Nozzle reinforcing questions- Section VIII Div. 1

[JStephen]

First Question: Section VIII defines the "limits of reinforcement", but the definitions become somewhat ambiguous when applied to curved plate. Are the limiting lines assumed to be parallel to each other, or normal to the tank surface? If normal, are the limits measured from one point, or from varying points depending on the radius? See the attached sketch to see what I mean. This won't affect calculated area so much, but does affect whether the reinforcing area extends into the knuckle of a head or not.

Second Question: Where the shell-to-nozzle connection is not a full-penetration weld, there will be some gap between the nozzle wall and the shell. Is this gap limited or accounted for in any way?

Thanks!

 

[unclesyd]

Here is a course outline by Whiteside on openings in PV's that might be of assistance. I don't recall whether its specifically address the area of reinforcement.

http://www.pdhonline.org/courses/m205/m205content.pdf

 

[jte]

Area of reinforcement is measured "parallel to the vessel wall."

You are not restricted from having a repad extend from the shell to the knuckle or vice versa.

See Interpretation VIII-1-89-83.

jt

 

[JStephen]

Well, if you assume "parallel to the vessel wall" means following the curve of the vessel wall, then you could take the extent of reinforcing as an arc measurement. Only you have several different radii at which the arc could be measured, and with a hill-sided nozzle, several different points from which you could start the measurement- see the second two figures in the sketch. I assumed there was some sort of accepted procedure here, maybe there isn't.

And of course the motivation in keeping the area out of the knuckle is to use the reduced crown thickness.

 

[prex]

Personally I have always used the arc length measured at mid wall like in your Figure B (not Figure C; Figure A is not according to code as noted above).
I think this one is in accordance to code because:
-d is to be taken at mid surface as specified by UG-37
-as the main limit for reinforcement is based on d it would be inconsistent to measure the limit on a different plane

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

 

[JStephen]

Thanks, sounds like I'm not totally out of the ballpark.

Where Fig. C came from- there is also a requirement that you have the same amount of reinforcing on each side of the opening. Originally, I drew the repad so it was centered over the intersection of nozzle and required-thickness, which made it obviously unbalanced. And I assume the repad would be centered on the neck when installed, anyway. But also, if all limits are measured from the center of nozzle/ center of required thickness, then the "excess plate thickness" will be unbalanced for a hillsided nozzle. So it would make sense to me if at each radius, the limit was taken from the nozzle axis, although this is not stated anywhere in the code.

 

[prex]

Carefully reread that code sentence. It doesn't tell you that you need the same amount of reinforcement on each side, but only that you need at least half of the required on each side. In other words you cannot compensate metal lacking for compensation at left with some excess metal on the right.
Another way of looking at it is that you should check the compensation on one side only of the opening at the most unfavourable angular position, using of course half of the required compensation.

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