Hello,

My question concerns the nozzle load / flexibility calculation as per API 650, Appendix-P, Figure P.4a

I would like to clarify how the graphic figure P.4a to be applied / interpreted in order to determine correct coefficients YF and YL, because:

1.) The logarithmic diagram does not match the intermediate values shown neither on the axis lambda (0.4 and between 0.5 and 1.0) nor on the YF/YL axis (0.8,0.9 and between 2 and 5 and 5 and 1), since the distances are NOT logarithmic.

2.) The stiffness and allowable values of the Appendix-P Sample in P.2.9.1 are different to those, if the sample tank is calculated with Caesar II (compared with Output -> Miscellaneous)

How are the coefficients YF/YL are calculated in Caesar II or how is the Figure P.4a implemented ?

Best Regards
Peter

There are a number of problems with the example in Appendix P.

1) The code was developed in Imperial units. The SI equations and examples were added much later. The code used "soft" conversions, so you can't really convert back and forth. CAESAR II works in Imperial units, and converts to user units for display purposes only. If you want to verify what the software is doing you should work in Imperial units.

2) The code rounded values, which introduces inaccuracies.

3) When interpolating on the stiffness curves, the code did not interpolate between various "a/R" values, they just pulled the closest value. CAESAR II will interpolate between "a/R" values.

4) CAESAR II also interpolates between graphs, between L/2a=1.0 and L/2a=1.5, using a linear interpolation. The code didn't do this either.

Dear Richard,

Thanks a lot,that helps understanding the different results, but anyway what's now the correct way to get YF,YL from figure P.4.a for intermediate points: linear or logarithmic graphical lengths ?

Best Regards, Peter

Both P.4.a and P.4.b are log-log graphs. Visual interpolation is shakey at best. We pulled values from the graphs where the line crosses the grids, and let the software interpolate, but you should be able to get close.

Dear Richard, ok, that means it would be better to ignore the shown interpolated points (those without written values) and determine the values as for a usual log-log graph, thank you very much!

Best Regards
Peter