Stellar Industries
Thermal Management Specialists

Direct Bond Substrates (DBCu) Inspection Requirements

This specification details the inspection criteria for Direct Bond Copper (DBCu) bonded substrates. This specification governs Final Quality Inspection for full faced DBCu ceramics substrates and for laser scribed or diamond sawed patterned circuit DBCu substrates.


The following documents form a part of this specification to the extent specified. Unless otherwise indicated, the issue in effect on the date of procurement placement shall apply. Later issue of these documents may be used at the option of the supplier providing no degradation of the product ensues.


Military Specifications
MIL-I-45208A Inspection System Requirements
MIL-M-38510 Microcircuits, General Specification for

Military Standards
MIL-STD-105 Sampling Procedures and Tables for Inspection by Attributes
MIL-STD-883C Test Methods and Procedures for Microelectronics
MIL-STD-45662 Calibration System Requirements

AMS-C-26074 (Replacement for Mil-G-45208): Coatings, Electroless Nickel
Requirements for SAE International
ASTM B488-01 (Replacement for MIL-G-45204): Standard Specification for
Electrodeposited Coatings of Gold for Engineering Uses
AMS-QQ-N-290: Nickel Plating
PC-10-1000: Ceramic: Alumina
PC-10-1001: Ceramic: Beryllium Oxide
PC-10-1002: Ceramic: Aluminum Nitride
PS-INSPECT-3043: Inspection of Ceramics

In case of conflicts between this document's visual inspection criteria and customer or military inspection criteria, this document shall govern final acceptance of material for the unique features of DBCu substrates. These unique criteria are specified in Table 1B. After these unique DBCu substrate criteria, the customers' documents shall take precedence.The order for governing documents is as follows:

  • PS-INSPECT-3032 (This Document): Inspection: Direct Bond Copper Substrates
  • Customer Purchase Order
  • Customer Drawings
  • Customer Specifications
  • Customer Magnetic Data
  • Military Specifications as specified in contract
  • Industrial Standards and specifications specified in contract
  • ST2000-10 (IS-30-1001): DBCu Visual Inspection Criteria
  • Stellar Industries Applicable Specifications
  • In-Process Quality Requirements

Substrates specified by this drawing shall meet all of the requirements specified in this document.

Burr: A burr is a nodule of substrate of foreign material which has become fired onto the substrate.

Pit: A pit is a small depression in the surface.

Chip out: A chip out is an area along an edge or corner where the material has broken off.

Color Spot: A color spot is a diffuse, discolored area sometimes accompanied by a foreign inclusion. DIRECT BOND COPPER
Blister: A raised area in the plated surface which is evidence of adhesion problems with the plating process. If electroplated, non adherent or blistered areas can occur due to poor contact from misalignment of opposite sides of the substrate. This is not a cause for lot rejection.

Burr: Extemporaneous attached material usually resulting from a diamond sawed edge or machined copper surface.

Cracks: A separation in the ceramic surface which is held together by the copper foil. Cracks appear as gray lines and normally (but not always) extend to the edge of the substrate. Cracks can occur due to the stresses induced on the ceramic due to the expansion mismatch of copper in which substrates are cambered and bowed during the rapid cooling cycle after bonding.

Filled Via: Copper particle filled vias in ceramic bonded with eutectic oxides. The via is porous from one side of the ceramic.

Gouge: Scratch or tear in the copper exposing underlying ceramic usually from a mechanical abrasion.

Nodules: Solid raised copper protrusions limited in smaller area sizes than tents due in part to high oxygen entrapment and high temperatures used in the DBCu bonding temperatures. Normal post bonding operations of sanding with controlled grit sand paper or lapping compounds are acceptable techniques to reduce these protrusions. Perforated Copper: Copper with a series of vertically and horizontally spaced holes. These holes allow gases to be released during the eutectic reflow and vaporization occurring during the bonding process. The holes are normally partially filled with copper eutectic during the copper bonding process.

Pit: A depression in the copper resulting from collapse of the copper foil during DBCu eutectic bonding.

Plating Flap: An overhang of plating, usually from nickel or gold, that extends over the edge of an etched copper conductor area caused by undercutting of the copper from the etching process. The flap is normally removed with mechanical brushing techniques after etching.

Pullouts: Areas where the ceramic surface is damaged due to lapping or copper bonding that pulls out ceramic grains from the substrate.

Stains: Discolored areas on the metal areas or exposed ceramic. The stains on exposed ceramic can occur from the oxide formed during the DBC process that can later show up when the copper is etched. These stains are considered part of the DBC process. Other stains can occur from chemicals or copper oxidation. Many of these stains are removable with solvents.

Tent: An expanded dome of copper resulting from gas pressure released during eutectic bonding stage. Although also called blister, the term tent specifically refers to un-bonded areas of copper that are quite evident from either a raised copper area or from a white circular area that appears on the reverse side of the ceramic. These areas do not represent degraded or reliability prone areas, but are areas that never were bonded to the ceramic surface.

