Pressure Testing

You may be interested in the post of "robsalve" and others at http://eng-tips.com/viewthread.cfm?qid=122175 .

"please let me know if there is a formula for calculating the energy that is stored in a given diameter/length of pipe work when pressure testing it with compressed air or hydrotesting,"

Compressed air is usually used for leak testing or finding pinhole leaks. Air or other gas should never be used for stress testing. It is a hazardous and unnecessary risk.

Hydrotesting will only result is spilled water or other test fluid in the event of failure. The stress and or stored energy in the pipe can be calculated but not usually done when hydrotesting as it has no bearing on the test results.

Please observe very carefully testing of pipe lines with air. May I add that besides testing there is a common method of multiple use pipelines particularly for fuels of all description and densities.

The method consist of using what is known as a " pig" . This is a metal piston shaped tool with adequate piston rings made of various material which is inserted at one end of the pipe through a box (which is fitted , similarly ,at both ends) enabling the tool to be inserted at either end of the common fuel pipe lines. Air pressure is then applied thereby moving the piston internally . This action cleans the pipe line and moves residual fuels to the opposite discharge point for disposal.

There have been many accidents some fatal with this technique when carried out by uninformed or : strictly disciplined personnel. Some of the causes can be attributed to work pratices and some due to internal faults of the pipe line.The latter generally caused by molten metal deposited by the high penetration of the external welds - thus blocking or jamming, the pig half way with corresponding build up of air pressure .This coupled to poorly trained attendants are a recipe for disaster. Yes , using air pressure without adequate protection is equal to dynamite !

Labor Omnia Vincit

I am aware about the dangers of testing with air and the reason we have to use air is that the product line cannot be "pigged clean" if we were to use water to test the pipeline, water would therefore contaminate the product and render it useless.

My question was and still is: is there a mathamatical formula for calculating the energy stored in the pipe while the "test" is on, for example: length x width x depth x 6.25 to give the gallonage??

If there is, could someone please reply with one for water please too.

m3bmw, the danger of testing with air/gas already was mentioned. Nevertheless there are equations for claculating the enclosed energy of course: have a look into the basic lessons of thermodynamic. First there is the famous equation for describing any gas' actual conditions:

p/ρ = m R T [ energy result in: kg m2/s2 = J ]

p = pressure in Pa (=kg/(m s2)); ρ = specific Weight of gas in kg/m3; m = mass in kg; R = gas constant= 287 J/(kg K); T = absolute Temperature in Kelvin (K).

Then Boyle-Mariott has teached us, that following rule is valid for gas under static conditions:

p1 V1 = p2 V2 [V = Volume im m3];

This way you can evaluate the expanded volume of an compressed gas (vice versa).

This I think should help you to do the evaluations wanted.

Albert

I used to leak test large [up to 60" diameter] flow control valves. The pressures on some units were as high as 1800 psi. The worst accident I ever saw was looking between 2 flanges & seeing the o-ring seal right before I got soaked from the resulting "leak"!

We would occasionally have a hard to locate leak & use air pressure & soapy water. We would never exceed 1/10th of the hydrostatic test pressure! worked like a charm.

If the writer has access to a fairly recent edition of 'Perry's Chemical Engineers' Handbook' , then in the chapter on 'Process Safety' he will find equations for estimation of damage effects for gas filled vessels and liquid filled vessels, including energy release and TNT equivalent.

I say fairly recent because I have just acquired an 8th edition and the relevant pages start at page 23-66. I think the 7th edition had this chapter in, but I know my old copy of Perry's didn't.

The following paragraphs are extracted from pressure piping code ASME B31.3, and I like to say that there is no problem with leaktest or hydrotest using gas or air if you take the necessary precautions as ASME noted:

Quote

The fluid used in hydrostatic test shall be water unless there is the possibility of damage due to freezing or to adverse effects of water on the piping or the process. In that case another suitable nontoxic liquid may be used. If the liquid is flammable, its flash point shall be at least 49°C (120°F), and consideration shall be given to the test environment.

A pneumatic test would present an undue hazard of possible release of energy stored in the system. A preliminary pneumatic test using air at no more than 25 psi gage pressure may be made prior to hydrostatic testing to locate major leaks. A leak test shall be maintained for at least 10 min, and all joints and connections shall be examined for leaks.

Pneumatic testing involves the hazard of released energy stored in compressed air or gas. Particular care must therefore be taken to minimize the chance of brittle failure during a pneumatic leaktest.

A pressure relief device shall be provided, having a set pressure not higher than the test pressure plus the lesser of 50 psi or 10% of the test pressure. The gas used as test fluid, if not air, shall be nonflammable and nontoxic. The test pressure shall be 110% of design pressure.

Hydrostatic test pressure, P_T = P (S_T / S), where

P_T = minimum test gage pressure, P = internal design gage pressure

S_T = stress value at test temperature, and S = stress value at design temperature.

Unquote

Thank you very much Abdel Halim and Albert Casper for the detailed answers. I am also thankful to other respected participants with their input. It was very helpful.

I always wanted to know but was afraid to ask. We do the flare line testing with Air and rest of the testing is done by water in the plant for Vessels and exchangers and I like to keep the duration to an hour. Once the ASME inspector told me during the inspection of the testing at the fabricator's facilities that if it does not leak within first 10 minutes, it will not. So it is not necessary to hold the pressure for 60 minutes or take notes for any decrease in pressure. I guess he was right to hold it for 10 minutes or he was in a hurry to go some where else so he terminated the test after 10 minutes. After that I was not very rigid in keeping the duration for 60 miniutes.

