Top rope fall analysis w/ strain gauge analyzer


Original Post
20 kN · · Hawaii · Joined Feb 2009 · Points: 1,348

I recently purchased a 500Hz. strain gauge analyzer and 5000 lbs load cell that I can use in the field with my laptop, so I decided to do some testing on TR falls today. This is just some quick non-scientific testing to familiarize myself with the equipment, so donÂ’t take anything in here too seriously. I took six falls of varying length and catch method (i.e. hard catch, soft catch, ect.). The graphs below represent the most serious of the six falls. Basically I wanted to determine what would be the highest impact force I could produce on the anchor while top roping without doing something extremely dumb, like taking a huge TR whipper on static rope.

Here are the specs:
-160 lb climber
-175 lb belayer
-Mechanically locking belay device
-9.9mm Maxim Glider, 9.5 kN impact force rating, 29% dynamic elongation, 5% static elongation
-Hard catch (the belayer just stood there and locked off)
50 feet of rope out, 12 foot free fall, 20 foot total fall distance (approximate, not measured)

To start with, some really crappy pictures of the set up:




Next, the fall graphed out:



This next graph represents the loading that was in the upper 50% of the fall range. The peak load was just shy of 700 lbs. so this graph shows the loading above 350 lbs. The numbers on the X axis represent the duration in milliseconds multiplied by ten.



This last graph captures what I am going to call the “high” or “mean peak” meaning it shows how long the force stayed in the upper 10% range of the fall range, or above 600 lbs. This graph would be useful to compare different falls of the same fall factor, but different fall lengths, because it shows how long the anchor experiences the peak force of the fall. As you can see the loading on the anchor was only in the top 10% for 160mS.



The peak impact force on the anchor was 3.09 kN. Keep in mind, I pulled out a fair amount of slack before I let go, my belayer gave me an intentionally hard catch, and I used a rope with a pretty high impact force. So your average TR fall will likely produce less of an impact force on the anchor, unless you weight a lot more than me.

This post marks the start of a series where I would like to compare a wide variety of falls using different ropes, catch methods, ripping gear, and so on. Tomorrow I hope to get some data on lead falls.

Josh Olson · · Durango, CO · Joined Mar 2010 · Points: 255

Awesome. Just awesome. Keep em coming.

Matt Hasenohr · · Unknown Hometown · Joined Nov 2010 · Points: 15

This is great! Would you take offense to me suggesting unit labels in the graphs next time? This would make interpreting the data easier. Thank you for this! Great work

20 kN · · Hawaii · Joined Feb 2009 · Points: 1,348
Matt Hasenohr wrote:This is great! Would you take offense to me suggesting unit labels in the graphs next time? This would make interpreting the data easier. Thank you for this! Great work
If you can tell me how to label them in Excel, sure. I dont use Excel very often so I only know the basics. I am trying to get Excel to use the millisecond time stamps that were generated in the .csv file that contains the load cell data set. However Excel wont let me change it, it will only allow me to use the cell numbers for the X axis. Its not a big issue because the cell number corresponds with the time stamp, so all you have to do is take the X axis value and multiply it by ten.
Jake Jones · · Richmond, VA · Joined Jul 2011 · Points: 1,420

Pretty cool device and software. Thanks for posting.

Ken Noyce · · Layton, UT · Joined Aug 2010 · Points: 2,067
20 kN wrote: If you can tell me how to label them in Excel, sure. I dont use Excel very often so I only know the basics. I am trying to get Excel to use the millisecond time stamps that were generated in the .csv file that contains the load cell data set. However Excel wont let me change it, it will only allow me to use the cell numbers for the X axis. Its not a big issue because the cell number corresponds with the time stamp, so all you have to do is take the X axis value and multiply it by ten.
Navy, are you using Excel 2003 or 2007? If you're using '07 when you plot the curves, the top bar will change to Chart Tools and will be on the Design tab, one of the sections on the bar will be called "Chart Layouts", you just choose a layout that has lables on the axis. Click on the lables in your chart (where it says Axis Title) to modify it.
J. Broussard · · CordryCorner · Joined Feb 2011 · Points: 50
This post violated Rule #1. It has been removed by Mountain Project.
Kurt Swanson · · Philadelphia, PA · Joined Mar 2010 · Points: 125

Cool!

