high pressure piping and Bourdon effect 2

[papiesz]

Hi all

I have to analyse feedwater pumps piping according to B31.1. Design pressure of discharge pipe is 23.4 MPa. Calcs during FEED was made by another company using Caesar II. They calculated standard operating cases W+P1+T1 but Bourdon option was set to "none" which results in no displacements due to pressure. While recalculating system in Autopipe I skiped pressure analysis to have the same as Bourdon effect = "none" in Caesar.

But I have doubts. B31.1 code is silent on including pressure elongation in analysis. Reason tells me to include bourdon effect, also my B31.3 experience incline me in that direction. But FEED calcs was made by big, international, German originated company with "thousands" years of experience in power. Maybe they are right.

The main problem is with pump nozzle load. When Bourdon is triggered on it is impossible to keep loads within allowable without significant routing change. I wonder what client will say as in this particular system only valve and strainer weight/length was changed since FEED.

Could someone give me a little piece of advice?

Regards

papiesz

 

[prex]

Of course, as I'm sure you are aware of, bourdon effect gives rise to strains that are analogous to thermal strains and must be treated in the same way.
In my (quite outdated) experience, it is customary not to include pressure strains in the analysis, though there is no rational reason for that.
See what a bibliographic reference (also quite oldish...) tells about that:

It may be remarked parenthetically that, while pressure effects discussed here could actually be taken into account in a flexibility analysis and in some cases might have a significant effect upon the final result, the author knows of no case in which this actually has been done.

prex

http://www.xcalcs.com

: Online engineering calculations

http://www.megamag.it

: Magnetic brakes and launchers for fun rides

http://www.levitans.com

: Air bearing pads

 

[zdas04]

While stress analysis is not my area of expertise, I have to comment that a company cannot have "thousands of years of experience". A company cannot have ANY experience. People gain experience (or not), companies can't. It could be that this old and revered firm gave your job to a tech with a high school education and 3 weeks experience who was checking boxes based on a script that no one ever reconsidered after it was written. Back when most computer programs simply generated reports we would say "if it's on green and white [paper] then it must be right". The same thing happens today, uncritical acceptance of program output is normal.

Could it be that in the slide-rule days of my youth, the Bourdon analysis was simply too time consuming and too rarely material to ever be done, so today when it is just a check box people still don't do it since it is rarely material?

It seems like if there was anywhere that piping would try to straighten out under applied pressure it would be on a feed pump discharge. If I was you I would get an authoritative answer from somewhere before ignoring your analysis that adding that analysis will over stress the pipe, but at the end of the day that answer should be in writing and signed by an engineer (from the German Firm I would think).

David

 

[CostasV]

Bourdon effect is valid for non circular cross-section pipes, isn't it?

 

[papiesz]

Varsamidis

I meant elongation of straight pipe and elbow centerline under internal pressure. As I know Bourdon effect is linked with tendency of pipe bends to open under pressure. Ovalisation (if exists as a result of fabrication)intensify that tendency.

Pipe stress software (autopipe, caesar) can consider both fenomenon: straight pipe elongation and bend opening. Options are called translational and rotational Bourdon effect. In caesar you can build operating case that consists pressure but Bourdon can be set in three different ways: none, translational, translational and rotational together.

In calcs mentioned in original post it was set to none. In autopipe if you build operating case that contains pressure you have to consider translational Bourdon effect. So in caesar you can have case W+P1+T1 without any pressure effect while in autopipe not.

I simply wonder if that Germany originated company made it intentionally or by mistake. I know what is more conservative but I'm fresh in power I dont know common practices, but I do not want to be too conservative or dumb in terms of money.

regards

Papiesz

 

[SNORGY]

Papiesz,

Have you tried contacting COADE / CODECAD with this question? Every time I talk to them they seem very knowledgeable. I am sure they could give good advice to get you going in the right direction.

Me, I would be inclined (if I had a non-run-limited key) to enable Bourdon effects and adjust the system until it passed, and then look at the ramifications in comparison to the system that passed without considering Bourdon effects. Then, once satisfied that both results were verified, the piping design I would propose or recommend would be the one with the Bourdon effects taken into account, and I would let someone else who might not like it prove to me that I was wrong.

Sometimes it's like the "project engineer's" approach to things (which I was in previous years). Sort of like, the best way to get the client to freeze the plot plan is to mobilize the piling contractor. (Did that once...worked like a charm.)


Regards,

SNORGY.

 

[JohnBreen]

Bourdon effect is usually less than most analysts imagine. Very large diameter pipe bends with relatively thin wall thickness may see some additional strain due to internal pressure, but internal pressure in pipe necessarily leads to thicker walls in pipe, bends and elbows. L.C. Peng discusses this effect on page 25 of his new book "Pipe Stress Engineering". The bourdon tube that is used in a pressure gauge is manufactured with a rather exaggerated oval cross section (not just "a little oval") to exacerbate the "opening effect" for use in the measuring pressure instrument. Real world piping elbows that are forged or made from straight pipe have significant thickening at the bend (elbow) intrados and this and the general heavy wall used in high pressure pipe tend to ameliorate the so called "bourdon effect". Peng points out that tests on mitered elbows showed that (in the elastic stress range) the component tended to CLOSE with pressurization!!! So opening or closing strains due to internal pressure are generally considered insignificant and are not included in the stress analysis. The degree of actual ovalization that you start with is the key - if the percentage of ovalization relative to the ideal round cross section is small the bend rotation due to internal pressure ("bourdon effect" is really a misnomer) will be similarly small (perhaps to the point where it is not a significant issue in stress analysis).

