ASNT Level Ill Leak Testing Method Chapter 1 Management of Leak Testing Introduction (LEAK TESTING is the branch of nondestructive testing that concerns the escape or entry of liquids or gases from pressurized or into evacuated components or systems intended to hold these liquids. = Leaking fluids (liquid or gas) can penetrate from inside a component or assembly to the outside, or vice versa, as a result of SIBRESSURSIMEEEHHE] between the two regions or as a result of peHnieation thYOUGH a somewhat extended bartier. C1 Leak testing encompasses procedures for one or a combination of theffollowing: = Locating (detecting and pinpointing) leaks. = Determining the rate of leakage from one leak(or from a system) = Monitoring for leakage. CO The applications of leak testing are very diverse because they are used In many industries, such as nuclear, aerospace, chemical, ¢lectronies and autom@tive, to name a few. complicates the subject because each field has its own special techniques and technical languages = Fortunately, the various techniques aréibasedion similar, familiar principles that provide a basis for understahding dbrevieWing the subject. CO Other nondestructive test methodspsuch a8 radiography, ultrasound, magnetic particle and penetrant testifig, are usually performed on raw materials or welds to ensure structural integrity, 1 Leak testing is frequently pétforitied on a finished assembly to ensure the leak tightness asa finished product. = The tecfitique does;not necessarily ensure structural integrity except to the extent that atmospheric pressure exerts force. Functions of Leak Testing Leak testing is a form of nondestructive testing used in either pressurized or evacuated systems and components for detection and location of leaks and for measurement of fluid leakage. Ol The word leak refers to the physical hole that exists (with some equivalent length and internal cross-sectional area or diameter) and does not refer to the quantity of fluid passing through that hole. in leak testing, the quantity used to describe the leak is the measured leakage rate. 1[Page= Aleak may be a crack, crevice, fissure, hole or passageway that, contrary to what is intended, admits water, air or other fluids or lets fluids escape (as with a leal roof, gas pipe or ship). 1 The word leakage refers to the flow of fluid through a leak without regard to physical size of the hole through which flow occurs. = Fluid denotes any liquid or gas that can flow 1 Like other forms of nondestructive testing, leak testing has a great impact on the safety and performance of a product. = Reliable leak testing decreases costs by reducing the number of reworked products, warranty repairs, and liability claims. Need for Leak Testing Cl Leaks are special types of anomalies that can have'tremendous importance where they influence the safety or performance of engineered systems. ity of many devices is greatly reduced if enough leakage exists. = The operational relial Leak testing is performed for three basic reasons: 1. to prevent material leakage loss that interferes with system operation; 2, to prevent fire, explosion and environmental contamination hazards or nuisances caused by accidental leakage; and 3. to detect unreliable comporients and thoselwhose leakage rates exceed acceptance standards Cl The purposes of J@ai testing are to,enstire reliability and serviceability of components and to prevent premature failure of systems containing fluids under pressure or vacuum. ed or evacuated systems and of leak tes Cl Nondestructive methods for tapi 1g of press sealed comp6fents are thus of great industrial and military importance. C1 Lealkltéting is\most efficient and cost effective WHEN IIs PeMorned ko meet alspecinie Leak Testing to Detect Material Discontinuities O. Many leaks are caused by material anomalies such as cracks and fissures.Some of these can be detected by measurement of leakage rates. Other leaks can be detected by discontinuity detection techniques that identify leak locations. = However, neither of these two leak testing technique categories will detect all anomalies. 1D Leak testing is therefore complementary to other nondestructive testing methods used to find and evaluate basic material anomalies. 2IPageBecause service reliability is not necessarily a direct function of the leakage in a system, itis. difficult to establish an acceptance level for leakage rate. The decision may be influenced by the fact that increased leak testing sensitivity may detect only a small number of additional leaks at considerable added cost. This is because most leaks in BIS)SFSSa a EERERESIIBIA tend to be relatively large. = This is partly due to the clogging of smaller leaks by water vapor and liquids that occurs in parts exposed to industrial processes or to the atmosphere. +The only case where lees 10" Pn (HOY are encountered is in parts that receive special clean room treatment during manufacture. Relationship of Leak Testing to Product Serviceability. tests are designed to aid in evaluating serviceability of Most types of nondestructi materials, parts and assemblies. = For most nondestructive test methods evaluation is indifSEb th@guantities measured have to be properly correlated to the serviceabilitysetiaifacteristics of the material in question. = Thus, the use of indirect tests depefids onithe interpretation of the test results. OF Leak testing procedures, on the other hah, facilitate directevaluation. ‘The measured leakage rate represéiits the physical effect of a faulty condition and thus requires no further analysisifor prattical assessment. Ensuring System Reliability through Leak Testing 1 One important reson for leak testi is te RieaBURS He eiability OF the Systehn UNUEE TES. = Leak testing is notaldirect measure Of reliability, but it might show a fundamental fault of the system by a higher than expected leakage rate measurement. O Ahigh rate offleakage from mechanical connections might indicate that a gasket is improperly aligned or missing. In the Same Manner, a high leakage value might show the presence of a misaligned or imisthreaded flange. ‘Therefore, it is possible to detect installation errors by high leakage values. However, the absen6&i6f high leakage does not necessarily indicate the absence of improperly installed components. 