Evaluation of printed-circuit boards materials for high temperature

Evaluation of printed-circuit boards materials for high temperature operation Oriol AVIÑO -S ALVADO, Wissam S ABBAH, Cyril B UTTAY, Hervé M OREL, Pascal B EVILACQUA Laboratoire Ampère, Lyon, France

10/7/17

1 / 27

Outline

Introduction Experimental Protocol Definition of the test conditions Test results Conclusions 2 / 27

Outline

Introduction Experimental Protocol Definition of the test conditions Test results Conclusions 3 / 27

Introduction Gate Driver for SiC MOSFET Power Module

I

Relatively complex board I I I

I

Need for prototyping solution I I

I

Several ICs Large magnetic components Not suited to power substrate (DBC) Easy procurement Cheap

Objective: eventually 200 °C I

Current SiC MOSFETs limited to 175 °C

Ü Use of Printed Circuit Boards 4 / 27

Objectives

I

Investigate behaviour of some PCBs at high temperature I I

I

Propose a test configuration for gate driver circuits I I

I

First tests at Ampère on PCB ageing Work on test protocol Thermal cycling not investigated Lab specific environment investigated (N2 atmosphere)

Also, investigation on solder I I

Not presented here Same approach (cheap and easy solution for functionnal testing)

5 / 27

Outline

Introduction Experimental Protocol Definition of the test conditions Test results Conclusions 6 / 27

Test Vehicles

I

6-layer PCBs, 70 µm Cu (2 oz), Ni/Au finish (6 µm/80 nm) I I

I

One set on FR-4 (Panasonic R-1755-V, TG >170°C) One set on polyimide (Arlon 85N, TG >250°C)

3 test patterns: I I I

Surface isolation resistance (500 µm spacing) 20x20 mm2 capacitance Daisy-chained vias (450 µm hole diameter) 7 / 27

Ageing conditions

Source: MMM

I I

Isothermal ageing (“storage conditions”, constant temperature) Two atmosphere compositions: I I

Air (standard, forced convection oven) Low-oxygen content (N2 supply, 02 content monitored to 250 °C) Solder resist degrades rapidly at 190 °C I

I

Atmosphere has a great influence I I I I

I

But this does not seem to cause much harm In air, even polyimide degrades, albeit slower than FR4 In nitrogen, no change observed Interesting for functionnal testing purposes (cheaper PCBs) Thermal cycling not considered here

Solder to be investigated I

I

SAC attach investigation at 190 °C failed (Pb contamination caused early failure) SAC attach at 175 °C worked fine after 1000 h

26 / 27

Conclusions – Regarding results

I

I

As expected, FR4 (Tg > 170 °C) more sensitive to temperature than polyimide (Tg > 250 °C) Solder resist degrades rapidly at 190 °C I

I

Atmosphere has a great influence I I I I

I

But this does not seem to cause much harm In air, even polyimide degrades, albeit slower than FR4 In nitrogen, no change observed Interesting for functionnal testing purposes (cheaper PCBs) Thermal cycling not considered here

Solder to be investigated I

I

SAC attach investigation at 190 °C failed (Pb contamination caused early failure) SAC attach at 175 °C worked fine after 1000 h

26 / 27

Conclusions – Regarding results

I

I

As expected, FR4 (Tg > 170 °C) more sensitive to temperature than polyimide (Tg > 250 °C) Solder resist degrades rapidly at 190 °C I

I

Atmosphere has a great influence I I I I

I

But this does not seem to cause much harm In air, even polyimide degrades, albeit slower than FR4 In nitrogen, no change observed Interesting for functionnal testing purposes (cheaper PCBs) Thermal cycling not considered here

Solder to be investigated I

I

SAC attach investigation at 190 °C failed (Pb contamination caused early failure) SAC attach at 175 °C worked fine after 1000 h

26 / 27

Conclusions – Regarding results

I

I

As expected, FR4 (Tg > 170 °C) more sensitive to temperature than polyimide (Tg > 250 °C) Solder resist degrades rapidly at 190 °C I

I

Atmosphere has a great influence I I I I

I

But this does not seem to cause much harm In air, even polyimide degrades, albeit slower than FR4 In nitrogen, no change observed Interesting for functionnal testing purposes (cheaper PCBs) Thermal cycling not considered here

Solder to be investigated I

I

SAC attach investigation at 190 °C failed (Pb contamination caused early failure) SAC attach at 175 °C worked fine after 1000 h

26 / 27

Thank you for your attention

[email protected]

This work was funded by the CORAC (French Research Council for Civil Aeronautics) as part of the GENOME-PREMICES initiative.

27 / 27