-
1
- #1
TomBarsh
Structural
- Jun 20, 2002
- 1,003
Several threads recently have asked about different results for "same" vessel analyzed in COMPRESS and other software (for example, see thread1259-202448).
Here is an example that came up today of differences between COMPRESS and another major commercial software application.
A vessel had been built in 2001 using a design produced by another major software. The vessel was being re-rated now for higher temperature and pressure. COMPRESS complained that a manway was not adequately reinforced for the design pressure. The manway was "integrally reinforced" per Code, and was offset from radial.
Because the offset nozzle was integrally reinforced COMPRESS reported calculations for the two planes cut through the nozzle, parallel and perpendicular to the vessel axis (ie: theta = 0° and 90°, where theta is defined in Figure UG-37). The nozzle had more than adequate reinforcing area in the plane theta = 90° with the long chord from offset nozzle (factor F = 0.5 in this case). But in the plane theta = 0° (which considers circumferential stress) there was grossly insufficient area available. COMPRESS warned of this case with a Deficiency message.
Our customer answered his own question in his description of the problem: the original calculations by the other software did not consider the cross-section with theta = 0°. Even though the other cross-section had a larger chord opening the critical cross-section had not been checked.
I have no way of knowing if this error was due to user-error or a program-error. Perhaps the other program requires the user to enter the chord opening and F factor? If so, this can lead to errors by an inexperienced user. COMPRESS calculates the chords automatically and would not produce such an error.
In fact, this week I heard anecdotally of a similar issue with a third software. In this, the program graphically drew an offset nozzle as offset, but the calculations did not reflect the longer chord due to the offset opening. This was a non-integrally reinforced nozzle. For this condition ASME Code does not permit the reduction in required area per the reduced F factor, and the required area is that based on the largest chord opening with F = 1.0. Again, the nozzle did not have sufficient reinforcement.
While these other programs likely provide accurate numerical results it seems that both require some care on the part of the user to make certain that the basis of the calculations are valid, whereas COMPRESS does this automatically.
Of course, software changes all the time with new features and perhaps these details have been tidied up in later versions of these programs. But users should always be familiar with any limitations or assumptions of the software they are using.
Tom Barsh
Codeware Technical Support
Here is an example that came up today of differences between COMPRESS and another major commercial software application.
A vessel had been built in 2001 using a design produced by another major software. The vessel was being re-rated now for higher temperature and pressure. COMPRESS complained that a manway was not adequately reinforced for the design pressure. The manway was "integrally reinforced" per Code, and was offset from radial.
Because the offset nozzle was integrally reinforced COMPRESS reported calculations for the two planes cut through the nozzle, parallel and perpendicular to the vessel axis (ie: theta = 0° and 90°, where theta is defined in Figure UG-37). The nozzle had more than adequate reinforcing area in the plane theta = 90° with the long chord from offset nozzle (factor F = 0.5 in this case). But in the plane theta = 0° (which considers circumferential stress) there was grossly insufficient area available. COMPRESS warned of this case with a Deficiency message.
Our customer answered his own question in his description of the problem: the original calculations by the other software did not consider the cross-section with theta = 0°. Even though the other cross-section had a larger chord opening the critical cross-section had not been checked.
I have no way of knowing if this error was due to user-error or a program-error. Perhaps the other program requires the user to enter the chord opening and F factor? If so, this can lead to errors by an inexperienced user. COMPRESS calculates the chords automatically and would not produce such an error.
In fact, this week I heard anecdotally of a similar issue with a third software. In this, the program graphically drew an offset nozzle as offset, but the calculations did not reflect the longer chord due to the offset opening. This was a non-integrally reinforced nozzle. For this condition ASME Code does not permit the reduction in required area per the reduced F factor, and the required area is that based on the largest chord opening with F = 1.0. Again, the nozzle did not have sufficient reinforcement.
While these other programs likely provide accurate numerical results it seems that both require some care on the part of the user to make certain that the basis of the calculations are valid, whereas COMPRESS does this automatically.
Of course, software changes all the time with new features and perhaps these details have been tidied up in later versions of these programs. But users should always be familiar with any limitations or assumptions of the software they are using.
Tom Barsh
Codeware Technical Support