Large openings in steel tanks 3

[Mboudreau]

Hi,
I would like to detail a 6' wide by 7' high opening at the base of a liquid storage tank (65' diameter by 45' high). I found guidelines in API 650 for the design of manhole with diameters of 36" and less. Where could I find guidelines for larger openings?

Thanks,

 

[aggman]

I design bulk storage tanks so your requirements may be different. We always just design a frame using an effective width of shell wall and typically angles to transfer the hoop tension and vertical compression around the hole. I think this is a pretty crude "approximation" of what is really happening which can only really be defined using finite elements. I would not resort to FEA for the solution unless I was struggling to make something reasonable work. John R. Buzek's book "Useful Information on the Design of Steel Bins and Silos" from AISI has some information on it also.

 

[theonlynamenottaken]

I asked our resident ME tank guru about this one....

From a code standpoint API makes no accomodations for openings/nozzles over 36" diameter other than "door sheets" which are covered under API 650. As aggman said you'll have to address the vertical compression and hoop tension, but you may have to also consider out-of-plane bending if whatever you're attaching to the tank shell has an eccentric center of gravity (use WRC 107 for that -

http://foreng1.securesites.net/cgi-bin/wrcbulletin/bulletin.pl

) This would apply if you were attaching a large flange to the shell.

As far as reinforcement, you'd probably want to use the "full area replacement" method that is common to this issue. It basically indicates that you must replace the total removed cross-sectional wall area at the vertical level of max calculated stress with an equal or greater cross-sectional area of reinforcement.

 

[JStephen]

Consider whether the tank really needs to comply with API-650. To put it simply, you can't put a 6'x7' opening in a tank and still have an API-650 tank. If local building codes or your insurance or other rules or regulations require an API-650 tank, you won't be able to comply. API-650 includes flush-type shell manways up to 48"x48", and prohibits anything larger than that.

From an engineering standpoint, it's just a bad idea all the way around. The main structural component of the tank is the shell, acting in hoop tension. You're proposing to cut out an entire course of shell plate, then try to reinforce what's left to handle the load. While it can be done to some extent, you're on your own as to how you figure the stresses and reinforcement required, and the required details and welding to make a suitable tank. Unlike a normal API-650 tank, you can't rely on past history of similar tanks to show that your product is going to be adequate.

Suppose you build a suspension bridge, and then decide that the cable needs to be cut at some point and offset to go around an obstacle. That is the situation you have here, on a smaller scale.

It is much better, if necessary, to put large openings in the roof.

On the reference to WRC-107 up there, I don't think it covers the thickness ranges you'd need, even if otherwise applicable.

 

[plasgears]

What do the aeros do when they design a baggage door opening? The opening is reinforced in the adjacent ribs and longerons, the door is reinforced in shear and framing, and they make provision for shear reinforcement in the surrounding panels.

Aggman has his finger on it.

 

[IFRs]

Airframes have vastly different stresses. A tank has it's extreme stresses where the shell meets the bottom. Failure of this seam, or any other part of the tank is catastrophic (not to minimize the failure consequences of an airplane). The airplane does not have the discontinuity a tank does - it is a cylinder under much different loading. It is also fabricated extremely carefully using special methods and materials that pass stringent QA/QC. Aircraft are analyzed to the max, and load tested to failure. Tanks are not and their design rules are conservative and by and large based on and proven by sucessful experience. I just don't see how you can absorb the hoop stress with a frame. There is also bottom settlement, bottom flex and the consideration that the materials and assembly is not aircraft quality - it's just welded steel. Bototm line: be careful!! One suggestion - leave 12" of tank shell at the bottom of the opening.

 

[JStephen]

"What do the aeros do when they design a baggage door opening?" I was under the impression that they analyzed the heck out of it with FEA, and have still had a few failures along the way.

I'm not saying it can't be done- I've actually done such a door, only somewhat smaller- but am saying it's just not a good idea and should be avoided in the first place. At some point, you have to decide whether to adapt the tank design to suit the operational requirements, or to adjust the operational requirements to suit the tank design.

Note that large doors are not uncommon in elevated tank pedestals- but that's mainly compression loading. And some of the silos will be governed by compression rather than hoop tension as well.