Overview of MIL-STD-883
MIL-STD-883 is the US Department of Defense test method standard that defines the requirements for screening, qualification, and quality conformance of microelectronic devices and hybrid microcircuits. It is not a single test -- it is a comprehensive framework of over 200 test methods covering everything from visual inspection to radiation hardness.
For hybrid microcircuit manufacturers and users, MIL-STD-883 serves two primary functions: (1) as a set of detailed test procedures that can be invoked by other specifications (MIL-PRF-38534, MIL-PRF-38535), and (2) as a quality reference that defines workmanship standards for the industry.
The standard is maintained by the Defense Information Systems Agency (DISA) and is updated periodically. It is available from the US Department of Defense at quicksearch.disa.mil.
Quality Classes
Class B
Class B applies to hybrid microcircuits intended for non-space, military ground and sea applications. Class B screening requires a defined sequence of tests including visual inspection, electrical testing at temperature extremes, mechanical tests (shock, vibration, acceleration), and hermeticity testing. Class B has historically been the "standard" military grade for hybrids.
Class K
Class K applies to hybrids intended for space and high-reliability military aerospace applications. Class K requires all Class B tests plus additional requirements: 100% die shear strength testing (MIL-STD-883 Method 2011), 100% wire pull testing (MIL-STD-883 Method 2010), acoustic microimaging (C-SAM), and more stringent hermeticity acceptance criteria. Class K is the highest quality class for hybrid microcircuits.
QML (Qualified Manufacturers List)
The QML system, managed under MIL-PRF-38535, qualifies manufacturers rather than individual part numbers. QML-qualified hybrid manufacturers have demonstrated their processes, quality system, and test capability to an accredited qualified testing laboratory (QTL). Procurement from QML-qualified manufacturers reduces the buyer's qualification burden because the manufacturer's processes have already been audited and qualified.
Visual Inspection (Method 2031)
MIL-STD-883 Method 2031 defines visual and mechanical inspection criteria for hybrid microcircuits. The method covers: substrate condition (cracks, chips, delamination), conductor pattern integrity (linewidth, spacing, registration), component placement (position, orientation, skew), wire bond geometry (loop height, wire sweep, wedge/ball placement), encapsulation and coating quality, marking legibility and permanence, and workmanship defects.
Method 2031 inspection is performed at the appropriate magnification using a stereo microscope. Critical features (bond pad area, wire bond geometry) require higher magnification than general workmanship items. The inspection criteria distinguish between "minor," "major," and "critical" defects, with critical defects requiring rejection of the unit.
Hermeticity (Method 1014)
MIL-STD-883 Method 1014 defines procedures for detecting leaks in hermetically sealed microelectronic packages. Hermeticity testing is critical because moisture ingress into the package cavity will cause corrosion of conductor traces, wire bonds, and die metallisation -- leading to early field failure.
Fine Leak Test
The fine leak test (typically using helium mass spectrometry) detects leak rates in the range of 10^-8 to 10^-5 atm-cc/s. The unit is pressurised in helium, then the helium decay rate from the package is measured. Fine leak testing detects small leaks that would allow moisture ingress over the application's lifetime but may not be detectable by gross leak methods.
Gross Leak Test
The gross leak test detects larger leaks (greater than 10^-5 atm-cc/s) using a fluorocarbon (FC-72 or equivalent) bubble test or a radioisotope (Kr-85) method. In the bubble test, the unit is immersed in heated fluorocarbon fluid after pressurisation -- escaping bubbles indicate a gross leak. Gross leak testing is faster and cheaper than fine leak testing but cannot detect the smaller leaks that fine leak testing finds.
Die Shear (Method 2011)
MIL-STD-883 Method 2011 measures the mechanical strength of the die attach bond. The test uses a die shear tool that applies a force parallel to the die attach interface, pushing the die off the substrate. The force at failure, divided by the die attach area, is the die shear strength. Minimum acceptable die shear strength is specified by the applicable detail specification -- typically 25 MPa for eutectic die attach and 10 MPa for conductive epoxy at room temperature.
Method 2011 is a destructive test (DTT: Destructive Test Technique) for non-QML products but can be performed as a nondestructive test (NDT) at reduced force levels for some applications. For Class K hybrids, 100% die shear testing is mandatory.
Wire Pull (Method 2010)
MIL-STD-883 Method 2010 measures the mechanical strength of individual wire bonds. A wire pull tool hooks under the wire at its midpoint and pulls upward until the wire breaks or the bond lifts. The pull strength is recorded in grams or grams-force. Minimum acceptable wire pull strength is defined in the applicable detail specification, typically 3-5 grams for 1-mil (25 micrometre) aluminum wire and 5-8 grams for 1-mil (25 micrometre) gold wire.
Wire pull testing is performed as a 100% production test for Class K hybrids and as a sampling test for Class B (performed on a sample of units from each lot). Wire pull failures can indicate poor bond formation (insufficient energy, contamination, improper pad metallisation) or wire damage from subsequent processing.
