Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Spreader Beam Design 7
- Thread starter WideMike
- Start date
Where are you designing this beam?
In the US the following may help guide guide you:
ASME B30.20a-2001 Structual and Mechanical Lifting devices. Section 20-1.2.2 Construction
CMAA crane requirements
AASHO 1.6.17
Steel Construction Manual-AISC
Occupational Safety and Health Administration (OSHA) standard 1926.753(e)(2)
American Society of Testing and Materials Specifications A391
Naval Facility Command NAVFAC-307
The design process for a steel beam can be summarised as follows:
a) Determination of all forces and moments on critical
section with appropriate safety factors included
b) Selection of UB or UC
c) Classification of section
d) Check shear strength; if unsatisfactory return to (b)
e) Check bending capacity; if unsatisfactory return to (b)
f) Check deflection; span/600 for vertical deflection and
span/400 for lateral, if unsatisfactory return to (b)
g) Check web bearing,unbraced length, flange ratio and buckling at
supports or concentrated load; if unsatisfactory provide
web stiffener
h) Check lateral torsional buckling; if unsatisfactory
return to (b) or provide lateral restraints
i) Size padeye for shear, include increased Kt stresses
j) Check shear tearout, bending and padeye welds, if unsatisfactory return to (i)
k) Summarise results
You must ensure the following items are met:
1. SF=3 to 5 depending on risk tolerence, enviroment, and application
2. "Registered professional engineer” to designed and sign a specific lifting fixture
3. Documentation of load calculations maintained
4. Rated capacity (working load limit) stamped on the fixture
5. Proof-tested, usually proof load I use is 2
Hope you find this helpfull
In the US the following may help guide guide you:
ASME B30.20a-2001 Structual and Mechanical Lifting devices. Section 20-1.2.2 Construction
CMAA crane requirements
AASHO 1.6.17
Steel Construction Manual-AISC
Occupational Safety and Health Administration (OSHA) standard 1926.753(e)(2)
American Society of Testing and Materials Specifications A391
Naval Facility Command NAVFAC-307
The design process for a steel beam can be summarised as follows:
a) Determination of all forces and moments on critical
section with appropriate safety factors included
b) Selection of UB or UC
c) Classification of section
d) Check shear strength; if unsatisfactory return to (b)
e) Check bending capacity; if unsatisfactory return to (b)
f) Check deflection; span/600 for vertical deflection and
span/400 for lateral, if unsatisfactory return to (b)
g) Check web bearing,unbraced length, flange ratio and buckling at
supports or concentrated load; if unsatisfactory provide
web stiffener
h) Check lateral torsional buckling; if unsatisfactory
return to (b) or provide lateral restraints
i) Size padeye for shear, include increased Kt stresses
j) Check shear tearout, bending and padeye welds, if unsatisfactory return to (i)
k) Summarise results
You must ensure the following items are met:
1. SF=3 to 5 depending on risk tolerence, enviroment, and application
2. "Registered professional engineer” to designed and sign a specific lifting fixture
3. Documentation of load calculations maintained
4. Rated capacity (working load limit) stamped on the fixture
5. Proof-tested, usually proof load I use is 2
Hope you find this helpfull
-
4
- #3
WideMike,
One trap for beginners is that the classical buckling solutions for lateral/torsional buckling do not apply to beams suspended from crane slings.
The point is that the classical solutions assume zero torsional rotation at the supports. However, for a crane suspended beam, any torsion must result in some end rotation, until the resulting eccentricity between the suspended load and the sling loads (applied at different heights) is sufficient to balance the torque.
For a long time I have made a practice of requiring that spreader beams are designed with sections with Iyy >= Ixx, so that they cannot buckle laterally (eg square or round hollow sections).
One trap for beginners is that the classical buckling solutions for lateral/torsional buckling do not apply to beams suspended from crane slings.
The point is that the classical solutions assume zero torsional rotation at the supports. However, for a crane suspended beam, any torsion must result in some end rotation, until the resulting eccentricity between the suspended load and the sling loads (applied at different heights) is sufficient to balance the torque.
For a long time I have made a practice of requiring that spreader beams are designed with sections with Iyy >= Ixx, so that they cannot buckle laterally (eg square or round hollow sections).
-
1
- #5
Boo1,
The basic problem was explained in detail in a copy of "Steel Construction" (Journal of the Australian Institute of Steel Construction), Vol 23, No 2 May 1989. Paper by Drs. Dux and Kittipornchai, "Stability of I-beams under self weight lifting".
In their conclusions the authors state :
The paper has investigated the stability of I-beams under self-weight lifting, when members are usually in their most slender state and when elastic flexural-torsional buckling forms an important design consideration. Existing classical and other published buckling solutions are not usually applicable...
An alternate reference (predating the 1989 paper, and not quite as detailed) is a paper by the same authors to the First National Structural Engineering Conference, Melbourne Australia, 26-28 August 1987. "Buckling of Suspended I-Beams".
