Do Serviceability Loads Need to Match ULS Loads
Do Serviceability Loads Need to Match ULS Loads
(OP)
For a long, long time, I designed stuff like this:
1) Get specified load from code.
2) Design strength based on specified load multiplied by appropriate factor.
3) Do serviceability checks based on specified load without load factor or, in some cases, based on a prescribed reduction factors.
So my question is, how appropriate is step #3? Does the the design serviceability load have to be the specified load?
A talented fellow that I used to work with called me out on this a while back for a stair that wasn't passing deflection/vibration checks. The ULS load was based on 100 psf so I based the deflection checks on that too. My colleague felt that 50 psf was more realistic and that, more importantly, we had the ability to decide that for our selves. Persuasive.
How about that? If you accept the premise, then the lid kinda comes off on a lot of things. Transfer slabs for example. They're usually governed by deflection concerns. Can I just design those for 50% LL when evaluating deflection? Obviously, I don't want to be the only guy in town not doing this if it's legit.
1) Get specified load from code.
2) Design strength based on specified load multiplied by appropriate factor.
3) Do serviceability checks based on specified load without load factor or, in some cases, based on a prescribed reduction factors.
So my question is, how appropriate is step #3? Does the the design serviceability load have to be the specified load?
A talented fellow that I used to work with called me out on this a while back for a stair that wasn't passing deflection/vibration checks. The ULS load was based on 100 psf so I based the deflection checks on that too. My colleague felt that 50 psf was more realistic and that, more importantly, we had the ability to decide that for our selves. Persuasive.
How about that? If you accept the premise, then the lid kinda comes off on a lot of things. Transfer slabs for example. They're usually governed by deflection concerns. Can I just design those for 50% LL when evaluating deflection? Obviously, I don't want to be the only guy in town not doing this if it's legit.
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.






RE: Do Serviceability Loads Need to Match ULS Loads
In principle, the SLS loading should be the maximum anticipated loading, so I'm not sure why you would use a reduced loading for a deflection/vibration check anyway.
Doug Jenkins
Interactive Design Services
http://newtonexcelbach.wordpress.com/
RE: Do Serviceability Loads Need to Match ULS Loads
Eurocode and Australian code base deflection design on a short term load condition which can vary from about .5LL (office/residential) to .9 - 1.0 LL (storage) depending on building/loading type.
In Eurocode crack control is based on various values of Permanent Live Load, Short Term Live Load and Characteristic Live load depending on building/loading type and building materials used.
But remember also that these are minimum values an intelligent designer should higher values if they consider that the real loading will be a higher value for their specific building and situation
RE: Do Serviceability Loads Need to Match ULS Loads
For code definition on SLS, I'm in the opposite boat. Codes are getting out of hand, and I see no reason to define SLS loads/combinations in the code. What happens if this stair has a perceived vibration? Nothing.
RE: Do Serviceability Loads Need to Match ULS Loads
Will throw it out there that for vibration you may want to actually look at a lower load despite what the IBC says. Having the load too high may be unconservative because it adds mass to the system, which helps reduce vibration. AISC Design Guide 11 has much lower live loads than the IBC for vibrations from both footfalls and rhythmic excitation, for instance.
If it's concrete, perhaps rather than playing with loads you play with stiffness instead? You can calculate out Ieff if you wish and for serviceability you can argue that Ieff should be based on unfactored loads.