WoodWorks Sizer - Built Up Column Design

[phamENG]

Bought Wood Works sizer and decided to test it out by spot checking some stuff I've done for built up columns. I was shocked to find that Sizer thinks all of my columns are undersized. I dug into a bit and discovered that the d/6 eccentricity (when the resulting moment is about the axis parallel to the plies) gets split up to each ply proportional to area, not applied to the section as a whole. In other words, it seems to ignore shear flow for the purposes of bending.

In one particular case, I have fc=256psi. For e/6, fb should also equal 256psi. But the program tells me I have 763psi fb. Well for (3)2x4, d=4.5" and A=15.75" so P=4032lbs, M=3024lb-in. For Sx=11.8125in3 for the full section, that works out to 256psi. If I divide that moment by three, I get 1008lb-in. For an Sx of 1.3125 (a single ply in flatwise bending), fb=768psi (0.65% error - close enough). So I'm confident I know what it's doing.

The question: why? Is there a rule that I've been ignorant of that says the connecting fasteners for built up columns can be used to augment Fc' but not contribute to shear flow for Fb'? Is the program just being hyper conservative? If so, is there a way to change it? (I have a tech support ticket in, but it's late in the afternoon on a Friday and I don't expect to hear back until next week.)

 

[Celt83]

yeah I'm seeing the same behavior it's using a single ply width in place of d in the M/s formula, they show the formula in the additional info field of the report.

I'm not aware of any provisions that say to do this either.

I'd guess this may be a bug and it's taking b from the main member definition and not the built-up section in this formula.

I'm making a thing:

http://www.thestructuraltoolbox.com

(It's no Kootware and it will probably break but it's alive!)

 

[skeletron]

It's a quirky program for sure. Admittedly, I've never looked into this or noticed the issue. I would be interested to hear what the tech support ticket comes up.

I think there is an option to put a custom eccentricity in the loads menu.

 

[phamENG]

Thank you both - glad to know I'm not crazy (or responsible for a bunch of undersized columns!)

Celt - thanks for pointing that out. I'm not impressed by the report format and I go cross eyed by the time I get that far down. I'm also curious how they're running the Eq. 15.4-3 check. I've been able to duplicate everything in the report by hand except for that result. Never mind. Was able to duplicate that one now.

skeletron - I'll let you know what they say.

 

[jayrod12]

Are there many instances where you're dealing with weak axis bending from eccentricity? In my experience the bulk of the eccentricity happens in strong axis. Maybe I'm wrong in that assumption.

 

[phamENG]

jayrod - the weak axis of the plies is often not the weak axis of the column as a whole. If I have 4 2x6s in a stud pack supporting a beam over a hall way that's in line with the wall, I'll be loading the column in strong axis bending but if try to distribute it to individual plies they'll be bending in their weak axis. If the beam is perpendicular to a wall, the column will be in weak axis bending but the individual plies will be bending about their strong axis. I have both in almost every house I design.

 

[jayrod12]

That's a fair point. I was really only envisioning beam perpendicular to wall. But you're completely right about beam in-line with wall.

 

[KootK]

It's a bit of a conundrum for a software vendor in that there is this dichotomy:

START

1) We know that conventional fasteners exhibit too much slip for us to be able to make truly composite sections of them. This, I believe is why Woodworks has made these two choices:

a) Multi-ply beams are not considered as composite for the purpose of calculating lateral torsional buckling resistance.

b) Multi-ply stud packs are not considered as composite when it comes to resisting a concentrated moment at the top.

2) NDS provisions do allow a partial degree of composite action to be considered for buckling checks when using conventional fasteners.

END

I can partially justify this as follows:

START

3) Bucking is a secondary effect rather than a primary effect like an directly applied, eccentric load. So the demand is less.

4) Buckling is mostly about flex at post mid-span. As such, you've got significant lengths of crap fastening available to you above and below the critical section. Conversely, to get composite resistance to an applied end moment, you'd need a zillion fasteners right at the end of the stud pack which, of course, isn't what we do.

END

In practice, I feel that this is often resolved by the bracing effect of the wall sheathing, intended or otherwise.

 

[KootK]

So, then, the interesting question becomes how should d/6 be applied parallel to the wall? I suggest this which would have the following advantages:

1) No assumption of questionable composite behavior for moments applied to the ends.

2) Buckling benefit from partially compososite mid-span flexural resistance accrued.