1 Leakage measurements to detect installation errors need not be extremely sensitive, because the leakage rates tobe expected fom serious error wil be relatively large (10? to 10° Pas or to 10stden3s%), 3 [Page= Thus, leak locations can usually be detected easily For practical discussions, alSHiallleak is often defined as having a low leakage rate, thatis, less than that which ensures water tightness, about 0° Pes 10" std ons). Leak Test Sen: ity Definition of Leak Detector and Leak Test Sensitivity 1 Aleak detector’s sensitivity is a measure of the concentration or flow rate of tracer.gas that gives imum measureable leak signal. jum detectable number of tracer gas molecuiles entering the = The sensitivity of a leak detector is independent of the pressuretin the system being tested, provided that time is ignored as a test factor. 1 Leak test sensitivity refers to the minimum detectable amount of leakage that will occur in a specific period of time under specified leak test conditions! + eis necessary to state both tH eSKSGSTa and the preig tet GSAGHS 0 property define leak test sensitivity in terms of teal physical se eal iat AE "To avoid confusion, a set of standard leak test conditions is required. Standard Conditions for Leak Testing 1 The set of conditions most cofiiinonly accepted as standard for pressure measurement is that of dry air at 25 °C (77°F), for a pressure differential between one standard atmosphere and a vacuum (a standard atmosphere is roughly 100 kPa or precisely 101.325 kPa). For practical purposes, the vacuum need be no better than 0.01 of an atmosphere or 1 kPa (0.15 Ibf.int?). D Whena leak is being described and only the leakage rate is given, itis assumed that the leakage fate referRlto leakage at standard conditions. D_ The sensitivity of a leak testing instrument is synonymous with the minimum detectable leakage or mifimum flow rate the instrument can detect. "These minima are independent of leak testing conditions. When the instrument is applied to a test, the leak testing sensitivity depends o + However, the leak test instrument should be more sensitive by at leasta factor of 2 than the ‘minimum leakage to be detected, to ensure reliability and reproducibility of measurements, 4|PageExample of Sensi and Difficulty of Bubble Leak Testing C_ Each modification of a leak testing procedure has anvOptimum Sensitivity Value at which it is most readily used. ie 1, ERSTE (61 OOaLON AFA ELON GT LEAD "Deviation from this optimum value of sensitivity cuca makes it more difficult to perform the measurement a and decreases confidence in the results. = Figure 1 shows the influence of leak testing sensitivity level on the ease of operation of test equipment. ¥ In most cases, after reaching a plateau, fur increase of sensitivity rapidly decreases the ease of operation. v isan example of how the optimum value affects the, teak Testing Sensitivity sis ‘ease of performing the test. The bubble testing sensitivity range extends from 10 tolG Palin’ s M(10" to 10 std cm3+s = In measuring for 10 Pavm?.s* (10- std em*sleaks, a component may be placed in water and observed quickly. = Bubbles may emerge from the pressurizedicomponent at such a rapid rate that there is no question of the existence of a leak. = When checking for leaks in the range of 10%ito 10%Pam?-s* (10 to 10° std cm?'s”), the operator must be sure that thetest object or component is submerged long enough for any bubbles coming from crevices to Rave a chance to collect and rise. = When locating leaks in the 10 *pa-m™é:' (10~ std crn”'s*) range, the component, after being immersed, has to be completely stripped of attached air bubbles so that the bubble formed by leaking gas may beldetected. = The 10 Parm?ss* (10+ std em*s~) leakage range jsihean the limitof detectability ofthe ‘BUBBIEFEchniqué, although longer waiting periods theoretically could obtain higher sensitivity. = Qlonger waiting Periods become impractical when the rate of bubble evolution approaches the rate at which tracer gas is dissolving in the test fh = Specifying sensitivity much greater than 10-* Pam?:s* (10 std cm*-s-) makes bubble testing exceedingly difficult. 1 For instance, bubble testing could be used at higher sensitivity by saturating the immersion liquid with the tracer gas used in leak testing. = However, it would be better to change to a different leak testing method that is more effective at that higher sensitivity. * Bubble testing to detect leaks greater than 10° Parm*s~ (10 std em*s") becomes difficult because of rapid gas evolution and rapid decay of pressure in the system under test. {ase of Opertion S| PageHowever, difficulties in the less sensitive test range are usually not so great as in the more stringent sensitivity range. Relation of Test Costs to Sensitivity of Leak Testing 1 Leak testing instrumentation costs increase as required Fly 2, ifct of requted senstivty on teak detection test sensitivity increases, as sketched in Fig, 2. = The test equipment investment for determining a leakage rate of 10“ Parm?s* (10° std ems") is negligible compared with that for a sensitivity of10" Parm®ss* (10 std cm''s*), whose cost is 10,000 times higher. 