Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Dynamic load of a free fall 7
- Thread starter SKJ25POL
- Start date
-
1
- #2
-
1
- #3
This is brought up time and time again. It has a ton to do with how stiff the steel platform is, how far the fall was. It's an iterative process because you have to assume a deflection of the steel platform when it has successfully stopped the mass from moving. That distance is used to determine the acceleration (in this case it's negative because the mass is slowing down) and then you solve for the force, then recheck the actual deflection under that load. Repeat steps until solution converges.
If you have an infinitely rigid platform, then you would in theory get an infinite force because you'd be stopping the mass from some velocity right before impact over zero distance meaning your acceleration would be approaching negative infinity.
And a 5000lb chunk of concrete isn't being stopped by much more than the ground, or a much larger chunk of concrete.
If you have an infinitely rigid platform, then you would in theory get an infinite force because you'd be stopping the mass from some velocity right before impact over zero distance meaning your acceleration would be approaching negative infinity.
And a 5000lb chunk of concrete isn't being stopped by much more than the ground, or a much larger chunk of concrete.
TehMightyEngineer
Structural
My previous employer had a project like this (unloading stand for a logging truck) and we spent almost a month on it. Just setup a test area with some load cells and test it to find the impact force, don't try to do it by calculations alone because it's going to be almost impossible.
Professional and Structural Engineer (ME, NH)
American Concrete Industries
Professional and Structural Engineer (ME, NH)
American Concrete Industries
- Thread starter
- #5
TehMightyEngineer (Structural),
I would really appreciate if you can explain more what test needs to be done (what the test is called)?
Is it like they drop the 4800 lbs to test it???
I have a canteleved platform in 120ft up in the air which is cantelevered by the open structure.
I have to check if the platform can withstand the drop of this conctere block (part of a counterweight system)
The free fall is 12.5 feet the dimension of the platform is 20 ft by 7.00 feet and covered by grating
Please if you have any documents pics, information on similar situation direct me.
I am unclue where to find some helpful information
Thank you so much
I would really appreciate if you can explain more what test needs to be done (what the test is called)?
Is it like they drop the 4800 lbs to test it???
I have a canteleved platform in 120ft up in the air which is cantelevered by the open structure.
I have to check if the platform can withstand the drop of this conctere block (part of a counterweight system)
The free fall is 12.5 feet the dimension of the platform is 20 ft by 7.00 feet and covered by grating
Please if you have any documents pics, information on similar situation direct me.
I am unclue where to find some helpful information
Thank you so much
I`ve always been curious about this type of problem and never reached a solution - iterative or otherwise.
Jayrod - you say you assume a deflection/distance, and that distance is used to determine the acceleration. Position, velocity, and acceleration are all related by time and if we're assuming the distance, we cannot also assume the time, right? How do you get time?
If you have a weight falling unrestricted, you could easily determine how long it takes the weight to pass through the deflection, but in reality, the beam pushes back and it will take longer to travel this distance. I think this change in time will depend on the stiffness of the beam, and hence, you can't easily calculate it.
What am I missing?
Jayrod - you say you assume a deflection/distance, and that distance is used to determine the acceleration. Position, velocity, and acceleration are all related by time and if we're assuming the distance, we cannot also assume the time, right? How do you get time?
If you have a weight falling unrestricted, you could easily determine how long it takes the weight to pass through the deflection, but in reality, the beam pushes back and it will take longer to travel this distance. I think this change in time will depend on the stiffness of the beam, and hence, you can't easily calculate it.
What am I missing?
this is an impact problem. The loads generated depend on the stiffness of the structures, the time of impact/deceleration; that's why it's very difficult to get a real answer from analysis.
We do similar analysis (bird strike) with FEA (LS-Dyna) but modelling the structures realistically for stiffness (particularly at joints) is time-consuming. And testing to validate the model.
I'd take a lot of care before I started lobbing 2+ tons lumps of concrete around.
it's easy to figure out the velocity of the concrete block prior to impact, right?
another day in paradise, or is paradise one day closer ?
