Return to index: [Subject] [Thread] [Date] [Author]

Shape properties: AISC or individual shape producers?

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Title: Shape properties: AISC or individual shape producers?

I've asked our legal counsel to review the issue of what legal rights AISC has in the protection of the data in the AISC Manual and Database (shape dimensions and properties). I'm only doing this to attempt to answer the factual question Bill Polhemus and others have raised. I'll report back what I find out when I find it out.

In the mean time, for your information, below is the "technical" synopsis I gave him.


A question has been raised as to whether AISC can hold and/or enforce a copyright on the shape dimensions and properties in the AISC Manual and AISC Database. Specifically, it has been asserted that this information is available from producers directly and other sources. Also, it has been asserted that the data can be calculated independently using equations and procedure s that are in the public domain -- or simply typed from the AISC printed text into a spreadsheet or similar program by hand.

It would be possible to obtain all shape information directly from a producer, at least if you only want information on the shapes that producer rolls. In many cases, the information they provide is obtained from AISC and reprinted by them with our permission. A few producers actually create their own data, but follow our procedures. A few do whatever they want to do.

As far as the AISC-published data is concerned, I believe there is a unique value that AISC adds to the data we produce for the Manual and the AISC Database. For simplicity, my explanation will be based upon W-shapes. A similar process applies to other shapes.

To create the data that goes into the AISC Manual and AISC Database, we collect physical dimensions from the mills as follows: depth, flange width, web thickness and flange thickness. This information is provided by us along with the cross-sectional area (determined using the properly considered fillet radius as explained later) to ASTM for inclusion in ASTM A6/A6M.

In addition to the dimensional information above, we perform a periodic survey of producers to determine the industry-wide maximum and minimum fillet radii that they use in the production of their shapes. The ranges we determine consider both producer-specific variations (i.e., how much do they let the rolls wear before they replace or redress them) and inter-producer variations (i.e., variations in what radius producer A and B use in their mill practices). We include Ameristeel, Bayou Steel Corporation, Corus (formerly known as British Steel), J&L Structural Inc., North Star Steel, Northwestern Steel and Wire Company, Nucor Steel, Nucor-Yamato Steel Company, Roanoke Electric Steel Company, SMI Steel Inc., TradeARBED and TXI Chaparral Steel Company in this survey. We are the only entity with which all those producers share this information. In fact, when they do share it with us, they insist upon anonymity and confidentiality as they each consider their radius practices to be proprietary. Said more simply, they do not want to tell each other how they make their shapes.

We then calculate the properties and dimensions we provide as follows. Design properties are based upon the smallest fillet radius applicable for a given shape. Detailing dimensions are based upon the largest fillet radius applicable for a given shape. In this way, no matter what mill you buy your steel from, the design properties will not be overstated, and the detailing dimensions, with proper detailing, will ensure proper fit-up.

There's been talk of reverse engineering and alternative engineering of the numbers -- even typing them all in on one's own time. But in the end, I believe that any solution intended to regenerate the AISC numbers would at some point require the use of information that AISC builds into the shape dimensions and properties it publishes. The design properties could be closely estimated by ignoring the fillet radii entirely in the calculations, but if this were done, the data would not conform to the dimensions and properties as they are standardized by AISC. I don't think the detailing dimensions could be recreated at all.