Leak Detection

Are you looking for more information regarding this and other subjects? Look no further than Semitracks' Online Training. Semitracks' Online Training contains short courses and other material useful for any Engineer trying to learn new subjects or brush up on old ones.

What is Leak Testing?

Leak Testing is a method used to determine the seal integrity of a hermetically sealed package. Hermetically sealed integrated circuit packages are commonly used for military, industry, and other high-reliability applications where penetration of moisture poses a threat to the operation of the integrated circuit.

Why Perform Leak Testing?

Leak testing is performed as part of the failure analysis process to determine if the hermetic seal of the integrated circuit package has been breached. Fine Leak and Gross Leak Hermetic Seal testing are both easy and cost effective methods for verifying a breach in the seal.

How is Leak Testing Performed?

Fine Leak Hermetic Seal Testing

Fine Leak Hermetic Seal testing is performed using commercially available helium-based or radioisotope tracer gas-based leak detection equipment. Equipment can be purchase from companies like Edwards, Varian, and Veeco Instruments. Tests using this equipment involve pressurizing the IC with the gas for a period of time (to drive the gas into the package cavity through the leak), placing the IC under vacuum, and allowing the gas to leak out and be detected. For specific instructions on how to perform a helium leak detection test, consult the MIL-STD Procedures for Leak Testing located in this document and the manual for your particular equipment.

Gross Leak Hermetic Seal Testing

Gross Leak Hermetic Seal testing can be performed fairly easily if you have access to pressurized air, a vacuum, a pressure vessel, and two different types of electronically benign liquids (fluorocarbon-based liquids work the best). Commercial systems are also available from companies like Triotech. The instructions for performing a gross leak hermetic seal test are in the MIL-STD Procedures for Leak Testing located in this document.

When is Leak Testing Performed?

Leak testing is normally a non-destructive technique. If the IC does leak however, a gross leak hermetic seal test will fill the cavity with the fluorocarbon. Normally, fine leak hermetic seal testing should always be performed before gross leak hermetic seal testing as the liquids used in a gross leak test can plug the leak sites on a fine leak test. An IC with a large leak may pass fine leak testing but fail gross leak testing.

MIL-STD Procedures for Leak Testing

MIL-STD 883C

METHOD 1014.5

SEAL

  1. PURPOSE. The purpose of this test is to determine the effectiveness (hermeticity) of the seal of microelectronic and semiconductor devices with designed internal cavities.

    1.1 Definitions.

    1. Standard leak rate. Standard leak rate is defined as that quantity of dry air at 25_C in atmosphere cubic centimeters flowing through a leak or multiple leak paths per second when the high-pressure side is at 1 atmosphere (760 mm Hg absolute) and the low-pressure side is at a pressure of not greater than 1 mm Hg absolute. Standard leak rate shall be expressed in units of atmosphere cubic centimeters per second (atm cc/sec).
    2. Measured leak rate. Measured leak rate (R1) is defined as the leak rate of a given package as measured under specified conditions and employing a specified test medium. Measured leak rate shall be expressed in units of atmosphere cubic centimeters per second (atm cc/sec). For the purpose of comparison with rates determined by other methods of testing, the measured leak rates must be converted to equivalent standard leak rates.
    3. Equivalent standard leak rate. The equivalent standard leak (L) of a given package, with a measured leak rate (R1), is defined as the leak rate of the same package with the same leak geometry that would exist under the standard conditions of 1.1 a. The formula (does not apply to test condition B) in 3.1.1.2 represents the L/R ratio and gives the equivalent standard leak rate (L) of the package with a measured leak rate (R1) where the package volume and leak test conditioning parameters influence the measured value of (R1). The equivalent standard leak rate shall be expressed in units of atmosphere cubic centimeters per second (atm cc/sec).
  2. APPARATUS. The apparatus required for the seal test shall be as follows for the applicable test condition:

    2.1 Test conditions A1, A2, and A4 1/ - Tracer gas helium (He) fine leak. Apparatus required shall consist of suitable pressure and vacuum chambers and a mass spectrometer-type leak detector preset and properly calibrated for a helium leak rate sensitivity sufficient to read measured helium leak rates of 10-9 atm cc/sec and greater. The volume of the chamber used for leak rate measurement should be held to the minimum practical, since this chamber volume has an adverse effect on sensitivity limits. The leak detector indicator shall be calibrated using a diffusion-type calibrated standard leak at least once during every working shift. In addition for test condition A4, the following apparatus is required:

    1. Fixture and fittings to mate the package to be tested to the leak detector.
    2. Surgical rubber gasket.
    3. Apeizon grease (type M or N) or equivalent.

