Plate design
Plate design
(OP)
I am designing weir in a cylinder. I want to make it out of steel plate. the diameter of the cylinder is 150" the height of the weir is 120". I know the thickness of the plate is 0.375". Assuming that one side of the plate filled to the top with water and on the other side is dry. How do I find out if the plate thickness is good and if not how would I go about finding stiffener thickness and spacing? Thanks in advance for the help.






RE: Plate design
the rest is QED ...
RE: Plate design
BA
RE: Plate design
RE: Plate design
Dik
RE: Plate design
RE: Plate design
no, each stiffening is designed for the particular case ... you may want a few large stiffeners, or lots of small ones.
RE: Plate design
The pressure on the plate (lbs./sq.ft.) increases with depth, as a function of the depth and the weight of water. Then the plate is designed as a continuous plate, spanning vertically, over a number of stiffeners; essentially acting in bending over the stiffeners and as a tension field btwn. the stiffeners. You can probably tolerate a fair deflection in the plate, except where it joins the main cylinder. And, you may want to add some sort of stiffener ring to the inside of the cylinder where it and the weir plate will join.
With the pressure increasing with depth, but the stiffener length decreasing as you move away from the 6.25' depth on the cylinder, I suspect you can pick one practical size stiffener and then maybe adjust the stiffener spacing a bit to stay within allowable stresses. Don't forget earthquake loading, etc. if that's appropriate. Make the end stiffeners out of channels with their legs down, so as not to collect junk. You might design these as simple beams, with a reduced center moment and a fixed moment at their ends. At the center height of the cylinder, the stiffener (12.5' + long) frames into the inside of a channel web at each end, and these end channels are toes into the cylinder and have a vertical web. These end channels should be long enough and lap the outside of the cylinder so as to distribute the end moments without stress concentrations in the cylinder plate. The other stiffeners have their length and end angles cut to fit end channels which still mate with the outside of the main cylinder, but now with sloped webs. These end details and the associated welding and weld designs are the important part of the project. Perfect fit-up may alter the exact stiffener spacing by a very small amount, but design for that in weir pl. and stiffener strength.
That's one of several possibilities, the design is your's, and if you can't figure it out you should be talking to a local mentor or your boss, so he knows what you don't know and can keep you out of trouble. Done wrong, this could cause some serious damage.
RE: Plate design
Then space your beams such that the allowable span of the plate is not exceeded throughout the height. When you are satisfied that you have found a workable solution, check the beam with the highest moment. For simplicity, you may wish to use the same beam size for every location or you may opt to design each one separately, i.e. use your engineering judgment.
BA
RE: Plate design
A 3/8 inch plate spanning 12'6" with a 10'-0" liquid height?
It's going to have to be stiffened with wide flange shapes.
I guess I'm even more scared if it's not a class project.
RE: Plate design
Your "scared" comment is exactly my point, about half the time.
These are some fairly serious projects, but some fairly basic problems, unless the guy slept all the way through college. I'm all for helping young engineers learn the ropes, but at the same time I wonder if an answer, without every possible detail covered in even more detail, just makes them all the more dangerous. The basic design and analysis question should not need to be asked, or I wonder if the guy isn't in way over his head; I wonder if the boss or client doesn't give a damn or doesn't know any better either; I wonder how these people have B.S'ed. their employer about their actual abilities; I wonder where all the senior engineers have gone at these companies and why there are no mentors around any longer, God knows we all needed them at the early stages of our careers too. Are they coming here because they don't want to admit to the boss what they don't know, when in fact most good bosses understand this situation with new young engineers, and are more than willing to help, if you didn't sell yourself for something you are not. The boss should want to help if for no other reason than to keep his own tit out of the wringer. Then, experience here might step in and help, even outside their own company, to discuss the question 'is there a better way to do this or is there a better detail than this?'
