Flat Plate Design 2

[mmaurigi]

I need to design a Rectangular tank that is holding water (sg=1). The tank dimensions are 28ft. long x 14ft. wide x 12ft high. When I analyze the tank wall without any stiffeners, I use a=144 and b=336 the a/b ratio =0.429. Using Roark Case 2d, the a/b ratio starts at 0.5, so my a/b ratio falls off the chart. What do you do in that situation.

Thanks for your help.

 

[rb1957]

a quick FE model would fix that ... make sure you include the hydro-static pressure of the water

 

[mechengdude]

add stiffeners

 

[dvd]

Look in "Design of Weldments" by Omer Blodgett available from Lincoln Welding for the story on how to cypher out those stiffeners.

 

[GregLocock]

Well, a/b =.5 for all practical purposes. You are aware that Roark's results have approximations in them? I haven't got a copy to hand, but whoever worked them out probably assumed perfect simple supports, for example.I suppose the idea of actually working the correct answer out from scratch is too much to ask? Whoever put that table together had to do it after all. Third year at uni, from memory.

Alternatively, extrapolate the curve, see how much difference it makes.

Or use FEA

Realistically, you need a stiffener, or a tie bar.










Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JStephen]

If this is a steel plate, you are going to calculate an unreasonably thick tank without using any stiffeners. Extrapolate the factor or use the value for 0.5 and see what you get.

 

[40818]

You could try ESDU 69018

 

[prex]

You can also go to the first site below, under Plates -> Simple bending -> Rectangular -> 3 clamped,1 free -> Hydro.load.
Of course your aim of calculating that flat wall under hydrostatic pressure without stiffeners appears unreasonable...

prex

http://www.xcalcs.com

: Online tools for structural design

http://www.megamag.it

: Magnetic brakes for fun rides

http://www.levitans.com

: Air bearing pads

 

[40818]

The problem you will have is the determination of simply supported and fully fixed corner restraints. Your design will have a rotational stiffness somewhere between them. A more elegant solution would be not to have stiffeners on the plates (or even just limited amount), but engineer a double skinned corner attachment which is semi flexible, which would allow a degree of rotation and allows some relief due to membrane effects. Otherwise you end up with a heavy tank due to heavy stiffeners in many places.

 

[GregLocock]

Given that this thing is the size of a house I doubt that any malarkey at the corners is going to prevent the top bowing outwards.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[BarryEng]

Use 'moments & reactions for rectangular plates' with three walls fixed & top edge free with a triangular water load.
Maximum moments are: -
* Top edge at side wall - horizontal - tension inside
* Centre of wall top edge - horizontal - tension outside
* Bottom centre of wall - vertical - tension inside
Coefficients for these three cases are: -
* 0.0715
* 0.0263
* 0.0984
Moment = coeffic x pressure x depth**2
Metric units with depth 3.7 m.
pressure = gamma x depth = 9.8 x 3.7 = 36.3 kN/m**2
moment = 0.0984 x 36.3 x 3.7**2 = 48.9 kNm
stress = M/Z
For a fy of 300 MPa & a stress of 2/3 fy = 200 MPa
t = 38 mm
That's 1.5 inches (in the old money).

As mechengdude said "use stiffeners".

Start with a square grid of stiffeners which is obvious for a uniform load but is very close even with a water (triangular) load. Near the bottom of the tank, the difference between the load at the top of the panel & the bottom of the panel, is becoming smaller.

Remember, that this analysis assumes that all edges of the plates are rigid (a good starting point & an extreme one). This concept is not acceptable in the 'real' world. There are rotations at the corners, the tanks sits on 'an elastic foundation' etc etc.

However, this concept design (rigid edges) is a safe design. It probably can't be any worse.

If you have access to a FEA model, use this but be very careful of your 'spring constants' (beam on an elastic foundation) that you use of soils or clay or rock etc.

 

[BarryEng]

I also should have mentioned, check for horizontal deflection at the top (centre) of the wall. For a single plate, this may be unacceptable depending on your connections to other structures.

 

[BarryEng]

I also meant to mention, the cantilever moment (for an infinitely long wall) at the bottom of the wall (vertical direction) increases as the cube of the water depth (BM = gamma x h cubed/6). Reduce the water depth from 3.7 m to 3 m, will reduce the moment by nearly 50 %.

 

[swearingen]

One thing to look out for with large, stiffened ducts and tanks with positive pressure is the negative moment issues at the corners. Your stiffeners must be designed not to buckle locally if you carry them around the corners - they are in compression in this zone, while in tension out in the middle of the plate.



If you "heard" it on the internet, it's guilty until proven innocent. - DCS

 

[JStephen]

If the top doesn't need to be open (as for dipping large items), then the simplest way to stiffen the top is to run members across the top, even if there's not a full roof.

Otherwise, on the stiffeners, you can work it a lot of different ways. Vertical stiffeners framing into a large beam at top, a series of box-shaped stiffeners around the sides, etc. Stiffeners can be inside or out, although outside is preferred. Stiffeners welded to the plate can be considered as composite sections using part of the plate. When you get done, it will make you appreciate a circular tank.

 

[HEC]

Just picking up on JStephen's comment, why does the tank of that size need to be rectangular!

Mark Hutton