Wind loading spreadsheets

[rn14]

I was wanting to add a bit more automation to some of my wind loading (envelope method) spreadsheets but I REALLY don't feel like doing all the data entry for the Ps30 tables. Has anyone done this and willing to share?

 

[dhengr]

To the best of my knowledge, there is no bldg. code requirement or PE licensure requirement that you must ‘wind load your spreadsheets.’

 

[CELinOttawa]

In my opinion that type of automation is dangerous and should be avoided.

I have a number of spreadsheets that do these calcs for you (sorry, none suitable for US practice), buy without exception, they require the user to transpose the answer from one block to another input.

Easiest thing the world to further automate with a cell reference, but I won't. I want to see the number BEFORE anyone (including me!) uses the results in a mad rush to get something out the door.

Black boxes are bad, even when we programmed them ourselves.

 

[rn14]

...appreciate the help, thanks.

CELinOttawa, I hear what you are saying but the method I am referring to is essentially a table lookup exercise in which case I think a spreadsheet is less likely to result in a mistake (although Enercalc actually has a mistake in there version of the same thing).

 

[medeek]

Instead of using the table lookup I've programmed a web app that calculates the design pressures (Envelope Procedure, Ch. 28 Part 1):

http://design.medeek.com/resources/wind/wind_calculator.pl

Unfortunately, I rarely use this calculator since most of the structures I've been looking at recently are not rectangular in plan.

For that I have created this spreadsheet (Directional Procedure, Ch. 27 Part 1):

http://design.medeek.com/resources/wind/WIND_ANALYTIC_MWFRS.xls

However, this spreadsheet specifically deals with enclosed and partially enclosed buildings (Fig. 27.4-1) or sections 27.4.1 and 27.4.2 of ASCE 7-10. All of the exceptions and footnotes of Fig. 27.4-1 have made this spreadsheet hard to program but for the most part I think I've got it right. One caveat is that I have not programmed it to deal with roofs with slopes greater than 80 degrees.





A confused student is a good student.

 

[medeek]

I agree that I don't like the black box mentality. I've been using Woodworks software recently to calculate my shearwalls since I can't match it in performance or the shear number of scenarios it analyzes, at least not in a reasonable amount of time. However, if there is a bug in the software I am at its mercy, it just makes me uneasy to trust it so completely or any software for that matter.

Case in point, on a recent job I setup a residential model in Woodworks and the roof lines of two blocks was slightly overlapping. I had a hunch that the software would not like this area of the model however I was trying to create as realistic a model as possible. When I looked at the loads they seemed a bit high to me, so I then compared the diaphragm and shear loads to a similar structure with all of the same design criteria within the software. Sure enough the loads were about 40% higher. I removed the offending geometry from the model and it then brought the loads into the expected range. My feel for what the loads should generally be was vital to getting the right answer in this particular case. Had I blindly followed the software I would have significantly over designed the shearwalls.

For wind loads calcs I think it is important to at least go through the full procedure a few times to understand how the numbers are arrived at, however on a practical level the calculations should be either tabulated or programmed into a spreadsheet and then manually checked to verify their accuracy. With my calculators I have checked them against different sample problems found in ASCE publications as well as more recently against the Woodworks software. Even so I cannot guarantee they are without error or my method for computing overhang forces (envelope procedure) may differ slightly from other software.

A confused student is a good student.

 

[jrbaus]

medeek - nice calcs on your website. Very impressive.

For what it is worth, I have found the avoiding vlookup functions in your spreadsheets greatly improves their accuracy since they rely on the data being sorted. Instead, use index(XXX,match(XXX,XXX,0)) which will work regardless of the data sorting assuming only 1 unique keys exists. Maybe there is a nicer way to do it, but this method hasn't let me down yet.

Jake

 

[rn14]

medeek - Ive looked at your calcs a couple times now and Im always impressed. One of these days Id love to pick your brain a bit about your report generation.