RE: Plate design
RE: Plate design
only if there was a standard stiffner !? (and a standard load, etc).
no, each stiffening is designed for the particular case ... you may want a few large stiffeners, or lots of small ones."
I am not asking for a standard stiffener but a standard to calculate the distance required between stiffeners.
Also I am not having troubles calculating the load or pressure.
RE: Plate design
In problems of this sort, it is best to start out with some rough approximations. For a steel plate spanning vertically between beams, assume that M = wL^2/12 where w is the water pressure halfway between beams.
Assume an allowable bending stress of f = 30,000psi.
The section modulus is approximately 12*0.375^2/6 = 0.281 so the allowable bending moment is f*S = 30,000*0.281 = 8,437"#.
Equating wL^2/12 t0 8437 gives the following maximum plate spans for various pressures:
w Lmax
psf in.
100 31.8
200 22.5
300 18.3
400 15.9
500 14.2
600 13.0
This gives you an idea about how to space your horizontal beams. So select a proposed pattern of beams and complete the analysis.
BA
RE: Plate design
You got jumped on a bit because of how you phrased your question, and you are not the first nor will you be the last. Also, for future reference, a sketch will be demanded most of the time unless it is a code/spec type question. Probably not necessary for your situation but keep that in mind.
"How do I find out if the plate thickness is good and if not how would I go about finding stiffener thickness and spacing?"
"I am not asking for a standard stiffener but a standard to calculate the distance required between stiffeners."
It would be the same as saying how do I figure out if this W12x14 works or not, and how far do I space them, and what span is it good for...
Dhengr's point is you should be able to attack this problem, break it down into pieces, and figure it mostly out on your own (engineering); or have someone in your office you can ask to get you started. There are no standard beam, column, stiffener, etc. sizes or spaces or a lot of structural engineers would be looking for other work. This is not residential floor design where you can pick a joist out of a table for a given span and spacing. (The guys' points above...)
Now with your response in mind, if you had presented the two formulas and why they are confusing, then you may have gotten responses more to your liking.
But some of the guys on this board get pretty PO'd, and understandably, if they think someone is fishing for a design and hasn't at least done their own homework or research. Hardly anyone knows each other on this board, so you can get judged pretty quickly and harshly based on what you write.
But don't let it discourage you from posting, just come with a strong post, full of details and a sketch, and take the problem as far as you can on your own.
FWIW I have never designed anything like you are describing in my 13 years, so I would also want to discuss this with another SE who has designed something like this before I gave it a go!
RE: Plate design
Reread your OP, and imagine how it might have prompted JedC's post; reread my posts and imagine how they might fit your OP, particularly my second post and if the shoe fits wear it. For the parts that don't fit, I apologize, but that post is all too often very fitting, it just showed up again in your thread, as a reminder to all. If you had asked your question more like your 6JAN11,17:56 post, you might have gotten completely different responses, because it at least shows you have done some digging and research on the problem. However, the book you reference has many equations in it, so we are still in the dark about your exact problem.
Get out your Statics and Strength of Materials books, much of what you need should be there. Look at Roark's book "Formulas for Stress and Strain," these books should be right at your fingertips if you are going to learn to do this kind of engineering work; and/or they should be available in your office or you shouldn't be doing this kind of work. Later you will want some advanced books on Mechanics of Materials and Theory of Elasticity: "Theory of Elasticity" by Timoshenko and Goodier, and "Theory of Plates and Shells" by Timoshenko and Woinowsky-Krieger, are two examples of the latter. You must, almost always, use some of your own basic reference materials to get started on a problem, and your own native abilities to formulate the problem, before you start asking question or you don't come across looking very competent. That is also a very good reason to develop a meaningful professional relationship with a very competent local engineer, maybe a mentor, if it develops that way. You can both be looking at the same specs., plans, sketches, etc. and ask questions or getting correction and feedback instantaneously. He can also tell you 'that's a pretty dumb question, you should know better, look in this book, and then tell me why your assumption or comment was wrong' and it won't hurt nearly as much because the relationship you have developed has your best interest in mind, and you know it by the smile on his face as he's calling you a dummy. And, you have to be brave enough to go find this senior engineer and ask him if he will be your mentor. It's a wonderful experience when it works. You'll be amazed at your progress and he'll be happy to watch your development.
