Web Crippling Question 1

[psmi8905]

All,

I'm a mechanical engineer at a steel mill and today I was asked to design a rack which will hold steel coils (see the attached picture). The coil rack shown in the picture is in use at a sister mill whose coils weigh only 50,000 lbs, whereas our coils weigh up to 90,000 lbs. As you can see, the beams are resting on the ground. I'm at a loss as to how to calculate the required beam size to support these coils. The only stipulation is that the overall height of the beam has to be 14", so it'll need to be a W 14 X (???). Another issue is that our coil outer diameters range from 56" all the way up to 77". Of course, a 56" O.D. coil will not weigh 90 kip, but they are substantially heavy as well (around 48,000 lbs). I think it would be most practical to forego web stiffeners and just weld a plate to the outside of the flanges on both sides of the beam; basically box in the I-beam. With that being said, do you guys have any advice for calculating the required beam section? Keep in mind that the weight of the coil will be shared over two I-beams, just as the attached picture shows.

-PS

 

[Hokie93]

I recommend you start with a beam on elastic foundation analysis to determine the internal shears and moments. Once a W14 is selected, checking for local effects such as web local yielding and web crippling would be the next step. The local effects to be checked are covered in Chapter J10 of AISC 360-05.

 

[JedClampett]

Side note that doesn't have much structural analysis value. W14's have more sizes (weights) available than any other depths. The heavier ones have a much larger depth than the lighter ones (up to 18 or 20 inches). So if you're trying to hold to a 14 inch depth, make sure you check the specific W14 dimensions.

 

[BAretired]

1) The near beam in the photo does not appear to be continuously supported on the grade. Is it? If not, what is the span?
2) Why is it important to maintain a 14" height provided the top elevation is maintained?
3) Are the two beams tied together periodically?
4) Is the soil adequate to carry the load without footings or other foundation? What are the properties of the soil?

BA

 

[dhengr]

Psmi8905:
Web crippling is only one of several problems which you must resolve for a best solution to your problem. Regarding web crippling, you can look at the AISC Manual and ‘allowable loads on beams,’ and ‘concentrated loads on a beam,’ pg. 2-30 in their green book, 9th Ed; and pg. 5-80, ‘Special Design Considerations, Local Web Yielding & Web Crippling.’ You certainly do not want to need any web stiffeners, you would not know where to put them, for starters. Thus, the web thickness and the ‘k’ values will influence your selection. It would seem that you need some spacing system to keep the two WF members aligned and spaced w.r.t. each other. It would also be neat if you could provide some sort of wooden chalks for each coil. It will treat your coils much better, and they are likely to be needed for shipment anyway, but I’m sure your management or yard people will complain about that extra feature or step in the process, let the transport guys worry about that. Hokie93 is right that these pairs of beams are ‘beams on elastic foundations,’ and in that respect you want a good uniform soil base to set these lines of beams on. It should be uniform and consistent, well compacted and well drained. There will be about 90kips on 6.5', on a pair of beams with some flg. width. Isn’t the real ‘stipulation’ that the beams all be the same general depth in any line of beams, although W14's may still be the best choice becuase for a given weight they offer a good flg./web/’k’ value for your type of concentrated loading, they are stocky w.r.t. the/your worst loading condition? Keep the beam sizes within your companies’ stock, and the least expensive grade, there is probably not much need for high strength unless its cheap too.

 

[BAretired]

psmi,

For web crippling, you need to know the length of bearing. When a circular coil rests on a hard surface, the length of bearing is theoretically zero but in practice the coil tends to flatten a bit. I like the idea of adding a continuous compressible material on top of each beam, perhaps wood chalks as dhengr suggested or neoprene pads which may have a longer life than wood and will spread the load over a calculable length of bearing if the properties of the neoprene are known.

BA

 

[psmi8905]

Thank you all for you replies. I've attached a picture of an unloaded coil rack, so please check it out.

Answers to specific questions:

BARetired:
1. Yes, both beams are continuously support by the ground.
2. This isn't a major concern for us, but we tend to have various W 14 beams on-site, and we are trying to see if our on-site stuff would suffice.
3. The beams in the picture I provided in my initial post are only tied together at the ends. Keep in mind that the example I've shown is at a sister mill and we will make all necessary modifications to our rack design so it will perform as well as possible. If that means tying the beams together more frequently throughout their length, then I have no problem doing that.
4. I have no geotechnical background whatsoever, but the coils currently rest on slag (a byproduct of steel making, so it's free to us) which has been pulverized into a gravel-like form. We'd like the new coil racks to rest on this slag as well.

dhengr & BARetired:
Ideally we would use a chock between coils, but it's not practical for us. Our customers order steel by the ton and specify a width and gage; the coil length changes to meet the theoretical volume of steel needed to meet the ordered tonnage. With that being said, because each coil's length can be unique, each coil's outer diameter can also be unique. A chock system would force us to place coils in pre-defined spaces, which could lead to wasted space on a coil rack that was chocked for the worst case (largest) diameter coils. An unchocked rack allows us to place several coils of varying diameters on the same rack, maximizing available storage space.

A continuous compressible material on top of each beam is also not feasible since the coils come out of the rolling mill between 900F-1100F. It goes without saying that any rubber or wooden "padding" would not stand up to those temperatures. Also, let me say that we have used other steel rack designs in the past for individual coils and the temperature didn't seem to affect the integrity of the rack itself.

-PS

 

[BAretired]

I think, at least as a first attempt, I would use a very simplified analysis for bending and neglect the elastic foundation. It can't be any worse than a concentrated load P on a beam with length D with uniform soil load acting upward. Dimension D is equivalent to the diameter of the coil.

For a coil weighing 90,000# and a diameter of 77" (6.4') the maximum moment would be 90/2 * 6.4/4 = 72'k for the pair of beams carrying the coil. The top flange should be considered laterally unbraced unless additional braces are provided.

The average soil pressure with coils stacked continuously in series would be 90/(2*6.4*b) where b is the flange width. This could be higher than the subgrade can withstand and needs to be confirmed by geotechnical testing.

For web crippling, I would follow the code using a conservative value for bearing length. This could be determined by direct measurement of the flat bearing length of a coil supported on a pair of beams.

BA