The generally accepted buckling formulae (incorporated in one way or another in most design codes), assume from the start that the torsional rotation at each end of the beam is zero, and remains so even when the beam deflects laterally within its span (thus generating some applied torque).
But the only way a suspended beam can resist such applied torque is by rotating sufficiently to move the lift points laterally relative to the CG of the suspended load, so that the couple formed by the sling loads and the suspended load is equal to the applied torque. ie the boundary conditions assumed in the 'standard' formulae are violated, and the formulae are no longer applicable.
The learned doctors give an example of a 35m span plate girder which would have a ‘safety factor’ of nearly 2 against self weight buckling if it were provided with the ‘standard’ end torsional restraints (ie placed on wide supports, with load bearing web stiffeners above them). However, when hanging from vertical slings at the ends, it would buckle under its self weight alone. i.e. the effect of suspending it from the crane hooks would halve the buckling moment. I have checked a typical 610mm deep universal beam on 22m span using their methods, with similar results.
I have tried modelling the suspended restraints as springs of equivalent stiffness, and found less drastic reductions. However, I do not entirely trust my own analysis, and would most certainly not set it up as being more soundly based than the Dux and Kittipornchai work, which compared within 5% with experimental test results.
Generally, a spreader beam would not be as prone to the 'suspended beam' effect as a beam under its own self weight alone, since the end restraining torques for a given degree of rotation will generally be greater, (assuming that the lifted load is supended from lifting lugs located below the beam, and hence the lever arm of the couple is greater).
My simple way of avoiding any possible buckling problems with spreader beams (and to avoid trying to do the fancy maths) has been to adopt a section which cannot possibly be subject to lateral-torsional buckling.
As a general comment, there is a possibility of 'unforeseen behaviour' in most lifting devices, so long as the designer treats them as if they were ground-based structures (with all of the artificially restrained dof that are required to make them suited to computer analysis), and ignores the possible effects of lack of straightness and 'minor' eccentricities.
For instance, C frames can be prone to large out of plane distortions if the vertical element has insufficient torsional stiffness (although any simple 2D analysis will conclude that the torque in the vertical member is zero). But I will keep that story for another day…
The basic problem was explained in detail in a copy of "Steel Construction" (Journal of the Australian Institute of Steel Construction), Vol 23, No 2 May 1989. Paper by Drs. Dux and Kittipornchai, "Stability of I-beams under self weight lifting".
In their conclusions the authors state :
The paper has investigated the stability of I-beams under self-weight lifting, when members are usually in their most slender state and when elastic flexural-torsional buckling forms an important design consideration. Existing classical and other published buckling solutions are not usually applicable...
An alternate reference (predating the 1989 paper, and not quite as detailed) is a paper by the same authors to the First National Structural Engineering Conference, Melbourne Australia, 26-28 August 1987. "Buckling of Suspended I-Beams".
The generally accepted buckling formulae (incorporated in one way or another in most design codes), assume from the start that the torsional rotation at each end of the beam is zero, and remains so even when the beam deflects laterally within its span (thus generating some applied torque).
But the only way a suspended beam can resist such applied torque is by rotating sufficiently to move the lift points laterally relative to the CG of the suspended load, so that the couple formed by the sling loads and the suspended load is equal to the applied torque. ie the boundary conditions assumed in the 'standard' formulae are violated, and the formulae are no longer applicable.
The learned doctors give an example of a 35m span plate girder which would have a ‘safety factor’ of nearly 2 against self weight buckling if it were provided with the ‘standard’ end torsional restraints (ie placed on wide supports, with load bearing web stiffeners above them). However, when hanging from vertical slings at the ends, it would buckle under its self weight alone. i.e. the effect of suspending it from the crane hooks would halve the buckling moment. I have checked a typical 610mm deep universal beam on 22m span using their methods, with similar results.
I have tried modelling the suspended restraints as springs of equivalent stiffness, and found less drastic reductions. However, I do not entirely trust my own analysis, and would most certainly not set it up as being more soundly based than the Dux and Kittipornchai work, which compared within 5% with experimental test results.
Generally, a spreader beam would not be as prone to the 'suspended beam' effect as a beam under its own self weight alone, since the end restraining torques for a given degree of rotation will generally be greater, (assuming that the lifted load is supended from lifting lugs located below the beam, and hence the lever arm of the couple is greater).
My simple way of avoiding any possible buckling problems with spreader beams (and to avoid trying to do the fancy maths) has been to adopt a section which cannot possibly be subject to lateral-torsional buckling.
As a general comment, there is a possibility of 'unforeseen behaviour' in most lifting devices, so long as the designer treats them as if they were ground-based structures (with all of the artificially restrained dof that are required to make them suited to computer analysis), and ignores the possible effects of lack of straightness and 'minor' eccentricities.
For instance, C frames can be prone to large out of plane distortions if the vertical element has insufficient torsional stiffness (although any simple 2D analysis will conclude that the torque in the vertical member is zero). But I will keep that story for another day…
-
2
- #6
austim is right on in his answer. Spreader beams have simple geometry, but are more complicated than they seem.
