Stress in Pipe at Support 1

[denniskb]

I would like to be able to calculate the stresses in a GRE laminate pipe where it passes over a pipe support and to design the reinforcing saddle to reduce the stresses to an acceptable level. This is a very important part of successful GRE pipe design but does not seem to be well documented.

I have an article "Overcoming the metal mindset in FRP pipe design" from Chemical Processing September 2000 (though my copy is hard to read) and have set this up in Excel. It assumes a full encirclement reinforcing ring, which is impractical to make without joins which will reduce its strength.

I have the ISO 14692 Appendix D method also set up in Excel but this does not allow me to design the saddle.

Also the results of the two methods vary widely and I suspect the published formulas may be wrong.

Any suggestions where I might find a valid solution would be appreciated.


Dennis Kirk Engineering

http://www.ozemail.com.au/~denniskb

 

[rd78]

Dennis-

If this is large diameter piping, you can use the concepts of horizontal vessels on saddle supports (Zick). On smaller diameter pipe, many support saddle manufacturers just go with 360° of support (JTI Manufacturing, for example). Do you have small diamter pipe (under 600 mm)?

Rick D.

 

[stanier]

If you go to

http://www.pipingdesigners.com

you will find the article that you refer to that your can download. Other articles are there also. It is also on the

http://www.chemicalprocessing.com/web_first/cp.nsf/ArticleID/DGRN-4R

site

I recommend that you look at the British Standards on this matter as they cover the topic.

BS 4994 Design and Construction of Vessels and Tanks in reinforced plastics

BS7159 Design and Construction of Glass Reninforced Plastics (GRP) Piping Systems for individual Plants or Sites

 

[denniskb]

rd78,

I am currently concerned about a DN200 pipe.

stainer,

The article has been removed from both of these locations. I have now been provided with a clear electronic copy which I can pass on if needed.



I have set up an Excel calculation using the method proposed in "Overcoming the Metal Mindset in FRP Pipe Design" and Roark's Table 17, Case 20.

The Excel file can be downloaded from my website at

http://www.ozemail.com.au/~denniskb

by following the link to "Design Calculations", submit the form (with name please) then selecting "Miscellaneous Files" at the bottom of the page.

I would like to get some feedback of whether this method "Ring Stiffener" is really appropriate.
- I have set up the calculation at top, side and bottom of the pipe as a rough means of determining how far round a partial saddle might need to go.
- I have serious doubts about the deflections calculated?
- I have been a little creative with the deflections for the stiffened pipe.

The "Saddle Stiffener" worksheet is the ISO 14692 method.

I will pass this on to a couple of GRE pipe manufacturers to see if I can get some input from them.


Dennis Kirk Engineering

http://www.ozemail.com.au/~denniskb

 

[stanier]

Dennis,

The link to your website worked however the file was not there. Web message was that the site was last updated April 2001. Is the link perhaps an old one?

 

[denniskb]

The file is in the Miscellaneous Files section at the bottom of the second page of file listings. The first page does not mention these files. When you find it the "here" word is the link.

Dennis Kirk Engineering

http://www.ozemail.com.au/~denniskb

 

[frost27]

The equations quoted in ISO 14692 relate to the axial stress induced at pipe supports. The formulas are specific to large diameter, low pressure systems. The formaulas are simple based on measurd test data by Strathclyde University (Prof. Bill Banks). ISO 14692 does not specify how to design a support but addresses the issue of what the effect of the support is on the pipework.

 

[denniskb]

Simon,

Thank you for the response, which should prove very useful.
Roark's formulas are providing me with hoop stresses in the pipe with GRE reinforcement bonded to the pipe (though only a full encirclement reinforcement case so far) but no axial stresses, other then from Poisson's ratio.

We can of course include the axial stresses due to bending from the free span and hoop and axial stresses from internal pressure.

Can you advise where the axial stresses from ISO 14692 are expected to occur. Are they within the reinforced section or in the bare pipe past the end of the saddle? Are they at the top, side or bottom of the pipe?

The location is important to how
i) the support axial stress
ii) the span axial stress,
iii) the ring bending stress and
iv) the internal pressure stress are combined
and so the acceptance criteria is established (via ISO 14692 criteria)

Obviuosly a simple criteria and method are required eventually but unfortunately the work needs to be done first to establish what simple criteria is acceptable. Can you advise if the ISO 14692 calculation method has sufficient background for it to be that simple method for the pipe wall itself beyond the saddle?

If so then it remains for the saddle reinforcement design method to be established.



Dennis Kirk Engineering

http://www.ozemail.com.au/~denniskb