Rectangular vessel 6

[Matrix_93]

Hello guys,
I am designing a pressure vessel with low pressure 15 psig but at high temperature 1025 °F in stainless steel 347H.
It would be very easy if the vessel would be circular, but it is a rectangular vessel Size 4 ft x 6 ft.
I started with 0.25 in but it seems to me the stress due to bending moment is very high with appendix 13 Asme code.
Do you have any suggestions in order to avoid to increase shell thickness?

 

[r6155]

@ Matrix_93
Remember, code is for design, fabrication and inspection.
How many longitudinal welds in each shell section?, corner joints?
Vertical or horizontal vessel?, manhole?

Regards

 

[Matrix_93]

Thank you at all for the answers, I am trying to find a solution but the problem is on the corner.
It is a vertical vessel and I don't need to use long. welds.
Basically I will have plates size 4 ft x 6 ft connected with butt welds on the corner.
I would want to use a butt weld corner joint and with FEM I used a fillet radius = 1 in, but the stress is very high even if I use a shell thickness = 1 in.
Probably I need to increase the fillet radius on the corner or I will propose a circular solution to the client.

 

[MJCronin]

It is a vertical vessel and I don't need to use long. welds.
Basically I will have plates size 4 ft x 6 ft connected with butt welds on the corner


I would keep the welds away from the corners. I do not think that this is wise

Is there some physical reason that you cannot bend the plates, slightly round the corners and limit the peak stresses ?

Rounded corners would be easy to fabricate. Then run the full penetration butt weld down the vertical centerline of each face.

Use a rolled angle and bolt or weld it to the flat plate on the ends.... Much like the ice cream container in your freezer

MJCronin
Sr. Process Engineer

 

[DriveMeNuts]

I agree with MJCronin,
Use a brake press to put a fold down the centre of your 4 ft x 6 ft plates.
This fold will form the corner of the vessel and be whatever radius you want.
Will possibly need heat treatment after forming, depending on the material.

Your vessel is so small that you could get two corners folded into a single 4 ft x 12 ft plate.
And then you only have two long seams along the mid-width of the flat plate (not the corner).

 

[r6155]

@ Matrix_93
Calculation without experience in fabrication and inspection is not a good idea.
You are very confused. See definition of butt weld.
Did you think in distortion after welding on flat plates?
Did you think in nondestructive examination?

Forget rectangular design for this example, and for design temperature 1025ºF ??!!

Regards

 

[IdanPV]

This is a regret post:

I am sorry for being smug and arrogant I learnt alot from this thread.
Sorry again.

 

[racookpe1978]

Many comments about the "middle straight length" of this long vertical squared pressure vessel." I thank the earlier writers for their contributions. (And their disagreements and various defenses of various assumptions.)

Yes, fold the side plates with a simple 90 degree bend, make the fabrication welds in a jig or fixture as butt welds in the least stress point on the wall. Not necessarily the middle of the vessel wall. Do NOT attempt to make the four corners as welds!

But! The pressure vessel has a finite length, and both ends will face greater vibration, heatup/cooldown thermal stress, and flow change stress from fluid vibrations and turbulence at both the inlet and outlet far greater than the (comparatively low) mid-vessel longitudinal stress.
Does the rectangle ends "cap" the vessel, or do they terminate in a flat plate or a dome or elliptical head? Are they "reinforced" with an external or internal bolted flange? How thick a flange, how many bolts in the bolt hole pattern and what are the dimensions of the flange(s)?

If welded, is the rectangular pressure vessel welded to another (larger) pressure vessel or pressurized tank or air/heated gas chamber like an exhaust duct that is itself subject to vibration and thermal stress at the joint(s)? What is its movement and changes in its welded joints? Large rectangular and round and round-to-rectangular Gas Turbine exhausts are notorious for cracking at low internal pressures but high temperatures and great vibration stresses! If welded, the end welds WILL restrain movement of the ends of the rectangular pressure vessel, and that restraint (on both ends) MUST BE accommodated by higher movements and higher stresses in the middle length of the pressure vessel.

My recommendation?
Yes, FEA is needed.
Yes, external ribs appear near-essential, each rib being equal in resistance to pressure vessel wall movement (expansion and "rounding" from the heated stress of the internal pressure) to the closest end point reinforcement (which may be different on each end.)
Internal ribs would be more expensive to install, QA, and inspect regularly against cracking. MUCH more expensive to repair too. (Confined space rules and limits, scaffolding, height/width/access limits.)

 

[steris]

I've done a fair number of large rectanglar vessels at much higher pressures and would very strongly encourage you to avoid welding the plates at the corners. Between strength, fatigue, distortion and other manufacturing issues, you are asking for a world of hurt. Not to mention, what's you NDE requirement? If you're doing UT or RT, examining a corner joint is much harder than a butt weld. Put a bend radius at the corners, as large of a bend radius as your design can tolerate. Many fabricators with a high tonnage press brake will either have the tooling needed to bend plate in one shot or they can do bend it progressively with multiple smaller angle bends.

Further, if possible, I would recommend putting stiffening rings around your vessel. You can use bar, angle, channel, or whatever shape your design and budget allow. If your stiffening rings are joined at the corners, think of them as picture frames, then they will pick up much of the bending moments. By sizing the 2nd moment of area of your stiffeners adequately and adjusting their pitch spacing, you'd surprised how much thinner you can make the shell. Considering that you've previously mentioned 1" thick plates, the use of stiffening rings may allow you to reduce material thickness enough that you may be able to avoid PWHT and/or impact testing requirements. Just something to consider...