Phill T · · Unknown Hometown · Joined May 2008 · Points: 80

this was effectively the same as taking a 6 foot fall above pro right? Fall 12 feet before the rope catches. I know its just semantics but what you did was not a typical TR fall unless your belayer got distracted by some boobies while you climbed 12 feet with a loop around your ankles. Not to say that doesnt happen...

I'd be interested in some peak forces involving a belayer that is anchored in (like on a multipitch climb), so they get pulled up some before coming taught to their tether. measure forces on both sides of that fall.

Ohhh you could have 'soft catch' competitions and see who could get the lowest peak force on the same fall...

Also recreating some of the DMM spectra/dynema static falls using a fleshy body and a harness rather than a rigid 150lb weight and steel cable. Granted don't go for the 12 foot ride on that, but comparing maybe 1-3 foot static falls or whatever and then compare it to a standard dynamic climbing rope.

Cool toy!

M Sprague · · New England · Joined Nov 2006 · Points: 4,878
Phill T wrote:Also recreating some of the DMM spectra/dynema static falls using a fleshy body and a harness rather than a rigid 150lb weight and steel cable. Granted don't go for the 12 foot ride on that, but comparing maybe 1-3 foot static falls or whatever and then compare it to a standard dynamic climbing rope. Cool toy!
I'd strongly suggest not doing that with a real person.

"Drop tests demonstrate the danger. DMM tested an assorted batch of Dyneema and nylon slings, using a 176-pound weight in fall-factor 1 (120cm drop on 120cm sling) and fall-factor 2 (240cm drop on 120cm sling) scenarios (www.dmmclimbing.com/video.asp?id=5). Even when the Dyneema slings did not fail, the impact force (18–22+ kN) delivered to the climber likely would have resulted in massive or fatal injury"
climbing.com/print/techtips...
Gunkiemike · · Unknown Hometown · Joined Jul 2009 · Points: 2,740
Phill T wrote:this was effectively the same as taking a 6 foot fall above pro right? Fall 12 feet before the rope catches.
Not the same unless the same length of rope is in play, which, as I understand the test fall, would not be the case. (The leader needs to have 6 ft of rope fed to him in order to fall 12 ft vs the OP pulled up 12 ft of rope then jumped.)
kBobby Hanson · · Spokane, WA · Joined Oct 2001 · Points: 1,265
20: Please don't take this as criticism, as it isn't meant as such and I think this is really cool. But I am curious about a couple things in your setup:

(1) Is there a lot of friction in the system between the climber and the anchor?
(2) Is the strain gauge properly calibrated?

I only ask because the steady-state load should be above 300 lbs (almost 320, really) and your data appears to have a steady-state below 300 lbs. Not really a big deal, and it certainly does not change the behavior of the system.

Anyway, well done and thanks for sharing!
JesseT · · Portland, OR · Joined May 2011 · Points: 110

Friction through a carabiner puts the static force experienced by the anchor at about 1.67 * climber weight as long as the belayer remains on the ground.

1.67 * 160lb = 267.2lb

If the belayer gets lifted up then static force at the anchor = climber weight + belayer weight.

20 kN · · Hawaii · Joined Feb 2009 · Points: 1,348
kennoyce wrote: Navy, are you using Excel 2003 or 2007? If you're using '07 when you plot the curves, the top bar will change to Chart Tools and will be on the Design tab, one of the sections on the bar will be called "Chart Layouts", you just choose a layout that has lables on the axis. Click on the lables in your chart (where it says Axis Title) to modify it.
I am using Excel 2010. Thanks for the info, I figured out how to label them now.
Jeffeos wrote:Side note: I think it's great that you're adding something of value to the database after seeing and despising your beta for The Moke. All of which was absolutely useless. Perhaps someone with a little more passion about the individual climbs should have had the honor of writing them up. As you were.
Well there is a reason why the descriptions on rc.com regarding the routes are very generic. Basically, before I came along there were not really any routes in the database, so I added them. At first I dident know much about the routes, so I just assigned the name and grade. Later I came back and added a description. Well one of the local climbers got all pissed at me for doing that. He did not want me to add the routes to the database at all because he is trying to write a guidebook for the area and he wants people to buy his book. He plans to use quite a bit of the money for rebolting and new bolting. Well I could just ignore him and do as I wish, but that guy happens to be my boss. So I just comply with his request. ThatÂ’s why the descriptions are shit, its intended that way, not because I am lazy or donÂ’t care. So I had to reach some type of middle ground with him, I tried to convince him to let me add the routes if I remained generic on their descriptions.

