The process to assemble a heatspreader package using the 2.16 inch square PGA 273-ld with 1.5 in square Cu-W heatspreader is certified. The process has met the reliability stress requirements. The heatspreader is being introduced as part of the conversion to the Pentium® Processor C-step.

Quality and Reliability Indicators

Temperature Cycle Condition C
Clean T/C results indicate the material's ability to withstand thermal fatigue stresses.

Stress Endpoint Goals (DPM/conf level) Sampling, 0 fails Result 10 cycles 500 cycles 1000 cycles Electrical & Hermeticity for info <0.5%@60% <1% @90% - 183 230 0/214 0/279 0/277 500 cycles Die Visual & Hermeticity for info - 0/102   Wirepull for info - 0/200 wires

Process Monitors
Process monitors indicate no major concerns with the process. Adhesion of the material is shown to be very strong from the studpull monitor. X-ray performed on a sample of laminated ceramic packages without heatspreaders showed the die attach voids to be of no issue. BLT and die tilt monitors indicated a stable process.

Monitor Endpoint/Spec Result Mark Permanency Mark Perm. test 0/174 Acoustic PIND 0/414 Centrifuge PIND, visual 0/414 Bond adhesion - Pre-seal - Post-seal Wirepull >3gm >2gm 0/800 wires 0/800 wires

Pentium® Processor Automated Adhesive Die Attach Qualification results The Automated Adhesive Die Attach process is currently used on a number of PGA packages at Intel including 168, 169, 208, 262 and 280 lead counts. All versions of the Intel® 486^TM CPU family have already converted to this assembly process step under previous customer notifications. This is a fully qualified, fully manufacturable process step. Data presented here is specific to the 273-ld PGA package with attached heatspreader and the Pentium® Processor.

Quality and Reliability Indicators

Electrical Characteristics
Comparisons done between Electrical Characterization lots manufactured using the Adhesive Die Attach process and control lots assembled using the current Eutectic die attach showed no significant differences in AC timings, DC levels, Icc and leakages.

Temperature Cycle Condition C
Clean T/C results indicate the material's ability to withstand thermal fatigue stresses.

Stress Endpoint Goals (DPM/conf level) Sampling, 0 fails Result 500 cycles 1000 cycles Electrical " <0.5%@60% <1% @90% 183 230 0/309 0/308

Process Monitors
Process monitors indicate no major concerns with the process. Adhesion of the material is shown to be very strong from the studpull monitor. X-ray performed on a sample of laminated ceramic packages without heatspreaders showed the die attach voids to be of no issue. BLT and die tilt monitors indicated a stable process.

Monitor Endpoint/Spec Goal DPM/confidence level Sampling, 0 fails Result Die backside damage Visual (no damage) <0.95% @60% 96.00 0/96 D/A thickness & die adherence BLT>0.5mils, <2.5mils Tilt<2.0mils X-ray* void<10% Studpull>200psi <0.95% @60% <0.95% @60% <0.95% @60% 96 96 96 0/96 0/96 0/96

* X-ray was performed on laminated ceramic packages without heatspreaders attached.

Preliminary Heat Spreader Package Specification For The Pentium® Processor This document contains the changes to the Pentium® processor, 273 pin PGA package, mechanical specification (chapter 9 of the Pentium® Processors User's Manual, 1993) necessitated by the addition of a heat spreader to the package for improved thermal performance of the die.

Customer Summary:
Spec. changes:

1. The A3, A4, D2, D3, D4 and F dimensions are added to the specification for the heat spreader package.
2. The weight of the heat spreader package increases to approximately 2X the weight of the standard PGA package. (70.6472 grams vs. 33.2003 grams)

Pentium® Processor Package Information Summary

Package Type Total Pins Pin Array Package Size Pentium® Processor PGA 273.00 21x21 2.16" x 2.16" 5.47cm x 5.49 cm

NOTE: See D.C. Specifications for more detailed power specifications.

Figures 1 and 2 show the package dimensions for the Pentium® processor. The mechanical specifications are provided in Table 2.

