Maximum hydrotest pressure

[miyanui]

thread378-182689

I have been reading the above referenced thread but can someone clarify how the maximum hydrostatic test pressure is determined.
I am planning to hydrotest a pipeline.
approx 2 miles long,
SMYS is X52,
ERW,
16" OD
Design pressure is 1460psig
MOP is 285psig
I have determined the min test pressure to be 1.25MOP (MOP is 285psig)but I am not sure how to calculate the max test pressure.

thank you

 

[SNORGY]

Check the Code of Construction and the local Regulations.

Otherwise, the maximimum test pressure will probably settle out at that pressure corresponding to approximately 100% of SMYS or 1.5 times cold flange rating, whichever is lower.

Unless you have significant elevation changes, in which case you have to achieve enough test pressure to prove your qualification pressure at the high point in the system.

 

[BigInch]

Something doesn't look right.
Your design pressure is EXTREMELY high for the MOP you have shown.
It is OK to do that, but it is generally a great waste of pipe wall thickness if you design a system like that. Maybe you need it for some reason that I don't see.

1460 is max design and max operating pressure for ANSI 600 class.
285 is max design and operating pressure for ANSI 150 class.
Normally with an MOP of 285 you would select ANSI 150 class, make the design pressure and max operating pressure both equal to 280 psig.

You also need to be sure that all pipe and fittings are designed for at least 280 psig and will withstand whatever test pressure you use.

The MINIMUM test pressure for everything in a liquid pipeline (ASME B31.4) system operating at less than or equal to 285 psig is 1.25 x 285 = 356 psig

Now check pipe, flanges and fittings.
The maximum test pressure for ANSI 150 class flanges is 1.5 x 285 = 427 psig
MAXIMUM test pressure for all pipes or fittings = 2 x WT x SMYS / OD and that should be higher than your selected test pressure.

I hate Windowz 8!!!!

 

[miyanui]

sorry for late reply, was away for a week on vacation,
Thank you BigInch, this is a pipeline that starts and ends at storage terminals but goes under the Houston Ship Channel, so we have ANSI 150 class flanges, thats why 285psig is our maximum operating pressure. I think that when the pipeline was built, it was designed as such without consideration of ANSI 150 within tank farm boundaries, hence the design pressure of 1460. but I am testing the entire system including the tank farm ANSI 150 and the #600 pound flanges outside of the tank farm.

"The maximum test pressure for ANSI 150 class flanges is 1.5 x 285 = 427 psig"
"MAXIMUM test pressure for all pipes or fittings = 2 x WT x SMYS / OD and that should be higher than your selected test pressure."

this 2 statements summarize my dilemma, during the actual test, we have always maintained pressure below 427 psig, why is this the case when "MAXIMUM test pressure for all pipes or fittings = 2 x WT x SMYS / OD and that should be higher than your selected test pressure."

the older hydrotest reports show this max test pressure to be at 100%smys but the actual test data max pressures never above 427psig.

so basically we have never tested this pipeline above 427psig but the reports keep indicating maximum test pressure criteria as 100% SMYS.

 

[zdas04]

I have never seen a test designed for 100% of SMYS. SMYS is Specified Minimum YIELD Stress. At SMYS the pipe can distort without any recourse (i.e., you can't go back to the pipe manufacturer and say "your pipe yielded at the SMYS, I want you to replace it").

Typically I see companies with limitations like "Normal operating pressure must be less than 20% of SMYS". With that I can use the AMSE B31 series equations to specify a wall thickness that won't exceed 20% of SMYS at normal operating pressure. Others set a maximum stress at MAWP (I've seen that number as low as 35% of SMYS). If my MAWP is 35% of SMYS and I'm testing to 150% of MAWP, then the max stress I'll see during the test is 52.5% of SMYS. I don't often see a "max stress during a test" number in company codes, but the ones I've seen have been around 70% of SMYS.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

 

[miyanui]

thank you Zdas04

I know what SMYS means, I do not mean that you test at 100% SMYS, its simply the criteria for calculating the internal design pressure
consider the equation below without any factors

P = 2 X SMYS X t/OD

100% SMYS simply means that if the pipe is X52 for example, the internal design pressure (P) was calculated using 52,000 or 100%SMYS

P = 2 X 52,000 X 0.312/16 = 2,028 psi (just an example of a 16" OD, 0.312 wall thickness, SMYS of 52,000)

 

[pennpiper]

According to the Flange Rating Chart (

http://www.scsenergy.com/scsdataflgrating.htm)

Class 150 flanges can only be Hydro tested to 450 PSIG.
(I hope this works, if not then go to the web site)

THIS INFORMATION IF PROVIDED FOR GENERAL INFORMATION PURPOSES ONLY. BEFORE USING ANY OF THIS DATA FOR ANY PURPOSE OTHER THAN FOR GENERAL INFORMATION, IT SHOULD BE REVIEWED AND CERTIFIED BY A QUALIFIED ENGINEER.

