Weld Flange To Dish Head Directly

[wcw1985]

Dear All,

I have 1 question which cannot be solve and would like to ask for dear forum members's opinions:

Design condition:
Internal pressure : 10 bar

Torispherical Dish Head: OD1300mm ( around 51")

SORF 20" Flange Ansi 150 -->OD about 700mm (27.5")

I would like to open an opening in center of the dish head and weld this flange directly on it using welding method just like Fig UW-16.1(c) without any nozzle.

How do i calculate to know if the opening is properly reinforced?

I checked UG-36 (b) (2) but i see none of the figure in Fig UG-36 similar to my case.
I checked Appendix 1-7 (LARGE OPENING IN CYLINDRICAL AND CONICAL SHELL) but it was for opening in shell not head.
I checked Appendix 1-10 (ALTERNATIVE METHOD FOR DESIGN OF REINFORCEMENT FOR OPENING S IN CYLINDRICAL AND CONICAL SHELLS UNDER INTERNAL PRESSURE) still was for shell not head.

I really hope for a reply that could help me. Thanks in advance.

The flange will be closed with blind flange in the end of the fabrication.

 

[prex]

This is not a large opening, as the opening diameter is the bore of the flange (513 mm). You just have to use the rules of UG-37 and following paragraphs.
It would be better if you provide a sketch of your setup just to make sure it is understood correctly.

prex
[URL unfurl="true"]http://www.xcalcs.com[/url] : Online engineering calculations
[URL unfurl="true"]http://www.megamag.it[/url] : Magnetic brakes and launchers for fun rides
[URL unfurl="true"]http://www.levitans.com[/url] : Air bearing pads

 

[SnTMan]

A sketch would be good. I infer that you will insert the hub of the SO flange into the head. If so, UG-37 will apply, you will most likely need to restrict the limits of reinforcement in the axial direction, see UG-40(e). You may not be able to take any reinforcing credit for the hub at all. In which case all reinforcement will have to be provided by excess thickness in the head, and perhaps a small amount from the fillet weld. You might consider a studding outlet, Fig UW-16.3

Regards,

Mike

 

[Fizza453]

I would suggest perform the reinforcement calculations per UG-37 and ignore the area contributed by the nozzle neck that is a little conservative approach. Since opening is well withing the limits of UG-36(b), you you can use UG-37.
Why I am asking to neglect the area contributed by the nozzle neck (that is hub of SORF is your case), because I am sure limit of reinforcement normal to the vessel wall falls beyond the hub length, so I am neglecting area A2.
Since your opening and its limit of reinforcement as set by UG-40 falls withing the spherical portion of the head tr (when used in reinforcement calculations) shall be the thickness of the sphere not that of tori-spherical head, this will help you in increasing area A1.
Please run the calculations and then share the results. It is better to perform manual calculations in this case.

Nasir
Welding Engineer
DESCON ENGINEERING LIMITED
PAKISTAN

 

[wcw1985]

I had uploaded 1 picture about the setup of the dishhead and the flange. For my case,i weld the OD of the flanges to the dish head therefore i see this as large head opening since the OD Flange > 0.5 (700mm > 0.5*1300).If UG-37 cannot be apply, which method shall i follow? I will try to upload my calculation later after i work it out to share with you all to see and check if i am correct. I will try the Mr Nasir method.

 

[wcw1985]

Inside diameter of manhole d = 513.1mm (ID of 20" flange)
Do = 1300mm (vessel/Torihead OD)
Minimum thk of vessel tr= 11.5184mm (As per UG-27)
Head wall thk t = 15mm (I use 15mm plate)
Nozzle wall thk tn = 93.45mm ((OD of flange - ID of Flange)/2)
Required nozzle wall thk trn = 4.548mm (Calculated as per UG-27) Corrosion Allowance = 1mm

F = 1
fr1,fr2,fr3,fr4 = 1
E1 = 1
Sn,Sv,Sp = 108000 kpa

A= 5910.7
A1= 1786.43
A2=6667.65
A3=Not Applicable (I does not thrust the flange beyond the ID of the flange)
A41=225 (leg^2)
A42= Not Applicable (since i assume no reinforcement pad)
A43= Not Applicable since no inward nozzle weld

As per Figure UG-37.1
A1+A2+A3+A41+A43 >= A
8679.07 > 5910 (Opening is adequately reinforced).


The reason i use A2 because currently i assume the flange as nozzle. My logical thinking tell me to include the A2. Feel free to comment on the calculation.

 

[doct9960]

wcw1985,

How are you going to seal the pressure at the bolt holes?

I suggest you use a studding outlet.

http://www.texasflange.com/Studding Outlets/Studding Outlets.html

 

[gr2vessels]

wcw1985,
Am I missing something? All you need in fact is a forged studded connection UG-43(d) and as per Fig UG-40, item a-2. Do you HAVE to use a SO flange? If so, hpw do you intend to bolt the cover and seal the bolt holes, from inside?
gr2vessels

 

[wcw1985]

Dear doct9960,

I got 2 ways to solve this:
1) Cut small Round plate around 6mmt~10mmt(for me i will see the what scrap metal i had) plate: fill the hole with the plate then GTAW the plate from outside.
2) Cut a plate that OD same as the flange OD and in the center, cut open an ID hole (probably a little bigger than the ID of the Flange) and weld from the bottom of the flange to seal off all the bolt hole.

I will do Dye Penetrant test afterwards to check for any visual defect.

