Elevator Machine Support Deflection Requirements

[bookowski]

I have a condition where my elevators do not have any columns immediately adjacent, i.e. they fall somewhere within a bay (steel framed building). The machine rooms are in a bulkhead that extends 22ft above the roof. See rough sketch attached showing the typical framing and posts for the bulkhead above. the machine roof framing is carrying by 4 posts which land on the typical framing.

My questions/concerns are:
- Looking at static loads and L/1666 deflection limitation, how far do I carry this and how do I apply it. Should I do all beams in the load path until the column as L/1666? That clearly doesn't guarantee any overall deflection limit
- Is there any overall deflection limit?
- Is the main concern a differential movement related to machine operation or is there also some component that is concerned with overall elevator movement that would be felt by the occupant (i.e. if all my beams collectively deflect 1" even though each is L/1666)
- At 22' the bulkhead has some wind on it, do I apply an l/1666 criteria for those supporting beams in combo with static elevator loads?

The columns were omitted here due to some existing conditions and the inability to place footings in the area.

The beams are huge to satisfy what I think is a safe interpretation of the above - I'm wondering if I'm being overly conservative.

I've also considered adding corner columns for 1 or 2 stories only to assist in deflection.

 

[canwesteng]

Shouldn't the vendor provide some absolute deflection limit, not an L/x value? Deflection as a ratio of span makes sense for finishes, occupant comfort, or crane rails, but I'm not sure how it applies here. Isn't the L/1666 a lateral deflection thing with elevator shafts?

 

[bookowski]

I've never seen an absolute value and the L/1666 is typical, it's from asme. It says sheave beams and "immediate" supports - not sure how far to take that.

2.9.5 Allowable Deflections of Machinery and Sheave Beams and Their Supports
The allowable deflections of machinery and sheave beams and their immediate supports under static load shall not exceed It1666 of the span.

 

[jike]

Why can't the 4 columns be carried all the way down to the footing level by burying them in the shaft walls? I assume there is a pit for the elevators with a mat foundation underneath. Why can't this support the 4 columns?

What am I missing?

 

[bookowski]

No columns because of some existing foundation issues in this area and there is no below grade pit, the elevator stops before the lowest cellar level and the pit is hung. Anyway it seems crazy to always put 4 columns around the elevator only for deflection control at the top story - for a 15 story building it can't make sense to install 700 linear feet of columns w/ splices, beam connections etc only for this reason.

Anyway my main question was if anyone knows what drives this criteria and how it should be applied. I can do it with large beams - I just don't want to get carried away.

 

[KootK]

Anyway my main question was if anyone knows what drives this criteria and how it should be applied.

When I dug into this previously, all I came up with was that the original code committee intent had been lost and was known as only hoping to "produce a stiff set of beams". I know, useless.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[bookowski]

Yeah I found the same language. Not much help. I'd guess that the main idea is to limit differential deflection at the machines themselves to ensure proper/smooth operation.

I can't get anything out of an elevator vendor. The requirement doesn't even usually appear on their shop drawings, it's more that if you ask 'does this need l/1666' then they say yes but they don't seem to know any more than that.

 

[calvinandhobbes10]

Very good question and one I doubt you'll find a direct answer for.
I would bet that L/1666 predates machineroomless elevators, like it sounds yours is.

I'd also expect that for sheave beam static loads, a good chunk of this load gets applied while they are aligning the elevator and then stays there. This makes me believe the L/1666 would be intended for any incidental deflection due to impact/dynamic/passengers/live-load/etc during operation, which is typically much easier to accommodate.
This may be an angle the elevator vendor could understand and comment on. If he can't, the vendor might have a manufacturer with an engineering department you can try.

That said, I think your best bet will be to find creative ways to add redundancy cost-effectively.
For example, if you do the corner "columns" at say the top 4 floors, you can distribute the sheave load by stiffness to the additional floor beams below. If these beams are composite you'll get good help from their composite-section stiffness.
Share the load enough, don't overkill the floor live load (I like ASCE 7-10 Table C4-2), and I think you can provide a good solution. I'd get my overall L/1666 from the long-axis (~18') of the hoistway.

If you get any push back about adding columns or upping beams, they will reconsider if you float the idea of a giant ugly bulkhead---35' x 24' X 22' high CMU/concrete bulkhead walls that span to the four on-grid columns (by the way, do these have footings?)