In service pipe thickness limits

Good answer.

Fully agreed.

In every case you should calculate minimum wallthickness of a pipeline.

Because 12.5% reduction is a approximatly reduction from NDT personnel who

don't like to calculate.

refer to pipe schedules and type of corrosive service. They will advise the wall thickness for the use of steel and other metals with nested curves for expected years of service with a safety factor at various pressures.

It is cheaper to use steel and schedule it than buying totally non corrosive pipes. Then you replace it in toto after the life is done

Engineer /Abdel Halim Galala,

is true when I say this :

the minimum acceptance thickness for in service pipe should not be lass than the Nominal thickness (as per the pipe sch ) minus Tolerance minus corrosion allowanace?

actually I used to take the actual thickness(in service pipe) and compare it to the Nominal thickness -0.125-C.A.

If the thickness of in service pipe Less than (Nom thick-0.125-CA) then I will go for cut and replace

is this true?

Thanks alot Eng Abdel Alhalim

Dear Mr. Galala and all other participants, hi

Its perfect what calculations you did present here for determining the wall thickness for pipe according to ASME B31.3.My question here is that can we perform the same steps [find tmin, then find t by using tm/(1 - Mill_Tol%) ] for determining the pipeline wall thickness for ASME B31.8? ASME B31.3 gives formula to find MINIMUM WALL THICKNESS, but in ASME B31.8, we see the formula for finding NOMINAL WALL THICKNESS (tnominal = PD/2*S*E*F*T ). Or simply, how can I calculate the Minimum Wall Thickness for ASME B31.8?

I used to do the same steps for ASME B31.8 which you mentioned here for ASME B31.3 but one of my colleague told it wrong for ASME B31.8. He told me not to include Mill Tolerance [t = tm/(1-c%) ] with ASME B31.8 as it gives formula for nominal wall thickness, not for minimum wall thickness. According to him, once we find tm, we only have to look at standard tables (ASME B13.10) for next larger standard available thickness as we have enough margin for accomodating the effects of Mill Tolerance. I am myself not convinced of his comment but unfortunately, i cannot just find the "Minimum Wall Thickness" formula for ASME B31.8.

Would highly appreciate if you give explanation like you did for ASME B31.3 case. I need example as ASME B31.8 does not give sample calculations like ASME B31.3 App. S.

Please help me, and Thanks in advance.

Dear AlBaigMughal,

The pipe mill tolerance is a request and demand by the code governing the pipe material specifications, and that tolerance is due to the inaccuracy of pipe production process at mill (manufacturer). And that tolerance is ±12.5% as found at codes of ASME or ASTM for material SA 106 or A 106. See the following page which extracted from ASME SA_106, where you can find the equation:

t_n x 0.875 = t_m

where, t_n = nominal (average) pipe wall thickness, and t_m = minimum pipe wall thickness

Therefore, when I use that material of ASME SA 106, whatever the applied code B31.3 or B31.8, I have to apply that mill tolerance to avoid the worst case in reduction of thickness when it becomes -12.5%.

Dear Sir,

I am sorry but I think I didn't make myself very much clear. Following is my real case which might help me make the experts clear my problem:

GIVEN CONDITIONS AND DATA:

Pipe Material = API 5L X52 Seamless

NPS = 18

Service = Location Class 1, Division 2

Corrosion Allowance = 4 mm

Mill Tolerance = +0% ~ -12.5%

SMYS = 52,000 psi

Design Pressure = 1885 psig

Design Temperature = 230 Deg. Fah

FORMULA:

According to ASME B31.8 - 2007, 841.11 Steel Pipe Design Formula,

P = 2 S t F E T / D

Re-arranging for t,

t = PD/(2SEFT)

where t = Nominal Wall Thickness, (as opposed to what ASME B31.3-2008 gives which is Minimum Wall Thickness tm = t + c, according to 301.1.1 Straight Pipe General)

SOLUTION:

Now, according to my case, as defined in "GIVEN" above,

P = 1885 psig

D = 18 in.

S = 52,000 psig

E = 1.00 (Table 841.115A)

F = 0.72 (Table 841.114A and B)

T = 1.00 (Table 841.116A)

t = (1885) x (18) / [ 2 x 52,000 x 1.00 x 0.72 x 1.00 ]

= 0.4531 in.

So the Nominal Pipe Wall Thickness is

t = 0.4531 in.

Adding the Corrosion Allowance,

t = 0.4531 + Corrosion Allowance

= 0.4531 + 4/25.4

= 0.6106 in.