Voids: Voids exposing underlying ceramic on circuit side. No voids allowed greater than 50% of the original metallization area.


The physical dimensions and tolerances of the substrate shall be in accordance with the procurement document or drawing.

The adhesion tests shall be considered a destructive test and when specified shall be performed either on test coupons or on sample substrates from the same metallization lot. Samples rejected for cosmetic or other non related visual inspection criteria may be used. The tests shall be performed using a standard 90 bend pull test in a dynamic peeling mode. Stellars standard test conditions are as follows:

1) Tools:

  • Chatillon DFIS50 force gauge
  • Computer
  • Interface software.
  • Force gauge stand with speed control
  • Backing plates
  • Loctite 454

2) Test Conditions:

  • Witness sample: Usually 1 inch coupons
  • Width of copper lead: 0.5 inches wide
  • Copper strip from coupon is attached to force gauge
  • Actuate motor to initiate peel testing
  • Strength is monitored for 15 seconds and average strength is calculated.
  • Peel speed is set at 1/minute throughout test.
  • Strength monitoring is started at minimum strength of 1 lb.
  • Readings are calculated in lbs/in

3) Strength Conditions:

Condition As Bonded
A > 20lbs/in
B > 28lbs/in
C > 35lbs/in
Condition After patterning and plating and baking at 350C for 15 minutes
D > 10lbs/in
E > 14lbs/in
F > 20lbs/in

Unless otherwise stated on the Customer purchase order or drawing the standard Strength Condition will be B (greater than 28 lbs/inch) for unplated and Condtion E ( greater than 14 lbs/inch) for plated parts after plating tape and bake test.

The physical properties of the ceramic shall be as specified in the applicable customer's specification or applicable Stellar ceramic PS-INSPECT-3043 specification.

The surface finish shall be as specified on the drawing. Surface roughness tests shall be performed with a Mitutoya Surface Profilometer Model 601, or equivalent.

Ceramic: Allowable visual defects on the ceramic shall be in accordance with Table 1A. Copper: Allowable visual defects for the Direct Bond Copper shall be as specified in Table 1B.

Voids on Plates
For full faced DBC master substrates the total voiding criteria shall not exceed 10% of the area bonded. Ultrasound testing at additional costs are required to insure inspection to this criteria.

Filled Vias: For substrates with copper filled vias the inspection criteria is specified in Table 1C

If required, and unless otherwise specified by the customer drawing, the plating shall be 50 uin minimum of electroless or electrolytic nickel in conformance with AMS-C-26074 or QQ-N-290 and 50 uin minimum of electroless or electrolytic gold in conformance with ASTM B488-01, Type I, Code A (formerly Mil-G45204C). Visual inspection requirements shall be as specified in Table II.


The copper electrical resistivity of the tough pitch or OFHC copper shall be equal to or greater than 2.5 micro-ohm-cm.

The withstanding voltage shall be a minimum of 500 v for any conductor to conductor spacing less that 0.015 inches. For conductor spacing equal to or greater than 0.015 the minimum withstanding voltage is 1000 volts.


Each substrate shall be capable of meeting the following environmental requirements:
a. Storage temperature range: -65 C to + 200 C. Un-plated and exposed copper or nickel plated copper shall be protected from oxidation in a non-oxidizing atmosphere.
b. Operating temperature range: -65 C to + 200C. Un-plating and exposed copper or nickel plated copper shall be protected from oxidation in a non-oxidizing atmosphere.
c. Short term high temperature exposure:

  1. Alumina and Beryllium Oxide DBCu bonded substrates: 30 minute exposure to temperature up to 1000C maximum in an inert atmosphere. Repeated exposure up to ten times for Cu-sil 870C brazing temperatures.
  2. Oxided Aluminum Nitride: DBCu bonded: Maximum 10 minute exposure to Cu-Sil 870C brazing temperatures in an inert atmosphere.
  3. Activated Brazed Cu bonded AlN: Maximum 10 minute exposure to In-Cu-Sil 770C brazing temperatures in an inert atmosphere.

d. Short term low temperature exposure: Indefinite -65C exposure. Lower temperature operation has not been determined.

NOTE: Repeated exposure to temperatures exceeding 300C in a reducing atmospheres (H2/N2) will degrade adhesion of the copper to ceramic surface. The DBC bond is an oxide bond which is affected by reducing atmospheres.