 

[gr2vessels]

wcw1985,
doct9960 beat me on the stroke of keyboard, but the question is not only how you make it, it is how (and why) would you design a such a complicated and questionable flanging method, when you have the code recommendation and standard studded flanges designed and made for your particular application.
By the way, you still have to design a revolutionary new bolting method and explain your AI why did you used that. I bet you also thought of welding some studs in the existing holes, then do the calcs for that arrangement, right? How you going to do the calculations for this flange, which does not comply any longer with any standard flange??
Cheers,
gr2vessels

 

[wcw1985]

Dear gr2vessels,

My area does not sell any "standard studded flanges", thats why i choose to use a more common sorf flanges. Yes, you are right, i will weld the stud in the existing hole. In my opinion, i will do some calculation for the opening reinforcement requirement as per what i do above. As for the blind flanges thickness, i will reconfirmed the thickness of the blind using UG-34(c)(2) and C equal to the value of Figure UG-34(p) which i found nearest to my design. For the thickness required for the flanges to be used, i am still scratching my head to work it out(still flipping the code book now ), in the same time hope fellow forum member to throw in some idea which we can discuss.

I start to think to use a plate which i cut it round shape same as flange OD, drill some bolt hole (not through hole), tapping it. For the thickness calculation i think i can use UG-27.

 

[prex]

Ouf! Your last sentence is much wiser with respect to the nonsense you stated before (weld the studs into the bolt holes?!?).
The thickness of your insert flange should be dictated by the depth of the bolt holes, just make sure that the flange encompasses the head wall as extended to the flange ID.
Your calculation seems correct in the principle, but I didn't check the figures. As a minor correction, you should take out the corrosion allowance on the inside diameter. Also t_r, as pointed out by Fizza453, can be calculated as the thickness of a sphere per 1-4(d) with M=1.

prex
[URL unfurl="true"]http://www.xcalcs.com[/url] : Online engineering calculations
[URL unfurl="true"]http://www.megamag.it[/url] : Magnetic brakes and launchers for fun rides
[URL unfurl="true"]http://www.levitans.com[/url] : Air bearing pads

 

[wcw1985]

Dear prex,

There is 1 question about the tr, shouldn't i use UG-32(e) for the thickness calculation? When shall i use 1-4(d) method? Is there any specific condition i missed out or is it 1-4(d)?

tr using UG-32 is 11.518mm
tr using 1-4(d) is 8.02mm

Oh ya, for the welding stud to flange issue, is it possible we use stud as the filling material to fill the bolt hole of the flanges for my case? After inserting the stud where the surface of the stud is same level as the bottom face of the bolt hole, and then we weld it(both top(fillet weld) and bottom (GTAW)of the flange. If that is possiblt, i may dictate the thickness of the flange by its thickness rather than the depth of the bolt holes.

My vessel is subjected to internal pressure only: Design pressure is 10 bar.
With Joint Efficiency = 0.85

 

[prex]

t_r stays per UG-32 as far as the whole head is concerned, but, only in the reinforcement calculation, you may use the t_r of a sphere.
Please do not refer any more to the studs welded into the holes, it is certainly not acceptable to code, and for more than one reason. If you want to do that, please don't tell us, we like to have relationships here with real and experienced engineers. If you insist I'll have to ask the maintainers of this site to delete this thread.

prex
[URL unfurl="true"]http://www.xcalcs.com[/url] : Online engineering calculations
[URL unfurl="true"]http://www.megamag.it[/url] : Magnetic brakes and launchers for fun rides
[URL unfurl="true"]http://www.levitans.com[/url] : Air bearing pads

 

[wcw1985]

Dear prex,
Thanks for the clarification. There is a lot of thing i am still learning, therefore i am keep asking for opinion and from more experience engineer in this forum. For my cases, i try to ask every possibilities i can think of with the hope that more experience engineer can provide a good guideline so that we all can learn from it. it is not that i want to influence other people to accept my method. From my reply you can read that i keep asking and giving out my idea with code to back me up. For the stud issue , i am asking for the opinion of the forum members to to see if any other member had been try the method or maybe some case study which we nvr encounter in our engineering life. You never know if you never ask. Please don't be too aggresive when you think other people is trying some new idea or asking for help.

 

[SnTMan]

wcw1985, first, your finished opening is the ID of the flange, not the OD, see UG-37. Second, in your calculation, A2 must be equal to zero, see UG-40(e).

Third, don't.

Regards,

Mike

 

[prex]

SnTMan, cannot agree with your position. This type of studded flange is not covered by UG-40(e).

prex
[URL unfurl="true"]http://www.xcalcs.com[/url] : Online engineering calculations
[URL unfurl="true"]http://www.megamag.it[/url] : Magnetic brakes and launchers for fun rides
[URL unfurl="true"]http://www.levitans.com[/url] : Air bearing pads

 

[wcw1985]

Dear All,

I found that UW-16(3) quite near to addressing to my cases, "the opening shall be reinfoced in accordance to UG-37 and the nozzle or other connection attached using a suitable detail in Fig-16.1,if welded." However, no matter how i find, none of the figure look as similar as Fig 40(a-2) just like gr2vessels told me. But from the UW-16(3) statement, i believe calculating the reinforcement required for my case using UG-37 should be correct.Just a matter of time before i can work out the correct way to find the real answer.

 

[SnTMan]

prex, what I was getting at is the exclusion of bolted flange material as opening reinforcement, except as permitted by UG-40(e), that is, studding outlets and LWN's. In which case A2 must be set to zero for the originally proposed construction.

Regards,

Mike

 

[doct9960]

wcw1985,

Out of curiosity...

Why can't you place a 20" pipe/plate neck on your 20" SORF flange?

Why do you need a manway at the bottom head? Why not on the cylindrical shell?

Is your vessel going to be U-stamped? In what part of the world will this vessel be installed?