I believe Modern Steel Construction magazine had a thorough discussion on this a while back. You may be able to search for it at their "Steel Interchange" online at .
I believe Modern Steel Construction magazine had a thorough discussion on this a while back. You may be able to search for it at their "Steel Interchange" online at .
Stephen, I see the pad eye safety fastor should be 5.
"The allowable capacities of prefabricated components, e.g. hooks, shackles, and wire rope, are established as required by OSHA in 29 CFR part 1926.251 with a factor of safety of 5 on ultimate strength."
"The allowable capacities of prefabricated components, e.g. hooks, shackles, and wire rope, are established as required by OSHA in 29 CFR part 1926.251 with a factor of safety of 5 on ultimate strength."
There is an excellent article in the AISC Engineering Journal written by David T. Ricker that addresses many aspects of lifting beam design. He also included some worked out examples as well. The article is in the EJ, fourth quarter, 1991. I used this article many times over the eyars.
I believe you can down load the article by visiting I am not sure there are fees associated or not.
Good luck.
I believe you can down load the article by visiting I am not sure there are fees associated or not.
Good luck.
- Thread starter
- #13
Seeking advise on 27T spreader beam construction.
25 feet span between padeye 1 and 2 each bearing a 10T vertically suspended load.
Single point crane lift using 2-legged wire-rope sling with 45 degree leg angle attached to top of padeye 1 and 2.
Will W8 x 67# beam or 8" Schedule 80 round pipe satisfy the design requirement within the buckling effect.
Any .xls spreadsheet calculation to assist on the beam/pipe and sling sizing?
25 feet span between padeye 1 and 2 each bearing a 10T vertically suspended load.
Single point crane lift using 2-legged wire-rope sling with 45 degree leg angle attached to top of padeye 1 and 2.
Will W8 x 67# beam or 8" Schedule 80 round pipe satisfy the design requirement within the buckling effect.
Any .xls spreadsheet calculation to assist on the beam/pipe and sling sizing?
Don't forget to proof test the lifting beam to 125% of the working load unless it will get used in Michigan where Michigan OSHA requires 200% (not more than 80% of the beam's maximum load). Also, mark the lifting beam with the manufacturer's name and address, lifting beam weight, serial number and working load. The proof test should be performed by a qualified company and witnessed by one or more qualified individuals.
I design all components to have a factor of safety of 5.
I design all components to have a factor of safety of 5.
There are two different systems used to lift large loads. They are spreader bars and lifting beams. Each has it's own design parameters and uses. The are too often interchanged.
ANSI has a specification on spreader bars. It is B30.20.
Here is site that might be of interest:
They have the hardware for the end caps for pipe type spreader bars.
ANSI has a specification on spreader bars. It is B30.20.
Here is site that might be of interest:
They have the hardware for the end caps for pipe type spreader bars.
I just designed a spreader beam as per Canadian Steel design codes. They don't really give any guidance on the design of a member with no end supports resisting rotation.
What I ended up doing was assuming that the beam is laterally unsupported along the entire compression flange. The beam would fail by laterally bucking sideways at each end. However, at the middle of the beam where the hookup to the crane is, the beam is not able to move out of its plane. For lateral buckling calculations, I treated the beam as having a "k" factor of 2.0 (rotation fixed at support, translation & rotation free at the loaded end) and used "L/2" of the beam as the unbraced length of the beam.
Therefore the effective unbraced length of the beam becomes 2.0*(L/2) = L = Full length of spreader beam.
I would also recommend using a closed "box" section to make the beam very strong in the lateral direction. For my application I used a W12X106 beam, with two continuous 3/4" thick side plates fillet welded to the beam flanges.
Karl T
What I ended up doing was assuming that the beam is laterally unsupported along the entire compression flange. The beam would fail by laterally bucking sideways at each end. However, at the middle of the beam where the hookup to the crane is, the beam is not able to move out of its plane. For lateral buckling calculations, I treated the beam as having a "k" factor of 2.0 (rotation fixed at support, translation & rotation free at the loaded end) and used "L/2" of the beam as the unbraced length of the beam.
Therefore the effective unbraced length of the beam becomes 2.0*(L/2) = L = Full length of spreader beam.
I would also recommend using a closed "box" section to make the beam very strong in the lateral direction. For my application I used a W12X106 beam, with two continuous 3/4" thick side plates fillet welded to the beam flanges.
Karl T
I am looking for info on working under a suspended load. I have a hoist like lifting device and need to have employees work under the product. 1: what is the definition of suspended load. 2: what are the conditional requirements that must be met to work underneath it?
DERLETS, OSHA PROHIBITS WORKING UNDER FREELY SUSPENDED LOADS UNLESS THE LOADS ARE SECURED SUCH AS BY CRADDLES. RIGGING SUCH AS HARNESSES,HEADACHE BALLS,HOOKS, SLINGS,SPREADER BEAMS ARE NOT CONSIDERED SUSPENDED LOADS BUT THEIR WEIGHTS MUST BE TAKEN INTO CONSIDERATION DURING LIFTS BY CRANES.
- Status
Similar threads