1 Evenafter a test technique has been selected, raising leak sensitivity requirements within this technique will result. in an increase in measurement cost. = This increase is usually caused by greater complexityof leak tests with increased sensitivity. ot et unt cup * Cost increases become particularly drastic whefithe Sao 707 > — required sensitivity is higher than the optimum oo oon 007) oo operating range shown in Fig. 1 Leokage Measurement Sensitvty, Pmt td mts) Determination of Overall Leakage Rates!through Pressure Boundaries Many leak tests of fAFBSWVESSEls OF BYSteMs are concerned with the determination of the rate at which a liquid, ga or vapor Will penetrate through their pressure boundaries. = Leakage may occur from any lotation within a component, assembly or system to points outside tHelBOURUAN Nr from external regions to points within a volume enclosed by a pressure boundary. Whena fluid flows through a snilailll@ak, the leakage flow rate depends on Af the geometty of the leak, 2. \the nature of the leaking fluids and 3. thelprevailing conditions of fluid pressure, temperature and type of flow. For purpotes of leak testing, an easily detectable gas or liquid tracer fluid may be used, rather than air or the system operating fluid. The flow of fluid through a leak typically results from a pressure differential or a concentration differential of a gaseous constituent that acts across the pressure boundary. The flow characteristics of a leak are often described in terms of theleonductance of thelleak. © The leak conductance Is defined both by EESISSRSSSS «nc 3SSSiie aaaeaT Sera the leak. 6 [PageThus, conductance or leakage rate at a given pressure for a particular tracer fluid should always be specified in reporting and interpreting the results of a leak test. The leak represents a physical hole with some equivalent length and internal cross-sectional area or diameter. + However, because a leak is not manufactured intentionally into a product or system, the leak hole dimensions are generally unknown and cannot be determined by nondestructive tests, = Therefore, in leak testing, the quantity used to describe the leak is the measured leakage rate. The leakage rate depends on the pressure differential that forces fluid throughithe leak passageway. = The higher this pressure difference, ERG BReatee the leakage rate throughis given leak, = Therefore, leakage measurements of the same leak under differing pressure Condition®can result in differing values of mass flow rate, O The term minimum detectable leakage refers to the smallest fluid’flow raté that ean be detected. = The leakage rate is sometimes referred to as the mags flow rate, "In the case of gas leakage, the leakage rate describes the number of molecules leaking per unit of time, ifthe gas temperature is constant, regardless of the nalyre Of the tracer gas used in leak testing. = When the nature of the leaking gas and the gas temperature are known, itis possible to use the ideal gas laws to determine the actual riiass of the leakage. Leak Tightness 1D Inindustry, the term i@aktight has taken on’a variety of meanings. = Awater buckets tight ifit dats noPallow easily detectable quantities of water to leak out. " Ahigh vacuundivessel is Hghtif tie, rateof apparent leakage into the system cannot be indicated with the,equipmenton ahd. = One might even consider that a gravel truck is leaktight so long as there are no openings in the ‘truck bed [@tge enough togllow the smallest nugget to escape. Cl The degree of leak tightness depends|on the individual situation. * Leak tightness requires that the leakage flow be too small to be detected. = However, leak tightness is a relative term. Therefore, it becomes a necessity to establish a practical level of leak testing sensitivity for any given component under test. Thus, nothing is teaktight IBY compaison toa standard or specification. * Even then, the measured degree of leak tightness can be ensured only at the time of leak testing and under specific leak testing conditions. "= Later operation at higher pressures or temperatures might open leaks. 7T[PageSpecifying Sensitivity of Leak Testing for Practical Applications CO NT on optimum leakage sensitivity value should be sought first. 2) Large deviations from this optimum value could increase the cost and the difficulty of ‘measuring the leakage rate. 2) Secondly, any increase in the sensitivity specified for a particular leakage test automatically increases the cost of leak testing, Therefore, a compromise has to be reactied between testing cost and leakage tolerance. 3) Thirdly, the sensitivity required in leak testing depends on the particular effects ofleakage ‘that must be controlled or eliminated, as illustrated in the following examples, 4) Finally, the language in which the leak testing specification is written shouild be Basy/ to interpret and to implement in testing, to ensure that management's goals arelachieved by the leak test. Tightness to Control Material Loss Cl The fist eonisideration in specifying the leak tightness required of a fluid containment system is. to ensure that the system does not leak Sifficiefit material to case system failure during the operational life of the system. (Then the largest leakage rate is the allowable total leakage system, 1 Ofcourse, conversion might have'to be made between numerical values for the tracer gas leakage during leak testing and those'for the Waterial leakage under system operation conditions. cd by the operational life of the Tightness to Control/Environmental Quality Contamination failure of a system might cause environmental damage, personnel hazard or degraded appearance. 