Bond Strength (Method 2011)
In addition to wire pull testing (Method 2010), Method 2011 bond strength testing encompasses wire shear testing -- a nondestructive alternative that uses a shear tool to push the wire bond off the pad without applying force to the die. Wire shear testing can be performed on 100% of bonds without damaging them (at appropriate force levels) and is widely used as a production screening tool in combination with statistical sampling of destructive wire pull tests.
Temperature Cycling (Method 1010)
Temperature cycling (MIL-STD-883 Method 1010) exercises the hybrid through alternating high and low temperature extremes to screen for infant mortality failures caused by thermal stress. The test conditions (minimum and maximum temperatures, number of cycles, dwell time, transfer time) are defined by the applicable detail specification. Common conditions for military hybrids are -55 degrees C to +125 degrees C for 100 cycles, with 15-minute dwells and less than 5-minute transfers between temperatures.
Internal Water Vapor (Method 1018)
MIL-STD-883 Method 1018 measures the moisture content inside the sealed package cavity. Moisture inside the cavity will cause corrosion failures during operation, particularly at elevated temperatures. The test uses a dew point or karl fisher titration method to determine the internal water vapor content. The maximum acceptable moisture level is 5000 ppm (0.5% by volume) for most military hybrids at the time of seal. This test is performed on a sample basis as part of qualification and periodic lot acceptance testing.
High Temperature Storage (Method 1008)
Method 1008 stores the hybrid at an elevated temperature (typically 150 degrees C or the maximum rated storage temperature) for a defined period (typically 168-1000 hours). The purpose is to accelerate failure mechanisms related to chemical reactions (oxidation, intermetallic growth, polymer degradation) and to screen for units with latent defects. High temperature storage is a standard screening and qualification test for military and aerospace hybrids.
Particle Impact Noise Detection (Method 2020)
PIND (MIL-STD-883 Method 2020) detects loose particles inside the sealed package cavity that could cause electrical short circuits if they move and contact active circuitry. The test uses a vibration table to excite the package and an acoustic sensor to detect the sound of particle impacts. Any particle detected above a defined threshold constitutes a failure. PIND is required for most military hybrid procurement and is performed as a 100% screen on Class K hybrids.
Building a Screening Tree
A screening tree is the sequence of inspections and tests applied to each hybrid during production. The tree must ensure that defective units are removed before they reach expensive or destructive later operations. A typical MIL-STD-883 screening tree for Class K hybrids follows this sequence:
Incoming inspection: Verify substrate, die, wire, and component identity against the BOM. Inspect die attach area for contamination or surface finish issues.
Pre-assembly inspection: Inspect substrate conductor pattern under magnification. Perform electrical test of conductor continuity on test vehicles.
Post-die attach: Cure verification, die attach inspection (die tilt, fillet, voiding -- voiding typically inspected by acoustic microimaging). Die shear test (100% for Class K).
Post-wire bonding: Wire pull and wire shear testing (100% for Class K). Acoustic microimaging (C-SAM) to detect wire sweep, die attach delamination.
Pre-seal inspection: Final visual inspection, cleaning verification, wire loop geometry measurement.
Seal: Hermetic seam sealing or glass frit sealing. Perform gross leak and fine leak testing on 100% of units.
Post-seal: PIND test (100% for Class K). Electrical final test over temperature. Internal water vapor test (sample).
Acceptance Criteria Table
| Test Method | Parameter | Class B | Class K |
|---|---|---|---|
| Visual Insp. (2031) | Workmanship defects | Per inspection criteria | Per inspection criteria |
| Die Shear (2011) | Min. strength | Sample (3 units/lot) | 100% |
| Wire Pull (2010) | Min. pull force | Sample (3 units/lot) | 100% |
| Hermeticity (1014) | Fine + gross leak | 100% | 100% |
| Temp. Cycling (1010) | Cycles | 100 cycles -55/+125 C | 100 cycles -55/+125 C |
| PIND (2020) | Loose particles | Not required | 100% |
| Water Vapor (1018) | Max moisture | 5000 ppm (sample) | 5000 ppm (sample) |
| C-SAM | Delamination/voids | Sample | 100% |
| Electrical Test | Parameters | Room + hot | Room + hot + cold |
Common Pitfalls
- Insufficient wire pull sample size: MIL-STD-883 requires a minimum sample size for wire pull testing on Class B hybrids -- 3 units per lot is the minimum, but higher sample sizes provide better detectability of process drift. Many low-cost assemblers skimp on sample size to reduce cost.
- Voiding in die attach: Void content above the die attach area exceeding 25% of the attach area can significantly reduce thermal conductance and create localised stress concentrations. Acoustic microimaging (C-SAM) is the standard method for detecting die attach voids.
- Wire sweep from mould compound: When the hybrid is overmoulded or encapsulated, the mould compound flow can displace wire bonds (wire sweep). Wire sweep can cause adjacent wire shorts or degrade wire loop geometry. Proper mould compound selection, mould tool design, and mould parameters are essential.
- Inadequate incoming inspection: Substrate surface finish, conductor adhesion, and die back-side metallisation must be verified before assembly. Assembling on non-conforming substrates consumes expensive dice before the problem is detected.
- Misapplying Class B vs. Class K: Class B and Class K have different screening requirements. Using Class B processing for a part intended for Class K application is a common non-conformance that results in customer rejection.