h/t_w for Channels

[HWDstorm]

AISC has an Excel file (AISC Shapes Database) on their website with shape properties from the steel manual. In the "Readme" section of the file it states that "the database contains some additional section properties that are not included in the Manual." One such property is h/t_w for channels. I'm having trouble figuring out exactly how they calculate this value in the spreadsheet. For wide flanges, h = d - 2*k_des. The database also provides a k_des value for channels (which isn't given in the manual either) but this isn't matching either. How is h/t_w calculated for channels?

 

[JoshPlumSE]

Let me just put this in terms of an actual shape. C15x50..... Per the AISC table:

d = 15.0"
k_des = 1.44"
tw = 0.716"
h/tw = 17.7

Whereas (15.0-2*1.44)/0.716 = 16.9 which is 4.4% off from the value given for the h/tw shown in the table.

It looks to me like they might be using the tw_det number for the calculation (11/16 = 0.6875") for this case.

(15.0-2*1.44)/0.6875 = 17.63 which is 0.4% off from the value given for the h/tw show in the table.

Honestly, I'm not fully confident that this is really what they're doing. But, it's my best guess after a quick run through the numbers.

 

[nutte]

In the spreadsheet, the kdes values are the same as the kdet values. This seems odd. I would expect kdes to be less than kdet, like it is for wide flanges. I don't expect that channels are rolled to such a tight tolerance and consistency that the minimum and maximum expected values of k are the same. So I wonder if AISC has more accurate kdes values they are using to calculate h, and thus h/tw, that they for some reason aren't reporting in the spreadsheet.

 

[HWDstorm]

Josh, Nutte,

Thanks for taking a look. I still can't see where it comes from. I looked closer at using h = d-2*k_des to calculate h/t_w. The error ranges from ~4 to 15 percent - generally increasing as the channel size goes down. I tried plugging in tw_det too. In most cases, the calculated values are not very close.

I think nutte is right; they are providing values based on information that is not reported in the spreadsheet.

 

[EDub24]

This is a bit late but my understanding is that all the section property values given in the Maunal are averages based off actual values taken from the top steel producers so I wouldn’t be surprised if the h/tw is the same. That could explain why it’s not exact.