Now, to "Steel Plate Engineering Data - Vol. 2," what page and which formulas? Why are you stuck on 3/8" pl. for the weir pl.? A thicker pl. might mean wider stiffener spacing, and those stiffeners are expensive. My earlier comments about your weir pl. conform more to Fig. 2, pg. 3 in my copy of that book. What's the thickness of the main cylinder pl. and where in the cylinder is the weir pl., at the end of the cylinder or at some distance interior from the end? There are probably many more questions like these which you must explain so that we have a full picture of your problem, and how we might analyze and detail a solution, if you want meaningful help. You can see it we can't, so you have to give all the info. needed, and ask a fairly specific question, to start to develop a meaningful solution.
RE: Plate design
Dik
RE: Plate design
Why .375" plate? How does the weir work in a cylinder (how is one side isolated from the other)? What supports the bottom of the weir? And how are the stiffeners tied into the cylinder (very important to carry shear)?
Now some supervisors (or mentors) believe in throwing a young engineer into the deep end and see what he or she does. And if that is the case, the questions adammilad had might just be an alternate way of researching the answer. But this problem has disaster written all over it.
RE: Plate design
Yes, I am treating the plate as a plate and not a membrane. If you wish to treat it as a membrane, you must consider large deflections. I do not think that is the proper way to treat the problem, but in the limiting case, it may be an acceptable solution to the problem provided the necessary connections can be provided at the boundaries.
BA
RE: Plate design
the stiffeners act like cantilevered beams, carrying the water pressure to the cyclinder.
water pressure varies with depth (obviously), so the stiffeners have a triangular loading.
you could start with a stiffener size (something typical, "standard"), and a spacing, (1' ?). what's the stress in the plate, say a 1" width close to the bottom ? can the plate carry the load ?? (bending, yes?) can the stiffener carry the load (from a strip 1' wide, 1/2 the spacing either side) ? again, bending (compression is going to be critical)
if not, change ... spacing (maybe inter-pitch some stiffeners in the heavily loaded section of the plate ?), stiffener size, ... my 2c, the load increases significantly towards the bttm, maybe two sizes of stiffener (careful how you join them), maybe two spacings (inter-pitching).
i'd also look into the codes you're reading, what seems contradictory ?? ask your lead.
does this construction seem odd to anyone ... why would you build a weir in a cyclinder ? presumably water is flowing in and out ?? doesn't this change things, maybe not significantly ??
RE: Plate design
dhenger-
I was looking at equation 3 on page 2. According to this formula it shows that there is no need for spacing and we all know that is not true. I used 18.8 ksi for f which is what the material is rated for at that temp.
Going to page 4 I decided to come up with a thickness that allow for 50" spacing so I used the rectangle section with uniform p, supported and B=120" I came up with a 0.5" plate. Now that I am thinking about it I should probably go to pg.466 in Roarks "Formulas for Stress and Strain" (3 sides fixed 1 free).
I will attach the pages I am referencing. This is part of a liquid seal I am designing for those that asked previously.
RE: Plate design
Attached is page 2 and 3 from "Steel Plate Engineering Data - Vol. 2,"
RE: Plate design
RE: Plate design
RE: Plate design
you don't have this. [depends on which way your stiffeners run !]
"the uniform load I used 2.2psi." ... you don't have a uniform load, you could use the load from 1/2 the depth ... makes me think i automatically assumed vertical stiffeners, but maybe everyone else is thinking of horizontal ones ??
also, distributed load should be lbs/in ...