We do similar analysis (bird strike) with FEA (LS-Dyna) but modelling the structures realistically for stiffness (particularly at joints) is time-consuming. And testing to validate the model.
I'd take a lot of care before I started lobbing 2+ tons lumps of concrete around.
it's easy to figure out the velocity of the concrete block prior to impact, right?
another day in paradise, or is paradise one day closer ?
- Thread starter
- #9
You guys keep saying we do tests. Can some body please explain what test? what company needs to be hired to do the test?
I appreciate somebody kindly explain to me what test?
Hope you are trying to help rather than mudding the water.
This is not a new problem and I thought it is a solved engineering subject
To test this - get a really strong scale, take it out into the parking lot, and drop a 4800 lb chunk of concrete on it from 12.5' up. Repeat the test a couple times, and look at the force that the scale measured.
This won't accurately account for the deflection of the beam, but will capture the force dissipation associated with the crushing of the concrete and is a pretty reasonable starting point.
If this is an existing structure that wasn't designed for this criteria, I'd bet a donut that it can't take the load.
If you're designing it, it's going to be some *serious* steel.
Either way - please share the scale readings. I think we'd all be interested.
-
1
- #11
They literally mean build a platform that is exactly the same as the one you want to know about, then drop masses on it from heights to determine what kind of force modifier to use.
In response to Once20036. You're right, it's a combination thing, I've never officially ran a number using this method, so I would have to spend more time fine-tuning it, but you have to assume something, a distance it is stopping over, the time it takes it to stop, something. If you have the distance it stops over and the speed prior to impact, you can determine the acceleration without caring about time.
I see it sort of the following procedure:
1) determine speed of mass right before impact
2) Assume a stopping distance (deflection) x
3) Use conservation of energy to determine the spring constant K that would result from said stopping distance x
4) determine force in spring using distance x and spring constant
5) apply force to model to determine actual deflection
6) Substitute calculated deflection for assumed deflection from step 2
7) repeat steps 2 through 6 until the solution converges.
Again that's just how I understand it being possible to determine, I don't know if it would ever converge, I also don't have a feel for how the localized damage from impact may affect the results.
SKJ25POL
In all seriousness, your platform is likely only designed for a total load between 7000 and 14000 lbs. That's gives you, at best, a dynamic impact factor of 2.8. Your 5000 lb hunk of concrete will be moving at 20 MPH by the time it gets to the platform, if you flipped your cantilever platform vertical would it stop a car driving 20MPH? My money is no chance. Or at least not without some serious deformation happening.
In response to Once20036. You're right, it's a combination thing, I've never officially ran a number using this method, so I would have to spend more time fine-tuning it, but you have to assume something, a distance it is stopping over, the time it takes it to stop, something. If you have the distance it stops over and the speed prior to impact, you can determine the acceleration without caring about time.
I see it sort of the following procedure:
1) determine speed of mass right before impact
2) Assume a stopping distance (deflection) x
3) Use conservation of energy to determine the spring constant K that would result from said stopping distance x
4) determine force in spring using distance x and spring constant
5) apply force to model to determine actual deflection
6) Substitute calculated deflection for assumed deflection from step 2
7) repeat steps 2 through 6 until the solution converges.
Again that's just how I understand it being possible to determine, I don't know if it would ever converge, I also don't have a feel for how the localized damage from impact may affect the results.
SKJ25POL
In all seriousness, your platform is likely only designed for a total load between 7000 and 14000 lbs. That's gives you, at best, a dynamic impact factor of 2.8. Your 5000 lb hunk of concrete will be moving at 20 MPH by the time it gets to the platform, if you flipped your cantilever platform vertical would it stop a car driving 20MPH? My money is no chance. Or at least not without some serious deformation happening.
a 5,000lb car ? ![[bigsmile]](/data/assets/smilies/bigsmile.gif "[bigsmile] [bigsmile]")
and I think a car will crumple more than a concrete block ! mind you the concrete will shear at some stage, so ... ??
I would not use a scale in a car park, though it would give you the worst case load (the ground is much stiffer than your structure, i'd expect.
