davidribeiro
Mechanical
- Jul 14, 2002
- 29
Hi all,
Anyone able to shed some light into my problem:
Pipe diameter: 2,4m
Design Temp: 200 deg C
Design Int Press: 20kPa
Design External Press: 85 kPa
Please see if I'm on the right track:
1. Calculated required pipe wall thickness from governing pressure case. Due to the long length of the duct, I based the wall thickness calc on using a max dist between stiffening rings.
2. From the structural layout of the pipeline, I am forced to support the pipeline at specific places. Thus, I did a check as to the max deflection and stresses induced if it is supported at these places.
3. Once the stresses were confirmed as being below the allowable, I tackled the problem of saddles.
4. Due to thermal expansion of this pipeline, sliding saddles are going to be used. However, I want to minimise stresses at the tie ins to other equipment (vessels etc) so at these supports I am opting for a fixed support, and the sliding supports intermittantly.
5. Using Codecalc's horizontal vessel support module, I am trying to check the induced stresses on the pipe, base, web etc.
Here's my problem (unless my methodology is also a problem!):
The Zick Analysis used by Codecalc for saddle design, uses dished ends on either side of the shell. In my case I have connecting flages & stiffeners instead. Can I use the distance from the tan line of the head to the centre line of the saddle as the if it was the distance from flange to centre of saddle?
What happens to the length of the pipeline beyond this distance? Apparantly this distance cannot be more than half the tan to tan distance?
Sorry this is so long, but I'd rather give a consise description of my problem.
Regards
David
Anyone able to shed some light into my problem:
Pipe diameter: 2,4m
Design Temp: 200 deg C
Design Int Press: 20kPa
Design External Press: 85 kPa
Please see if I'm on the right track:
1. Calculated required pipe wall thickness from governing pressure case. Due to the long length of the duct, I based the wall thickness calc on using a max dist between stiffening rings.
2. From the structural layout of the pipeline, I am forced to support the pipeline at specific places. Thus, I did a check as to the max deflection and stresses induced if it is supported at these places.
3. Once the stresses were confirmed as being below the allowable, I tackled the problem of saddles.
4. Due to thermal expansion of this pipeline, sliding saddles are going to be used. However, I want to minimise stresses at the tie ins to other equipment (vessels etc) so at these supports I am opting for a fixed support, and the sliding supports intermittantly.
5. Using Codecalc's horizontal vessel support module, I am trying to check the induced stresses on the pipe, base, web etc.
Here's my problem (unless my methodology is also a problem!):
The Zick Analysis used by Codecalc for saddle design, uses dished ends on either side of the shell. In my case I have connecting flages & stiffeners instead. Can I use the distance from the tan line of the head to the centre line of the saddle as the if it was the distance from flange to centre of saddle?
What happens to the length of the pipeline beyond this distance? Apparantly this distance cannot be more than half the tan to tan distance?
Sorry this is so long, but I'd rather give a consise description of my problem.
Regards
David
