Malone's Irregular Diaphragms

[MILRAD]

I have to design my first irregular diaphragm and have been going through Terry Malone's example in "The Analysis of Irregular Shaped Diaphragms" from WoodWorks. (Attached)

I think I am getting it. Seems like a series of FBD's and superimposing them(?).

The thing I don't get is where the 20' width of the transfer diaphragm came from. He doesn't mention it and there is nothing in the plan to suggest it has to be 20'. It seems like it could be anything from 0 (which can't work) and 80' or maybe 40' where the shear is 0. It seems like the width changes the plf results.

What am I missing? Is there a rule of thumb? Do you play around with the distance to get the final shear in plf under the capacity of a typical nail spacing (6", 4", 3", 2")? Malone gets a max shear of 400 plf and assuming allowable and other values looks like a 4" spacing is needed.

Thanks

 

[bones206]

I think it's dictated by where you can conveniently locate a chord/collector line. So maybe in that example there are roof beams along grid 2 that could be used for chords/collectors.

 

[JoshPlumSE]

I had the same impression as Bones.

FWIW, I tend to think Malone over does it a bit with his calculations. Don't get me wrong, I learned a lot from his presentations and his book. But, once I understand the basic mechanics and the way to configure the load path, I think the calculations can be greatly simplified over what he does.

 

[bones206]

A while ago I tried benchmarking the transfer diaphragm examples from the book using FEA and semi-rigid diaphragms. The results did not agree very well with the hand calcs.

 

[JoshPlumSE]

With Malone's Hand calcs being overly conservative. Right? I did some of the same with his shear wall calculations.

 

[bones206]

It was almost 2 years ago, so I don't remember the details of the results. I just remember that it was difficult to see any correlations between the FEA version and the hand calcs.

 

[STpipe]

Funny enough Bones, I tried the same thing last week. I didn't get a chance to delve too deeply into the 'why' with the lack of correlation, but what I found in the example I looked at was the model exhibited the general behaviour and mimicked the distribution of forces, however the magnitudes and deflections were off significantly.

 

[lexpatrie]

It's like FTAO, folks, the analysis produces safe design but isn't all that literal versus say testing or FEA. Look at some of those papers, you're satisfying statics here.

Evaluation of Force Transfer Around Openings – Experimental and Analytical Studies, APA M410, 2011.

As to the 20' it's a working dimension, it could be a different value and you analyze for that.

 

[MILRAD]

Thanks for the answers. I plowed through the design. As typical the real world problem was much more involved than the simple example.

Is there a consensus that the Malone approach is conservative compared with FEA?

 

[lexpatrie]

I don't think a blanket statement can be made like that. These aren't exactly energy methods, but I'm not aware of any attempts to compare FEA to testing, there's been attempts to compare the mechanics-based approaches on FTAO (Diekmann, etc.), but there's spread in those results where they are sometimes conservative, sometimes not, which would put them in the category of energy methods because they don't follow the lower bound theorem.

https://lsc-pagepro.mydigitalpublic...662539&article_id=3690977&view=articleBrowser

 

[JoshPlumSE]

I've made some attempts to compare the Mechanics-based FTAO hand calcs to an FEM based approach back when I worked with RISA. I think it worked pretty well. It felt a good bit more rational and reliable than the hand calc methods.

One of these days, I really need to do a proper write up of the procedures I came up with (that the current RISA method is based on).