I am currently in the design phase of a project where the client is wanting to hang an operable partition wall over a clear span of approx 90'. Seems like quite the challenge with the restrictive deflection limits of operable partitions. But, I came across this thread (thread507-498793) where it was mentioned that someone had recently designed a frame to span 75' with restrictive deflection tolerances, so maybe there's hope for this 90' span as well. Does anybody have any insight on how this could be achieved? Would it be some sort of super deep, stiff Joist Girder/"SP" Joist? Or a Moment Frame of some kind (similar to what you would see in a pre-engineered Metal Building)? Thanks in advance!
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 JAE on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Hanging Operable Partition Support Over LONG Span
- Thread starter Stephen_S
- Start date
Harbringer
Structural
For a 90' span I'd be looking at designing a steel truss. What does a moment frame have to do with it? Do you need this for lateral also?
- Thread starter
- #3
@Harbringer The use of this beam/truss would be only to support the gravity loads from the operable partition. No lateral loads, or other gravity loads like roof or floor loads will be going into this member.
A moment frame would provide some extra restraint at the ends of the beam, creating less deflection along the length of the beam as the operable partition is opened and closed, whereas bar joists or joist girders (trusses) typically have pinned end connections, allowing for more rotation at the joist supports and therefore more deflection along the length of the member. With a moment frame, bringing those end moments down to the foundation would be something I would need to address, but that seems doable.
Perhaps designing a custom steel truss with fixed ends (or partially fixed ends) would be the way to go here. This would require some extra analysis of the truss that I'm not necessarily used to, but it may be the only way to get the job done.
A moment frame would provide some extra restraint at the ends of the beam, creating less deflection along the length of the beam as the operable partition is opened and closed, whereas bar joists or joist girders (trusses) typically have pinned end connections, allowing for more rotation at the joist supports and therefore more deflection along the length of the member. With a moment frame, bringing those end moments down to the foundation would be something I would need to address, but that seems doable.
Perhaps designing a custom steel truss with fixed ends (or partially fixed ends) would be the way to go here. This would require some extra analysis of the truss that I'm not necessarily used to, but it may be the only way to get the job done.
I think you need to visit with the applicable partition manufacturer. Many of these partition systems have variable skirts or buffers that allow for a degree of deflection.
This may allow you to stiffen your system only to the degree where the partition demands it vs. some pie-in-the-sky L/480 limit or something.
This may allow you to stiffen your system only to the degree where the partition demands it vs. some pie-in-the-sky L/480 limit or something.
jerseyshore
Structural
How long is the moveable portion? The entire 90 ft?
- Thread starter
- #7
@JAE Thanks for the input. I didn't realize these "variable skirts" or "buffers" existed and thought it was just a clean cut deflection tolerance of 0.5"~1" for partitions. I will followup with the manufacturer and see what they say.
@jerseyshore Yes the moveable portion will be along the entire 90' span.
@phamENG A floor supported partition has been discussed briefly, but the client would like us to explore the possibility of a hanging partition first. The potential ROI of the project is rather high so the client is willing to shell out quite a bit of money for this support, even if it ends up being a ton of steel. Any idea on the cost differences between floor supported and hanging partitions? Would the install effort/labor be the difference?
@jerseyshore Yes the moveable portion will be along the entire 90' span.
@phamENG A floor supported partition has been discussed briefly, but the client would like us to explore the possibility of a hanging partition first. The potential ROI of the project is rather high so the client is willing to shell out quite a bit of money for this support, even if it ends up being a ton of steel. Any idea on the cost differences between floor supported and hanging partitions? Would the install effort/labor be the difference?
Harbringer
Structural
Fixed ends or not you'd need a very deep beam, probably a plate girder would be more appropriate and constructable. Instead my approach would be a regular pinned end steel truss with cord splices so it can be transported. Depth would be dictated by deflection. The more difficult thing to get right will probably be floor flatness tolerances. The skirt at the bottom will only accommodate for so much floor irregularity.
jerseyshore
Structural
I would think a big truss would be best for this. Easily spliceable as Harbinger said. And I agree with JAE that speaking to the manufacturer is the best idea. There is always some flexibility to their partitions that allow a little more than the footnotes may seem to indicate.