    2.2 Test condition B - Radioisotope fine leak. Apparatus for this test shall consist of:

    1. Radioactive tracer gas activation console.
    2. Counting equipment consisting of a scintillation crystal, photomultiplier tube, preamplifier, ratemeter, and krypton-85 reference standards. The counting station shall be of sufficient sensitivity to determine through the device wall the radiation level of any krypton-85 tracer gas present within the device. The counting station shall have a minimum sensitivity corresponding to a leak rate of 10-9 atm cc/sec of krypton-85 and shall be calibrated at least once every working shift using krypton-85 reference standards and following the equipment manufacturer's instruction.

      1/ A3 was intentionally omitted.

    3. A tracer gas consisting of a mixture of krypton-85 and dry nitrogen. The concentration of krypton-85 in dry nitrogen shall be no less than 100 microcuries per atmospheric cubic centimeter. This value shall be determined at least once each 30 days and recorded in accordance with the calibration requirements of this standard (see 4.5.1 of MIL-STD-883).

    2.3 Test condition C - Fluorocarbon gross leak. Apparatus for this test shall consist of:

    1. A vacuum/pressure chamber for the evacuation and subsequent pressure bombing of devices up to 75 psig up to 10 hours.
    2. A suitable observation container with provisions to maintain the indicator fluid at a temperature of l25_C and a filtration system capable of removing particles greater than 1 micron in size from the fluid.
    3. A magnifier with a magnification in the range between l.5X to 30X for observation of bubbles emanating from devices when immersed in the indicator fluid.
    4. Sources of FC-72, FC-84, or PP-1 fluorocarbon detector fluids, and FC-40, FC-43, PP-7, or PP-9 fluorocarbon indicator fluids.
    5. A lighting source capable of producing at least 15 thousand foot candles in air at a distance equal to that which the most distant device in the bath will be from the source. The lighting source shall not require calibration but the light level at the point of observation (i.e., where the device under test is located during observation for bubbles) shall be verified.
    6. Suitable calibrated instruments to indicate that test temperatures, pressures, and times are as specified.
    7. Suitable fixtures to hold the device(s) in the indicator fluid.

    2.4 Test condition D - Penetrant dye gross leak. The following apparatus shall be used for this test:

    1. Ultraviolet light source with peak radiation at approximately the frequency causing maximum reflection of the dye (3650 A for Zyglo; 4935 A for Fluorescein; 5560 A for Rhodamine B, etc.,).
    2. Pressure chamber capable of maintaining 90 psig.
    3. Solution of fluorescent dye (such as Rhodamine B, Fluorescein, By-check, Zyglo, FL-50, or equivalent) mixed in accordance with the manufacturer's specification.
    4. A magnifier with a magnification in the range between l.5X to 30X for dye observation.

    2.5 Test condition E - Weight gain gross leak. Apparatus for this test shall consist of:

    1. A vacuum/pressure chamber for the evacuation and subsequent pressure bombing of devices up to 75 psig up to 10 hours.
    2. An analytical balance capable of weighing the devices accurately to 0.1 milligram.
    3. A source of FC-72, FC-84, or fluorocarbon fluid.
    4. A filtration system capable of removing particles greater than 1 micron in size from the fluorocarbon fluid.
    5. Suitable calibrated instruments to measure test pressures and times.
    6. A source of Freon TF.
  3. PROCEDURE. Fine and gross leak tests shall be conducted in accordance with the requirements and procedures of the specified test condition, Testing order shall be fine leak (condition A or B) followed by gross leak (condition C, D, or E). When specified (see 4), measurements after test shall be conducted following the leak test procedures. Where bomb pressure specified exceeds the microcircuit package capability, alternate pressure, exposure time, and dwell time conditions may be used provided they satisfy the leak rate, pressure, time relationships which apply, and provided no less than 30 psig bomb pressure is applied in any case. When test condition A4 is used, gross leak testing is not required. However A4 shall not be used in lieu of the required seal testing of lidded packages. When batch testing (more than one device in the leak detector at one time) is used in performing test condition A or B and A reject condition occurs it shall be noted as a batch failure. Each device may then be tested individually one time for acceptance if all devices in the batch are retested within one hour after removal from the tracer gas pressurization chamber.

