PE stub flange and steel backing ring calculation

[stanier]

Traditionally piping in PE and other thermoplastics uses a stub end with a steel backing flange. The backing flanges are drilled to a standard to match valves, instruments and equipment.

There appears to be no calculation method to determine the design rating of such an assembly in any code I have seen. Obviously the flange cannot be rated to the full design rating of a steel flange assembly.

Has anyone any experience of a design approach for this type of flange where the stub end is in PE that can behave in a non linear fashion and steel would be linear in the temperatures and pressure to suit PE. Use of FEA is problematical unless verified by testing. As these products are used in mining and low risk industries there is not the demand for intensive scrutiny.

ôThe beautiful thing about learning is that no one can take it away from you.ö
---B.B. King

http://waterhammer.hopout.com.au/

 

[LittleInch]

Stanier,

ISO 4427-3 relates to PE fittings. Section 6.8 refers to ISO 9624, see

http://www.strataplastics.co.nz/files/file/285/ISO_9624.pdf

ISO 4427, does say in appendix B that the manufacturer is responsible for design and pressure rating and to demonstrate conformity to the declared PN rating and then lists a number of tests which need to be performed to show compliance.

I don't understand your position of "Obviously the flange cannot be rated to the full design rating of a steel flange assembly" If the steel flange assembly is rated to e.g. PN 16, then why isn't a PN 16 rated backing flange and stub assembly not equal to it?

An issue I've seen is where the supplier just drilled the backing plate to match an ASME #150 flange and the whole plate assembly bowed due to the larger PCD...

What this what you were after?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[stanier]

LittleInch,

the reason the flange cannot be rated to PN16 using a flange drilled to ASME B16.5 or AS 2129 or some other standards is that the flange is design for a steel pipe. The PE stub end results in a larger bore for the backing ring, the PE does not behave linearly when stressed, and the gasket diameter is different. Using practices for flange design set out in ASME VIII or AS 1210 do not work as the factors used are based upon empirical data. These data have not been produced by the manufacturers or standards authorities.

PE is used in industries that do not have the resources of the petro-chem or oil and gas industry. Standards were developed as thermoplastics were introduced as competition to ferrous products. The manufacturers had to provide design guidance for PE and other products. They neither possessed the technical expertise to undertake the extensive testing done by Markl et al nor the desire.

Most standards for thermoplastic pipe design, supply, testing and installation have been developed based upon specific materials NOT as piping codes.

ôThe beautiful thing about learning is that no one can take it away from you.ö
---B.B. King

http://waterhammer.hopout.com.au/

 

[shorion]

Afternoon Stanier

I do a bit of FEA work and if you have a couple of cases where you already know the experimental results and want to see if simulation agrees with them let me know.

Keep the experimental results under your hat, I love seeing how close I can get without knowing the answer.

 

[LittleInch]

I'm a little confused now.

The question posed at the start was about experience of a design approach. My response was that if you follow a published international specification / code, then you're using the combined experience, testing regimes etc from a lot of companies and people. This means that you need to use the flange ratings and system design, i.e. the PN rated flange dimensions.

The key issue normally is that you need to use a smaller PE pipe than its steel comparison unless you use something like this -

http://www.gpsuk.com/uploads/docs/124.pdf

The company mentioned here I've always found to be very knowledgeable about the theory as well as the practice of PE - worth trying to find someone there to talk to- my contact retired a few years ago.

You're making fairly sweeping statement here that I'm not sure I agree with.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[rconner]

I have seen many attempts over the years, and for many perhaps practical reasons where fusions could or would not be done in the field, to apply backing ring flanges for polyethylene systems beyond the "mining" industries, and some (e.g. marine construction applications) certainly not in the "low risk" realms. Some designs have historically been some problematic, witness e.g.

http://www.eng-tips.com/viewthread.cfm?qid=305046,

and I seem to recall an account a few years ago of another one that somehow cut loose in a float/sink operation of a very large sized hdpe ocean outfall installation in the Pacific Northwest, requiring the contractor(s) to somehow fish out the pipes/ends out there and somehow re-connect etc?) While these are not the only problems, I guess at least some of the reasons for problems with some such designs may be as explained in the website at

http://www.ipp-inc.com/selectiontools/deltaflexperformancehdpeflangeadapters.aspx

. It leads into a discussion of these technical issues why such are not equivalent to traditional metal designs with the heading, "No reliable flange standards for the thermoplastic piping industry currently exist." I noticed another paper that gets into even more nitty gritty in the context of perhaps some similar connections of also non-metallic/composite pipes at

http://wiredspace.wits.ac.za/bitstream/handle/10539/4833/AsumaniO_Chapter 7.pdf?sequence=9.