Phill T wrote:this was effectively the same as taking a 6 foot fall above pro right? Fall 12 feet before the rope catches. I know its just semantics but what you did was not a typical TR fall unless your belayer got distracted by some boobies while you climbed 12 feet with a loop around your ankles. Not to say that doesnt happen... I'd be interested in some peak forces involving a belayer that is anchored in (like on a multipitch climb), so they get pulled up some before coming taught to their tether. measure forces on both sides of that fall. Ohhh you could have 'soft catch' competitions and see who could get the lowest peak force on the same fall... Also recreating some of the DMM spectra/dynema static falls using a fleshy body and a harness rather than a rigid 150lb weight and steel cable. Granted don't go for the 12 foot ride on that, but comparing maybe 1-3 foot static falls or whatever and then compare it to a standard dynamic climbing rope. Cool toy!
Pretty much, it was closer to a lead fall than a TR fall, but like I said, I wanted to get a worst case scenario, or close to it anyway. However there are some differences between the fall I took and a real lead fall, such as the point Gunkiemike made.

kBobby wrote:20: Please don't take this as criticism, as it isn't meant as such and I think this is really cool. But I am curious about a couple things in your setup: (1) Is there a lot of friction in the system between the climber and the anchor? (2) Is the strain gauge properly calibrated? I only ask because the steady-state load should be above 300 lbs (almost 320, really) and your data appears to have a steady-state below 300 lbs. Not really a big deal, and it certainly does not change the behavior of the system. Anyway, well done and thanks for sharing!
That is a very important point. I know this, the cell did come with a load cell calibration certificate. I used the mV/V value on the calibration cert to program the software I am using. I weighed myself and another person on it and it was within 2 lbs of what the bathroom scale said for the sum of both of our weights.

Now, as another person said, the load on the top anchor should only be the sum of the climberÂ’s and belayerÂ’s weight if the belayer is completely off the ground. Although she was completely off the ground in some falls, she was not in other falls. Additionally, before I set the load cell up, I lead up the route, clipping five bolts. I left those draws clipped to the rope when I was conducting my falls, so there was some drag on the belayerÂ’s side of the rope. Furthermore, the route was not completely vertical, it was maybe 80-85 degrees. So even both the climber and belayer was completely off the ground, some weight would be lifted from the rope on both sides from the climberÂ’s and belayerÂ’s feet being in contact with the rock. I weigh 160 lbs, the belayer was 175. So thatÂ’s about 338 lbs. with gear. Now in the case of the fall I listed above, I do know the belayer was off the ground because it was a 20 foot fall. But looking at the chart, the data hovers around 280-300 lbs. It bounces up and down because after I fell I hand over handed my way up the rope. But with the drag of the draws, and both of our feet on the rock, I believe that the 40-60 lb difference between the sum of our weight and the weight the anchor saw could have been transferred to the bolts through drag or to the rock through our feet. Anyway, I know its accurate at the weight of both of us because we weighed ourselves free hanging. Tonight I am going to try to weigh us with our feet on the wall and see how much of a difference it makes.
Phill T · · Unknown Hometown · Joined May 2008 · Points: 80
M Sprague wrote: I'd strongly suggest not doing that with a real person. "Drop tests demonstrate the danger. DMM tested an assorted batch of Dyneema and nylon slings, using a 176-pound weight in fall-factor 1 (120cm drop on 120cm sling) and fall-factor 2 (240cm drop on 120cm sling) scenarios (www.dmmclimbing.com/video.asp?id=5). Even when the Dyneema slings did not fail, the impact force (18–22+ kN) delivered to the climber likely would have resulted in massive or fatal injury" climbing.com/print/techtips...
yeah I'm not saying do the full length falls. Start with like 6"-12" (15-30cm). See what kind of forces are generated. Repeat with a dynamic rope. Repeat with a steel cable and rigid weight. While I'm sure the data gathered by DMM was accurate, I do not think it is a good real world example at all. I dont think they meant it to be, but I think they did the climbing community a disservice by not adding a few more parameters. I am speculating, but I feel like the elasticity and impulse added by a real world climbing harness and a real human body with all its fleshyness will have a HUGE decrease in peak forces vs the steel cable/rigid weight used by DMM. I could be wrong, but thats why I'm so curious as to someone actually testing it, and now 20 shows up with this fancy new toy!
Jim Titt · · Germany · Joined Nov 2009 · Points: 490
Phill T wrote: yeah I'm not saying do the full length falls. Start with like 6"-12" (15-30cm). See what kind of forces are generated. Repeat with a dynamic rope. Repeat with a steel cable and rigid weight. While I'm sure the data gathered by DMM was accurate, I do not think it is a good real world example at all. I dont think they meant it to be, but I think they did the climbing community a disservice by not adding a few more parameters. I am speculating, but I feel like the elasticity and impulse added by a real world climbing harness and a real human body with all its fleshyness will have a HUGE decrease in peak forces vs the steel cable/rigid weight used by DMM. I could be wrong, but thats why I'm so curious as to someone actually testing it, and now 20 shows up with this fancy new toy!
The 80kg standard test weight is chosen because it represents the impact of a (slightly over) 100kg climber in a harness after considerable research and live-body testing by Troll (one of the pioneers of climbing harnesses and a world leader in industerial harness design) back in the 60´s when the UIAA decided to start rope testing.
There is plenty of research done into this and the factors are fairly well established, for a waist belt the multiplier is 1.4, for full body harnesses 1.1 and for sit harmesses the UIAA weight works out at 1.275.