Pressure - Temperature Ratings

For Steel Pipe Flanges and Flanged Fittings
American National Standard ANSI B16.5 - 1988

CLASS 150 lb. 300 lb. 400 lb. 600 lb. 900 lb. 1500 lb. 2500 lb.

HYDROSTATIC
TEST
PRESSURE, PSIG

450 1125 1500 2225 3350 5575 9275
TEMPERATURE, F MAXIMUM ALLOWABLE NON-SHOCK PRESSURE PSIG
-20 to 100 285 740 990 1480 2220 3705 6170
200 260 675 900 1350 2025 3375 5625
300 230 655 875 1315 1970 3280 5470
400 200 635 845 1270 1900 3170 5280
500 170 600 800 1200 1795 2995 4990
600 140 550 730 1095 1640 2735 4560
650 125 535 715 1075 1610 2685 4475
700 110 535 710 1065 1600 2665 4440
750 95 505 670 1010 1510 2520 4200
800 80 410 550 825 1235 2060 3430
850 65 270 355 535 805 1340 2230
900 50 170 230 345 515 860 1430
950 35 105 140 205 310 515 860
1000 20 50 70 105 155 260 430

Ratings apply to materials.

SA-1051.2 SA-515-702 SA-516-702 SA-181-701.2 SA-350-LF2 SA-537-C1.13 SA-216-WCB2

NOTES:

1. For service temperatures above 850 F it is recommended that killed steels containing not less than 0.10% residual silicon be used.

2. Upon prolonged exposure to temperatures above 800 F, the carbide phase of carbon steel may be converted to graphite.

3. The material shall not be used in thickness above 2-1/2 in.

Flanges of ANSI B16.5 shall not be used for higher ratings except where it is justified by the design methods of the Code.

Ratings are maximum allowable non-shock working pressures expressed as gage pressure, at the tabulated temperatures and may be interpolated between temperatures shown.

Temperatures are those on the inside of the pressure-containing shell of the flange. In general, it is the same as that of the contained material.

prognosis: Lead or Lag

 

[BigInch]

zdas, you're just not using all that you can use. You don't see "100% SMYS" because it is still not an absolute maximum. 100% SMYS is not written down anywhere. And pipe distortion, as long as any distortion is limited to 2% strain, needn't be a concern. In fact, if you use the 2% strain criteria limitation given by B31.4 and hoop stress alone as the basis to calculate the test pressure via 2 * SMYS * WT / OD, then you might theoretically reach a test pressure that would equate to a stress = 102% of yield stress.

SMYS is not theoretical burst stress, it's the yield stress. Burst strength actually occurs after yield; when ultimate stresses are reached.

Miyanui,

"MAOP" under CFR Title 49, is calculated as the minimum test pressure reached at any point in the system / test factor. If the test factor is 1.25 and you hydrotested and maintained a minimum test pressure of 500 psig for a minimum of at least 4 hours, then MAOP cannot be greater than 500/1.25 = 400 psig.

"Design pressure" of a system is something engineers use only to target the design of a pipeline system to allow it to reach a certain working pressure, which would allow the system to be operated at least as high as the required (desired) MAOP. That working pressure must be limited to 2 * SMYS * WT * SF / OD for straight pipe or fittings (provided that the pipe and fittings were produced to a specification recognized by the design code, such as API 5L for pipe and ASTM A234 WPB for fittings) OR, if you have any flanges, or flanged valves in the system, then the design pressure would be additionally limited by the lowest rating pressure of any flange (in your case 285 psig). So you cannot have a design pressure above 285 psig, if you have ANSI#150 flanges. 1480 psig, if you have any ANSI#600 flanges, or 2160 psig if you have ANSI#900 flanges, etc.

That is why MAOP should be specified as THE performance criteria for a system to be designed, constructed, tested and delivered as a completed turnkey pipeline system, in lieu of (wrongly) specifying just a "design pressure". The design pressure might not be equal to the finally permitted maximum allowable operating pressure MAOP, if the minimum test pressure was not reached and held for a minimum of 4 hours.

