi- terminal failure at spine,gate mech stayed
>intact.Gate pin quite bent would have failed soon after.
>
O.k., this thing is cracking me, I can't let it go.

Let's go back to the assumptions

Your assumptions;

1) The rig pulls 10 tons as per specifications

2) The minimum surface area for the piston must be X, as required for the rig to perform as per assumption 1)

My assumptions;

3) The rather large body of previous pull testing is more reliable than these new tests

4) Manufacturers (of biners and pull-testers) like to talk their shit up as much as possible

Conclusion

Either my assumptions or yours are wrong. Despite how it might look from all the equations, the calculated results hinge entirely on things which haven't been measured + the claims of the manufacturer

I think that the piston must be smaller than calculated, meaning that the force is lower than calculated for any given pressure. The likely consequnce of this is that the max pull of the rig isn't as high as the manufacturer claims (I think this is more probable than all those biners being so much stronger than the manufacturers claim)

I was hoping that an imperial ton was 1000lbs, it would answer the problem nicely. Unfortunately it is 2000lbs which is 8.9ish kN (this could knock 10% off your values if the thing maxs at 10 imperial tons rather than 10 metric tons)......still not enough to make me happy

Its not a problem which can be solved by clever thinking, the rig just needs calibration


voodoo

15-Oct-2010
11:52:54 PM

I know this might seem like a silly question, but I'll ask it anyway...

When you're calculating the 'pressure surface' you're using radius in pi*R^2, not diameter, yes? Because that would account for an error of significant magnitude.

As obvious as it may sound, I figured I'd give Occam's Razor a spin... :)

patto

16-Oct-2010
1:03:46 AM

Everything does seem to be breaking 100% more than predicted. At the same time it seems you have been doing things correctly.

How about you just calibrate the damn thing. Hang 100kg or just body weight on the ram and check the pressure read out.

I presume the external diameter of the ram is about 2inches?

GPage

16-Oct-2010
10:28:08 PM

I'm going to try and find someone local with a set of crane scales and see if they will let me do some pull test so as to calibrate the thing. It's going to be the most sure fire way of knowing. Probably also go see the engineering dept at the tafe or uni see what they got to play with.

davepalethorpe

7-Dec-2010
4:50:06 PM


IdratherbeclimbingM9

18-Sep-2012
12:18:20 PM

On 7/09/2010 J.C. wrote:
>Just some input if anyone is looking at buying one I have used the Hilti
>ones (a couple generations of it and tested thousands of anchors with em)
>and I'm not the biggest fan, thought the Hydrajaws one was much nicer to
>use.

Hmm.
A belated milestone achievement congratulation is finally recognised, with this being JC/mousey's 3,000th post to Chockstone... , though he is up to 3,105 posts now!
~> Foreezajollygoodposterandallthatstuff! ... and it is always good to have a reason to celebrate!
Way to go J.C.!
☺

TonyB

18-Sep-2012
3:16:52 PM

On 11/10/2010 One Day hero wrote:
>On 11/10/2010 davidn wrote:
>>Also worth noting that for those manufacturers using 3sigma (DMM, Black
>>Diamond, some others), they have confidence that 99.7% (or more) of products
>>won't break within the noted range. That doesn't say when it will break.
>
>More psychobabble. There is no "noted range", there is a single strength
>value. Almost all the biners will break above the stated value by a little
>bit..........double is not a little bit!

It's an old post but worth a comment. "3 sigma" is not "psychobabble" and is probably a reference to this from BD http://info.rockrun.com/articles/3-sigma-rating.html. It's not psychobabble but BD do show a lack of understanding of quality in manufacturing.
"A little bit" is just ODH's usual psycho babbling nonsense.

Lots of wonderful articles here, from the world's leading statistician in quality in manufacturing:
http://www.qualitydigest.com/read/content_by_author/12852
Here's a relevant one to start:
http://www.qualitydigest.com/inside/quality-insider-article/myths-about-process-behavior-charts.html
This one relates to BD's 10% comment
http://www.qualitydigest.com/inside/twitter-ed/part-spec.html

Jim Titt

19-Sep-2012
4:17:15 AM

The biggest Hilti Tester at the moment is a 30kN model (HAT 28E) and you really don´t want to know the price! The other problems with the Hilti tester are the limited pull-out distance (though you can get around that) and that if something breaks the seals blow under the shock loading, we´ve been there and done that!

