API 650, Appendix P.3 - Not a clear cut!!! 1

[ANK72]

Hello all,
Appendix P.3 (2003 addendum) now allows to evaluate stresses at nozzle-to-shell intersections due to external nozzle loads.
Tank parameters (d/tn), (t/tn) and (u) determine equations (or curves) for calculating the stress factors.
The three sample problems given in the code deal with cases only when value (d/tn) lies between two fixed curves (say, d/tn = 10 and d/tn = 30). Solution is achieved by interpolation between the two. Easy enough!

Now what if values (d/tn) and (t/tn) both lie between the fixed values. Say (d/tn) = 17 (between 10 and 20) and (t/tn) = 1.6 (between 1 and 2). What do you do then? Does one have to interpolate twice, first for (d/tn) and then for (t/tn)? Will it be acceptable to just solve two equations:
1) d/tn = 10, t/tn = 1 and
2) d/tn = 30, t/tn = 2.

Any help will be greatly appreciated!
Many thanks

 

[jte]

ANK72-

Glad to see the new appendix being used!

You may interpolate twice, once for T/t and once for d/t.

Remember that you are one of the first users of this appendix and there may be some bugs to work out. I'd suggest that you use the equations for the interpolations and then check them against the graphs. If you have any issues, please let me know.

jt

 

[ANK72]

jt
Thanks a lot for advice. That's exactly what I did - interpolated twice. As soon as I had a good look at the graphs it all made sense.
However, I found that the “u” value for one of my nozzles was marginally outside the upper limit of the graph (one of the two). The graph ends at u=1 and my calculated value is 1.24. I had no choice but extrapolate. Hope it is acceptable.
I’m putting together an Excel spreadsheet at the moment (with lookup and macro loops) and I’m afraid that cases like this might screw up the logic. The upper limit of “u” will determine the applicable nozzle-to-shell geometry range.
JT, thank you again. I needed reassurance.
ak

 

[ANK72]

Someone was quick to insert a (commercial) hyperlink into my thread. So annoying!!!
Anyway, I've just found some inconsistencies in Appendix P.3.
I hope jt will be able to comment.

The "Upper limit (u)" presented in Tables P-2 through P-4 contain some errors. Some of the graphs representing the equations given in these tables have totally different limits.
For example Equation # 30 shows the upper limit value of U=5. The graph for this equation (Fig P-8E) ends at U=0.3.
The divisions on the U-axis of all graphs seem to have an error as well. The value 0.11 is supposed to be 0.1. This last one may be due electronic distortion though.
Our company subscribes to API standards and we have internet access to electronic copies. Can someone please check their hard copies to see if it is different?
Regrads
AK

 

[jte]

AK-

Can you send the details of your problem to esto2 at aol dot com? I don't usually check mail on that address, but I'll keep an eye on it for the next few days. I'm transitioning into a period of being extremely busy, but I'll try to get some answers for you. If I can't help, I'll get in touch with someone who can. Again, the appendix is brand new and likely needs to have some bugs worked out.

jt