Browse Prior Art Database

High-Resolution Measurement System for Package Thermal Resistance

IP.com Disclosure Number: IPCOM000007941D
Publication Date: 2002-May-07
Document File: 3 page(s) / 234K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method for cooling packages during high-resolution thermal resistance tests. Benefits include higher precision in measuring junction-to-case thermal resistance, the ability to dissipate more heat, and a shorter testing time.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 50% of the total text.

High-Resolution Measurement System for Package Thermal Resistance

Disclosed is a method for cooling packages during high-resolution thermal resistance tests.

Benefits include higher precision in measuring junction-to-case thermal resistance, the ability to dissipate more heat, and a shorter testing time.

Background

Currently, thermal resistance ("T-res") metrology is carried out by using a conductive wax to bond a thermo-couple and active heatsink to a package. This solution is unable to resolve junction-to-case resistance variations less than 0.012 oCcm2/W (large die), and to handle more than 160 watts.

General Description

The disclosed method uses impingement cooling to cool a package during a test (see Figure 3). Water from a temperature-controlled reservoir is pumped into a pipe which releases high velocity jets onto an integrated heat spreader (IHS) surface. A rubber seal presses against the IHS

surface and prevents any water from escaping. After flowing over the IHS surface, the water falls back into the reservoir.

Experimental verification

Impingement cooling has been verified at a low impingement velocity. Seventy four watts were removed from a large die package (2C11L = 800mil x 800mil die) at 2.4 m/s water velocity, with only 0.50 oCcm2/W junction-to-liquid thermal resistance and less than 1oC rise in water temperature (see Figure 1). Figure 2 shows how the convection thermal resistance would decrease at higher impingement velocities. Above 50 m/s, the convective resistance is relatively constant, and does not vary by more than 0.001 oCcm2/W per 1 m/s velocity variation.

Details of invention

Figure 4 shows the thermal resistance network for the disclosed method. The total resistance (Rtotal) consists of the package resistance (Rjc) and the convection resistance (Rcf). During a test, the total resistance is found by measuring the incoming water temperature (Tf0) and the die junction temperature (Tj). Dividing this temperature difference by the die power (Q) gives Rtotal. For high impingement velocities, the water temperature at the case (Tf1) is almost identical (within 0.06 oC) to the water temperature in the pipe (Tf0). Also, the convection resistance (Rcf) is nearly constant for impingement velocities greater than 50 m/s (see Figure 2); t...