Figure 1. Pentium® Processor Package Dimensions

Top View of Package

Figure 2. Pentium® Processor Package Dimensions

Table 2. Pentium® Processor Mechanical Specifications
Family: Ceramic Pin Grid Array Package

Symbol Millimeters   Inches     Min Max Notes Min Max Notes A 2.84 3.51 Solid Lid 0.11 0/14 Solid Lid A1 0.33 0.43 Solid Lid 0.01 0.02 Solid Lid A2 2.51 3.07   0.10 0.12   A3 3.66 4.72   0.14 0.19   A4 3.33 4.29   0.13 0.17   B 0.43 0.51   0.02 0.02   D 54.61 55.11   2.15 2.17   D1 50.67 50.93   2.00 2.01   D2 37.85 38.35   1.49 1.51 Spreader Size D3 40.34 40.95   1.59 1.61 Spreader Size D4   8.38     0.33 Ref. to Pin 1 E1 2.29 2.79   0.09 0.11   F 0.13   Flatness of spreader measured diagonally   0.01 Flatness of spreader measured diagonally L 3.05 3.30   0.12 0.13   N 273.00   Total Pins 273.00   Total Pins S1 1.65 2.16   0.07 0.09

Thermal Specifications: Pentium® Processor, with and without Heat Spreader
This document contains the changes to the Pentium® Processor thermal specification (chapter 10 of the Pentium^TM Processor's User's Manual, 1993) necessitated by the addition of a heat spreader to the package for improved thermal performance of the die.

How does the heat spreader improve package thermal performance?
Since the Pentium® Processor requires an external heat sink in order to maintain the junction and case temperatures below the acceptable levels, the main contributors to the total junction to ambient thermal resistance are junction to case (Theta_JC), case to heat sink (Theta_CS), and heat sink to ambient (Theta_SA) thermal resistance's. Theta_JC is mainly a function of internal construction of the package and packaging material thermal properties such as the die size, die attach and ceramic thermal conductivity. Theta_CS is a function of the thickness and thermal properties of the interface material between the package and heat sink, package and heat sink flatness and surface finish and effective heat transfer area between the package and the heat sink . Theta_SA is a function of heat sink design and the airflow type and rate.
Using a heat spreader in the package lowers the overall thermal resistance in two ways:

1) It increases the effective heat transfer area between the package and the heat sink and as a result lowers Theta_CS. The actual reduction in Theta_CA depends on the magnitude of Theta_CS  without a heat spreader. The larger a value of Theta_CS without using a heat spreader, the larger will be the reduction in the value of JA if a heat spreader is used.

2) A heat spreader may also improve the heat sink thermal performance by increasing the effective heat transfer area in the heat sink and making the fins away from the die more effective.

Lower case temperature with a heat spreader package
In the case of the heat spreader Pentium® package, the case temperature is defined as the center of the package top surface on the heat spreader. To measure the case temperature, the procedure outlined in chapter 10 of the Pentium® Processors User's Manual, 1993 has to be followed. The thermal specification of the heat spreader package calls for 5C lower case temperature compared to the non spreader package for both 60 and 66 MHz versions. The reason for the lower case temperature of the heat spreader package is its lower junction to ambient thermal resistance compared to a non spreader package with the same heat sink. For example, at 66 MHz, the heat spreader package will have .4 x 16 = 6.4C lower case temperature with the same heat sink design and grease interface. This implies that in a system designed for a non spreader package, if the non spreader package is replaced with a heat spreader package, the measured case temperature will be lower by 6.4C for the 66 MHz and 5.8C for the 60 MHz versions. The actual reduction in the case temperature will be slightly higher or lower depending on the efficiency of the thermal interface. Therefore, a more conservative value of 5C is used as the difference between the case temperature specifications of the two package types for both frequency versions. The expectation is that the ambient temperature in the system will be maintained while gaining the benefits of lower junction and case temperatures when the heat spreader package is added to an existing system with the same airflow and unmodified heat sink.

Thermal Specification For The PGA Package Without Heat Spreader

1. T_C (case temperature) 0°C to 85°C @60 MHz .
2. T_C (case temperature) 0°C to 75°C @66 MHz.

Theta_CA** Vs Airflow (ft/min)

ThetaJC   0   200 400 600 800 1000 With 0.25" Heat Sink 0.6 8.3 5.8 3.9 3.0 2.5 2.2   With 0.35" Heat Sink 0.6 7.9 5.0 3.4 2.8 2.2 2.0   With 0.65" Heat Sink 0.6 6.4 3.4 2.3 1.8 1.5 1.3   Without Heat Sink 1.2 11.6 9.4 6.7 5.4 4.6 4.2

* Heat Sink: 2.05 in2 base, omni-directional pin AI heat sink with 0.050 in. pin width, 0.143 in pin-to-pin center spacing and 0.150 in. base thickness. Heat sinks are attached to the package with a 2 to 4 mil thick layer of typical thermal grease. The thermal conductivity of this grease is about 1.2 w/m c. * Theta_CA values shown in this table are typical values. The actual Theta_CAvalues depend on the air flow in the system (which is typically unsteady, non uniform and turbulent) and thermal interactions between Pentium® CPU and surrounding components through PCB and the ambient.