3) Still a reasonable penalty for accidental eccentricity.

This is similar, yet subtly different, from an applied end moment.

 

[phamENG]

Interesting. So you're suggesting that rather than induce a flexural stress and a compressive stress, we should consider a uniform compressive stress in each ply that varies linearly from one side of the stud pack to the other? So then what should the effective unbraced length of each ply be?

 

[phamENG]

This would be a good paper to have:

Malhorta, S.K.; Sukumar, A.P. 1989. A simplified procedure
for built-up wood compression members. Annual confer-
ence. St. John’s, Newfoundland: Canadian Society for Civil
Engineering: 1–18 (June).

A quick google search didn't turn up a freely available copy. Anyone have it lying around?

 

[phamENG]

I feel the darkness of this rabbit hole enveloping me...

Nail Laminated Column Design - ASAE EP 559

 

[Celt83]

phamENG:
some more references for you:
J.A. Scholten, Built-Up Wood Columns Conserve Lumber, Engineering News-Record, pp. 333-335 (August 27, 1931)
H.Y. Rassam and J.R. Goodman, Building Behavior of Layered Wood Columns, Wood Science, Vol. 2, No. 4, pp. 238-246 (1970)
S.K. Malhotra and D.B. Van Dyer, Rational Approach to the Design of Built-up Timber Columns, Wood Science, Vol.9, No.4, pp. 174-186

I recently got a copy of "Wood: Engineering Design Concepts",AbeBooks. I bought it for the module by Diekmann on Diaphragms but it also has a good module on columns. The column module doesn't specifically address this condition but speaks briefly to built-up columns and notes the above references.

On a long shot I've emailed AWC to see if they can provide copies of the above, if they do I'll follow up here.

Look familiar KootK:

I'm making a thing:

http://www.thestructuraltoolbox.com

(It's no Kootware and it will probably break but it's alive!)

 

[KootK]

Look familiar KootK

Quite... that will need to become CeltWare at some point in the new future.

 

[Celt83]

Pham: did you ever hear back from the WoodWorks folks on this one?

I'm making a thing:

http://www.thestructuraltoolbox.com

(It's no Kootware and it will probably break but it's alive!)

 

[phamENG]

They did eventually get back to me (took almost a week), but by the time they did I had moved on and had a bunch of deadlines stacked up. Thanks for the reminder. I'll follow up this week and see if we can get to the bottom of it.

 

[skeletron]

Was there any follow-up discovery from this @phamENG? I'm going through the same process of examining some post calculations and trying to correlate to what I thought was correct hand calcs. It does seem to consider each ply separately for bending, Peuler, etc. Makes me wish the calc report was a little more thorough.

 

[phamENG]

Right. "this week" turned into "next week" turned into a long forgotten, very red item on my outlook to-do list. Sorry.

I put together my inquiry and sent it this afternoon. I'll let you know what they say.

 

[Mike Mike]

d/6 eccentricity is not a code requirement, and is probably too harsh for this situation. if your studs are braced by wall sheathing, don't apply the eccentricity.

Small correction, woodworks does give the user the option to use the full cross-section of multi-ply beams for lateral torsional resistance. I would imagine it has a similar check-box for columns.

Koot K. are you suggesting applying a different axial load and no moment to each ply? Seems reasonable. What should the effective unbraced length of each ply be, pretending we didn't have any wall sheathing?

Woodworks applying a fraction of the moment to each ply makes no sense from a mechanics standpoint, but at least it's wrong in the conservative direction

 

[skeletron]

@Mike Mike:
I was going to contend with your post (call me a recreational opposition-ist or a PITA), but I had to look it up for myself. A couple of threads have discussed this:
Hall-of-fame thread: thread507-435497
Also a good thread: thread507-456187 (e/6 is equal to the axial compressive stress)

My version of WoodWorks (2020 Canadian) allows for checking of the Lateral Stability factor (KL) based on either full member width or single ply width. From my recent review, this would rarely be an issue unless the post was free-standing. In that case, I would probably default to a solid post, so the difference is moot for my practical purposes.

I do think the eccentricity is a good thing to check in most situations. I think WoodWorks does the moment-to-each-ply method because there isn't a lot of info in the code (CSA O86) specifically dealing with weak-axis moments. Even if you run a header in Forte, the program won't analyze the weak axis load effect for built-up beams.