LRFD Design

[SKJ25POL]

I have heard some DOTs push for LRFD design. I am familiar that AISC has Steel Design based on LRFD. What about concrete part? Latest I knew was ACI is based on Ultimate Strength Design, so LRFD design of a bridge means Steel with LRFD and concrete portions with Ultimate Strength???
Just I am a bit confused!!!

Thank you for helping me
Skj

 

[bridgebuster]

All DOT's have to use LRFD for Federally funded projects; that's been in effect since 2008(?). AASHTO LRFD is now the governing bridge design specification, as modified by each DOT. In AASHTO LRFD, concrete design (as well as steel, timber, aluminum) is based on ultimate strength with the appropriate load and resistance factors.

I believe ACI has been using ultimate strength design since 1963. AASHOadopted LFD concrete design in the 1974 spec but retained the ASD specification through the last version of the Standard Specifications. LFD was USD. In college in the 70's we were taught both methods in concrete design.

 

[BridgeSmith]

Generally, current bridge design in the US is in accordance with the AASHTO LRFD Bridge Design Specifications, (it's required for federally-funded bridge projects) with the various design sections fairly similar to the corresponding ACI, AISC and other design specifications for other components of bridges, culverts, retaining walls, etc. and other transportation-related structures.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[SKJ25POL]

bridgebuster (Civil), Thank you for your response. So please correct me if I am wrong, the "Limit State Design/LRFD" that AASHTO prescribes for the concrete design of bridge components is exactly the "Ultimate Strength Design" in ACI?
So I can use the ACI Ultimate Strength design method and claim is AASHTO LRFD?


BridgeSmith (Structural), thank you for your response too. I appreciate you opinion on the questions I asked above too.

Thanks
Skj

 

[bridgebuster]

SKJ25POL - The short answer: No.

While the bridge doesn't know if you used ACI or AASHTO the owner will know. If you're legally bound to use AASHTO then use AASHTO. ACI does have a bridge design spec, 343R-95 but it's not AASHTO. If something goes wrong, even if it's not serious, the owner's lawyer will ask why you didn't follow the nationally recognized bridge code? AASHTO & ACI are both ultimate strength but they're not an apples to apples comparison in terms of design requirements.

Parts of the AASHTO concrete spec are straight out of ACI 318. A few times I've used ACI because AASHTO didn't address particular design issue.

 

[SKJ25POL]

bridgebuster (Civil),
Thank you very much greatly appreciate it.
Skj

 

[bridgebuster]

you're welcome Skj

 

[Guest090822]

SKJ25POL - Most of the steel and concrete capacity calculations in AASHTO are the same as AISC and ACI, but the load factors and limit states are different. AISC and ACI use ASCE7 load combinations, while AASHTO uses their own. As bridgebuster points out, AASHTO is federally mandated if using federal funding. However, in my state we rehabilitate most bridges/structures in whichever code they were originally design in due to incompatibility with mixing codes.

 

[IDS]

Can someone summarise the N American usage of the terms Ultimate Limit State design and Load and Resistance Factor design? My understanding, based purely on reading here, was that LRFD was the usual N American term for ULS design.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[BridgeSmith]

In bridge design, allowable stress design (ASD) was used for most all aspects of design until the late 1970's, I believe. The next approach was termed load factor design (LFD), which started with fairly limited use, but continued to expand in its breadth of use, until it was mostly replaced by load and resistance factor design (LRFD), beginning in the late 1990's. Some areas of design remained ASD until the transition directly to LRFD (sign and signal structures, and the like, for instance).

ASD generally uses unfactored (expected or nominal) loads and provides a factor of safety (FOS) with a single value reduction on the material strength.

LFD, which I think is similar to ULS, uses factored loading, where the expected loads are increased by load factors based on the variability of the source of the load. Material strengths are generally input at their nominal minimums (typically 5% exclusion values, i.e. 95% of specimens exceed the strength).

LRFD, as the name implies, incorporates load factors in similar fashion to LFD, but also incorporates resistance factors that reduce the material strengths, based on the variability in that strength in a particular application (resistance factor for concrete in compression is greater than in shear).

Rod Smith, P.E., The artist formerly known as HotRod10

 

[IDS]

Rod - in my experience (UK, Europe and Aus/NZ), ULS design (which is a sub-set of Limit State design) incorporates factors on both the load and the materials, in the same way as what is known as LRFD in the USA.