= 15.51 mm

My QUESTION starts from here: Do we need to divide this value of t (0.6106 in or 15.51 mm) by (1-Mill_Tolerance%) ? i.e., (100% - 12.5% = 87.5% or .875 ) keeping in view that we have already found the Nominal Wall Thickness, not the Minimum Wall Thickness?

t = 0.6106 / 0.875

= 0.6978 in.

= 17.72 mm

Is the above step of dividing the Nominal Wall Thickness by 0.875 really required? Once we have calculated the Nominal Wall Thickness using formula, what we are going to calculate by dividing it with 0.875?

According to my colleague, we simply have to look at standard tables of ASME B36.10M for the very next larger value after 0.6106 in. or15.51 mm, which is .625 in. or 15.88 mm. We don't have to include the Mill Tolerance Step and so, we don't have to see the next larger value after 0.6978 in. or 17.72 mm which is 0.75 in. or 19.05 mm. This is the point which I want you experts to clear whether the understanding of my colleague is right or we should get the mill tolerance step included in our calculations which leads to the greater wall thickness and so more cost. He seems to be right because the standard ASME B31.8 gives NOMINAL WALL THICKNESS FORMULA, not the MINIMUM WALL THICKNESS as ASME B31.3 does.

Yes, the method of calculation is OK.

But note that the tolerance ± 12.5% is based on ASME material SA 106, but for material API 5L X52 the tolerance is differ, please refer to the following Table 9 of API 5L "Spesifications for Line Pipe":

..........................................

From Table 9, for NPS 18, the tolerance shall be + 15% & - 12.5%.

Note. The worst case for reduction in thickness in both materials (ASME SA 106 & API 5L X52) are - 12.5%, which indicates that your method of calculation is correct.

Thank you very much Mr. Galala for the correction of Mill Tolerance Values.

Its mean we should take into account the Mill Tolerance to calculate the Wall Thickness for Pipeline too, just like process piping according to ASME B31.3?

Yesterday, I also went through the Canadian Standard CAN/CSA-Z184-M86 Gas Pipeline System which gives the same formula to determine the pipeline wall thickness as ASME B31.8 does give, but the Canadian Standard defines the t as "Design Thickness" not "Nominal Thickness", as ASME B31.8 does. I think the definition of "t" should be corrected in ASME B31.8 as at present, it makes a lot of confusion by telling us it is "Nominal Thickness".

Anyways, Thanks alot for your kind help and time.

AlBaigMughal

You are welcome AlBaigMughal,

And pipe wall mill thickness tolerance is a request and demand by the standards and codes of material, and not necessary to be included at code of design.

you can go to www.standardshop.org
there sale a lot of asme aws api standard and so one .can immediately download,and very cheap low price!

regarding ASME B31.8 (and ASME B31.4) for underthickness tolerance, isn't it already covered by the design factor?

in ASME B31.8, state in para. 841.1.1 (a) regarding F= design factor "... due consideration has been given and allowance has been made for the various underthickness toler ances provided for in the pipe specifications listed..."

ASME B31.8 para. 804.5, "... nominal wall thickness, t: the wall thickness computed by or used in the design equation in para. 841.1.1 or A842.2.2(a) in Chapter VIII. Under this Code, pipe maybe ordered to this computed wall thickness without adding allowance to compensate for the underthickness tolerance permitted in approved specifications."

ASME B31.4 para. 403.2.2, "... Design factors in this Code were established with due consideration for underthickness tolerance and maximum allowable depth of imperfections allowed by the referenced standards; no additional allowance is necessary."

But for ASME B31.3, underthickness tollerance still need to be consider in determine minimum thickness, as per para. 304.1.1 (a)

The 12.5% is typically a manufacturing tolerance for some pipes and therefore would be allowed as such for new installation, relative to the nominal thickness. Strictly speaking, however, not all manufacturing tolerance is 12.5%. It depends on the applicable ASTM specification of the pipe. Refer to the proper ASTM standards applicable for the pipe and verify the manufacturing tolerance. Although 12.5% is considered conservative.

The criteria for allowable thickness during its service life is the calculated net thickness at corroded condition. Therefore the service corrosion allowance must be known. The minimum allowable service thickness therefore would be calculated as the nominal thickness minus the manufacturing tolerance minus the corrosion allowance, for butt welded piping. For pipes that are threaded and grooved ends or similarly prepared end conditions, the mechanical allowances are also subtracted to get the net thickness. Each pipe size has a different mechanical allowance. Consult the piping engineer who has Jurisdiction. It would be proper to confer with him with this advice you are getting. It is like getting a second opinion, at least.

