LRFD Wood Design

[Perception]

I am a faculty member that teaches wood design as part of an advanced structural design course.

In the past I've primarily taught ASD with a short introduction to LRFD wood design. My understanding has been LRFD wood design is not commonly used in practice given the extra factors and with the reference design values being in terms of stress already. I'm curious, is this still the case? For those that do regular wood design, how often are you utilizing LRFD vs ASD?

 

[dik]

It likely depends on the jurisdiction. LRFD or the equivalent has been used for wood in Canada for the last couple of decades. In 1965, our engineering class was taught Limit States for steel and concrete; we were the first year on that program. Other than for checking 'stuff', I've not used Working Stress Design in over 50 years.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[ChorasDen]

ASD is the prominent mode of analysis for wood design in the US. I very rarely see analysis considered via LRFD. With that said, AWC recently hosted a webinar that detailed CLT shear wall design, and their example was solved using LRFD analysis, because the equivalent lateral force was established at a strength level, and they probably wanted to simplify their consideration for wall strength performance and deflection calculations by keeping the lateral forces at strength level, rather than switching between strength/ASD loading.

 

[Lomarandil]

I still see ASD used as the default (in the US). More often so when the timber design interacts with other elements (foundations, manufactured products) that are also called out in ASD, and less so when interacting with elements (concrete structures, bridges) that are designed by LRFD at strength level.

 

[Aesur]

I agree with ChorasDen, rarely is LRFD used for wood when doing by hand design, however I do find myself switching back and forth when using software like EnerCalc as occasionally one versus the other provides better results. Loading in the software is typically entered at ultimate level now and therefore switching back and forth is as simple as pressing a buttom many times, factors are automatically applied to combinations now days in many software applications.

 

[sauce_man]

Much of the mass timber research, fabrication and connector products are coming out of Europe and Canada. And mass timber is only gaining steam. I am almost 100% LRFD for wood design now. There is effectively little difference in the calculation methods, but I envision NDS having to move toward LRFD to catch up with the rest of the world they are way behind.

 

[271828]

I teach wood design in one of my classes also. All of my real-life experience has been in ASD. The design criteria (f vs F) looks like old-school ASD. I don't see any reason to make the subject even a tiny bit more complicated. One or two extra factors -- forget it.

If the AWC wants LRFD to be used, then the NDS needs to be like the AISC Specification, with there being little or no difference between the two. Or just go LRFD (fat chance). LOL

 

[ChorasDen]

One thing to note, is that I have been hearing from my connections inside the ASCE 7 committees that there is a push to eliminate ASD load combinations in chapter 2 of ASCE 7 going forward. I expect these ideas to get shot down by the appropriate design committee, but interesting to hear that there are limits trying to be made by some players at the code level.

 

[phamENG]

I've never met anyone that uses LRFD for wood. Only way it'll happen is if reference design values are published at the strength level.

Wood structures also tend to be lighter, which means that ASD designed structures will tend to be a little more on the lean side as the effective factor of safety tends to be lower for variable load dominated structures using ASD.

The purest in me loves the statistical uniformity of LRFD/Limit States, but the inertia of standard practice is a bit too much for me to change my day-to-day just yet.

 

[JoshPlumSE]

I still tend to use ASD for wood. Though it's been a couple years since I've had a good wood project. Those are the tables and capacities I'm more familiar with. That being said, it's pretty much the same code now. So, it's just a matter of which factors and which loads I use.

I've switched over to LRFD for everything else. Even masonry. I imagine that I would switch over for wood design as well. I just haven't done much wood design lately. I'd have to make sure I the shear wall capacity tables are all available in LRFD. Same for hold downs, Simpson ties, Truss-joists and such. But, not much is keeping me attached to ASD.

 

[phamENG]

Josh - good call on Simpson.