    3.1 Test condition A1, A2, or A4 - Tracer gas (He) fine leak. Test condition A1 is a "fixed" method with specified conditions per table I that will ensure the test sensitivity necessary to detect the required measured leak rate (R1). Test condition A2 is a "flexible" method that allows the variance of test conditions in accordance with the formula of 3.1.1.2 to detect the specified equivalent standard leak rate (L) at a predetermined leak rate (R1). Test condition A4 is a method that will detect the required measured leak rate (R1) of an unsealed package.

    • 3.1.1 Test conditions A1 and A2 - Procedure applicable to "fixed" and "flexible" methods. The completed device(s), shall be placed in a sealed chamber which is then pressurized with a tracer gas of 100 +0, -5 percent helium for the required time and pressure. The pressure shall then be relieved and each specimen transferred to another chamber or chambers which are connected to the evacuating system and a mass-spectrometer-type leak detector. When the chamber(s) is evacuated, any tracer gas which was previously forced into the specimen will thus be drawn out and indicated by the leak detector as a measured leak rate (R1). (The number of devices removed from pressurization for leak testing shall be limited such that the test of the last device can be completed within 60 minutes for test condition A1 or within the chosen value of dwell time t2 for test condition A2.)
      • 3.1.1.1 Test condition A1 - Fixed method. The devices(s) shall be tested using the appropriate conditions specified in table I for the internal cavity volume of the package under test. The time t1 is the time under pressure and time t2 is the maximum time allowed after release of pressure before the device shall be read. The fixed method shall not be used if the maximum equivalent standard leak rate limit given in the procurement document is less than the limits specified herein for the flexible method.

        TABLE I. Fixed conditions for test condition A1.

        Volume of Bomb condition R1 package (cc) Psig Exposure time hours Maximum dwell hours Reject limit (atm cc/sec)
        V < 0.40 60+2 2 +0.2, -0 1 5 x 10-8
        V > 0.40 60+2 2 +0.2, -0 1 2 x 10-7
        V > 0.40 30+2 4 +0.4, -0 1 1 x 10-7
      • 3.1.1.2 Test condition A2 - Flexible method. Values for bomb pressure exposure time, and dwell time shall be chosen such that actual measured tracer gas leak rate (R1) readings obtained for the devices under test (if defective) will be greater than the minimum detection sensitivity capability of the mass spectrometer. The devices shall be subjected to a minimum of 2 atmospheres absolute of helium atmosphere. If the chosen dwell time (t2) is greater than 60 minutes, graphs shall be plotted to determine an R1 value which will assure overlap with the selected gross leak test condition. The chosen values, in conjunction with the value of the internal volume of the device package to be tested and the maximum equivalent standard leak rate (L) limit (as shown below or as specified in the applicable procurement document), shall be used to calculate the measured leak rate (R1) limit using the following formula:

        R1 = (LPE/P0) (MA/M)1/2 { 1 - e-[(Lt1/VP0) (MA/M)1/2] } e-[(Lt2/VP0)(MA/M)1/2]

        Where:

        R1 = The measured leak rate of tracer gas (He) through the leak in atm cc/sec He.

        L = The equivalent standard leak rate in atm cc/secAIR.

        PE = The pressure of exposure in atmospheres absolute

        P0 = The atmospheric pressure in atmospheres absolute. (1)

        MA = The molecularweight of air in grams. (28.7)

        M = The molecularweight of the tracer gas (Helium) in grams. (4)

        t1 = The time of exposure to PE in seconds.

        t2 = The dwell time between release of pressure and leak detection, in seconds.

