Reinforcing ring 1

[SJMoore]

I need an education concerning a reinforcing ring around a a waste water tank. The current design calls for a rolled angle with one toe welded to the outside of the tank wall and one leg vertically down. Common enough design. My question is, what does the vertical leg bring to the party? Wouldn't a flat bar rolled on edge that had an equal or higher section mod. do just as well? I'm not an engineer, just a steel guy trying to understand this.

Thanks

 

[fegenbush]

A common question among steel guys. To start, think of an I-beam. The idea is to have as much material as far from the center as possible. This makes the shape better able to handle a bending load. This is the idea of having the angle welded leg-out to the side of the tank. The shell of the tank acts as one flange, and the leg of the angle acts as the other. The horizontal leg of the angle acts as the web of the new combined shape. This allows the stiffener-tank shell combination to resist the most load with the least material.

If you have any further questions, please do ask.

 

[MJCronin]

I believe that this is a standard detail for an API-650 style stiffening ring - at the top of the shell.

The top of the shell must be stiffened in fragile, flat bottomed tanks to prevent excessive deflection at the roof/shell juncture due to snow or wind loadings.

 

[SJMoore]

Thanks for the explanation, but....

If you use a flat bar you still have the tank wall's contribution to the equation. Then if you use a bar with a high section modulus, but a lighter weight than the angle, don't you achieve the same thing and also save material at the same time?

 

[IFRs]

You can do what you suggest but the flat bar will not be lighter for the same strength. Without the other vertical leg, the contribution of the shell is much reduced because it's distance to the neutral axis is reduced. So, you need more flat bar. Plus, the free end of the flat bar will react to the stress by moving vertically, which equals structural failure. If the flat bar is thick enough, this effect can be mitigated. Again, more steel for the same strength. Keep trying!!!

 

[LittleInch]

A Regen bush says the ideal is an I beam shape. This would require a rolled T bar shape using the tank as one end of the I. T bars are expensive compared to simple angle and more difficult to roll. However the angle is still lots better than just a simple bar at providing rigidity due to the immense additional moment from the far face. Also on top of a tank an angle won't gather water on the upside and rust which aT section would. So an angle is a sensible compromise between the ideal of T bar and a flat bar at 90 degrees to the tank surface.

To understand why a T section is much better than a simple bar, look up how to calculate I, the second moment of area, which is essentially a measure of how stiff a particular section is. Formula exist for standard shapes which can easily demonstrate why particular shapes are better than others.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[IFRs]

I don't think fegenbush was saying to use an I-beam on the tank - cost is only one reason why it would be a bad choice. He was trying to illustrate why a flat bar is not an efficient use of metal. Rolled Tees are not hard to get or roll, being symmetric they twist less. But, they are ill suited to the top of a tank for a number of reasons. API 650 and AWWA D-100 have suggested roof-to-shell configurations that have stood the test of time - pick the least expensive for your situation. Those codes also have formulas that determine the length of shell you can include in the strength calculation.

 

[LittleInch]

I didn't say that an I beam should be used, just that that is the most efficient way to increase rigidity. A T bar may be able to be used, but an I beam on it's side would create a situation whet this creates a pool of water, product, leaves etc and simply is not recommended. An angle is simple and cheap and so long as you weld it toe down, won't collect water and become an area of corrosion.

In essence the simple answer to the op is that weight for weight an angle is simply much more efficient at providing stiffness than a flat bar due to the second moment of area calculation.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[SJMoore]

Ok, you've convinced me that an angle may indeed be the best compromise, but let me give some back ground as to why I'm seeking an alternative.

We've rolled these angles successfully for a customer for about twenty years. Life was all good when everything was just A-36 material. Then along comes material with multiple certification, higher yields and higher tensiles. Now an angle that I may have rolled down to a 3' or 4' radius starts fracturing the heel at 30' and 40' radii.

We've had a discussion with a metalurgist and he confirmed that when the tensiles and yields go up, we lose the ductility. My customer is not wanting to stop using angle, but we can't predict the fracturing from one batch of material to the next. My customer also doesn't want to absorb the added cost to repair all of these fractures, and understands that we can't either. We're just trying to find a way out.

 

[LittleInch]

Glad we could provide you with some info, but your issue is a different question and may be better posted in a differnt forum in metallurgy or something like that. Other than heating it to get you more ductility when you bend it, I don't have the knowledge to help you further or see any easy way around your problem other than using less high strength material.



My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[fegenbush]

There is little benefit to using higher strength materials for a stiffening ring. The common failure mode of a stiffening ring is buckling. Classic Euler buckling is controlled by geometry and modulus of elasticity. Since carbon steels have a fairly consistent modulus of elasticity, higher strength does not add much value to the design.

If your company only does rolling, this puts you in quite a bind. If you can competitively burn and break plate, perhaps that would be the more effecitve solution (think plan view octagon).

 

[dhengr]

SJMoore:
Try grinding a 1/16" - 1/8" radius on the heel of the angle before you roll form it. The fracture is starting at the sharp edge/corner or little nicks in that corner. Don’t use a real course grinding wheel either. Also, a little heat would help, maybe not even heating the whole section, but just the heel area. Play a rose bud torch or ribbon burner on that area to put a few hundred degrees of heat into the heel. Experiment a bit with these ideas. Also, the mill certs. on some angles and bar stock can be kinda loose unless you order it under tighter control. Otherwise, I agree with what’s been said above about an angle being pretty efficient for this detail. Also, a higher strength Fy & Fu to match the tank sheets may not be needed, rather something higher than A36, but not Fy = 50ksi or more.

 

[SJMoore]

dhengr,
Thanks for the suggestions, but my fracturing doesn't occur at the corner of the heel. The angle is 5" x 3 1/2" x 3/8" with the 5" leg rolled in. The fractures occur 3/8" down from the true heel of the angle, straight across from the inside face of the leg. Anytime your rolling angles with the leg in, that is where the most stress is placed.

As to heat; while it might help some, it's just not practical to try and heat 40' lengths.

Pertaining to the chemistry; we tend to do these in lots of 20 to 40 pieces per order. All of the warehouses around us only stock the multiple certified material that we're having the problems with. The straight A-36 only is just not available unless you want to buy the product of a heat at the mill, and that just can't happen. The mills aren't going to run plain A-36 anymore when they can cover a broad spectrum of certs. with the same effort.

 

[IFRs]

Sound like a lack of ductility. Does A-36 have specifications for this?
Does it help to roll it 2 or 3 times, each to a smaller radius?

 

[SJMoore]

IFRs

It is indeed a lack of ductility, due to the elevated yields and tensiles, but we just can't get the A36 material anymore.

Whenever we're rolling a large angle like this, it's always an incremental process that will require several passes.If you try to roll a bunch of angles like this in one pass, they'll be all over the place as far as radius goes.

 

[hokie66]

The law of unintended consequences, often attributed to Murphy. Have you made the AISC Steel Solutions Centre aware of this problem?

 

[SJMoore]

hokie66

I have not. We're not AISC members, so I'm not familiar with a Solutions Centre. How does that work?

 

[hokie66]

SJMoore,
I don't know either and am not in the US, but I found a reference to that resource in this paper which I found by googling.

http://www.modernsteel.com/Uploads/Issues/January_2009/012009_steelwise_martin.pdf