1. T_C (case temperature) 0°C to 80°C @60 MHz .
2. T_C (case temperature) 0°C to 70°C @66 MHz .

Theta_CA** Vs Airflow (ft/min)

ThetaJC   0   200 400 600 800 1000     With 0.25" Heat Sink   0.6 7.9 5.4 3.5 2.6 2.1 1.8 With 0.35" Heat Sink   0.6 7.5 4.6 3.0 2.4 1.8 1.6 With 0.65" Heat Sink   0.6 6.0 3.0 1.9 1.4 1.1 0.9 Without Heat Sink   1.2 10.5 8.2 5.5 3.8 2.8 2.4

• Heat Sink: 2.05 in2 base, omni-directional pin AI heat sink with 0.050 in. pin width, 0.143 in pin-to-pin center spacing and 0.150 in. base thickness. Heat sinks are attached to the package with a 2 to 4 mil thick layer of typical thermal grease. The thermal conductivity of this grease is about 1.2 w/m c.
• Theta_CA values shown in this table are typical values. The actual Theta_CAvalues depend on the air flow in the system (which is typically unsteady, non uniform and turbulent) and thermal interactions between Pentium® CPU and surrounding components through PCB and the ambient.

ESD-MIL (Mil-Spec Notice 8)
B-1 Stepping

Date Fab Lot Quantity Failures 5/6/93 53070218-A 4 0 5/7/93 9306142FNA 1 0 5/7/93 9306411FNA 1 0 5/15/93 C304954FNA 2 0 5/15/93 53070223-A 3 0 Totals   11 0

C-0 Stepping

Date Fab Lot Quantity Failures 7/26/93 53229040NF 1 0 7/26/93 53229060-C 2 0 Totals   3 0

LATCH-UP (Vcc & I/O)
B-1 Stepping

Date Fab Lot Quantity Failures 5/17/93 53070218-A 10 0 5/17/93 C304956RNA 4 0 5/17/93 9305143FNA 3 0 5/17/93 53120170-A 1 0 5/17/93 C304433F-A 1 0 5/17/93 9306411FNA 1 0 Totals   20 0

HVTOL (7 Volts, 125 Degrees C)
C-0 Stepping

Fab Lot # 168 Hrs 500 Hrs 1K Hrs 53229060NA 0/26 0/26 0/25 53229060NA 0/93 2/93 a 0/91 53229040NC 0/24 0/24 0/24 53229070NA 0/22 0/22   53229070NA 0/26 0/25   Net Reject/Total 0/190 2/189 0/140

Analysis Update
a) These two units are in analysis.

System Assurance / Compatibility Validation: The purpose of System Assurance is to test the Pentium® Processor in a system level environment with emphasis on compatibility and validation to the published specification.

Compatibility Validation
A large number of standard software programs are being used to verify compatibility - including all major operating systems, network environments and over 200 application and benchmark programs. Additional testing has been performed by an external compatibility lab.

Thermal Compliance
The thermal compliance stress monitors the component package, system case interior and ambient temperatures. The thermal stresses are run at case operating specification limits. This verifies that the test systems and the test components are operating within thermal specifications.

Performance Verification
Industry standard benchmark tests are used to verify performance enhancements.

Beta Site Testing
Beta sites thoroughly evaluate the devices, with emphasis on code compatibility. Beta Sites fill out and return a report form when the evaluation is completed.

System Assurance Skew Testing
Process-skew units are run at extreme voltages and temperatures to find approximate pass/fail break points in a system environment.

System Validation Lab Coverage
Several internally designed systems are used to verify all new features, run regression tests, an exercise the device using strings of instructions paired with various external events. All testing is performed at the highest frequency of each flavor. Testing is done at different environmental conditions within the datasheet spec.

This applies to:

Intel® Pentium® Processor