        V = The internal volume of the device package cavity in cubic centimeters.

        • 3.1.1.2.1 Failure criteria. Unless otherwise specified, devices with an internal cavity volume of 0.01 cc or 1 ss shall be rejected if the equivalent standard leak rate (L) exceeds 5 x 10-8 atm-cc/secAIR. Devices with an internal cavity volume greater than 0.01 cc and equal to or less than 0.4 cc shall be rejected if the equivalent standard leak rate (L) exceeds 1 x 10-7 atm cc/secAIR. Devices with an internal cavity volume greater than 0.4 cc shall be rejected if the equivalent standard leak rate (L) exceeds 1 x 10-6 atm cc/sec.
    • 3.1.2 Test condition A4 - Procedure applicable to the unsealed package method. The fixture and fittings of 2.1a. shall be mounted to the evacuation port of the leak detector. Proof of fixturing integrity shall be verified by sealing a flat surfaced metal plate utilizing the gasket and grease of 2.1 and measuring the response of the leak test system. Testing shall be performed by sealing the package(s) to the evacuation port and the package cavity evacuated to 5 x 10-7 atm cc/sec. Care shall be taken to prevent contact of grease with package (seal ring not included) to avoid masking leaks. The external portion of the package shall be flooded with Helium gas either by the use of an envelope or a spray gun, at a pressure of 30 psig.
      • 3.1.2.1 Failure criteria. Unless otherwise specified, devices shall be rejected if the measured leak rate (R1) exceeds 1 x 10-8 atm cc/secHe

    3.2 Test condition B - Radioisotope fine leak test.

    • 3.2.1 Activation parameters. The activation pressure and soak time shall be determined in accordance with the following equation:

      QS = R/(skTPt) (1)

      The parameters of equation (1) are defined as follows:

      QS = The maximum calculated leak rate allowable, in atm cc/secKr, for the devices to be tested.

      R = Counts per minute above the ambient background after activation if the device leak rate were exactly equal to QS. This is the reject count above the background of both the counting equipment and the component, if it has been through prior radioactive leak tests.

      s = The specific activity, in microcuries per atmosphere cubic centimeter, of the krypton-85 tracer gas in the activation system.

      k = The overall counting efficiency of the scintillation crystal in counts per minute per microcurie of krypton-85 in the internal void of the specific component being evaluated. This factor depends upon component configuration and dimensions of the scintillation crystal. The counting efficiency shall be determined in accordance with 3.2.2.

      T = Soak time, in hours, that the devices are to be activated.

      P = Pe 2-Pi 2. where Pe is the activation pressure in atmospheres absolute and Pi is the original internal pressure of the devices in atmospheres absolute. The activation pressure (Pe) may be established by specification or if a convenient soak time (T) has been established, the activation pressure (Pe) can be adjusted to satisfy equation (1).

      t = Conversion of hours to seconds and is equal to 3,600 seconds per hour.

      NOTE: The complete version of equation (1) contains a factor (P0**2 - (delta P)**2) in the numerator which is a correction factor for elevation above sea level. P0 is sea level pressure in atmospheres absolute and alpha P is the difference in pressure, in atmospheres between the actual pressure at the test static and sea level pressure. For the purpose of this test method, this factor has been dropped.