That round thing you invented of is called the wheel!
Copperhead · · Unknown Hometown · Joined Aug 2011 · Points: 0
Jim Titt wrote: The 80kg standard test weight is chosen because it represents the impact of a (slightly over) 100kg climber in a harness after considerable research and live-body testing by Troll (one of the pioneers of climbing harnesses and a world leader in industerial harness design) back in the 60´s when the UIAA decided to start rope testing. There is plenty of research done into this and the factors are fairly well established, for a waist belt the multiplier is 1.4, for full body harnesses 1.1 and for sit harmesses the UIAA weight works out at 1.275. That round thing you invented of is called the wheel!
You are assuming the response to all of these systems is linear, when in fact it will not be. As we approach the harsh conditions of DMM's tests with a high factor fall on non-stretchy dyneema, the body will deform an immense amount (maybe enough for severe injury) and greatly reduce the force at the anchor by factors much greater than you've stated.

It would be very interesting to see. Aid climbers have been taking daisy falls for years, some of them on dyneema, and I haven't heard about results like DMM published, so there is a discrpency, but how big is it?
Jim Titt · · Germany · Joined Nov 2009 · Points: 490

Good that you know what I assume, presuming I assumed anything at all!

Phill T · · Unknown Hometown · Joined May 2008 · Points: 80

do you know where this is published? I was sure someone had looked at these factors before, but I certainly didnt see anything referenced in the DMM video. Seems a bit too simplistic to just make the 'climber' fatter and calling it good. I believe you that its been done, just be interesting to give er a read.

they really based UIAA falls on a 220lb climber?? I guess its better to err on the heavy side XD.

Copperhead · · Unknown Hometown · Joined Aug 2011 · Points: 0
Jim Titt wrote:Good that you know what I assume, presuming I assumed anything at all!
I'm really only sayting that your simplistic assertions are grossly inaccurate when used to predict the response of this system to a factor 1 or 2 fall onto dyneema.
Cody M · · Tucson, AZ · Joined Oct 2010 · Points: 90
20 kN wrote: If you can tell me how to label them in Excel, sure. I dont use Excel very often so I only know the basics. I am trying to get Excel to use the millisecond time stamps that were generated in the .csv file that contains the load cell data set. However Excel wont let me change it, it will only allow me to use the cell numbers for the X axis. Its not a big issue because the cell number corresponds with the time stamp, so all you have to do is take the X axis value and multiply it by ten.
The behavior you describe is unique to "Line" charts in Excel. Using chart type "Scatter" gives you a lot more freedom to manipulate how the data is displayed, especially regarding the x-axis.
Guideline #1: Don't be a jerk.

Post a Reply

Log In to Reply