Steel I Beam under Glulam 1

[gfounds]

Dealing with a 20' span in basement. Existing 6x10 glulam circa about 1970 spans 20' with one support post in the middle. Need to remove the middle support post and span the entire distance. A W8x28 steel i beam would do the trick,but, due to existing joists, plumbing, electrical, ductwork, etc., I am thinking about simply placing the steel beam directly under the old glulam and attaching with bolts and supporting on each end with steel posts. Has anyone ever dealt with a similar situation/approach? Seems much simpler than ripping out existing beam, joists, plumbing, etc., and we can spare the loss of headroom. Thoughts?

 

[MegaStructures]

I don't see any reason why this wouldn't work. Steel beam will be stiffer than glulam and should support it across the entire length. I assume the foundation is more than sufficient for the two posts.

 

[gfounds]

Thanks, MegaStructures. My thoughts also. There will be proper footers for the posts, so I don't see why this wouldn't work. I've just never done that before myself. Now, it might be a trick to get it trimmed out so that the beam(s) are covered well and look good, but should be doable.

 

[Craig_H]

I would anticipate that the glulam has likely crept a bit and will no longer be perfectly planar on its bottom surface. This may impose some challenges when connecting the two together. It may be wise to plan for bearing plates on top of the new steel beam, to act as support points for the glulam. This would keep load transfer predictable. The glulam would be a 3-span beam with cantilevers both sides, and the steel beam would be exposed to two-point loading. That approach might rely on the glulam being EX grade.

How slender is the glulam? How about the slderness of the new steel beam? I might worry a bit about LTB of the combined section, or the relative LTB of the two beams (hinging at the interface between them).

 

[MegaStructures]

Interesting Craig_H, clearly I don't work with wood often, or I would have thought of the sag. perhaps the glulam can be straightened during the installation of the steel beam with the jack posts? Would the section really be combined and be able to buckle as a composite shape? I guess it depends how they are connected, but I'm not picturing much moment being transferred between the two.

Edit: I see now that you mentioned there might be hinging between the two, which I suspect would be the case.

 

[gfounds]

Craig H - good points. I will consider the bearing plates idea.

On the issue of LTB, the steel beam would be about 8" x .285" x 6.5". I think that the number and spacing of the attachment points to the glulam would help control to some extent. A 6"x.32"x6" would also be sufficient and might reduce the likelihood of LTB. The precise measurements of the glulam are 6.75" wide and 10.5" deep. I should also point out that the glulam continues for another 10' past one endpoint of the planned end of the steel beam. If the glulam and steel beam are securely connected, I think that the joists on each side of the glulam and the complete structure of the floor system will help control LTB.

I am not too concerned about this issue, but I will give it some more thought, and any suggestions for additional control would be appreciated.

 

[MegaStructures]

A 6"x.32"x6" would also be sufficient and might reduce the likelihood of LTB

I trust that you know the exact LTB strength of your steel beam

 

[dik]

Are the joists flush framed to the glulam to provide lateral restraint? Be prepared to have some cracking.

Dik

 

[dhengr]

Gfounds:
There may actually be some advantage to sistering two large LVL members, one each side of the 6x10 glulam beam, assuming the fl. jsts. rest atop the glulam, another question you haven’t answered. The glulam and the LVL’s take the load on the new span in proportion to their relative bending stiffness and are attached (screwed and glued) together to transfer the load accordingly. Since the fl. jsts. are already working and applying the fl. loads to the glulam, leave them alone. Set the top of the LVL’s 3/8” or ½” below the top of the glulam, so the fl. jsts. still rest on the glulam, and a slight joint movement doesn’t change this. You will likely regain some headroom. Jack up the glulam near the existing col. and at the mid spans to lift it off the col. and start to remove the glulam mid-span deflections. You actually want to jack a bit higher (a slight camber in the glulam). Watch and listen upstairs for any movement, sht. rk. cracking and the like during this jacking operation, and do this jacking slowly to allow the structure to adjust to the movement. Then apply the LVL’s, screw them off to the glulam and support them at the two ends. Then release the jacks.

Stretch a line below the existing glulam beam to see how it has deflected due to load and creep over time, and study this to understand the beam action. Also, note that with the existing three span cont. beam condition which comes to light late in the doling out of important engineering design info., you might get some uplift on the untouched back span when you releases the jacks on the 20’ span. You have to study this beam system as the three span condition which exists now and the two span condition which you want to achieve, and see more exactly what actually happens.

 

[gfounds]

Dik, the joists are 2x6 but span only 7 feet. A bit undersized for my liking, but within acceptable span range and there appears to be very little flex/bounce. These rest on a notch in the glulam, back slightly less than 2”. No hangers, just nailed (ommon practice for 1969). There appears to have been nearly zero lateral movement or twist to any of the joists over time. They are generally in excellent condition.

Regardless, I will likely rip lvls to fit the space and sister each joist for good measure.

 

[structSU10]

Does the glulam only support one floor, or is there a bearing wall on it? how will you handle to construction condition when the post has to be gone before the steel beam is in? If there is no wall it's pretty easy with shoring walls on each side, but if there is a beam there might not be a safe setup to carry that temporary condition so having LVLS or steel channel on either side, attach it, then remove the post to eliminate shoring may be a better option.

 

[gfounds]

structSU10, I have wrestled with the same "timing" issue for a while now and came up with the same solution that you suggested for the temporary support while the steel beam is lifted into place. There is a wall above, so I want to ere on the side over overbuilding temp supports.

The plan is to have a temp wall on each side of the existing glulam, about a foot or two back on each side to allow working room. Then, jack post(s) near the center of the glulam to remove any sag that might be present currently. Then attach 1/2" x 6" steel plates on each side of the glulam with bolts. Then remove the existing post and asap lift the i beam and bolt to the glulam. The side plates and temp walls should provide enough support to be safe while the post is out and beam is lifted into place. Thoughts?

 

[SlideRuleEra]

Dealing with a 20' span in basement.
Need to remove the middle support post and span the entire distance.
A W8x28 steel i beam would do the trick...
I am thinking about simply placing the steel beam directly under the old glulam...
Thoughts?

gfounds - You need to take a step back from your proposal and look at the entire existing 30' glulam, just as dhengr suggested. Removing one support from a 3-equal span-continuous beam and turning it into a 2-unequal span-continuous beam causes profound, counter intuitive changes that affect the supports, moment, stress, and deflection.

A question: Can a 20' beam weighing over 500 lb. be taken to the basement?

BTW, despite what software may "say". Any W8 beam with a 20' span has a span-to-depth ratio of 30, (240"/8"). Sure, the beam will "work" (for both allowable stress and deflection) but the supported floor will be noticeably "springy" (especially compared to how it was before the modification).

http://www.SlideRuleEra.net