    • 3.2.2 Determination of counting efficiency (k). The counting efficiency (k) of equation (1) shall be determined as follows:
      1. Five representative units of the device type being tested shall be tubulated and the internal void of the device shall be backfilled through the tubulation with a known volume and known specific activity of krypton-85 tracer gas and the tubulation shall be sealed off.
      2. The counts per minute shall be directly read in the shielded scintillation crystal of the counting station in which the devices are read. From this value, the counting efficiency, in counts per minute per microcurie, shall be calculated.
    • 3.2.3 Evaluation of surface sorption. All device encapsulations consisting of glass, metal, and ceramic or combinations -thereof, including coatings and external sealants, shall be evaluated for surface sorption of krypton-85 before establishing the leak test parameters. Representative samples of the questionable material shall be subjected to the predetermined pressure and time conditions established for the device configuration as specified by 3.2.1. The samples shall then be counted every 10 minutes, with count rates noted, until the count rate becomes asymptotic with time. (This is the point in time at which surface sorption is no longer a problem.) This time lapse shall be noted and shall determine the "wait time" specified in 3.2.4.
    • 3.2.4 Procedure. The devices shall be placed in radioactive tracer gas activation tank. The activation chamber may be partially filled with inert material to reduce pumpdown time. The tank shall be evacuated to 0.5 torr. The devices shall be subjected to a minimum of 2 atmospheres absolute pressure of krypton-85/dry nitrogen mixture for a minimum of 12 minutes. Actual pressure and soak time shall be determined in accordance with 3.2.1. The R value in counts per minute shall not be less than 600 above background. The krypton-85/dry nitrogen gas mixture shall be evacuated to storage until 2.0 torr maximum vacuum exists in the activation tank. The storage cycle shall be completed in 3 minutes maximum as measured from the end of the activation cycle or from the time the activation table pressure reaches 60 PSIA of a higher bombing pressure is used. The activation tank shall then be backfilled with air (air wash). The devices shall then be removed from the activation tank and leak tested within 1 hour after gas exposure with a scintillation-crystal-equipped counting station. Device encapsulations that come under the requirements of 3.2.3 shall be exposed to ambient air for a time not less than the "wait time" determined by 3.2.3. In no case will the time between removal from the activation chamber and test exceed 1 hour. This exposure shall be performed after gas exposure but before determining leak rate with the counting station. Device encapsulations that do not come under the requirements of 3.2.3 may be tested without a "wait time." (The number of devices removed from pressurization for leak testing shall be limited such that the test of the last device can be completed within 1 hour.) The actual leak rate of the component shall be calculated with the following equation:

      (ACTUAL READOUT IN NET COUNTS PER MINUTE) X QS

      Q= ------------------------------------------ -----------------------------------

      R

      Where Q = Actual leak rate in atm cc/sec, and QS and R are defined in 3.2.1.

      NOTE: CAUTION - discharge of krypton 85 into the atmosphere must not exceed limits imposed by local and Federal regulations.

    • 3.2.5 Failure criteria. Unless otherwise specified, devices that exhibit a leak rate equal or greater than the test limits of table II shall be considered as failures.

      NOTE: CAUTION - devices which do not exhibit a leak rate sufficient to fail seal test, may retain radioactive tracer gas in sufficient concentration to cause soft errors in complex, small geometry devices.

      TABLE II. Test limits for radioisotope fine leak method.

      Volumen of package cc Calculated Q
      %lt; 0.01 1 x 10-8
      %gt; 0.01 5 x 10-8
    • 3.2.6 Personnel precautions. Federal, some state and local governmental regulations require a license for the possession and use of krypton-85 leak test equipment. In the use of radioactive gas, these regulations and their maximum permissible exposure and tolerance levels prescribed by law should be observed.

    3.3 Test condition C1 or C2 - Fluorocarbon gross leak. Test condition C1 is a fixed method with specified conditions that will ensure the test sensitivity necessary. Test condition C2 is a flexible method that allows variance of test conditions in accordance with the formula in 3.3.1.2.

