-
Freelance Work at Height and Lifting Instructor.
There are currently three active cavers working with SpanSet as Freelance Work at Height and Lifting Instructors – are you interested in joining them?
Due to the continued expansion of our busy Training Department, we are always looking for freelance Instructors to join our team delivering training at our head office in Middlewich and at clients sites throughout the UK.
Wanted FEM analysis
- Thread starter Bob Mehew
- Start date
Steve Clark
Well-known member
Does it need to be FEM?
If I understand your problem correctly, a pin-jointed tetrahedral frame is statically determinate and you can calculate the forces in the nodes and members from simple statics. Member capacity likely from Euler buckling (or compact). Happy to send you a calc with notes if I can dust my brain off.
(If it’s a digging tripod built from scaffold, I’d hazard a guess the limiting factor will be the shear capacity of the swivel couplers and exactly how the load gets into the top of the frame)
If I understand your problem correctly, a pin-jointed tetrahedral frame is statically determinate and you can calculate the forces in the nodes and members from simple statics. Member capacity likely from Euler buckling (or compact). Happy to send you a calc with notes if I can dust my brain off.
(If it’s a digging tripod built from scaffold, I’d hazard a guess the limiting factor will be the shear capacity of the swivel couplers and exactly how the load gets into the top of the frame)
Bob Mehew
Well-known member
Sorry to disappoint you Ian, but Steve answered my question albeit using a static approach. Simply put and quoting their blurb, it is a freeware program providing "tools to produce, export and edit solid, full-precision models, export them for 3D printing or CNC machining, create 2D drawings and views of your models, perform analyses such as Finite Element Analyses...". FreeCAD does have a forum if you wish to learn more about it.
My interest was getting an understanding of what sort of structure I would need to put together to build a temporary drop test rig to do some experiments to obtain an understanding of the impact of dynamic drops on cows tails. Steve's work has given us that insight.
My interest was getting an understanding of what sort of structure I would need to put together to build a temporary drop test rig to do some experiments to obtain an understanding of the impact of dynamic drops on cows tails. Steve's work has given us that insight.
Your interest is interesting!
Steve Clark
Well-known member
It's not just the rig stiffness, the actual force in the rope will be totally dominated by the behavior of the squidgy human attached to the end. In a serious fall, parts of the bag of bones will be deflecting dozens of cm's different to the attachment point. The momentum for a given fall is a fixed input but the time element of the impulse varies with the behavior of the body and reduces the peak force.
DMM did a video on this a decade ago. Impressive snapping of dyneema slings with a big steel weight. Teaches a lesson about dyneema I suppose, but it doesn't acknowledge that a human isn't a steel lump [edited - I'd confused a later video where they explained this]
If you compare it to a car crash test - the important thing is what happens to the body. The acceleration experienced by different parts of the body relative to each other is what breaks things. Engineers could model the behaviour of the steel bits of the car a century ago. It what happens to the human that's the important bit and where modern engineering with airbags, pre-tensioners and crumple zones goes.
Avoiding sharp edges, I'd postulate that it's impossible to break a properly tied cowstail or even a dyneema sling with a falling human.
The cowstail or sling may easily kill the human though, and the only way to investigate that is to test it on a human (or a realistic crash-test model of one).
DMM did a video on this a decade ago. Impressive snapping of dyneema slings with a big steel weight. Teaches a lesson about dyneema I suppose, but it doesn't acknowledge that a human isn't a steel lump [edited - I'd confused a later video where they explained this]
If you compare it to a car crash test - the important thing is what happens to the body. The acceleration experienced by different parts of the body relative to each other is what breaks things. Engineers could model the behaviour of the steel bits of the car a century ago. It what happens to the human that's the important bit and where modern engineering with airbags, pre-tensioners and crumple zones goes.
Avoiding sharp edges, I'd postulate that it's impossible to break a properly tied cowstail or even a dyneema sling with a falling human.
The cowstail or sling may easily kill the human though, and the only way to investigate that is to test it on a human (or a realistic crash-test model of one).
georgenorth
Well-known member
I’m pretty sure there was a climbing accident a few years ago where a dyneema sling broke at a belay. I can’t find the report now unfortunately. As I remember the climber only climbed a tiny distance above the belay and then slipped.