1 The enivironmental damage to a system may be caused by material leaking either into or out of the system. = Fotexample, system damage may be caused to a liquid rocket motor when the oxidizer leaks out offthé storage tank and reacts with parts of the motor. = On the other hand, electronic components can fail when air or water vapor enters a hermetically sealed protective container. 1 tis sometimes difficult to calculate the very small amount of material necessary to cause a contamination failure to occur, = However, in most cases, such calculations are not impossible if the failure can be defined. 8 [Page"For example, ifsome decision can be made as to the allowable amount of reaction between the oxidizer and the rocket engine parts, the maximum acceptable rate of total leakage of, oxidizer from the storage tank can be defined. = Similarly, in an electronic component, if failure results from adsorption of a monolayer of leaking molecules on the surface, then knowing that 10" molecules form one monolayer on a square centimeter of surface makes it possible to calculate the allowable leakage rate for this, particular component. 1 Iffailure results from a pressure rise, then fIRiaiiiuni alGWaIe|ESSUe) Re) pIARAEE SVS (Speration time land system Volume are all that are necessary for calculation of thelallowable leakage rate, Tightness for Safety Material leakage can cause personnel hazard during system operation, = Ifthe tolerable concentrations are known, and these are often reported in literature, again quite easy to calculate the maximum tolerablelequipment leakagerate. Tightness for Appearance O_Anappearance specification is\alspeeification for maxinuni|leakagelthatisimade because leakage ofa higher value will spoil the appearance of thesystéim. = Appearance is often specified when fo midre stringent Specification is necessary. = A specification for leakage of pif BUt of the olpan of a new car is a good example. This leakage specification may notbe called by concern that too much oil wil be lost or that damage to the car motor'will Gecur; instead, itis specified because the prospective buyer would not be inclined to buy a car that iStdripping oil onto the showroom floor. Tightness for System Operation 1 When appeafaice sets theallowable leakage of the system, the leakage is often only @ nuisance. However, even leaks that are largely a nuisance may alter the effectiveness ofthe total system. 1 “For example), during the East Coast power blackout in the United States on Novernber 9, 1965, a large'steam generator failed during the shutdown because the auxiliary steam supply used for lubrication purposes was not available. = This steam supply had been shut off earlier by workers who were bothered by excessive leakage of steam through some valve packing. + This steam leakage was not critical, but it was enough of a nuisance that the system was shut down for repair. * The repair did not take place in time and the bearings of the generator burned out during. emergency shutdown of the system. 9|PageSpecifying Leak Testing Requirements to Locate Every Leak C_ Occasionally itis desirable to locate every existing leak irrespective of size for the following reasons. 1. SRESTRSkS HAVES RABIEOFEFEWINE, Le, very smal leaks may become very troublesome later, after repeated stressing, 2. High temperature leaks may be very small at test temperature but may have higher leakage 3. Temperature cycling to either high or cryogenic levels usually creates stress that results in 1D The criterion whereby a decision is made whether or not to seek greater reliability should be Be ratio of cost ofthe leak testing procedure to the number of leaks found, = For example, improving leak testing reliability from 10 Parm?ysstmm(10-Pstd em?'s*) to a reliability of 10°’ Pam®:s~ (10 std cm*'s~) may notbejustified. = The cost of obtaining the small increase in reliability may be protibitivln relation to the value of the increase in detection reliability. D_ The expected leak tightness of sealing operations that will be used to isolate the system during leak testing must also be considered, The leak testing specification should be wiitten with aeviee from aivexperieniced engineer = Factors to be considered include the leak testing tiiethod and technique; Avoiding ImpracticalSpecifications for Leak Tightness 1D Aiming at absolute tightness is an academic endeavor. "In practice, all that can Be asked for is a more or less stringent degree of tightness selected according to’the application requirements, 1D Nothing made by man can truly be considered to be absolutely leaktight. = EVBn in the absence of minute porosities, the permeation of certain gases through metals, crystals, polymers and glasses stil exists. "Thus, itis necessary to establish a practical leakage rate that is acceptable fora given component under test. 1D Apreliminary decision has to be made concerning the definition of leak tightness for the particular situation. = Because leak tightness is a relative term and has no absolute meaning, the sensitivity of the available leak testing equipment is a practical guide to attainable levels of leak testing+ Anyincrease in required sensitvity of leak testing [GES Ns SSE ‘= This increase in cost of leak testing reaches a maximum when the leakage specification is. such impractical terms as: tno detectable leakage, 2, no measureable leakage, 3, no leakage and A zero leakage. o _Imorctaeak testing spctations are expensive to implement. They are also very confusing unless the leak testing method is precisely described. = With specifications