    • 3.3.1 Procedure applicable to fixed (C1) and flexible (C2) methods. The devices shall be placed in a vacuum/pressure chamber and the pressure reduced to 5 torr and maintained for one hour except that for devices with an internal volume >0.1 cm3, this vacuum cycle may be omitted. A sufficient amount of FC-72 or equivalent detector fluid shall be admitted to cover the devices. When the vacuum cycle is performed, the fluid will be admitted after the one-hour period but before breaking the vacuum. The devices shall then be pressurized per 3.3.1.1 (fixed) or 3.3.1.2 (flexible). When the pressurization period is complete the pressure shall be released and the devices removed from the chamber without being removed from a bath of detector fluid for greater than 20 seconds. A holding bath may be another vessel or storage tank. When the devices are removed from the bath they shall be dried for 2 +1 minutes in air prior to immersion in FC-40 or equivalent indicator fluid, which shall be maintained at 125_C +5_C. The devices shall be immersed with the uppermost portion at a minimum depth of 2 inches below the surface of the indicator fluid, one at a time or in such a configuration that a single bubble from a single device out of a group under observation may be clearly observed as to its occurrence and source. The device shall be observed against a dull, non-reflective black background though the magnifier, while illuminated by the lighting source, from the instant of immersion until, expiration of a 30-second minimum observation period, unless rejected earlier.
      • 3.3.1.1 Test condition C1 - Fixed method. Devices with an internal cavity <0.1 cm3 shall be pressurized at 60 psig for a duration of two hours minimum. Devices with an internal cavity volume >0.1 cm may be subjected to 30 psig (45 if the vacuum cycle was omitted) for ten hours if they cannot withstand the 60 psig, two hour condition.
      • 3.3.1.2 Test condition C2 - Flexible method. Devices shall be pressurized at 30, 60 or 90 psig for a minimum time determined by:

        TP = (0.1VFt)/(6 x 10-4 cm3)

        Where: TP = Time of pressurization in minutes

        V = Internal volume of device under test

        Ft = Filling time (from table III)

        TABLE III.

        Pressure Ft Minutes
        30 45
        60 15
        90 10

        The pressurization time used shall not be less than one Ft.

    • 3.3.2 Failure criteria. A definite stream of bubbles or two or more large bubbles originating from the same point shall be cause for rejection.
    • 3.3.3 Precautions. The following precautions shall be observed in conducting the fluorocarbon gross leak test:
      1. Fluorocarbons shall be filtered through a filter system capable of removing particles greater than 1 micron prior to use. Bulk filtering and storage is permissible. Liquid which has accumulated observable quantities of particulate matter during use shall be discarded or reclaimed by filtration for re-use. Precaution should be taken to prevent contamination.
      2. Observation container shall be filled to assure coverage of the device to a minimum of 2 inches.
      3. Devices to be tested should be free from foreign materials on the surface, including conformal coatings and any markings which may contribute to erroneous test results.
      4. A lighting source capable of producing at least 15 thousand foot candles in air at a distance equal to that which the most distant device in the bath will be from the source. The lighting source shall not require calibration but the light level at the point of observation (i.e., where the device under test is located during observation for bubbles) shall be verified.
      5. Precaution should be taken to prevent operator injury due to package rupture or violent evolution of bomb fluid when testing large packages.

    3.4 Test condition D - Penetrant dye gross leak. This test shall be permitted only for destructive verification of devices (see 3.5). The pressure chamber shall be filled with the dye solution to a depth sufficient to completely cover all the devices. The devices shall be placed in the solution and the chamber pressurized at 90 psig minimum for 3 hours minimum. For device packages which will not withstand 90 psig, 45 psig for 10 hours may be used. The devices shall then be removed and carefully washed, using a suitable solvent for the dye used, followed by an air-jet dry. The devices shall then be examined under the magnifier using an ultraviolet light source of appropriate frequency.

    • 3.4.1 Failure criteria. Any evidence of dye penetration into the device cavity shall constitute a failure.

