Live load reduction on a transfer girder

[chrislsnider]

I have a transfer girder that is supporting 4 floors of load-bearing walls. It seems to me that this still would still be something that we could reduce the live loads on, but how should we consider the Area that this transfer girder is supporting for the reduction calculation?

Say for example that we have 40 psf residential live load. Tributary width onto the load bearing walls for all stories above is 50 ft. Our span is 30 ft. What is the area we can reduce the live load by? Conservatively I would say just 50x30 = 1500 sq ft, but is that too small?

Or am I completely off base?

 

[AELLC]

You are already almost to your maximum floor live reduction area anyway, but did you calculate the usual dead and live maximum reductions? ie ratio dead to floor live?

 

[AELLC]

Back to OP, I always take maximum reduction possible. It is more realistic, and wood has a large FoS anyway - comfort level.

 

[AELLC]

How can 1500 be too small if you have no more than 1500 sf to begin with, like 10 point something ft trib x 4 floors for a total of 50'?

 

[Archie264]

Allowable live load reduction is listed in Section 4.8 of ASCE 7. It should be fairly straightforward to chase it out. That said I'm not sure I would want to use it in the situation you've shown even if allowed. There's a lot going on in the system shown and I would be concerned about deflection if nothing else. In fact, I'd be inclined to bump the loads up a bit, even if it is only residential loading. Designed to take only 40 psf I'm betting it would be a very flexible system.

But, having said that, is it really a 50' tributary width? That would imply 50' to the next supports in each direction, which is a very long (read "bouncy") span. Are you sure it's not 25' in each direction and therefore 25' tributary width? Just asking...;-)

 

[AELLC]

Archie,

All you do is keep a close watch on the deflection.

 

[Archie264]

I guess so. It still makes me a bit nervous though.

 

[AELLC]

To me it is easy-peasy. I deal with loads easily up to 20K in custom residential. I use old-fashioned ASD design methods, and all the old (alternate method) load combination and reduction methods, but my calcs are highly automated on Excel.

 

[chrislsnider]

I was trying to get in my head around being able to count the trib of the 4 floors above (+ roof maybe) just to get the area higher. And, now that I look at my example, I am way out of scale for my exact conditions. I was just looking for round numbers to put into the example and messed it up. AELLC - you're right, that example gives reduced L to be around 25.5. So reality looks like this per floor:

Trib width = 24'
span = 23'
At = 552 sq ft
Lo = 40 psf
KLL = 1.0 (all other members not identified)

L = Lo (.25 + 15 / ((KLL * At)^.5)
= 40 (.25 + 15 / ((1.0 * 552)^.5)

L = 35.5 psf

However, if I can count on the other floors contributing area, then the reduced L is around 23 psf.

AELLC - does that change your answer?

 

[chrislsnider]

And Archie - you are right. This isn't a pretty situation necessarily, but it is the problem I have to solve. For what it's worth, the spans on either side of this wall are 19' on one side and 24' on the other. As you surmised, this is residential construction, so I'm looking (currently) at a pretty beefy glulam under these walls. Again, not ideal, but it's what I've got in front of me at this moment.

 

[AELLC]

@chris

LOL, I don't even use that method of live reduction. I have used the older method prior to that, which is still legal.

Either way if you have BOTH Lr and L, you can reduce basic load by a percentage, up front.

If you have S, I don't know.

The old method of reducing L is R=0.8(A-150) for floor. Also governed by maximum R=23.1(1 + D/L)

Also R=0.40 max for one level only, and R=0.60 max for members or posts receiving load from more than one level.

 

[AELLC]

23 psf is still within the upper bound limiter 0.60 in my above post, for more than one level. Except I did not check the math on R=23.1(1 + D/L).

 

[chrislsnider]

Gotcha - thanks for the help. I think I may have found my comfort level...

 

[AELLC]

Well, good.

Just be careful if you have a beam carrying 3000 sf of L and 1 sf of Lr, do not reduce L and Lr by 25%.

LOL

 

[chrislsnider]

Yeah - that'd be a bit excessive for sure.

 

[SAIL3]

I would be enclined to agree with Archie on this one for the following reason. I always try and consider the redundency of the structure in my design especially when it comes to live load reduction which relies on a reasonable probability that a max live load will not exist on the entire area at the same time.All well and good for a typical structure, but, when there is a single key load path in the structure then I adjust my reductions accordingly to address the importance of that member in the integrity of the structure as awhole. From the OP's post and the info supplied, it looks like that girder would be one of those cases, IMO.

 

[AELLC]

SAIL, that is you policy, but it varies widely according to company or engineer.

Any hi-rise with the usual transfer girder made of steel or concrete is going to support 100's of kips, yet be designed with maximum live reduction - it is an economic necessity in most cases..

On the other hand, I was directed to design a steel OMF 11-story with a full 100 psf on each floor, zero reduction. That was the developers' criteria.

 

[WillisV]

A transfer girder qualifies as a member supporting two or more floors and is therefore subject to up to a 60% LL reduction per ASCE 7-05 4.8.1 depending on the total tributary areas of all floors supported.