in impractical terms, the leak testing operator is always Working against background instrument noise, = He must then decide whether the leakage reading obtained is@lsed by the ‘abdom fluctuations of test instruments or by the actual des@étion of specifielleakage. = tls much easier to discriminate whether a measured{ leakage rates above or below a given standard than to discriminate leakage from random instfumen@o}e. = Itis therefore suggested that, when specified, zéro leakage be defined as a measurable quantitative value of leakage rate that fs insignificant in the operation of the system. = Such a definition allows the system or the fjeasulément sensitivity to be compared with a flow through a standard physical leafét = inthis way, 2 qualification of theV&tem petformatice acceptability can be made during the test operation, nChapter 2 Selection of Specific Leak Testing Technique for Various Applications 1 Figure 3 provides a graphical guide to selection of leak testing methods and techniques for various applications. = Itshows a decision tree with which the choice of a leak testing method becomes a step-by- step process. = The selection processes suggested by Fig. 3 serve as a basic guide. "Further consideration of specific leak testing requirements may suggest other Methods or techniques for test selection or cause the test engineer to modify leak testing protedures. See also Table 2. = The final selection of the leak testing method will typically be made from perhapsiohly three or four possible test methods. "The special conditions under which tests must be male can becomela major factor in this final test selection, D_ The first question to be asked when choosing the best leak testing method, or technique of a method, is “Should this test reveal the presence ofia suspected leak) or is its purpose to show the location of a known leak?” "The second question to be answered is, “(Sit necessary to measure the rate of leaki@eat they Soi LLA ier: mennganee Taoit 2. Leak testing methods and techniques, essential, use of calibrated or feferencelleaks oP Methods Techniques other means to provide guanra"N@lesK#8® uct solution immersion: fm solrion ieee dakunvind Uttrasonic/acoustic _sonicimechanical flow: sound generator * Inthe decision tree of Fig. 3, the first branch (or yotiage discharge _voltage spark: color change decision point) answers the preceding questions: Pressure hydrostatic; hydropneumnatic; pneumatic and determines if the purpose or requirements of _lonization photo loniation me ion zation the testlead to the uppéfbranch of leak location Conductity thermal conductivity catalytic combustible eniy Rp ower Minch oF leakage rate Radiation absorption — infared ultraviolet; ser neasurttant’ Chemical based chemical penetrants; chemical traar gases Helogen detector halide torch: electron capture; halogen diode Radioisotope krypton-85 Pressure change absolute reference: pressure rs aw measurement; pressure decay; volute Mass spectrometer helium or argon; tracer probe ecaton hooding total leskager detector probe locaton: sealed objects: resdul gas analyzer Z| Page5 a SSS Fioure 3. Graphical decision tre for step-bystep selection of leak testing methods Halogen sbcton captuhaogon hosted anode eur wa pone ned pa detection age reson ranean ete sty ry eu mas stort eomtoar [avr ten son mas pcre erro esa as rae Gein [hes icon Detector ry inrared ‘Compare these factors in robe. 2 agen tater anode ene Pressutivedd 2 nine seen} ve PM [sage rpone =e [Pressuro measuroment assed fPeseemcserenins | ome eauomen cx stem Innoren dotoctor | Arbome ulasonie eer pe racctaty sum mss specter sie pee Voge rated ete ~ Lon sn tote sex prosuang {are 9 [teem nas gecroret] Sarhgh z : Ses tot ge [row meet E |rainynronowne | £ Evacuated g seacaaniy —§ [irra cange Mutiple Messspectometer [icin mas © [Row measrement se fests lnc teckage jes Ieaogoncicron re eae togn eptracn essrement eee haste arode ested ance acoactvty vas — Wighsnitiy _JOvtetacecton | | ae” |B tae el 2 aa 2 mtorr | § [re f insti a [Peter masons |B wemeasurement | & [Subba dow secity = Hs [Prevaure measurement women Gageinpace lrevararad tee um peace pense Sti lancet agence eho tore 13 [PageSelection of Technique for Leakage Measurement C. The lower half of the decision tree diagram of Fig. 3is a guide for step-by-step selection of optimum techniques for leakage measurements. Leakage measurements can be divided into two different types based on the nature of the test objects whose leakage is to be measured. The first decision is based on the accessibility of test surfaces on the pressure boundaries of the test object. Y Test objects are classified by accessibility into two groups. 1) Open units are accessible on both sides of the pressure boundaly, foF'tracer probes or detector probes. 2) Sealed units are accessible only on external surfaces. Inthe lower portion of Fig. 3, this choice is indicated first on the @ecision path for leakage measurement. 