glulam splitting water damage large commercial hardware building

[bdruehl]

5 1/8 x 18 glulam splitting at hanger bolts due to water damage... must remove entire splitting protion and replace with new glulam... (too far away from post for a cantilvered saddle splice... dont want to break into floor for new foundation directly under splice)

any thoughts as to the best fix?

current solution is to sandwich the splice with steel plates, lag em in with how many it takes to counter moment at the splice point (splice is about at 1/3 point).. splitting problems? bolting into side of glulam issues... load is say 400 # ft and span is 30 feet (splice at say 10)

 

[msquared48]

Are the hanger bolts in a vertical pattern or a horizontal pattern. Vertical can cause splitting too, in addition to the water delamination.

Must be an older glulam as most now are made with a waterproof glue.

Mike McCann
MMC Engineering

 

[bdruehl]

hanger for the glulam that is rotted is on the wall side on masonry with 2 bolts in vertical pattern - but this hanger is not the problem - the new beam will sit in old hanger (the water damage rotted the bottom of the glulam, and in turn the beam settled and crushed and the 2 bolts ((which were at the top of the beam seat hanger and vertical)) propogated nice cracks about 8 feet into the glulam near the top).

The Design issue is the splice point of the old and new glulam beam after the entire portion of the existing glulam that is cracked is removed and replaced.

 

[bdruehl]

.

 

[RARWOOD]

The first thing you should do is make sure that the leak has been fixed so that your new beam does not get wet. Also you may need to replace your hanger. If the bolts are near the top of the support beam you may getting splitting even in a dry condition. You could consult the the AITC Manual for examples of proper detailing and examples of bad detailing. My advice is to add a column or replace the entire beam.

 

[bdruehl]

yes. the leak has been fixed. replacing the entire 30 ft span is not really an option because there are too many roof members being supported by the beam. hanger should be ok. - it has a seat

A column is being added, but ofcourse its only for backup because we dont want to have to prove that the 4" slab can take the load - wont work. SO! the question is really the plate...

as of today... design is for sandwiched 1/4" plate (16"x80") with THIRTY TWO 5/8" thru-bolts to develop the required moment.

additional comments? hasnt anybody every spliced a glulam like this?

thanx for replies!

 

[GT1878]

I think I'd contact an engineer that specializes in glulam design. The complication(for me) is that each layer in the glulam may be a different species of lumber. For example, you may have SYP DSS for the tops and bottom laminations and progressively get to SYP no 3 at the center lamination. If that is the case, you want most of your load development to be in the top and bottom portions of the beam....or you may have to develop the entire load into the top and bottom lamination.
If so, 5/8" holes are going to kill your net section checks....
Just my concerns to try and get this thread going. Its interesting, but I'd go with a new beam....or bust that concrete up and pour a footing to support a new post.

 

[bdruehl]

ARgH!

well, i read last post ... good concerns...

if i am thinking about this correctly, actual load analysis gets pretty hairy involving calculating net section reduction and implications.. but, in my mind i have replaced the missing wood holes with certainly an equivalent strength of (2) 1/4" plates and the bolts?? so, as long as the compression developed by the bolts into the wood does not exceed the allowable for the lams then shouldnt it be ok... !!?? the plates can take the shear im sure. all that has been done is a simple how many bolts does it take about the splice to resist the moment in the beam at the splice....

The acutual loading on the bolts is hard to visualize.. but i think the direction of the loading slowly changes as you start from the splice and go to the end of the plate, as well as as you go up or down the hieght of the beam... at any case i think it is not directly up and down... which might blow out my simple moment calculation....

which brings another question... i have the plate in the middle of the beam... maybe it would be better to have it skewed to the tension side of the beam? does it really matter?!!

off to the overthinking races!!!!!????

 

[bdruehl]

got this job on the 25th! sure is good they put a temporary post in the interim eh?!!

 

[GT1878]

I'm kinda treating it like a boxed beam sort of problem, where the extrme laminations resist all the moment and the interior(web) portions resisting the shear. Just due to the fact that I have no idea what those interior lumber grades are. I'd think this would be conservative???
Get your moment at the splice and resist with a couple from a row of bolts in the extreme laminations? Then your connection from bolts is direct tension/compression(theoretically at least). You'd have to consider group action factor per nds if you go over a 1/4" diameter connector.
Back to the net section checks, your load doesn't fully develop into the splice plates till you get to the inner most row of bolts. So the top and bottom laminations still have to carry alot of the load for the first couple of bolts or so, which you are removing a significant amount of lumber from predrilling the bolt holes.
This would be my best guess. No expert with this sort of problem, I'm just trying to get some talk going.
I'd love to hear what others think about this line of reasoning so don't hesitate to criticize.

 

[bdruehl]

client's gonna come-a-knockin soon!!!

 

[civilperson]

Replace with similar beam, temporary supports for all the intersecting beams are simple even if numerous and expensive. Repairs with steel assumes you know the extent of the failure for attachment and shear strength purposes.