50kN is really a bit low for guaranteed removal of glue-ins since there are plenty which go above this, I tested some Raumer ones today for a customer and they hung on to 58kN, the typical Salewa bolts get up to 75kN and our 8mm ones 79kN. And then we do the ones that get over 100kN. Even a 12mm wedge anchor should be getting well over 50kN.
I use a smaller portable tester which pulls 80kN with a struggle and weighs 15kg all up and a bigger one which gets well over 120kN (its rated at 20 tons) which is 35kg with the load cell, an unpleasantly heavy lump to screw around with on the cliff!

PDRM

19-Sep-2012
11:06:00 AM

I thought the 'standard' quality system metric/objective was six sigma. Thee sigma looks to me like 3 in every 1,000 units/events will be outside your control bands. This isn't great odds when you're mass manufacturing rated gear.

P

One Day Hero

19-Sep-2012
7:55:04 PM

On 19/09/2012 PDRM wrote:
>I thought the 'standard' quality system metric/objective was six sigma.
>Thee sigma looks to me like 3 in every 1,000 units/events will be outside
>your control bands. This isn't great odds when you're mass manufacturing
>rated gear.

God, some people are shit at working out simple stuff.........or they're just pussies, or maybe both.

Look, you don't really know anything unless you know what the standard deviation is, and they don't tell you that. It wouldn't be that high for most climbing gear, how about we say 1kN? So, your biners rated to 25kN would have their mean at 28kN, and 3 in 1000 will fail between 24 and 25kN, and 1 in 5000 (or whatever the 4sigma percentage is) will fail between 23 and 24Kn........etc. By time you get to 6sigma, it'll still be over 20kN which is still plenty strong enough.

Do you understand that going from 3sigma to 6sigma doesn't alter the actual gear? It just changes the number which is printed on the side. If the biner you're holding is exactly the same, who gives a shit what number is on the side?

I might have said it somewhere else on this thread, but what you want with climbing gear (at least, what a smart person wants) is high levels of inspection and a pull test before your bit of gear leaves the factory. Individual testing is an ironclad guarantee of some strength (probably more than you'll ever need), whereas statistical methods don't pick up rare yet critical flaws.

Petzl worked this out a couple of years ago when someone died due to a one off failure in stitching a daisy. It left the factory rated to 20kN and failed at bodyweight........no doubt it had quite a few sigmas attached, but that probably wasn't a comfort to the dude who splatted. I'm pretty sure that Petzl's response was to beef up their individual testing, rather than trying to add even more fuching sigmas!

Jim Titt

20-Sep-2012
7:21:06 PM

On 16/10/2010 GPage wrote:
>I'm going to try and find someone local with a set of crane scales and
>see if they will let me do some pull test so as to calibrate the thing.
>It's going to be the most sure fire way of knowing. Probably also go see
>the engineering dept at the tafe or uni see what they got to play with.
>
>
I´d also get it checked because we have also used this exact type of pull cylinder for a tester we built for someone else.
The 10ton rating is for the strength, not what you will achieve unless you use a different pump, if you back calculate from the pump piston to the cylinder piston you´ll find you have to jump up and down on the pump handle to get anywhere near 10 tons.
We could just get 3500kg on the tester we built using both arms (and I´m 6ft and big).
If you can still use a 10mm stainless maillon after 75kN you must have some good ones, we break 12mm ones at around 60kN and don´t generally use them for bolt testing anyway.

PDRM

21-Sep-2012
8:19:43 AM

On 19/09/2012 One Day Hero wrote:

>God, some people are shit at working out simple stuff.........or they're
>just pussies, or maybe both.

I have process design and control quals so yes I understand this. If you did, you would understand I was talking about process controls (manufacturing process QA systems) not product testing (QC controls). Individual testing is not the basis for quality management in any sophisticated manufacturing system; healthcare (cGMP), electronics (6-sigma usually) or climbing gear. You cannot test attributes into a product, it has to be designed and manufactured in, in accordance with a quality system.

Edit: I see now with a bit of searching that this 3-sigma applies to their batch QC procedures only. I would still hope that their manufacturing systems have tighter controls.

P

There are 72 messages in this topic.