1 The second category usually consists of mass produced items such as transistors, relays, ‘ordnance components and sealed instruments. Fluid Media in Leak Testing Ol Types of Fluid Media Used in Leak Testing) = Leak testing can be divided into threé!fiain categories 1) leak detection, 2) leak location and 3) leakage measurement "Each technique'inall categories involves a fluid leak tracer and some means for establishing a pressure differential or othef mean®for causing fluid flow through the leak or leaks. = Possible fluid media Include g38es, vapors and liquids or combinations of these physical states of fluid probing media. = _selectidh of the desired fluid probing medium for leak testing depends on operator or engineering Judgment involving factors such as: 1) type and sizeof test object or system to be tested; 2) ‘typical operating conditions of test object or system; 3) environmental conditions during leak testing; 4) hazards associated with the probing medium and the pressure conditions involved in testing 5) leak testing instrumentation to be used and its response to the probing medium; 6) the leakage rates that must be detected and the accuracy with which measurements must, be made; and 1] Page7) compatibility of test probing medium with test object and content (to avoid corrosion etc,). + Gases and vapors ar generally preferred quid media where highsensitty wo eakage must be attained; however, liquid probing media are used for leak testing in many specific Detector Probe versus Tracer Probe One of the most difficult and important decisions is the choice of which leak testing method should be used. * Acorrect choice will optimize sensitivity, cost and reliability of the leak testing procedure. = Choice of an incorrect test method makes leak testing less sensitive aflless Peliable) While adding to the difficulty of testing. "One simplified way to choose is to rank various leak testing methods by means of their leakage sensitivity. = Ifthis were sufficient, the test engineer would only feed to decide what Uegree of sensitivity is required and then to select the test method from among those offering adequate sensitivity for the specific test application. However, each leak testing technique can have adifferent test sensitivity under different ‘operating condition: = For example, a mass spectrometer léak détectotis 10,000 times more sensitive than a heated anode halogen vapor detection instrument vihen used for leak location in the tracer probe leak location test of an evacuated vessel Y However, if these/tw@linstriiments are used for leak detection on a pressurized test ‘systent, the halogen leak detector is 100 times more sensitive, ¥ The reason fonthis apparent discrepancy becomes obvious on close examination of the operating charatteristics of these two instruments. Thelmass spectrometers ‘designed for operation under vacuum conditions, whereas the halogen leak detector is designed for operation in air at atmospheric pressure. * As another example, 2 helium mass spectrometer leak detector may have a leakage sensitivity ‘6f 107? Parm®s* (10 std cm*-s~) during routine leak testing with dynamic leakage measurement techniques. ¥VOn very small systems, this optimum sensitivity may be increased to 10" Pam (10 std cm?s), a gain of 1000x, by using the static accumulation leakage measurement technique. However, the static leakage measurement technique is not the standard method of using the mass spectrometer leak detector. Therefore, the last sensitivity stated above is subject to some question. It must be recognized that each method of leak detection or measurement is usually optimized for 15| Pageone particular type of leak testing, Therefore, it can be a mistake to compare sensitivities of various leak testing methods under the same conditions, if each test is not designed to operate under these same conditions. Tracer Gas Technique for Leak Location Only C_ Asshown on the upper branch of the decision tree of i is leaklocation electronic leak detecton nstrumonts (tracer prove Fig. 3, tracer gas tests whose purpose is leak location eohrfane detector probe technague. only can be divided into a tracer probe technique and a detector probe technique (see Fig. 4). 9 Fiouee 4. Tracer gas probing for locating leaks with sensitive When choosing either technique, it is important that Soo — ——I leak location be attempted only after the presence of cle a leak has been ascertained. D1 _The S@eBHDFOBE technique is used when the test ices system is evacuated and the tracer gas is applied to isles ‘the outside of the pressure boundary of the test © roe system, mA, The BgHSEESEIBFOBE technique is selected when thé) fine Snser ‘test system is pressurized with gases including the: tracer gas (if used) and the sniffing or samplinglof the! leaking gas is being done at atmospherie pressure ity soucn of the ambient air. * This selection corresponds to the second decision point in the upper branch of the decision tree of Fig. 3. CO Leak Location Techniquewith Trater Probe outside an Evacuated System = When testing anlevacuated System that has in-leakage from the ambient atmosphere or from a tracer probe, the first contideration in selection of a test method is whether there is an IREFEREDEREEBE within the system. ™ the inherentéetector might be a pressure gage of an electronic type or, more ‘desirably, a gage that is specifically responsive to the partial pressure ofa specific trater ga. %_ Vacuum systems often contain one or more types of vacuum gages. v™ Ii Fig. 3, this point appears in the second main line from the top, for tracer probe testing of evacuated systems, and is labeled inherent detector. = Ifa vacuum gage does not exist within the evacuated system under test, other test methods must be examined individually to determine their limitations and advantages for leak testing of this system. v The tracer probe leak testing methods, in order of increasing leak sensitivity, time and costae ultrasonic, pressure change gage response, high voltage electrical charge, 16|Pace"These methods are listed vertically at the right end of the second horizontal branch in Fig. 3. = The methods shown in the upper half of Fig. 3 for leak location are those in primary or most common usage, Y Other methods, such as those using fadioactive tracer Bases, are not generally used because of safety and other operating problems associated with