L.C. Peng points out however that pressure stiffening is significant and that is why the B31 Codes have provided a "adjustment equation" to modify the flexibility factor for this effect.

Regards, John.

 

[SNORGY]

That does it...I have to get this book.

Thanks so much for this explanation, JohnBreen. Do you happen to know if L.C. Peng's book is available in Canada?

I'm heading straight from here to a Google search...

Regards,

SNORGY.

 

[SNORGY]

I just ordered my own personal copy from the ASME website. If it looks as good as everyone is saying, I will have my employer order another copy or two for the young mechanical engineers in my department.

Amazing the things I do on my day off...(Canada Day)...

Regards,

SNORGY.

 

[JohnBreen]

SNORGY

Do you happen to know if L.C. Peng's book is available in Canada?

Hey, if you order it on "Canada Day" how could ASME possibly turn you down. You Canucks better not try asking them to pay the shipping just because of "Canada Day" though - bunch of tight hosers eh?

John.

 

[SNORGY]

It's the cost of delivery by dog sled that hurts the most. Plus the impact on our financial credibility brought on by the striking colours of our often bi-metallic money. Hey, at least we *still have* a two-dollar denomination.

Regards,

SNORGY.

 

[papiesz]

Lucky you, gentlemen

You can buy this book directly from ASME store. I come from Poland and I would have to buy it via English resseler which is 1.5 times more expensive and in addition he does not respect Polish credit cards. World more and more resembles jungle to me.

By the way I read Mr Peng old paper about interpretation of pressure elongation from

http://www.pipestress.com

(I think he is one of the authors of book mentioned by John). As I understand him, his opinion is the same as John's.

On the other hand Mr Peng provides equation from which pressure elongation can be calculated. For my parameters: 23.4 MPa, 150 C, NPS 8" Sch 140, pressure elongation is about 13% of thermal. It is rough calcs but indicates that pressure could play significant role when assessing nozzle loads.


Regards

papiesz

P.S.

Thanks to all for valuable posts

 

[papiesz]

Hello again

Pressure elnogation is not 13% of thermal. I made mistake, it is roughly 1/13 which is 7%. So it is less significant.

I'm sorry. This is because here is 11 PM, and I focus too much on my English instead of the problem.

papiesz

 

[BigInch]

Take a closer look at primary stresses,

Bourdon pressure "expansion" results in contraction in the axial direction, which, if restrained, produces axial tension. As longitudinal stress is the result of Poisson's ratio x hoop stress, its much less than the hoop stress. Plotting these principal stresses on Mohr's diagram results in hoop stress (tension) far to the left and axial stress (also tension) between hoop stress and 0. The result is a lesser maximum shear stress than when Bourdon axial stress is not considered at all. Burdon "None" therefore results in a conservative design for a straight pipe, and going by what's been said above, for a curved pipe too. I would guess, probably from being the result of secondary effects.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[BrianCNTFM]

Papiesz I would support the argument for analysis. It may not be an issue if the pump asssembly has been designed robustly however you may find as I did that the restrictive limitations imposed for pump loads had not been taken into account.

10's of newtons as opposed to the hundreds exterted on chain blocks to align pipework! Check out pump manufacturers requirements.

Regards
Brain

 

[papiesz]

Big Inch,

That is not true.

There is not any contraction due to pressure. Such situation exists in uniaxial tansion. You can observe it on tensile test machine where you can observe cross section area reduction.
In case of pipe you have membrane stress and strain state in which both streses are tensile. And both strain try to lengthen material. If you dont believe just put untied expansion joint in your piping.

By the way read your first sentence:

"...pressure results in contraction in the axial direction, which, if restrained, produces axial tension."


Regards

papiesz

 

[papiesz]

BrianCNTFM

I agree with you. Although pressure forces are rather in magnitude of hundreds they are far lesser than thermal and from earthquake and of course those which can arise during not carefull assembly. But you dont know Germans they are more strict than API 686.

Regards

papiesz

 

[C2it]

Papiesz,

Strictly speaking, Big Inch is correct ... in that an element of internal pressure causes axial contraction. It is the poisson effect of hoop stress, which logically reduces axail length. This effect is however tiny in comparison with the the direct stain associated with axial pressure stress, which is of course tensile in the pipe wall with positive internal pressure.

Also w.r.t. Bourdon effect, I am fairly sure that Caesar II assumes an ovality ratio of 0.9 if the effect is specified. This is quite severe and would only normally be associated with pulled bends. The usual assumption with forged bends is that the Bourdon effect is aero in my experience.

 

[BigInch]

Poisson's ratio is 0.3, so contraction is 30% of hoop stress. End Cap Effect is 50% of hoop stress in tension - 30% of hoop stress, leaves 20% of hoop stress net axial expansion. Since axial stress is tension and hoop stress is tension, Mohr's maximum shear stress is reduced.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[MBlackman]

I think you are justified in running the analysis with pressure elongation switched on and using the maximum operating pressure rather than design pressure (for nozzle loads only), and running it with design pressure for the stress check.