IPC9592 Power Standard Enhancements
Power conversion modules are key hardware building blocks critical to electronic products’ quality and reliability. Power continues to evolve rapidly with increasing current and power densities. There are many different qualification processes at both the supplier and OEM (Original Equipment Manufacturer) level. At the same time products encounter a variety of harsh applications world-wide that are not accounted for in all qualification processes.
IPC-9592, Requirements for Power Conversion Devices for the Computer and Telecommunications Industries, was developed by a group of major OEMs and suppliers to provide a common set of test requirements to proactively complete a comprehensive test regimen that will satisfy both OEM and their end customer requirements.
The IPC-9592 standard was officially released in September 2008, and sets the requirements for the design, qualification, and manufacturing quality / reliability processes of power conversion devices for the computer and telecom industries. This standard provides guidance to help reduce the risk of known quality and reliability problems in power conversion devices.
Revision A of the IPC-9592 standard was officially released in May 2010. This is an enhanced version of the original standard, which provides additional guidance on corrosion, MSL (Moisture Sensitivity Level), module preconditioning, and HALT (Highly Accelerated Life Test) testing.
Corrosion has become a significant issue in harsh applications where the proper mitigation steps have not been addressed in the design. Figure 1 is an example of creep corrosion and Figure 2 exemplifies sulfur corrosion post MFG (Mixed Flowing Gases) testing. Standard qualification tests will not expose these potential problems before the product is released to the customer.
There are a number of corrosive agents that have to be accounted for, including moisture, sulfur-containing chemicals (e.g. SO2, H2S, elemental sulfur), and chloride/chlorine-containing chemicals (e.g. salts, HCl, Cl2), the top three causes of corrosion related field failures according to a study done by one major OEM. Applications in harsher environments and longer life expectations are also factors in the corrosion problem. IPC-9592A provides mitigation steps to take to prevent failures due to these different corrosive agents during design and manufacture of the power conversion devices.
The DfR (Design for Reliability) section provides approaches to mitigate these problems before they occur, along with some references for further information.
Sulfur Creep Corrosion on a PCB (Printed Circuit Board), courtesy of HP
Evidence of corrosion on Gold Plated Pins following MFG test, courtesy of ALU
MSL is another important factor in achieving high reliability in electronic products. There is a new subsection in the DfR section to specifically address this. This is more critical in products using Pb-free solder with higher reflow temperatures than for SnPb solder. Some surface mount components absorb moisture over time. When these parts go through reflow, the high temperature may vaporize the moisture inside these parts, causing separation and either immediate failures or latent failures that don’t show up until the product has been out in an end application for some unpredictable time.
IPC-9592A outlines the steps to take to prevent this problem from occurring. MSL levels for components are explained based on J-STD-020, and an MSL rating process for surface mountable PCDs (Power Conversion Devices) was added. There was no industry standard for an agreed upon method to rate MSL for a surface mountable PCD, and this provides a frame of reference to compare different products.
IPC-9592A defines MSL for PCDs as follows: “Surface mount PCDs shall use an MSL rating equal to the worst case MSL rated component contained in the bill of materials (BOM) and for PWBs (Printed Wiring Boards) on the BOM, a default of MSL-2a shall be assumed”. The default value for PWBs was assumed because there was no agreed upon approach in the industry to determine this value for PWBs.
Packaging and labeling requirements consistent with J-STD-033 for the surface mountable products were also added. Proper packaging and labeling is important to ensure that moisture is kept out of these PCDs before being opened at the end application factory, and to ensure the factory personnel know the MSL rating to follow the proper storage conditions between when the PCD is removed from the packaging and when it’s reflow soldered onto the end application.
Moisture absorption is highly variable and not precisely defined for all environments and assemblies, for example PWBs. In addition standard bake out procedures may be impractical for BMPM (Board Mounted Power Module) assemblies; so additional attention needs to be made to control time out of the moisture barrier bag.
Module preconditioning was added as a test requirement for surface mountable board mounted products, to simulate the environmental stresses that a product like this may see after assembly at the supplier and when it’s reflowed into the end application. The preconditioning consists of a humidity soak and exposure to two reflow cycles, to simulate the potential situation where it’s reflowed on one side of an assembly and then other parts are reflowed on the opposite side of the assembly. The humidity soak requirements come directly from J-STD-020, using the MSL rating of the surface-mountable module.
An appendix was added for HALT, with the process and conditions for HALT testing that were not in the original IPC-9592 Standard. This is not a comprehensive treatment of HALT overall but rather a detailed course of action to perform HALT on a power conversion device; those new to HALT should contact a HALT subject matter expert before running this test. This has seven HALT procedures, including the Low Temperature Limits Test, the High Temperature Limits Test, the Random Vibration Step Test, the Random Thermal Cycle Test, the Input Voltage Test, the Output Load Test, and the Combined Stress Test.
The Input Voltage Test and the Output Load Test were added to the other five more traditional HALT tests, and the Combined Stress Test has more combinations than the traditional HALT test. Each test should be repeated until no more failures occur after fixes have been incorporated, or the practical design limitation has been reached. IPC-9592A allows reuse of BMPM when they don’t fail previous HALT tests. Revision A provides added details on input voltage requirements and output load requirements for some of the HALT procedures.
The upcoming Revision B has a number of enhancements planned, which is currently being developed by the IPC Subcommittee. These include updating the component derating guidelines and referencing JEDEC tin whisker mitigation recommendations. Figure 3 shows one example of tin whiskers. The temperature cycling test is being refined to better define the process, in terms of the ramp rate, and evaluation of the test units during and after the test to define more quantitatively the status of the test units during and after the test.
The shock and vibration operating and non-operating tests will be reviewed to align the tests better. There are plans to make the BI (Burn-In) reduction plans more statistically significant.
All of these changes will improve the IPC-9592 Standard as it continues to evolve. It should be noted that inputs on the standard are welcomed, and are reviewed by the subcommittee on where they can be utilized.
About the authors
Don Gerstle is Director of Global Design Quality at Murata Power Solutions (MPS). Gerstle has been employed by MPS and C&D Technologies Power Electronics Division prior to being acquired by Murata, for 14 years.
Neil Witkowski is a Senior Reliability Engineer at Alcatel-Lucent. Witkowski’s 28 year career with Alcatel-Lucent has been focused on electronic component selection, quality and reliability.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
