Peer-to-Peer Standards Exchange

Alex Thomas posted 02-08-2024 12:54 PM

For the four legs of a table on a RCC slab , shall we sure about whether the legs be over Re bar or on concrete alone. Live or dead the load is taking by slab. What's the action of the slab on punching shear. I think it may take a whole year to adjusting the reaction of the slab. It's hardening more and more. 

No difference in type of loading.

Alex Thomas BSc FIE MASCE 

Stephan Demers posted 02-27-2024 04:03 PM

Thank you Alex for the response. Is there anyone else from ASCE that can provide a response to my two questions above?

Alex Thomas posted 02-28-2024 02:47 AM

Neutral Axis will be shifting with respect to live load that is the beam or slab shall be comes under reinforced or over reinforced. For design LL is converted to equivalent DL. I think it is enough for decision. 

Alex Thomas BSc AMIE civ FIE MASCE CEng 

Ronald Hamburger posted 02-28-2024 08:50 AM

Stephen

I believe there are a number of reasons that server racks should be considered live, as opposed to dead load.  First, they are subject to change in configuration from year to year and from tenant to tenant, as new technology evolves and tenant needs change, resulting in chnage of weight.  Second, their weight, even when first installed, has uncertainty in magnitude similar to that of other live loads.  Finally, as buidling uses change, over its life, the servers may not even be present, and therefore, should not be relied on for resistance of overturning demands.

Please note that although I am past chair of the ASCE 7 standard, the above are my personal opinions, not official positions of ASCE.  ASCE does maintain a process by which such questions can submitted for official interprertations.  The process typically takes several months to complete as it requires developing a committee consensus.  You can find more information on the ASCE 7 page on teh ASCE web site.

Stephan Demers posted 02-29-2024 09:05 PM

Ronald, thank you for the response. Agreed that if those unknowns exist then live load is the appropriate classification. For data center projects, SEs work closely with the owner to establish the loading criteria since ASCE7 does not provide a prescribed load for server racks. Specifically, projects establish a maximum server rack weight and tightly control their locations, and include these sustained loads in the seismic mass. Overturning considers both the presence and removal of these racks following ASCE 7, Section 3.1.3. With these provisions in mind, footnote 'a' in Table C4.3-1 notes that for computer equipment, the actual weight should be considered and that this fixed service equipment is treated as a dead load.

 

Ultimately the question I'd like to answer - is whether something in ASCE7 would prohibit server racks from being considered as dead load on a project-specific basis?

Christian Parker posted 03-10-2024 06:54 PM

Hello Stephan,

I don't think I have a satisfying answer, but I tend to agree with the direction you're heading: a project-specific agreement with your owner and AHJ.  A safe but conservative approach would be to consider the weight as a sustained SDL for seismic demands but use a LL factor on the gravity system.  Less conservative approaches may be admissible, but your client should know what they're getting, i.e. a smaller margin for future changes than the code provides for typical buildings.  As I understand it, data center clients can be very stringent on design criteria; they may balk at the idea that swapping in heavier equipment could result in overstress.  Or they may not.

It's also worth giving some thought to what kind of change could occur within the life of the building that wouldn't trigger reanalysis under IEBC.  We have half a century of parking garage decks designed for lower live load than a full complement of EVs could impose, and that's in the relatively slow-to-change automotive industry.  I would have to really think before hanging my hat on an assumption that the needs of tech industry facilities won't change within a 50-year design life.

For a more informed opinion than mine, this article is on design criteria for data centers and has insights relevant to this discussion:

https://www.structuremag.org/?p=22372

Stephan Demers posted 03-20-2024 01:30 PM

Thanks Christian for the response. In regards to IEBC triggers for reanalysis, it shouldn't change if the load is classified as dead load or live load. For example: if you design for 2000lb racks - the 5% IEBC triggers means you can put in 2100lb racks without reanalysis - whether the rack is live or dead load.

 

Agreed that treating racks as live load and sustained SDL for seismic demands is a conservative approach. However, classifying racks as dead load can be rational for a data center project where the owner is tightly controlling the location and weight of all equipment. SEs work with the owner to establish a maximum loading allowance that allows for future flexibility as technology evolves in data centers.

 

If anyone knows of language in ASCE7 that prohibits server racks from being considered as dead load on a project-specific basis, please comment. From the conversation above, my understanding is that treating server racks as dead load is not prohibited by code.

Ronald Hamburger posted 03-21-2024 10:22 AM

ASCE 7 does not prohibit classificaton of server racks as Dead Load and could be interpretted as specifically permitting it.  ASCE 7 Chapter 2 defines "Dead Load" as "Dead loads consist of the weight of all materials of construction incorporated into the building including, but not limited to, walls, floors, roofs, ceilings, stairways, built-in partitions, finishes, cladding, and other similarly incorporated architectural and structural items, and the weight of fixed service equipment, including cranes and material handling systems."  Servier racks could easilty be classified as "fixed service equipment"  As in all things, design, the engineer of record has the authority to make this determimation subject to approval of the building official.

Christian Parker posted 03-30-2024 08:19 PM

Regarding IEBC, I was actually referring to the change of occupancy and classification of work.  I realize that I did not make this clear; my mistake!  The crux is that there are a lot of Level 1 alterations with no change of occupancy where the project doesn't even have a structural engineer on board.  A well meaning architect or O&M team might note that there is no fundamental change in the use of the space and not think to have someone check the framing.  They don't hire an engineer when moving a filing cabinet, and the distinction between code-prescriptive live load and bespoke equipment weights might not be apparent to non-engineers.  The extra 60% on live load doesn't guarantee that future occupants don't exceed the design load, but it means when they do they really have to push it by a large margin to get anywhere near failure.  I've struggled with the sometimes excessive conservatism of floor loads in ASCE 7, but in a way it reflects the unpredictability of human behavior and future modifications compared to natural hazards.

But no, I am not aware of any prohibition in the code on classifying data center loads as dead.