The allowable loads published in this catalog are for use when utilizing the traditional Allowable Stress Design methodology. A method for using Load and Resistance Factor Design (LRFD) for wood has been published in ASTM D5457...When designing with LRFD, reference lateral resistances must be used. Contact Simpson Strong-Tie for reference lateral resistances of products listed in this catalog. For more information, refer to the 2015 NDS Appendix N, which contains a conversion procedure that can be used to derive LRFD capacities.

This produces a pretty complicated back and forth of conversions and manipulations that are prone to error and not necessary if one chooses to stick to ASD.

Shear wall nailing is irrelevant, though - those tables are methodologically neutral and require a factor of safety for ASD (v/2.0) or a resistance factor for LRFD (0.80v).

 

[CrabbyT]

I've never used LRFD for wood design, but I use LRFD for pretty much everything else.

 

[Perception]

Interesting discussions. I'm going to stick to teaching ASD with a small intro to LRRD. My wood design examples are by hand so I don't have the issue of trying to utilize the results with something else that is LRFD.

This discussion reminds me of when I worked at a firm doing steel connection design. My boss who was an old ASD guy brought up the point that all they did was move the area from one side of the equation to the other (P instead of P/A). Also, it made little since to have the strength reduction factor be multiplicative while using division for the safety factor. Make it so you either multiply or divide by both sets of factors.

 

[dik]

I'm not big on wood, but usually the Standards, in transition, recognise WSD and USD. Looking at my latest wood code (CSA O86-14, there are newer editions, there doesn't seem to be a WSD approach). I'll check with my wood guy to see what's up...

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[XR250]

For me it comes down to speed and therefore my bottom line. Most of my calcs are by hand and I usually check deflection first on longish members as that is generally what controls. Since I already have the demand in ASD, I might as well use it for the capacity check.

 

[phamENG]

My boss who was an old ASD guy brought up the point that all they did was move the area from one side of the equation to the other (P instead of P/A)

This is valid for Allowable Stress Design vs. the newer Allowable Strength Design used in steel design...but it's fundamentally incorrect for ASD (either stress or strength) compared to LRFD. LRFD is a completely different statistical model for comparing demand and capacity.

 

[SJBombero]

LRFD is a completely different statistical model for comparing demand and capacity.

True, but fundamentally you still comparing stress by means of safety factors. The only real difference is that the safety factors used for LRFD are based on a different statistical model. The AISC Design Guide for torsion for example, shows you how to compare demand stress to capacity stress with the various phi factor for each type of stress. Then use the sum of the ratios to determine if the section is adequate.

 

[phamENG]

True, but fundamentally you still comparing stress by means of safety factors.

Depends on the material. For wood - yes, 100%. We use the same allowable stresses and just modify that end to work with LRFD load cases.

Steel, though, it's not the same thing anymore. For stable sections, the old AStressD set a hard and fast limit on stress. For compact sections, it was 0.66F[sub]y[/sub]. In other words, the factor of safety was applied to the onset of yielding. With the new(er) AStrengthD and LRFD, we look at a capacity based not on the onset of yielding but at the formation of a plastic hinge and then back it off from there (either with a factor of safety for ASD or a resistance factor from LRFD). If you look at that moment capacity and back out the actual stress in the material (f=M/S[sub]x[/sub]), you'll find that it varies. Perhaps not widely, but it's not the same. For some sections the section will still be partially yielded, for others we'll still be short of M[sub]y[/sub]. Generally speaking, it'll vary based on the shape factor or ratio of Z[sub]x[/sub]:S[sub]x[/sub], which is usually between 1.05 and 1.2 for a wide flange. Yes, there remains a relationship between allowable capacity and stress, but it's a correlation rather than a causation.

 

[dik]

Just heard from my wood guy... ASD hasn't been in Canadian Codes for years.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[MotorCity]

Its very interesting to hear the regional practice of using LRFD vs ASD.

I have never used (or worked with anyone who used) LRFD for wood or masonry.
It seems that for our neighbors to the north its just the opposite.
Who knows, maybe the US will convert to LRFD once we convert to metric