Is this thing 90 ft long that will fold up on one side? Half to each side?
Is this thing 90 ft long that will fold up on one side? Half to each side?
People really miss the boat on this. The deflection requirement should be confirmed with the manufacturer, but a 90' span is plausible.
While you can get a Joist girder up to 120' long ( it shouldn't be a joist girder (joist girders are for intermittent large point loads), 48DLH or 48LH joist or so would be the ballpark figure, and these joists all have inherent camber (based on a circular radius/arc, so it's not the quartic deflection that actual loads produce), then decide how many you want to engage to control deflections and detail accordingly. The dead load deflection is at least somewhat already in-place on the steel joist by the time the partition guy appears, (joist weight, metal deck, accessories, roofing, insulation, probably sprinklers if needed, mech/elec and miscellaneous), and there is some consideration for that, or there should be.
It's more a coordination item than anything else. Plus how the partition deploys can create odd loading conditions that would lean toward a special joist, particularly because of the odd load distribution that may be needed, splits to both sides, all runs to one site, etc. Stacked load at point of storage, etc. If you are going to engage multiple joists then the loads from the transfer/bracing would also show up, and a note that the panel points should align is perhaps a good idea so whatever detailing you put between the joist works out, rather than trying to drag down an adjacent joist with a connection at the top chord without a panel point being there.
The weight of the various models can vary as well, so get a basis of design down early with the architect (if applicable) picking it. The weight isn't typically that much, unless it's a particularly tall partition, but the deflection requirements can be onerous, particularly if you misconstrue the load cases it applies to.
If you end up needing a xxDLH17 or 18 keep in mind the seat depth changes.
Regards,
Brian
P.S. a "separate steel beam" is not the way to go, they are far less stiff than the joist will be due to shallower depth and continuous webs, even with the added deformations from shear. There is also an ASTM on the subject, although it's more about acoustical shorts than anything we get involved in.
See also (wood, but discusses panel load patterns)
While you can get a Joist girder up to 120' long ( it shouldn't be a joist girder (joist girders are for intermittent large point loads), 48DLH or 48LH joist or so would be the ballpark figure, and these joists all have inherent camber (based on a circular radius/arc, so it's not the quartic deflection that actual loads produce), then decide how many you want to engage to control deflections and detail accordingly. The dead load deflection is at least somewhat already in-place on the steel joist by the time the partition guy appears, (joist weight, metal deck, accessories, roofing, insulation, probably sprinklers if needed, mech/elec and miscellaneous), and there is some consideration for that, or there should be.
It's more a coordination item than anything else. Plus how the partition deploys can create odd loading conditions that would lean toward a special joist, particularly because of the odd load distribution that may be needed, splits to both sides, all runs to one site, etc. Stacked load at point of storage, etc. If you are going to engage multiple joists then the loads from the transfer/bracing would also show up, and a note that the panel points should align is perhaps a good idea so whatever detailing you put between the joist works out, rather than trying to drag down an adjacent joist with a connection at the top chord without a panel point being there.
The weight of the various models can vary as well, so get a basis of design down early with the architect (if applicable) picking it. The weight isn't typically that much, unless it's a particularly tall partition, but the deflection requirements can be onerous, particularly if you misconstrue the load cases it applies to.
If you end up needing a xxDLH17 or 18 keep in mind the seat depth changes.
Regards,
Brian
P.S. a "separate steel beam" is not the way to go, they are far less stiff than the joist will be due to shallower depth and continuous webs, even with the added deformations from shear. There is also an ASTM on the subject, although it's more about acoustical shorts than anything we get involved in.
See also (wood, but discusses panel load patterns)
milkshakelake
Structural
I did 100' with movable partitions by designing a 60" deep custom joist with steel double angles. Loading was given by others, and they took responsibility for it so I didn't question it. No special deflection criteria on that particular one beyond typical code requirements, though I'd definitely check it for your system. I had to do the connection design as well, because it was a custom item.
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question