    3.5 Test condition E - Weight gain gross leak.

    • 3.5.1 Procedure. The devices shall be cleaned by placing them in a container of clean Freon TF at 25_C and allowed to soak for 2 minutes minimum. The devices shall then be removed and placed in an oven at 125_C for 1 hour minimum, after which they shall be allowed to cool to room ambient temperature. Each device shall be weighed and the initial weight recorded or the devices may be categorized into cells as follows. Devices having a volume of <0.01 cc shall be categorized in cells of 0.5 milligram increments and devices with volume >0.01 cc shall be categorized in cells of 1.0 milligram increments. The devices shall be placed in a vacuum/pressure chamber and the pressure reduced to 5 torr and maintained for 1 hour except that for devices with an internal cavity volume >0.1 cc, this vacuum cycle may be omitted. A sufficient amount of FC-72 fluorocarbon or equivalent fluid shall be admitted to the pressure chamber to cover the devices. When the vacuum cycle is performed, the fluid shall be admitted after the 1-hour period but before breaking the vacuum. The devices shall then be pressurized to 60 psig except that 75 psig shall be used when the vacuum cycle has seen omitted. The pressure shall be maintained for 2 hours minimum. If the device will not withstand the 60 psig test pressure, the pressure may be lowered to 30 psig with the vacuum cycle and the pressure maintained for 10 hours minimum. Upon completion of the pressurization period, the pressure shall be released and the devices removed from the pressure chamber and retained in a bath of the fluorocarbon fluid. When the devices are removed from the fluid they shall be air dried for 2 +1 minutes prior to weighing. Transfer the devices singly to the balance and determine the weight or weight category of each device. All devices shall be tested within 4 minutes following removal from the fluid. The delta weight shall be calculated from the record of the initial weight and the post weight of the device. Devices which were categorized shall be separated into two groups, one group which shall be devices which shifted one cell or less and the other group which shall be devices which shifted more than one cell.
    • 3.5.2 Failure criteria. A device shall be rejected if it gains 1.0 milligram or more, If the devices are categorized, any device which gains enough weight to cause it to shift by more than one cell shall be considered a reject. A device which loses weight of an amount which if gained would cause the device to be rejected may be retested after it is baked at 125_C for a period of 8 hours.

    3.6 Retest. Devices which fail gross leak (test conditions C and E) may be retested destructively. If the retest shows a device to pass, that was originally thought to be a failure, then the device need not be counted as a failure in the accept number of LTPD calculations. Devices which fail fine leak (test conditions A1, A2, A4, or B) shall not be retested for acceptance unless specifically permitted by the applicable procurement document. Where fine leak retest is permitted, the entire leak test procedure for the specified test condition shall be repeated. That is, retest consisting of a second observation on leak detection without a re-exposure to the tracer fluid or gas under the specified test condition shall not be permissible under any circumstances. Preliminary measurement to detect residual tracer gas is advisable before any retest.

  4. SUMMARY. The following details shall be specified in the applicable procurement document:
    1. Test condition letter when a specific test is to be applied (see 3).
    2. Accept or reject leak rate for test condition A or B when other than the accept or reject leak rate specified herein applies (see 3.1.1.1, 3.1.1.2, 3.1.2, and 3.2.4).
    3. Where applicable, measurements after test (see 3).
    4. Retest acceptability for test conditions A and B (see 3.6).
    5. Order of performance of fine and gross if other than fine followed by gross (see 3).
    6. Where applicable, the device package pressure rating shall be specified if that rating is less than 60 psig.

Photographs

This photo shows an example of an IC that is failing the gross leak test. (Photo courtesy Sandia National Labs)

References on Leak Testing

  1. A. DerMarderosian and V. Gionet, "Water Vapor Penetration Rate Into Enclosures with Known Air Leak Rates," Proceedings 16th Annual Reliability Physics Symposium, 1978.
  2. D. Stroehle, "On the Penetration of Gases and Water Vapor Into Packages with Cavities and a Maximum Allowable Leak Rate," Proceedings 15th Annual Reliability Physics Symposium, 1977.
  3. D. J. Santeler and T. W. Moller, "Fluid Flow Conversion in Leaks and Capillaries," 1956 National Symposium on Vacuum Technology Transactions, Pergamum Press, pp. 29-36.
  4. J. G. Davy, "Calculations for Leak Rates of Hermetic Packages," IEEE Transactions on Parts, Hybrids, and Packaging, Vol. PHP-11, No. 3, 1975 5. D. A. Howl and C. A. Mann, "The Back-Pressurizing Technique of Leak Testing," Vacuum, Vol. 15, No. 7, pp. 347-352.
  5. Leakage Testing Handbook, NASA document N69-38843, General Electric Co. (obtainable from National Technical Information Service, US Department of Commerce, 5285 Port Royal Road, Springfield, Virginia 22161).
  6. Failure Analysis Techniques, A Procedural Guide, eds. E. Doyle Jr. and B. Morris, IITRI, Section III-C, 1980.