Natural Gas Pipeline Hydrostatic Pressure Testing

[bonkers]

Does anyone know the reasoning behind natural gas companies hydrostatically testing main gas lines with the maximum test pressure being the lower of 103% SMYS of the lowest strength pipe or weld fitting, 1.5 times the flange rating or the valve shell test pressure? I've also seen 110% and 112% SMYS. I thought you were to never exceed 100% SMYS. What am I missing here? Any and all ideas will be appreciated.

Thanks,
Bonkers

 

[bimr]

Refer to page 4 in the link:

http://kiefner.com/downloads/apihydro.pdf

http://www.pipelineandgasjournal.com/pipeline-hydro-test-pressure-determination?page=show

 

[zdas04]

I think you are a bit confused. SMYS is used in the calculation of MAWP. Test pressure is a multiple of the MAWP that you select, NOT the SMYS. Pipe will generally have a calculated MAWP higher than flanges, so you use the flange MAWP. If you put a component on the system with an MAWP lower than the flange rating, then that becomes the system MAWP. It is up to the Engineer to set the system MAWP at an appropriate and safe level.

Once the MAWP is defined then the codes have rules for the multiple of the MAWP that you use to set the test pressure. For cross-country piping (B31.8 for natural gas) the multiple is based on population density--the more people affected the higher the test pressure.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[BigInch]

The code B31.8 does not actually place any limitation on maximum test pressure, provided that the test medium is not gas or air although the link provided by bimr goes through the logic of why some limit must obviously be introduced and what it should be.

zdas, If I can offer some clarification, your, "Pipe will generally have a calculated MAWP higher than flanges" isn't logical. There is (usually, maybe always) no reason that pipe should have a design pressure greater than flanges. Flanges and Pipe both have the same scheduled wall thickness and can be bought with the same material strengths, so why buy something different. It would only waste material of one or the other. Most efficient use of material would be at the point where both have equal wall thicknesses and strengths.

With very long pipelines it is possible to order a custom wall thickness directly from a pipe mill that could be different than those of the standard schedules, however standard flanges of std scheduled wall thicknesses would (normally) still be used, which of course must be of an equal or greter pressure rating than the pipe. So, it would be the flanges that would usually (if not always) have a higher pressure allowable than the pipe, not the other way around as you have said.



If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[zdas04]

BigInch,
What the heck are you talking about??? If I do the B31.8 calculations for a given location I always get a number that is well under the actual wall thickness. For example, if you look at the next to last column of Cameron for Welded and Seamless Wrought Steel Pipe then for 6-inch standard wall you get an Allowable Working Pressure of 1210 psig at 100°F. If I do the B31.8 calcs for Grade B in a rural location I get a number very close to that when I solve the thickness equation for pressure and put in actual wall thickness (less a corrosion allowance).

The vast majority of engineers in the world will mate 6-inch S40 pipe with ANSI 300 flanges without hesitation. ANSI 300 flanges in Carbon Steel Material Group 1.1 are rated at 51.1 bar(g) for temperatures below 38C and above -29C, so call that 740 psig. That is less than 1210 psig according to my cheat sheet, but no one would ever question a decision to put Standard Wall 6-inch with ANSI 300 and call the whole system MAWP 600 psig. I see analogs of that decision all over the world every single time I look.

Yes, custom pipe is available. There are even jobs that warrant the extra lead time. They are rare and really not germane to most discussions.

"Most efficient use of material" is a stupid bugbear that almost never makes sense. What makes sense almost always is "Most efficient use of capital and time". I can purchase standard wall ERW 6-inch (for example) anywhere in the world in just about any quantity that I want for commodity prices--there just isn't much difference in price from one supplier to the next in a given geographic location. If I want to match pipe capability to flange capability I have to go to the mill and pay published prices. I don't know what your experience is, but I have found that the big supply houses get a discount off published prices and I can almost always purchase standard stuff from them for less than I'd pay the mill--very poor capital efficiency to try to get material efficiency.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[BigInch]

If you are using ANSI 300 you don't really need more than a DP of 740 psig, do you, yet you have selected pipe for 1210 psig based on the Cameron book. 39% used wt, 61% overkill and wasted.

I'm talking about 38.2 MM$
DP = 1000 psig
OD 36"
Length = 320 miles
DF = 0.72
API 5L X60
Min Req'd wall thickness for pressure continment = 0.417"
Closest scheduled pipe is 20, wt = 0.500
Mill ordering saves me 26,400 tons of steel, 38.2 MM$
pipe design pressure for 0.417" is 1000 psig.
pipe design pressure for 0.500 is 1200 psig
BOTH LESS THAN the ANSI 600# flange rating of 1460 psig.

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[zdas04]

If I'm ever doing a $150 million, 36 inch, 320 mile, 1000 psig operating pressure job, I'll pay attention to that, saving 20% or so is worthwhile. There are 0-2 of those in any given year somewhere in the world.

Do the same arithmetic on 6-inch for 10 miles operating at 350 psig. Or even 12 inch (Cameron book calls standard weight 860 psig).

My point is that few people who are asking basic questions where they confuse SMYS and MAWP will be designing systems of the magnitude where a special mill run makes sense. When dealing with standard stuff, the trade off is obvious--getting S20 6-inch is a special mill run, getting S40 is off the shelf, I'm going with S40 every single time and call the extra steel a cost of doing business.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[BigInch]

On our pipelines we find it more efficient to reduce pipe design pressure to at or below a flange class pressure rating rather than oversize a pipe wall thickness and pay for it mile after mile. Obviously our suppliers are more accomodating. In your case we would buy higher class flanges ratings and increase the design pressure to maximum allowed by the pipe wall.

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[bimr]

Inside of a plant battery limits, the piping systems may be designed so that the flanges are the lowest rated element but it's not always true.

When operating at higher pressures and for cross country pipelines, limiting the pipeline pressure based on the flanges may be an expensive option and doesn't make any sense.

This is also off topic to the original poster.

 

[Latexman]

Yes, but when the elephants dance the mice scatter.

Good luck,
Latexman

 

[rmw]

I think you are both right - in the universe that each of you operate in. For me, I don't operate in a world where projects require pipe in mill run quantities so I have to deal with commodity pipe available at Joe's bait shop and pipe yard. Often that means I will specify std wall (in CS pipe) when the DP would call for S10 or S20. Involved with that same project may be fittings and flanges that number in quantities measured by the dozen but less than two dozen. S10 and S20 fittings in those quantities are "unobtanium".

I can be a pedantic engineer and insist on S10 or S10 because that is all the MAWP calls for, and pay for a mill run to get the few pieces I need or expect a supplier request to supply what can be obtained in his world - did one of those just this week for S10 to std wall for a project designed by a purist.

Alloy pipe now, well we adhere to the schedule required and no more, but guess what - so does everyone else, so S10 and S20 schedules in pipe and fittings are generally more available at competitive prices in our area.

So I agree with both of you.

rmw

 

[BigInch]

I admit that find it difficult to let wide, general statements just pass on by when I know that there is at least one other important prespective that isn't being considered.

Bonkers, sorry for straying from your topic.

If it ain't broke, don't fix it. If it's not safe ... make it that way.