There’s some more testing from Edelrid here: https://edelrid.com/gb-en/knowledge...7eCcOkJOoQltYCxBh5ezE0U9uVs1v3CkOfIUqAQKOE9HO
There’s some more testing from Edelrid here: https://edelrid.com/gb-en/knowledge...7eCcOkJOoQltYCxBh5ezE0U9uVs1v3CkOfIUqAQKOE9HO
TheBitterEnd
Well-known member
Perhaps instead of FEM it needs a CFD model of the squidgy human with an inertial model of the bag of bones 
Good point, Steve. I made the comment in the context of the rig not the test as whole but you're quite right, for any test of cowstails to be meaningful it needs a realistic representation of the dynamic behaviour of the human. Once you include that, the compliance of the rig and even cowstail become minor factors. So Bob is probably right just to consider the strength.
I wouldn't expect FEM to be necessary or even useful. The way we did things before FEM will work fine. I'm wary of these little FEM programs, as I suspect there's a lot they don't model, like buckling and connection offsets, so people think they have an answer but it's unsafe.
pwhole
Well-known member
I used to work in computer graphics, and worked for a software company developing the initial applications of soft object dynamics in animation software (3DS MAX, Maya, etc.). All the test animations done with rigid bodies or simple structures were fairly predictable, but once human or animal forms got involved, the calculations went through the roof, and every bug in the code would then start to manifest, as it was just so damned complicated to model. And that was just assuming objects were made of stiff jelly - once bones or organs were included internally, even crudely, the software would collapse. I assume they've fixed that problem these days, but I bet it was just by 'buying more computers', as it's still just as complicated to model.
Steve Clark
Well-known member
Yes, I am very wary of relying purely on anything FEM. It is normally quite easy to get an upper and lower bound on the 'correct' solution using normal statics/mechanics and these can be close together if you refine it manually. And it's usually in the details - exactly how the load gets into the structure - connections are mini-structures in themselves that can't be modelled with 6 numbers at a point in an FEM model. It also very easy to just make a mistake in a model by setting one member's properties, boundary condition or a constraint just wrong.
I would encourage anyone getting involved with FEM to play Polybridge. One missing member and the bridge fails under it's own weight. One extra member and it totally changes the local stiffness. My son, aged 8 at the time, was far better at it than me!
Poly Bridge on Steam
Unleash your engineering creativity with an engaging and fresh bridge-building simulator with all the bells and whistles. Enjoy hours of physics-based puzzle solving in the Campaign and then jump in the Sandbox to create your own bridge designs and puzzles!
Bob Mehew
Well-known member
I guess that I should add that we are well aware of the impact of a squidgy body as well as beams not being stiff enough on reducing peak forces. What I was originally after was a feel for what steel work might make a reasonably rigid structure for dynamic work, which is what Steve's calculations gave me. Our work over the past couple of years (thanks Glyn) using a static rig (thanks Paul) has been focused on the energy needed to stretch a cows tail and related forces. I won't go into the detail save to say it is a moderately complex area. What we now want to look at is comparing that static work to a dynamic situation where whilst in theory things are the same, in practice there are differences in detailed areas. Which is why we are on the hunt for a drop test rig or even a high rated lifting beam. Our objective is to publish a paper on the work so forgive me if I don't go in to further detail.
I will add that I think Paul Seddon's paper for the HSE still remains the best summary of impact forces on the human body. When I talked the topic over with him more than a decade ago, he did recall some practical work done in eastern Europe with people jumping off bridges. Alas I never found any report on it. So I suspect Riches paper for the HSE entitled "Analysis and evaluation of different types of test surrogate employed in the dynamic performance testing of fall-arrest equipment" is the best available linking real and test data. (PM me if you wish to read it as alas the HSE seem to have removed a range of useful but old reports so original web links no longer work.)
I will add that I think Paul Seddon's paper for the HSE still remains the best summary of impact forces on the human body. When I talked the topic over with him more than a decade ago, he did recall some practical work done in eastern Europe with people jumping off bridges. Alas I never found any report on it. So I suspect Riches paper for the HSE entitled "Analysis and evaluation of different types of test surrogate employed in the dynamic performance testing of fall-arrest equipment" is the best available linking real and test data. (PM me if you wish to read it as alas the HSE seem to have removed a range of useful but old reports so original web links no longer work.)
Similar threads