All the above additionally assumes that you have no operating temperatures greater than whatever temperature the SMYS is given at for your pipe and fittings, or temperatures greater than whatever temperature the ANSI rating is determined at for any flanges in the system and that SF = 0.72 (as it would be for liquid pipelines, or gas pipelines in area class 1)

I hate Windowz 8!!!!

 

[miyanui]

A very hearty thank you gentlemen, for the education. Much appreciated.

 

[LittleInch]

miyanui,

My points echo the other posters, but for clarity, your system, like any other under test or operation, is limited by the lowest pressure rated item on that connected system - in this case your #150 flanges. If you could de-couple the pipeline from the lower rated section then you could have two pressure tests, but if this is not possible then you have got your maximum of 427psi

ZDAS04 - you haven't worked with National Grid designs where they often specify a "high level" test which sets test pressures at 105% of SMYS. I've never really liked it, but the theory is that it "blunts" cracks which otherwise could grow under repeated pressure cycling seen in gas transmission lines. They also put a stop if the pressure reaches half slope of the pressure versus volume of water in graph before the test pressure is reached as some part of the pipeline is starting to get a bit bigger. It kind of relies on the actual steel being supplied being of a slightly higher yield strength than the minimum and also being thicker than the minimum manufacturing tolerance which is added in these codes unlike the B31 codes. But it exists in a written code.

BI - the latest version of 31.4 437.4.1 states that during testing "special care shall be used to prevent overstrain of the pipe". This could be interpreted to be your 2% strain, but personally I would work to 100% SMYS at the lowest point myself. PD 8010 states that apart from this "high level" test the hoop stress should not exceed SMYS at the lowest point.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[SNORGY]

My reference to a maximum test pressure corresponding to 100% x SMYS was extracted from CSA Z662 Table 8.1 and Clauses 8.8, 8.9.

The flange rating usually limits for smaller diameter piping; for an MOP of 285 psig, I offer the bold assumption that the corresponding flange rating might possibly be that for ANSI B16.5 150#. In this instance, I did not perform the hoop stress calculation for 16"-OD x 0.375" (assumed) WT x 52 ksi SMYS to see if the test pressure is limited by flange rating or hoop stress, but I assume the former. So, my post kind of aligns with BigInch's, although his explanation is better and more thorough than the one I offered.

 

[weldstan]

We routinely hydrotested pipelines at a maximum of 105% of the SMYS on 72% SMYS designs taking elevation changes into account.
When designing to 80% of the SMYS the minimum test pressure is 100% of the SMYS. Taking elevation changes into account, one should expect to test at 110% of the SMYS max.

 

[LittleInch]

weldstan - Did you routinely end up with a slightly bigger pipe at the end of the test?

Joking aside did you ever see half slope on the test pressure versus test water volume chart? I'm interested to know if in reality it rarely happens for your max 110% SMYS test.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[weldstan]

LittleInch,
No.

In the USA, take a good look at the elevation differences in mountainous locations, availabilty of hydrotest water and disposal thereof, and costs of numerous hydrotests.

 

[miyanui]

I can imagine very long pipelines with significant elevation differences. differences in pressures at the lowest points will be significant. Thank goodness our pipelines are only few miles long and less than 150ft elevation change.

 

[LittleInch]

Yup, elevation changes are a complete b*tch wen you start testing in the mountains alright... I've got one now which has something like 40 sections in 800km due to it. Maybe I'll try the pneumatic test angle and see what happens...

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

At Northern Natural Gas back in the 70-80s we used to test to 110%, while keeping both eyes on the pressure vs water volume plot. I had to suggest to them that they stop doing that for offshore pipelines, which they did.

Lil"
In section 403.3.3 Strain Criteria for Pipelines, 2%, for longitudinal strains is mentioned. True it says it applies to longitudinal stress, not hoop stress. But I've applied that to hoop stress as well, probably for no particular reason, other than to limit it to something. Otherwise weldstan might keep bumping us up higher and higher . As before, there is no specified limit. As far as the code is concerned, you can do a burst test, if you don't mind cleaning up a big mess.

Take care as well. As always, there are (at least) two schools of thought, one, as you mention, that overstress is actually beneficial, but the other says that all strains are additive and some amount of permanent damage was done. Know what the client's preference is ...beforehand.

I hate Windowz 8!!!!

 

[miyanui]

http://www.ogj.com/articles/print/v...ctiveness-in-line-inspection-complements.html

this article looks interesting