All DBC laminations of copper and Alumina or BeO substrates shall be capable of 200 cycles of -65C to +125C without evidence of degradation. The test condition shall be in accordance with Mil-Std 883C, Method 1010. Copper laminations thinner than 0.005 on Aluminum nitride shall also be capable of 200 cycles of -65C to +125C without evidence of degradation

TABLE 1A: Ceramic Inspection Criteria

Dimensions Length, Width Tolerance 1% or 0.005" none   Caliper
  Thickness Tolerances 10% or 0.002" none AQL II 2.5% Micrometer
  Squareness 90 1 none AQL II 2.5%  
  Blisters >0.001 inch high none AQL II 2.5% Microscope 10 - 30 x with reticle
  Burrs >0.001 inch high none AQL II 2.5% Microscope 10 - 30 x with reticle
  Chips - Edge >0.040 inch none AQL II 2.5% Microscope 10 - 30 x with reticle
  Chips - Corner >0.030 inch none AQL II 2.5% Microscope 10 - 30 x with reticle
  Cracks   none AQL II 2.5% Microscope 10 - 30 x with reticle
  Holes, Pits, Pocks >0.010 inch none AQL II 2.5% Microscope 10 - 30 x with reticle
  Inclusion >0.010 inch none AQL II 2.5% Microscope 10 - 30 x with reticle
  *Color Spots >0.0.30 inch none AQL II 2.5% Microscope 10 - 30 x with reticle

*Note: Some discoloration or staining of the ceramic occurs due to the oxide bond.

TABLE 1B: Direct Bond Copper Inspection Criteria

Copper Thickness 0.005", 0.008",0.012" none AQL II 2.5% Caliper
Cu Thickness Tolerance 0.002 inches none AQL II 2.5% Micrometer
Camber < 0.005 inch/inch none AQL II 2.5% Parallel Plate
Tents /Voids >0.100 inch diameter none AQL II 2.5% Microscope 10 - 30 x with reticle
  >0.050" but <0.100" dia <1/inch 2 * AQL II 2.5% Microscope 10 - 30 x with reticle
  >0.010 but < 0.050" dia <5/inch 2 * AQL II 2.5% Microscope 10 - 30 x with reticle
  >0.001 but <0.010" dia <10/inch 2 * AQL II 2.5% Microscope 10 - 30 x with reticle
Burrs >0.003 inch height none AQL II 2.5% Microscope 10 - 30 x with reticle
Pits (in copper) >0.100 inch diameter none AQL II 2.5% Microscope 10 - 30 x with reticle
  >0.050" but <0.100" dia <1/inch 2** AQL II 2.5% Microscope 10 - 30 x with reticle
  >0.010" but <0.050" dia <5/inch 2** AQL II 2.5% Microscope 10 - 30 x with reticle
  >0.001 but < 0.010" dia <10/inch 2** AQL II 2.5% Microscope 10 - 30 x with reticle
  An aggregate area of pits or Pinholes greater Than 10% of pad Or greater than 50% of width
Nodules >0.001" height none    
Grain Size Not Controlled      
Surface Roughness <200 -in rms none   Surface Profilometer
Conductor Width/Spacing >150% design width* none   Or as specified
  >50% design width* none   Or as specified
Conductor Area >70% design width* none   Or as specified
Stains Surface copper oxides Acceptable    

*but not reducing a conductor width by more than 50% of the design width
**but not reducing a conductor area by more than 25%

TABLE 2: Plating Criteria

Thickness Within Specifications none Xray or Dektac or Beta Backscatter
Adherance Bake test 350C for 15 min no evidence of blisters  
  followed by tape test no flaked or peeled plating  
Coverage No exposure of copper except for sidewalls    

TABLE 1C: Copper Filled Via Inspection Criteria

Chart Guidelines


The manufacturer shall establish, implement and maintain an inspection system in accordance with MIL-I-45208A to assure compliance wit the applicable requirements and quality standards of this specification. QUALITY CONFORMANCE INSPECTION
The substrates shall be subjected to 100% in-process inspection to the visual inspection criteria specified in this specification. Final Inspection performed at Quality Control will be in accordance with the sampling plan established by MIL-STD-105 or as specified in the customer specifications.

Lot traceability shall be mintained for the following:
a. Substrate material
b. DBCu metallization run
c. Plating Lot

Optical inspection shall be in accordance with MIL-G-45204C and this specification, and shall be capable of 10X to 30X magnification. A micrometer eyepiece disc calibrated with a stage micrometer shall be used to make linear measurements. High intensity incident lighting shall be used to illuminate the substrate during inspection.



Surface finish shall be in accordance with ANSI B46.1

The thickness measurements shall be made using calibrated micrometers. Each substrate shall be measured near each of the four corners and near the center. The thickness shall be the average of the five readings.

5.1.3 CAMBER
Conformance to the camber requirements shall be determined as follows on substrates:
a. Determine the average thickness.
b. Prepare two spacers. The spacers thickness shall be equal to the substrate thickness plus its tolerance plus a thickness equal to the specifice camber to measure times the substrate length. Spacers = Sub Thickness + Tolerance + Camber*Substrate length
c. Clamp the prepared spacers between two flat (flatness within 0.0002 inch) plates so that the distance between the spaces is 1 time the length of the substrates to be measured.
d. Mount the clamped, separated plates so that they are 45 +- 5 degrees from the vertical axis.
e. Place each test substrate at the top edge of the prepared slot and let it free-fall.
f. Any substrate that fails to pass through the slot of its own weight has excessive camber.

Reject Chart


Stellar Industries Corp. 50 Howe Avenue Millbury, MA 01527 | 877.840.1884 | 508.865.1668 |