Unstayed Flat Head Query

[tph216]

UG-34 equation (1) for thickness seems to me to give a very conservative value when 2UTS/7 is used as the basic allowable stress level SE.

As far as I can see, the major stress type in the head is general primary bending, with general primary membrane being extremely low, almost absent.

As such, how would I go about justifying the use of 1.5SE in this case to set thickness, and still meeting VIII div 1 requirements?

Other notes:
* The setup relies on a retaining ring Fig UG-34(m), to hold the head in place.
* This calc is to be applied to a range sizes and pressure ratings using the same method.
* FEA verification has been undertaken (and will be on all size / rating variations), and backs up the assumption of very low membrane stress (seen to be no more than 10% of bending stress).
* The head is in effect stayed, by further bolted on components, but these are not modelled for purpose of the ASME calc.

In my mind I think this is justifiable in a number of ways, but just wondered if anyone else had come up against the same thing in relation to flat head thickness, which really does seem to be over-kill?

Any help or prior experience with the same issue, would be greatly appreciated.

 

[bernoullies123]

May be of interest: Interpretation: VIII-1-07-02

1) Can u use 1.5S in UG-23(c) in lieu of S in equations UG-34 to determine the thickness
Reply: No

2) Can FEA be used instead of rules in UG-34?
Reply: No

 

[bernoullies123]

The stress in the middle of the flat head is primiary bending stress hence the allowable of 1.5S is inline with the allowable primary bending stress in VIII Div. 2.

 

[tph216]

Reading UG-23(c), to me reads as though by splitting down primary stresses into membrane and bending, that Om < SE, and Om + Ob < 1.5SE.

How does this relate to the equation from UG-34:

t = d*(C*P/S*E)^0.5

I think I really need to know what stresses or combination of stresses this formula is based on. I assume bending is in there, but is membrane also accounted for?

I am wanting to justify putting 1.5SE in there as S, which does fall in line with my FEA findings.

But how does this stand in terms of then certifying the vessel as a Div 1 vessel? UG-23(c) to me justifies this. Thoughts?

(Thanks for the prompt response, by the way bernoulli).

 

[bernoullies123]

I'm assuming that you actually want to use 1.5S in addition to C=0.3 which breaks down to 4.5S with C=1. As noted in the code, these formulas satisfy stress, but they may not in regards to deflection (i.e. leakage).

The maximum stresses will be at the edge of the plate either in the shell or the head depending on the thichness ration of each.

Based on the reply from ASME, you cannot satisfy Div. 1 unless the rules of UG-34 are used.

For a discussion of the stresses in Flat heads refer to Moss (pg 67, 3rd ed.)

 

[LSThill]

tph216 (Mechanical

do you have a drawing as a reference?

L S THILL

 

[prex]

The derivation of that formula is quite straightforward.
For a circular plate supported at the edge the bending stress at center is
[&sigma;]=((3+[&nu;])P(d²/4)/16)(6/t²)
Primary stresses should be evaluated with [&nu;]=0 and you end up with
t_req=d[&radic;]((9/32)P/S)
that justifies C=0.3 (with some margin).
So the factor 1.5 is not included in the formula and t is overestimated by a factor of more than 20%.
However I'm afraid that you can't overcome that formula if you are bound to Div.1 compliance, as noted by bernoullies123.

prex

http://www.xcalcs.com

: Online engineering calculations

http://www.megamag.it

: Magnetic brakes and launchers for fun rides

http://www.levitans.com

: Air bearing pads

 

[tph216]

Ok I think this is a pretty conclusive 'no' in terms of Div 1 compliance (which our client has specifically requested).

bernoullies123 - I've got Moss's design manual, and have been reviewing the treatment of flat heads from that (which again supported my thinking that primary bending stress was the main consituent in flat heads).

LSTHill - No dwgs, client wouldn't be happy.

prex - thanks for the formulation. I'd attempted that, but hadn't set v = 0, so it wasn't tieing up for me. All seems clear now - that formula is limiting by bending stress at plate centre.

It's just a little frustrating because the difference in thickness between 1.0S & 1.5S adds a lot of extra weight to the head. I am still intrigued though, by the wording of UG-34(b) "S = maximum allowable stress value in tension, psi, from applicable table of stress values referenced by UG-23". And then UG-23(c) "...the induced maximum general primary membrane stress does not exceed the maximum allowable stress value in tension" - it doesn't, and "Except where limited by special rules... shall not induce a combined maximum primary membrane plus primary bending stress across the thickness which exceeds 1.5 times the maximum allowable stress value in tension". By my reasoning, the latter should be the case for flat heads. But is this not actually the case?

Is this a limitation or oversight on the way flat heads are dealt with, or am I missing something? I can see, for example, how primary membrane stress would be detrimental for a hemispherical head, for example, but not a flat head.

Thanks all for the advice, much appreciated.

 

[tph216]

One other question. Is it acceptable to treat a circular flat head as a braced / stayed surface if it has auxilliary components bolted on, which provide a bracing function.

I.e. falling in line with UG-47, rather than UG-34)?

The reason I ask, is that my application is a quick-actuating door, in which part of the hinge assembly, bolted to the door, doubles the effective door thickness and halves the effective door diameter. Which in reality would have a massive effect on stresses / deflections, in reference to my earlier queries on calculating flat head thickness.

 

[LSThill]

tph216 (Mechanical)

BELOW IS TWO (2) REFERENCE

STAYED FLAT PLATE RE: MINIMUM THICKNESS PER (PFT-31.2 IN
ASME SEC. I)

DIAGONAL STAY (GUSSETS STAYS) RE: (PFT 32.2)

 

[prex]

tph216,
UG-23(c), if you read it carefully, is applicable to combinations of loadings from UG-22; as far as pressure only is concerned, you are bound to UG-34.
And I think it would be hard to classify a bolted external stiffener as a brace. You could count on the brace for limiting deflections, but IMHO not to contribute to strength against pressure.

prex

http://www.xcalcs.com

: Online engineering calculations

http://www.megamag.it

: Magnetic brakes and launchers for fun rides

http://www.levitans.com

: Air bearing pads