their use. However, if none of the leak location methods described for detector probe or tracer probe leak tests in the preceding discussion is satisfactory for a specific application, ‘more complicated leak testing methods may be considered dating selection ofan appropriate leak testing test. 1 Leak Location Technique with Detector Probe Operafifig at Atmospheric Pressure "When testing a pressurized system that is leaking into the atmosphere, the next decision point is whether or not the leaking fluid can be used as a tracer (this décision point lies along the top branch of the tree of Fig. 3). * For example, most refrigeration and air conditioning systems are charged with a refrigerant gas (efrigerant-22 or -134a) that.isa fluorocarbon to Which the heated anode halogen vapor detector is specifically Highly Sensitive. Y When searching for leaks{in operating’systemns of this type, the inherent tracer dictates the use of the halogen leak testing method. ¥ Because of potential@nvirdtimental effects from fluoracarbons, some current systems are being charged with refrigerante1343 ga GF Sul REXATUBHIASTOFIKE, respectively, with madified residual gas analyzer halogen leak detectors or electron capture halogen leak detectors. Ifthe presStirized test system contains ammonia gas, a Chemical type of leakdetector might prove to be optimum Rig Main cases where ihe IAAI s to be used, the presence of specific gas (such as Stony helium|GF eon) within the system provides an excellent inherent tracer, ¥ Uiternative procedures involve pressurizing the test system with such #itfaeergasior/a ‘minture of air with tracer gas. = Some other methods for leak location do not depend on the specific nature of the leaking gas; among these are the ultrasonic leak detector and bubble'testing. ¥ In-some cases, the tracer gas might be suitable for use with more than one testing method, e.g, helium could be used for bubble testing for large leaks or for mass spectrometer testing for small leaks or quantitative leakage measurements. 17 [Page+ The detector probe leak testing methods, in order of increasing leak sensitivity, time and costs, are ultrasonic, bubble, chemical, pressure or flow gage response, infrared gas detector, ‘mass spectrometer leak detector and halogen vapor detector. These relative sensitivity ratings apply for detector probes searching with the detector inlet probe or sniffer searching in air at atmospheric pressure. = These alternative leak test methods are listed vertically at the right end of the top branch of the decision tree of Fig. 3. The lowest cost, highest speed, simplest leak tests are at the bottoM of this list. ‘The slower, more costly, higher sensitivity test methods appear at the'top of the list shown to the right of the top branch of the decision tree of Fig. 3. Leakage Measurement with Tracer Gases 0 Principles of Leakage Measurement = Allleak detection with tracer gases involved their flow from the high ipresstre side of a pressure boundary through a presumed leak to the Wower.pressufe side of the pressure boundary, = When tracer gases are used in leak testing, instfiments sensitive to tracer gas presence or concentration are used to detect outflow from the low pressure side of the leak in the pressure boundary, = Where leak tests involve measureménts Ofichane in Pressure or change in volume of gas within a pressurized enclosure, the loss of internal gas pressure or volume indicates that leakage has occurred through the presstire boundary (or temporary seals placed on openings of the pressure boundary). = When evacuated or low pressureltest systems or components are surrounded by higher pressure media Such asthe earth’s atmosphere, or a hood or test chamber containing gases at higher pressures, leakage can be detected by loss of pressure in the external chamber or by rise in pregsure within thelower pressure system under test. O Classification of Teghniques of Leakage Measurement with Tracer Gases © Leakage tate measurement techniques involving the use of tracer gases fall into two other Classifications known as 1) static leak testing and 2) dynamic leak testing. = Instatic leak testing, the chamber into which tracer gas leaks and accumulates is sealed and is not subjected to pumping to remove the accumulated gases. 18] Pose= In dynamic leak testing, the chamber into which tracer Fines” vomage wemuncmon syeame ae _gas leaks is pumped continuously or intermittently to rents) prs draw the leaking tracer gas through the leak detector instrumentation, as sketched in Fig, 5.5 The leakage rate measurement procedure consists of first placing ‘tracer gas within or around the whole system being tested + Apressure differential across the system boundary is. established either by pressurizing the one side of the pressure boundary with tracer gas or by evacuating the other side, = The concentration of tracer gas on the lower pressure side of the pressure boundary is measured to determine leakage rates, Leakage Measurements of Open Test Objects Accessible on Both Sides 1 when test objects have pressure boundaries.accessible on both sides, the second decision in the selection of a leakage measurement test metholhis whether the unit can or should be evacuated during leak testing. * This decision will determine if the leak test is Performed with the tracer probe or detector probe. 