Roark's is a good source ... look into his reference, google something like "moody plates" to get to the original doc, it includes more loadings, more relevent to your case, and a lot more detail.
RE: Plate design
RE: Plate design
but if you have vertical stiffeners, then they don't have uniform load and probably aren't fixed at both ends. running an angle along the top makes sense, and you can probably could this as a pinned support [but i suspect that it'll see some pretty big loads]. with this, the base (connecting to the cyclinder) can be pinned or fixed.
horizontal stiffeners would work just as well. a grillage (horizontal and vertical stiffeners) would work too; in you case you could have continuous members running in both directions (on either side of the plate).
many ways to skin cats ...
RE: Plate design
RE: Plate design
The stiffeners could be placed vertically or horizontally. If they are placed horizontally, the plate spans vertically with a variable pressure. Without introducing significant error, if the pressure is taken half way between two beams, the maximum moment is roughly wL^2/12.
Since fmax = 18,800 and S per inch = 0.375^2/6 = 0.0234 in^3, we have the relationship that wL^2/12 = fS.
L = (12*f*S/w)^1/2.
If H = 9', w = 562 psf or 3.9 psi and L = 36.8" (say 36")
If H = 6', w = 374 psf or 2.6 psi and L = 42.6" (say 42")
So use 3/8" plate throughout with three stiffeners at Elevation 36", 78" and 120" respectively.
Design the three beams for their tributary load.
BA
RE: Plate design
Well, you did say 'some rough approximations to start with,' so, only Forty Lashes for you, or a few beers if we ever meet. Ah, yes, using ft.-lbs. when we should be using inch-lbs. might lead to an error on the order of (12)^½ to small a span length. We disagreed so seldom that I usually wouldn't check your every calc., and you do such a nice job of laying it out that I don't ever have much to add to your presentation. Actually, I generally worked with large structures and loads, so ft.-kips were the norm until I got to the stress calc. at which point I made the conversion to inch-kips with a 12 in./ft.; and kips and ft. were the norm and 1728 showed up in every deflection calc. Of course, this very feel for and comfort with the units you normally work in is the cause of the whole resistence to the conversion to SI units, isn't it? I do hope the younger people make the change, but it's still fraught with industry conversion problems. I actually have some trouble working in inch-lbs., just no feel for the magnitude of the numbers. I looked at the equations, they looked right, and the proportions of the spacings looked about right, and I really didn't give the exact magnitudes a second thought because I hadn't done the calcs. and tabulation.
Even so, and maybe more importantly, you showed the younger engineers that everyone makes a mistake from time to time, so we must check our own work. Otherwise another fine example of a top notch experienced engineering simplifying the problem, paying attention to his work, by checking his own work, always checking the units in your calcs. or having a gut feeling that something doesn't look quite right, as he thinks through the problem. The difference is that you have the experience to do this, and I wonder if us older guys are the only ones who check their units in calcs. any longer, out of force of habit, since a slide rule wouldn't do that for us. But, so many of these young people are floundering around understanding the very basics of the problem, to the extent that they wouldn't know a wrong answer if it bit them in the butt. Your suggestion to make some quick first approximations on a problem like this, to start to get a feel for it, is certainly the approach I usually take. We may never have seen this problem before, so we have to hone in on it, to get a feel for sizes, thicknesses and proportions; then we can start to add the complexities of deflections, connections, welding and any stress concentrations btwn. the stiffeners, weir pl. and the cylinder plate, etc.
Just before 7JAN11 12:32, I was writing a piece extolling the virtues of your posts and approach, and I looked up and found that he forgot to mention that he was considering vert. stiffeners, so you and I weren't even in the same discussion as he and Rb; at which point I dashed off the 12:32 bit in exasperation. And, now I learn that you went from virtuous to sinner in one fell swoop, so consider yourself taken to task, we don't use much coal around here anymore, so no hauling over the....
RE: Plate design
RE: Plate design
RE: Plate design