In a nutshell, you will see there are higher localized stresses than may be obvious (or comfortable?), and with materials that respond to same (and in the long-term) some differently than traditional metals.
It appears some major manufacturers of hdpe pipe are attempting to insulate themselves from problems that may occur with some quite restrictive verbiage in their handbooks that appear to be handled differently than their marketing literature and standards e.g. as one will see in e.g.

http://www.gajeske.com/documents/resources_docs/Field Handbook - Performance Pipe.pdf

that includes such requirements as

"Surface and above grade flanges must be properly supported to avoid bending stresses."

"Where pipe is connected to rigid devices such as fabricated directional fittings or where flanges or other rigid connections are employed, the pipe must be protected from shear, flexing and bending."

In practical application it may alas be some hard (and/or expensive in supports/restraints?) to protect plastic flanged joints from "shear, flexing, and bending". All have a good weekend.

 

[stanier]

Hi Shorion,

Thanks for the offer. I have excluded this requirement from my scope and have recommended that the end user do some physical testing. If I get meaning full results I will be in contact. The non linear behaviour of the PE makes for interesting modelling.

Hi Littleinch,

To the contrary you generally have PE pipes "larger" in nominal; size than the ferrous pipe you are connecting to as the PE has a much larger wall thickness. You need to preserve the bore for hydraulic reasons. That results in a non standard size flange on the metallic pipe. Hence bending stresses on the flange are greater as the moment arm has increased.

I have served on Australian standards committees for thermoplastic pipe for decades. The committees are dominated by Water Authorities and pipe manufacturers/suppliers. The question of flange design is always skirted around as alluded to by rconnor.

If you know of a standard that covers the design of a flanged joint PE to PE or PE to steel, stainless steel, copper or other material then please enlighten us.

rconnor,

The issue fusion welding of PE is not related to this topic! We are all aware of your role in life in promoting DI pipe and denigrating thermoplastics. PE has been welded with fusion welding for many years and thousands of kilometres of pipelines. Like steel, stainless steel and other materials there have been failures that we all learn from. So thanks for the links.

Good try.

I quite agree with you about manufacturers catalogues not being a good design guide and gloss over essentials. But where are the standards? Perhaps the risks are perceived to be too low as the pressures are low compared to other systems. Life and assets, along with environmental contamination, are still as risk!

ôThe beautiful thing about learning is that no one can take it away from you.ö
---B.B. King

http://waterhammer.hopout.com.au/

 

[LittleInch]

Stanier,

You keep changing the issue. Now we have "To the contrary you generally have PE pipes "larger" in nominal; size than the ferrous pipe you are connecting to as the PE has a much larger wall thickness. You need to preserve the bore for hydraulic reasons. That results in a non standard size flange on the metallic pipe. Hence bending stresses on the flange are greater as the moment arm has increased."

True, if you want to keep the same bore (not really vital when using PE to steel) when you have two widely different wall thicknesses, then you need to do something strange. However that could easily be, use one of the fittings I noted above which mates similar ID PE pipe to its equivalent in steel. Or you use a larger steel flange at the normal dimensions and then have a reducer. Clearly if you make something non standard then you'll have a non standard rating

I have listed two ISO codes which cover the design of flanged joints above. Please let me know why these don't meet what you're looking for as it is not clear to me why you've seemingly ignored them. (ISO 4427 and ISO 9624).

Some of the responses seem to be more about the impact on the pipe rather than the flange joint. Clearly with any backing ring type joint, rotational forces are lower than you would find for a WN flange, but in pure bending it should be the same.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.