In Australian codes there is a single reduction factor applied to the section capacity (which may vary depending on the type of load), in the same way as US codes, whereas the Eurocodes apply partial reduction factors to each material, with a further overall factor to cover uncertainties in the analysis, etc; but both approaches are referred to as Limit State Design.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[BridgeSmith]

Interesting how the terminology differs, IDS. Thanks for setting me straight on ULS.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[bridgebuster]

@THeRick109 - I do a lot of bridge rehab work too; typically we go with the Standard Specifications. Recently on two projects we had to mix and match codes. Gets a bit weird, for example, on both we designed a bunch of pier replacements but the footings were retained. The footings were checked using the Standard Specs and everything else was LRFD.

 

[Guest090822]

Yup - same here. When I first started my career the state I'm in had not adopted LRFD for substructures. We used to design the superstructure in LRFD then switch to ASD for the substructure. It was a pain, but I can say that at least I now know ASD, LFD, and LRFD like second, third and fourth languages! I learned LRFD in college so I actually had to work backwards. I understand why we went LRFD, but I do like ASD better since it was simple and you knew your factors of safety!

 

[bridgebuster]

TheRick109 - my first LRFD project was 4 bridges in 1999. The guys doing the substructures were coming up with a bizarre rebar quantity; maybe 4 times what would be expected. It turned out that the crack control provisions were governing the design. I called one of the guys who wrote the code but he was no help. Then I found a professor at Rutgers who did some work on the code. His response was "well, the substructure provisions haven't been calibrated yet." What I proposed to the client was to design the substructures using LFD but using LRFD loads; they agreed. we ended up with a normal design.

Also back then, there wasn't any consensus on loading rate a bridge using LRFD. About a year or so after we finished the project the client provided us with a methodology.

 

[r13]

with a bizarre rebar quantity; maybe 4 times what would be expected

Well, I don't know how he got that much of steel, and I don't know the way you used to correct the problem, but to my knowledge, the intent of crack control is encouraging the use of smaller steel with a closer spacing to minimize the crack width. Yes, typically the quantity would be higher, but 4 times....! Also, the fs in the calculation is service load stress, I don't know how AASHTO address this.

 

[wiktor]

Bridgebuster - comparing the load rating using ASD and LRFD or LFD is simple - one have a $100 in the left pocket, and move it into right pocket, and out of sudden has $120.... It's giving up on the safety factor - the border line (ASD versus LRFD) is approximately 60'-80' span, DL to LL ratio, and particular code being used - but the trucks (and tanks) are similar worldwide, so all bridge codes are very similar. Typical bridge design should be roughly 1.55 SF for steel, and 1.8- 2.0 SF for concrete. For RC trusses some sources are recommending 2.5 minimum - so the Miami bridge was doomed as specified - LRFD Highway Bridges Specs - 1.5 using Strength V ( not checked), so they ended up with 1.35 as designed, with errors, not even making 1.0.

I had a chance to compare Japanese Code versus AASHTO on the executed design, and found these to be almost identical. Same goes with the Eurocode.
On one positive side the LRFD design provide more efficient design, but complexity of it, and lack of basic understanding what all this factors means, leads to the gross errors, ending up in catastrophes.

The bizarre rebar quantities are the result of the lack of basic understanding of the difference in between crack control reinforcement, and minimum reinforcement required for the element to be considered "reinforced concrete" - typically at the substructure, and on most projects given to the compulsory subs - as these are the less risky elements of the project...

As my old Professor teach me - design a safe structure, and then check it if it complies with the code. It appears that common practice this days, is to comply with the code using computer, without even knowing what's the result.

The best example - the discrepancies in between the models for the Miami Bridge - a simple supported truss, which a student of the fourth semester should be able to solve by hand in few minutes - as it's just the total weight of the bridge and angle of the diagonal.

 

[bridgebuster]

You issue an new edition of the Code, and charge $395.

Early editions of LRFD were focused on superstructure design. In the early 2000's FHWA began offering courses in substructure design.

Wiktor - Let me clarify my remark about the rating. The AASHTO Maintenance Manual addressed load ratings. There wasn't any equivalent LRFD publication.