AASHTO LRFD Bridge Design Specs, 9th Edition 2

[OSUCivlEng]

PDF download is available now. Hard copies begin shipping in June. I believe AASHTO has went to a 3 year cycle with no interims.

https://store.transportation.org/item/collectiondetail/202

 

[BridgeSmith]

Anything is better than how they did the interims for the last few editions, where they put the interims at the end instead of incorporating them. I got so annoyed having look 3 different places to be sure I was using the correct provisions - look in the original version, then look at the next year's interim to see if they change it, then look at the interim from the year after that to see if changed it in that one. Ugh!

I really wish they'd split the pages with the spec at the top and the commentary at the bottom, so they could eliminate much of the white space by just dividing it wherever they needed to - One line of spec and 4 paragraphs of commentary? Then the dividing line goes right near the top; if it's the opposite, the line is at the bottom. We could maybe get back to 2 binders that are only 4" thick instead of 6".

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

 

[Guest090822]

Can't wait to see what fundamental engineering concepts no longer apply!

 

[r13]

Yeah, your bridge design is outdated every three years

 

[BridgeSmith]

Can't wait to see what fundamental engineering concepts no longer apply!

It's not really like that. There are new ways to analyze some components, like concrete shear, for instance, but we never really had a "fundamental" understanding of how concrete reacts to shear stresses anyway, so the strut & tie model is as good or better at approximating the real behavior as previous models.

Yeah, your bridge design is outdated every three years

If sure feels that way, but it's always been like that. For the most part, the newer versions are making the designs more precise and less conservative by being more detailed.

The glaring exception in the last 2 versions was much simplified and much more conservative bolted field splice design, which according to the writer of the previous provisions, was already massively conservative. I had several conversations with Dr. Ibrahim about the splice design when I was reviewing and helping rewrite WYDOT's BRASS Splice design software.



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

 

[MIKE_311]

It really seems like AASHTO is really embracing computer analysis and design. The design code has become so precise because its become trivial for a computer to rip through the code checks. The problem is that it's become a chore to make sure the software is doing the proper code checks and tracking down when tings aren't correct. I remember being a bridge engineer not doing software QC and filing bug reports...

rant over.

 

[bridgebuster]

I have all the AASHO/AASHTO's from 53 onward (44 and 49 grew legs), anyway the 53 is 5"x8"x about 3/4" thick. 8th Edition printed is 8 1/2x11x4" thick; a bit overwhelming to say the least. Thankfully, they put in those flow charts.


@MIKE311 - Around 2000 several of us did a 27-hour LRFD course, quite boring, anyway John Kluicki, who was one of the original writers, brought up your point. He said because of computers we can add all of these checks. Sometime I wonder if it's all necessary.

 

[Guest090822]

We can’t help ourselves as engineers. We go and invent computers to speed up designs based on an easy, reliable code; then turn around and complicate the code so we are right back where we started.

As long as structural engineering has unpredictable materials like concrete and soil, nothing will be “precise”. I can guarantee that 90% of the bride decks that called for 3,000 psi concrete were actually supplied with 4,000-6,000 (contactors don’t want to risk not meeting strength requirements) but the computer says it’s “precise”. Half of those “precise”
designs wouldn’t pass the ductility checks using the actual concrete strength, but again, it’s more “precise” (sarcasm).

I hope I’m around to see this 75-year life LRFD promises. 75 years in Arizona maybe, but you aren’t going to get 75 years in the Northeast!

 

[bridgebuster]

Some food for thought...

"Structural engineering is the art of molding materials we don't wholly understand, into shapes we can't fully analyze, so as to withstand forces we can't really assess, in such a way that the community at large has no reason to suspect the extent of our ignorance." ...Jim Amrhein

 

[MIKE_311]

I think what bothers me most about all this is the expectation of computer help. Black box effect aside, in my experience software design only works for structures that can be analyzed by line girder or using distribution factors. 3D FEM program only do AASHTO design checks, working for a state (like PennDOT) that has a supplementary bridge spec, makes software checks a chore and you end up doing it by hand anyway or trying to simplify the analysis so you can use a software solution, which defeats the purpose of the refining nature of the code!