1 _Ifone side of the pressure boundary cariibelevacuated so that leakage occurs to vacuum and the leak detector is placed in the va€uum system, more sensitive leak testing will usually result. = In vacuum, the tacerigaies Canyreach the detector quickly, particularly with dynamic tests in the evacuated test Valume is pumped rapidly and continuously. Y Inftfils case, thereiis little possibility of stratification of tracer gases. = gHBWiever, evacuatior{does not always produce the most sensitive and reliable leakage measurements) Dif the test volume is extremely large, high pumping speed is necessary to reduce response time. = Suchlauxiliary pumping will cause split flow, thus reducing the amount of tracer gas reaching the leak detector. ‘turn, can reduce signal levels and leakage sensitivity. may prevent evacuation of the test system to a sufficiently low absolute pressure. Conventional helium mass spectrometer leak detectors, for example, should be operated at vacuum levels of 0.1 Pa (1 mtorr) or lower. 19] PageY Conventional helium mass spectrometers can operate with manifold vacuums of 2 Pa {20 mtorr) or lower whereas counterflow helium mass spectrometers can operate with manifold vacuums of 10 Pa (0.1 tort) and higher. * The structure of the equipment under test (particularly if thin walls not intended to withstand external pressure are involved) may prevent use of leakage rate measurement techniques in Which the leak detector must operate within a vacuum, = In Fig. 3, the lowest branch leading to the junction of the leak to vacuum path and the leak to atmosphere path represents the point of decision discussed in this paragraph 1 Selecting Specific Method for Leak Testing of Evacuated Test Units or Systems = As indicated along the next-to-bottom decision path at the center of Fig23, the first approach ‘to selecting leak test methods for units that can be evacuated is to determitie whether or not ‘there is an inherent tracer in the test system while in operatioft! ¥ For example, ifin normal operation the syst under testicolitains\one of the specific ‘tracer gases such as helium or halogenated hydrocarbons, a test method sensitive to ‘that specific tracer gas might be preferred. ¥ inthis way, considerable savings in testime and cost can be realized if there is no need to fil the system under test with a tracer gas, «If there is fio inherent tracer gas within the system|under test, the next decision step might be ‘to determine if there is @lpressurelOF floW/Bage already present in the evacuated system to be leak tested YIfso, this age might be used for leakage measurement ifplae® of some additional type of leak detector, ¥ This internally available gage might be a simple vacuum dial, thermocouple or ization gageor, insome fortunate cases, a mass spectrometer that is incorporated into the system as apart of its analytical instrumentation or controls, Y Céiisideration need not be limited to those types of gages commonly used for leak testing. ¥ “Any gas concentration measuring equipment that happens to be available may be used for leakage measurement and is accurate enough and sensitive enough for the required results. “his decision point is that labeled gage in place in the two bottom decision pathways shown in Fig. 3. Methods of Leakage Measurement Tracer * there s IIS or SEEERIII vithin on KEN test system, other vacuum mode leak testing methods must be considered. Evacuated Systems with No Inherent 20| Pace= Methods for leak testing of @vaeated systems, in order of increasing leak sensitivity and cost of leak testing equipment, include gas flow measurement, pressure change measurement, heated anode halogen vp ea ete ar EE ‘These methods, listed vertically at the end of the next-to-bottom decision line in Fig, 3, should each be considered individually and evaluated in terms of their advantages and limitations, ¥ In most cases, all of the possible leak testing methods should be considered. "Selection depends on pertinent factors. For example, a more sensitive leak testing method might involve highePiinitial costs for equipment and test setups but, on the other hand, it might result in greatscost savings during testing programs or provide greater reliabilityinileak testing reSits "Once the basic vacuum leak testing method has been selected, a second colisideration involves selection between static and dynamic test techniques! Its usually preferable to perform leak tests{using a dynamic testing technique (tests involving pumping of the vacuum system thtoughout the test petiod). However, static techniques of leakage rate measuremeltt shuld also be considered. ‘Static tests involving rise or loss in pressbire, or accumulation of tracer gases over prolonged leak periods, are slower than typical dynamic leak tests. However, higher sensitivity can be athieveth in static tests if the volume under test is not excessive; this may be worthithe eXtra effort. Selection of Test Methods for Systems Leaking to Atmospheric Pressure 1 The choice of pressure mode testing methods Le., for test systems leaking to atmospheric pressure should belmadelbyfollowing the same type of decision pattern as for leak testing of evacuated systems. = The decision path for thisiéase appears at the bottom of Fig. 3. = giTHBleak testing metfidds applicable to testing of systems leaking to atmosphere, in order of increasing testensitivity, ae flow measurement, pressure measurement (for larger volume Systems), immersion bubble testing, infrared gaseous leak testing, heated anode and electron capture halogen leak testing, mass spectrometer helium leak testing and Jeak testing Using ‘radioactive tracer gases = Aldyinaimic\leak testing imethiod should be used wherever possible. After various dynamic leak test methods have been considered and those whose limitations are unacceptable have been rejected, a static leak testing method should also be considered. Y Although a static technique will i testing sensitivity. \crease leak testing time, it will also increase leak 21[ Page