After determining the minimum allowable net thickness as above, this would be the basis of comparing the measured in service thickness, such that all measured service thickness greater than the minimum allowable net thickness is still acceptable for continued service. Otherwise, the pipe needs to be replaced, since it has reached the end of its service life.

"The 12.5% is typically a manufacturing tolerance"

We have to differentiate between Mill Tolerance and Manufacturing Tolerance. Mill tolerance is that tolerance at Pipe Mill (Mill is that factory for producing pipes), where manufacturing tolerance may be means that tolerance while piping fabrication at workshop.

Please refer to the following CR4 Thread: Significance of "-Mill Tolerance" in the Piping Stress Analysis.

it depends also on the Corrosion Allowance of the pipe. If the pipe is in carbon steel then it should have a corrosion allowance, have a look on the piping class on the P&ID. as a first criteria if you have a corrosion above the corrosion allowance then you should change the pipe. or you could do an RBI analysis or a fitness for service analysis (API 579)

corrosion prevention & protection

Mill tolerance is differs from corrosion allowance, and in any case we have to take the both (mill tolerance & corrosion allowance) into consideration, and in no way they depend on each other.

As already stated, the minimum wall thickness has to be calculated before purchasing the pipes by the authorized personnel.

When you make any wall thickness calculation (regardless pipe, vessel, valve...) it's really unusual you get just a "standard available thickness". Normal case is to round off to the nearest high standard thickness. But in any case the minimum calculated wall thickness remain the same and it's the value against what you must make comparisons of UT measurements results.

If measurement is lower than minimum calculated wall thickness then repair or replace the pipe.

There are even software to store the successive measurements and to predict the remaining lifetime of the pipe before it reaches the minimum thickness.

BTW, and don't take this as agressive... What thickness acceptance value do you use for UT measuring?

Kind regards

Hello Pitting depth is the concern, as well as wear. The defect will be convex meaning under neath the surface. 12.5% is hugh limit. I used the inspect to .0030. You can visual the wear pattern and or make a wax impression. Remember 100 %,75%,50%,25%, then half that to 12.5%. Former UT inspector of aircraft engine parts.

The problem with steel piping and corrosion is the formation of small local cells that create deep pits. Once they hit the surface you can get a leaker or a squirter.

When I was in the plany business, we had reams of corrosion schedules for different pipes, fluids and temp/pressure that took a conservative approach. There were parametric lines for 5, 10,15,20 years that you had to stay within.

Perry's Chemical Engineers handbook has some of them and points you to where to find more at the various societies that have accumulated tyhis data.

First of all thank you guyz for replying in my post, before when i was evaluating my the data in UT testing my basis for minimum wall thickness is -12.5% from the nominal thickness of the pipe.. So I think it is in correct for in service pipes...

After reading API 579-1/ASME FFS-1, to assesed the if the pipe is still serviceable you should compare the original design pressure to the new maximum allowable operating pressure that have been computed base on the new wall thickness from the UT thicknesss gauging you have conducted..

So do you think is this correct?.. Tnx again

For general wall loss you should still look to the original code of construction, whether that is B31.1, B31.3, ................

FFS (API 579) is more suitable for other defects such as localized corrosion, buckling, cracks, operating in creep range, ....... that are not explicitly covered in the original codes - and is generally used as a "last resort" to justify running equipment longer until it it can economically be replaced (i.e. not having to shut the entire plant down for a single vessel/pipeline).

(Having said that, and opening up the conversation for debate, have you specifically reviewed ALL of Section 4 and Appendix A of API 579?)

Section 7 "Inspection Data Evaluation, Analysis and Recording" of API 570 "Piping System Repair, Alteration, Rerating and Pressure Testing", is very interesting in evaluation of inservice piping systems. I have a solved example for determining the remaining life of a piping system, can be derived upon your request.

the final decision, in my opinion, shall be based on a risk assessment. It depends on what are the consequence of a failure, what is the fluid.... sometimes the leak before break criteria is also used.

What kind of fluid is inside the pipe?

S

corrosion

@ Abdel Halim Galala

Tnx again for replying...

about the example can u email them to me.

@ Guest.

I have read Section 4 of FFS but the problem is the appendix A, my e copy is corrupted so i cant read some parts of it, if you can send a copy of appendix A(Equations for determining MAWP, thickness, and stress in components), i would appreciate that or u just can brief me.. tnx again