 

[BridgeSmith]

...makes software checks a chore and you end up doing it by hand anyway or trying to simplify the analysis so you can use a software solution, which defeats the purpose of the refining nature of the code!

The supposed reason for refining the code to make it more detailed, is to narrow down the magnitude of the loading effects and more closely approximate the capacity of the system. It doesn't always get accomplished, and even when it is accomplished, whether the savings justify the increased design burden is certainly debatable.

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

 

[OSUCivlEng]

Around 2000 several of us did a 27-hour LRFD course, quite boring, anyway John Kluicki, who was one of the original writers, brought up your point. He said because of computers we can add all of these checks. Sometime I wonder if it's all necessary.

About 10 years ago a now coworker of mine was designing what was supposed to be a cable stayed pedestrian bridge over an Interstate. The DOT brought Kulicki in from Modjeski and Masters as a third party reviewer. In subsequent discussions my coworker had with Kulicki he stated something to the effect of "did you check it with 2D statics"? My coworker came to the conclusion that was how Kulicki had done all of his design. I don't think he ever used computer modeling.

Interestingly enough he is the only non DOT/government agency person mentioned in the AASHTO LRFD bridge specs. For most run of the mill bridges we have learned some things about fatigue and serviceability, but the basic design concepts have not changed.

The additions may be justified for highly complex bridges, but we are increasingly running the risk of blindly trusting what a black box spits out.

 

[BridgeSmith]

The additions may be justified for highly complex bridges, but we are increasingly running the risk of blindly trusting what a black box spits out.

It appears that was a central part of the failure of the Miami/FIU pedestrian bridge collapse - multiple black boxes were used, and weren't verified with hand calcs.


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

 

[Settingsun]

Surely there are >75 year old bridges all across the country? What is the context of the 75 years: basis for determining return periods for environmental loads, or maintenance-free period?

(100 is standard in Australia and 120 was/is? in the UK.)

 

[MIKE_311]

Surely there are >75 year old bridges all across the country? What is the context of the 75 years: basis for determining return periods for environmental loads, or maintenance-free period?

Interesting question, one i don't know the answer to. I do know some agencies in the states will increase the lifespan to 100yrs.

 

[OSUCivlEng]

Design life should not be confused with service life. The design life of 75 years has nothing to do with how long the bridge will actually last. It's about how accurate the load the bridge was designed for will be in 75 years.

Straight from the definitions in Chapter 1 of AASHTO LRFD:
Design Life - Period of time on which the statistical derivation of transient loads is based: 75 years for these Specifications.

Commentary of Section 1.3.2.1:
The Strength I Limit State in the AASHTO LRFD Design Specifications has been calibrated for a target reliability index of 3.5 with a corresponding probability of exceedance of 2.0E-04 during the 75-year design life of the bridge. This 75-year reliability is equivalent to an annual probability of exceedance of 2. 7E-06 with a corresponding annual target reliability index of 4.6. Similar calibration efforts for the Service Limit States are underway.

 

[BridgeSmith]

Design life should not be confused with service life. The design life of 75 years has nothing to do with how long the bridge will actually last. It's about how accurate the load the bridge was designed for will be in 75 years.

Not quite. Design life is the length of time used to determine how many cycles of loading at various levels the bridge is expected to see while it's in service. For instance the number of cycles of fatigue level loading takes the number of trucks of a certain size (75% of the HL-93 truck) that the bridge is expected to carry every day, multiplied by 365 day and multiplied by 75 years.

Service life is how long the bridge is expected to last before major components fail.

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

 

[OSUCivlEng]

BridgeSmith you are referring to the Fatigue Design Life.

The load and resistance factors in the AASHTO LRFD specifications are based on statistical models. 75 years was chosen as the return period and the live load factors were calibrated based on that. Therefore the maximum live load effect the bridge is designed for is the maximum expected load in a 75 year period.

You can read more about it here: Link

Agree with you on the service life. Things may wear out faster or slower depending on many factors.

 

[NateBloss]

Is anyone aware of a good summary of the changes incorporated in the 2020 AASHTO LRFD Bridge Design Specification?

Thanks,
Nate Bloss, PE

 

[bridgebuster]

Here's some info

on concrete design
Link

from AASHTO; not much. You would think they'd have a comprehensive list.
Link