Regards;

Nadeem

04182008

Thank you Nadeem0430,

Please don't hesitate to ask any question.

For flare system, it is a gas line in a process plant, and the ASME piping code has no objection to proceed the test pressure by air or gas. To the extent, in many cases where it is difficult to proceed the hydrotest by a liquid, you can proceed a pneumatic test in lieu of hydrotest as per code recommendations. The test of rest piping work at the plant is carried out by hydrostatic.

With respect to test duration, the ASME code states a 4 hours min. as mentioned at ASME B31.4 for pipelines. But, depending on the inspector, the duration can be extended. In pipelines, we carried out hydrostatic tests for 24 hour, using a pressure-time recorder.

For plant piping like as ASME B31.3, I agree your inspector with the duration not to be less than 10 minutes.

Please refer to the following CR4 Threads related with hydrostatic test and the difference between: hydrostatic mill test duration for pipes (before fabrication) and field hydrostatic test for a plant piping : Hydrostatic Tests & Piping Pressure Design.

Sorry, there is type writing mistake for the eqn. P_T = P (S_T / S), the correction is: P_T = 1.5 P (S_T / S). [Reference: ASME B31.3, Para. 345.4.2(b]

I have question for hydrotest pressure from ASME B31.1 code.

As per ASME B 31.1 code,Pt=1.5*Pd

Pt-Hydrotest Gauge pressure

Pd-Design Gauge pressure.

We know as per code,

Maximum allowable stress at test condition increase upto 90%of yield stress at test condition.Maximum allowable stress can be taken as 2/3 of yield stress at design temp.

For example if we consider carbon steel A106B pipe ,Design temperature-200 degree celcius,as per ASME B 31.1 ,Appendix A,maximum allowable stress for design and test condition is same,Since same allowable stress values available for temp upto 343 degree celcius.Also yield stress value will be same for temp upto 343 degree celcius.

So if we take pipe thickness say"t" used for design and test condition,the relation between Test Pressure and Design pressure come as

Pt=1.35*Pd(which is less than code requirement)

(Here maximum allowable stress at test condition will be 0.9*yield stress and maximum allowable stress at design condition will be 2/3*yield stress )

But ASME B31.1 code suggest minimum hydrotest pressure as 1.5*Pd.So in such condition pipe thickness will not be sufficient to hold test pressure.

So should I do recheck for each pipe thickness for hydrotest condition???

As per ASME B 31.1 code, P_t = 1.5 * P_d * (S_{at test temp.} / S_{at design temp}.) see Para. 137.4.5 Required Hydrostatic Test Pressure. And by substituting for S at different temperatures from Mandatory Appendix A (Tables A-1 up to A-9) of ASME B31.1, the P_t shall at least = 1.5 P_d.

But as per Para. 102.3.3(B), during hydrostatic test there is a limitation to be taken into consideration where the hoop stress shall not exceed 90% of the yield strength at test temperature. In this case the P_t may be less than 1.5 P_d.

102.3.3(B). During pressure tests performed in accordance with para. 137, the circumferential (hoop) stress shall not exceed 90% of the yield strength (0.2% offset) at test temperature. In addition, the sum of longitudinal stresses due to test pressure and live and dead loads at the time of test, excluding occasional loads, shall not exceed 90% of the yield strength at test temperature.

137.4.5 Required Hydrostatic Test Pressure. The hydrostatic test pressure at any point in the piping system shall not be less than 1.5 times the design pressure, but shall not exceed the maximum allowable test pressure of any nonisolated components, such as vessels, pumps, or valves, nor shall it exceed the limits imposed by para. 102.3.3(B). The pressure shall be continuously maintained for a minimum time of 10 minutes and may then be reduced to the design pressure and held for such time as may be necessary to conduct the examinations for leakage. Examinations for leakage shall be made of all joints and connections. The piping system, exclusive of possible localized instances at pump or valve packing, shall show no visual evidence of weeping or leaking.

But as per Para. 102.3.3(B),you have mentioned Pt may be less than 1.5*Pd.

Also mentioned as per para 137.4.5 Required Hydrostatic Test Pressure. The hydrostatic test pressure at any point in the piping system shall not be less than 1.5 times the design pressure.

As code does not allow Pt to be less than 1.5*Pd.We will not hydro-test any pipe below this pressure.

My question:-In case of Pt less than1.5*Pd,Does Pipe thickness have re-check for Pt=1.5*Pd.

The hydrostatic test P_t shall be 1.5 P_d per the code unless the calculated hoop stress during hydrostatic test condition exceeds 90% of the yield strength at test temperature. If you found that hoop stress exceeds 90% of the yield strength of the material, hence you can use P_t less than 1.5 P_d.

Hi all,

I agree with all considerations made about safety issue of pneumatic testing, but all codes allow this test when Hydrostatic test is impracticable and taking all safety measures (relief devices, smooth pressure rising, thorough inspection at lower than test pressure....)

But, as explained very well in post nº 6, the energy stored is the difference between initial and final status. If you assume the piping system is constant, then initial status is 0 bar (psi) gauge pressure and final is test pressure.

Energy stored is (PV)_f-(PV)_i → P(N/m²) x V(m³)→ E(N.m)= E (J)

and as V_i=V_f → E (J)= P_f-P_i (Test pressure)

Regarding the equivalence with explosives, here you have a formula:

W_e=4,63.10^-8 .P_i.V.log_e(P_i/14,7), in which

P_i is the internal gauge pressure (psia)

V is the system volume in cubic feet

W_e is the equivalent weight of TNT in tons

To calculate the stord energy is as follows;

SE = VOLULME X PRESSURE