The best test is build a copy of your structure and drop your concrete block on it ... but, of course, no one's going to do that ! (way too expensive, and way too risky with some analysis to give you confidence that it'll work)
Personally, I'd go with analysis that assumes an impact load factor ... 2, 3, ... and goes from there. Remember the impact force is only part of the problem, what about the impact area (is the block landing face on, or corner first ?) and where ? (the middle of the panel is probably the most compliant, at the support the structure is very stiff)
another day in paradise, or is paradise one day closer ?
and I think a car will crumple more than a concrete block ! mind you the concrete will shear at some stage, so ... ??I would not use a scale in a car park, though it would give you the worst case load (the ground is much stiffer than your structure, i'd expect.
The best test is build a copy of your structure and drop your concrete block on it ... but, of course, no one's going to do that ! (way too expensive, and way too risky with some analysis to give you confidence that it'll work)
Personally, I'd go with analysis that assumes an impact load factor ... 2, 3, ... and goes from there. Remember the impact force is only part of the problem, what about the impact area (is the block landing face on, or corner first ?) and where ? (the middle of the panel is probably the most compliant, at the support the structure is very stiff)
another day in paradise, or is paradise one day closer ?
rb1957,
Couldn't you just assume some elastic deformation of the structure, and work out the resulting accelerations, forces and resulting stresses? Obviously, elastic deformation will be very small, accelerations and forces will be high, and there is an excellent chance this thing will not work. The OP can iterate if he wants more accurate results. It should not be that difficult to estimate roughly the maximum elastic deformation.
I would wonder about pieces of shattered concrete falling one hundred twenty feet to the ground? Given that the deformation is elastic, what is to stop the concrete block from bouncing off in some random direction?
--
JHG
Couldn't you just assume some elastic deformation of the structure, and work out the resulting accelerations, forces and resulting stresses? Obviously, elastic deformation will be very small, accelerations and forces will be high, and there is an excellent chance this thing will not work. The OP can iterate if he wants more accurate results. It should not be that difficult to estimate roughly the maximum elastic deformation.
I would wonder about pieces of shattered concrete falling one hundred twenty feet to the ground? Given that the deformation is elastic, what is to stop the concrete block from bouncing off in some random direction?
--
JHG
I'm just throwing darts here:
Start getting a handle on the deflection, do a static analysis of the platform support structure, find out at what load it will yield, and what the deflection is at that load. Then figure out what de-acceleration will produce that force, then figure out what time that process takes over the deflection available. Somewhere in there you may find out it either works or does not work. Your end result may be that the structure has to deflect over a long period of time, you may be able relate the strain energy of the structure to this time frame, maybe not. If the time frame is an hour, that's unrealistic. If it is on the order of a millisecond, I think you may be in the ball park. I'd be interested in seeing the data. Use MathCAD or some other software package so you can iterate, change things, make plots, etc...
Start getting a handle on the deflection, do a static analysis of the platform support structure, find out at what load it will yield, and what the deflection is at that load. Then figure out what de-acceleration will produce that force, then figure out what time that process takes over the deflection available. Somewhere in there you may find out it either works or does not work. Your end result may be that the structure has to deflect over a long period of time, you may be able relate the strain energy of the structure to this time frame, maybe not. If the time frame is an hour, that's unrealistic. If it is on the order of a millisecond, I think you may be in the ball park. I'd be interested in seeing the data. Use MathCAD or some other software package so you can iterate, change things, make plots, etc...
-
2
- #16
GregLocock
Automotive
"of course, the easy (and inaccurate) way is to use an impact factor of 2 (so that the force on the structure is, near enough, 10,000 lbs)." No don't do that. Check out how that factor of 2 is derived. For a start you can see it gives an odd answer, since it ignores the drop height.
You could build an accurate scale model of the setup and instrument that and test it. Scaling it is not going to be straightforward.
