Continous LVL beam

[woodman1967]

I have been given a project by a local modular home company. They have been using 9.25" LVL for the intermediate floor support.

Basically, the floor is built in two sections (left side and right side). The rimrail that is supported by the exterior concrete wall is constructed with 2"x10". The rimrail where the two floor sections come in contact is constructed with a 9.25" LVL. Once the floors are in place the LVL becomes the beam supporting the floor sections down the middle of the floor. This now 2 ply LVL beam will be supported by support columns.

Most homes are 48' or less. The maximum length the LVL is delivered is 48'.

I've been asked to design a splice detail so that the LVL will continue to act as a continuous beam but be able to work on homes with floors longer than 48'. Up to now they have been splicing the LVL over support columns but they would like to have more flexibilty. They would like a splicing detail so the splice can be located anywhere within a span.

The main problem is that most homes are only a few feet longer than 48' so the splice would be located within the first or last span of the continuous beam. Anyway, I want to locate the splice within the span at the point where the moment is zero so the connection would be shear only but would require the LVL to be cut to length, this is something they want to avoid.

Anyway, do any of you see a problem with locating the splice at the point the moment is zero (approx. at the quarter span) and design a splice using shear only? The splice material will be LVL as well.

Thanks for your help and hope you understand my request.

 

[COEngineeer]

You should only do splicing at moment 0 when your DL takes a lot of the loading. How do you know that the liveload will be distributed along the beam? My answer is, design it as simply supported at each support. Problem solved!

Never, but never question engineer's judgement

 

[MiketheEngineer]

Making a wood moment connection is very hard. Splice it over the supports.

The reason they only come 48'long is because they are delivered by truck. They can make them 1000 miles long if wanted. Train delivery - if available - would let you go to 80'. Pretty much same idea with I-beams.

 

[JAE]

If you design a splice somewhere off an interior support, and your splice is a shear splice only (allows a bit of rotation through the splice with no moment transfer) then you've effectively forced a pin at that location, regardless of where a continuous LVL would want to place zero moment.

You essentially are forcing a pin (zero moment) point along the span. As long as you design your LVL's to support the resulting moment gradient along the span, then you should be OK.

 

[COEngineeer]

I knew an engineer who did this shear splice at 0 moment. The continuous steel beam with splice at inflection point supports living room and a big balcony. There was a lot of snow so the beam was only loaded on one side, the beam had upward deflection and they cant operate their sliding door. I think he got his inflection point with all the snow and live load working together.

Never, but never question engineer's judgement

 

[FishingFrank]

There are LVL manufacturers that make and ship LVL's longer than 48ft. Making a moment splice in wood for an application where it will be subject to the stresses of a mobile modular structure would be tough.

 

[mark1234]

Modular companies need to stop thinking about flexibility and get their construction suitable to adequately support a multi-story home.

Alot of fundamental problems with modular construction currently.

MDJ

http://www.windspeedbyzip.com

http://www.groundsnowbyzip.com

 

[kslee1000]

I wouldn't bother to locate zero moment location. Sure you may if you wish. But I would try something below:

If you plan to have more than 2 columns, then follow JAE's suggestion - place a pin within a span (Ex. For 3 columns, 1st span = 25', second span = 3' cantilever + 22' simple span, total length = 50'). You should try couple different arrangement to get a feel (better to figure out maximum cantilever length first). Keep an eye on deflection, as it likely to be the governing factor.

 

[MaddEngineer]

Just a thought, could you not offset the adjacent runs,

First run 48' + 12' = 60'

Second run 12' + 48' = 60'

Third run 48' + 12' = 60'

etc...

And just specify a lapped splice in the LVL's. It is assumed the column spacing will be between 8' to 12' so the lap should occur close to the support of the continuous run, where moment is reduced (approaches zero). The eccentricity on the single continuous LVL should be able to be restrained by the floor joists. You could also specify some adhesive between the LVL's to enhance composite action. If analysis shows more capacity is needed I would provide metal straps along the top and bottom.