A non linear FEA model such as LS Dyna would be used in the automotive industry to simulate crashing cars into concrete blocks to some tremendous accuracy. I think that is overkill, and you probably haven't got the budget and you certainly don't have the correlation data.
jayrod12 suggested this is an impact problem, which he's simplified down to a single degree of freedom spring mass system. That is the way I'd approach it. You know the stiffness of the platform, you know the velocity at initial impact, you know the mass. First year dynamics. k*x2=m*v2 gives you x, then k*x is your maximum force. This assumes linear elastic, in real life you might prefer to absorb the energy in a plastic hinge in which case you need to know the moment of plasticity of your platform.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376
You could build an accurate scale model of the setup and instrument that and test it. Scaling it is not going to be straightforward.
A non linear FEA model such as LS Dyna would be used in the automotive industry to simulate crashing cars into concrete blocks to some tremendous accuracy. I think that is overkill, and you probably haven't got the budget and you certainly don't have the correlation data.
jayrod12 suggested this is an impact problem, which he's simplified down to a single degree of freedom spring mass system. That is the way I'd approach it. You know the stiffness of the platform, you know the velocity at initial impact, you know the mass. First year dynamics. k*x2=m*v2 gives you x, then k*x is your maximum force. This assumes linear elastic, in real life you might prefer to absorb the energy in a plastic hinge in which case you need to know the moment of plasticity of your platform.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376
It's a simple energy problem. mgh = spring energy of platform. You'll have to figure the spring constant of the platform. From this, you compute the deflection of the platform and the corresponding forces in the platform.
As a check, a suddenly applied load applied to a non-deflected platform will cause 2g's as it deflects the platform.
As a check, a suddenly applied load applied to a non-deflected platform will cause 2g's as it deflects the platform.
I had a much smaller, but illustrative project. It was a metal frame with and electronic instrument inside; part of a stack that clipped like LEGO bricks. One of the tests was a 3 foot onto plywood. The G forces recorded for bottom down were less than for top down. At first the test guys were puzzled, but it turned out the engaging pins on the bottom embedded themselves in the plywood, increasing the impact duration and lowering the peak load.
tl;dr - prediction is very inaccurate because collision deforms and damages the items in ways that are expensive to evaluate. Test it in the condition you expect and see if works afterwards.
tl;dr - prediction is very inaccurate because collision deforms and damages the items in ways that are expensive to evaluate. Test it in the condition you expect and see if works afterwards.
MikeHalloran
Mechanical
A 4800 lb concrete block is probably much smaller than 7x20 ft in planform, so it will likely punch right through the grating and keep on going, impacting on whatever is under the grating.
If there's any structure between grating and ground, e.g. sway bracing of the cantilevered platform, and/or perimeter and intermediate support of the grating, and that structure is at all efficient, it will probably buckle, the block will fall farther, and progressive buckling or subsequent impacts by the block in its now deflected trajectory will have a fair chance of bringing down the whole damn thing.
Could be an expensive test.
Could also be an expensive analysis.
Analysis gets interesting, even more so when you try to account for 35+ years of neglected maintenance and corrosion of the tower and uncertainty about the properties of the material actually used.
Have you considered a new tower, designed from the start for the subject load case?
Or perhaps with redundant load paths for the counterweight, to mitigate the chance of it falling?
Mike Halloran
Pembroke Pines, FL, USA
If there's any structure between grating and ground, e.g. sway bracing of the cantilevered platform, and/or perimeter and intermediate support of the grating, and that structure is at all efficient, it will probably buckle, the block will fall farther, and progressive buckling or subsequent impacts by the block in its now deflected trajectory will have a fair chance of bringing down the whole damn thing.
Could be an expensive test.
Could also be an expensive analysis.
Analysis gets interesting, even more so when you try to account for 35+ years of neglected maintenance and corrosion of the tower and uncertainty about the properties of the material actually used.
Have you considered a new tower, designed from the start for the subject load case?
Or perhaps with redundant load paths for the counterweight, to mitigate the chance of it falling?
Mike Halloran
Pembroke Pines, FL, USA
- Status
Similar threads
- Locked
- Question
- Locked
- Question