What is the relation between VITA 51.1 and revisions to MIL-HDBK-217?
ANSI/VITA 51.1 is an industry standard that provides consensus based modification factors for use with MIL-HDBK-217 Revision F, Notice 2. ANSI/ VITA 51.1 was developed prior to the launch in 2008 of the DoD-led effort to revise MIL-HDBK-217. It provided uniformity of practice for use of MIL-HDBK-217 Rev F, Notice 2 and consensus based modification factors for the models.
MIL-HDBK-217 Rev F, Notice 2 was published in 1995 and has become obsolete for many technologies. Most companies that use the methodology have supplemented it with internally developed modification factors based on product testing and/or field failure data.
The effort to revise MIL-HDBK-217, led by the Naval Surface Warfare Center (NSWC) in Crane, IN, was organized into two phases. First, a Revision G was developed to provide modifications to the existing models but no new models or modeling methods. The second phase will develop further revisions and include advanced methods such as physics of failure methods. Many of the same people worked on both ANSI/VITA 51.1 and MIL-HDBK-217 Rev G. All the data collected and used for ANSI/VITA 51.1 was provided to the MIL-HDBK-217 revision working group as a starting point, but there was further development of the modification factors using new data.
What is the current status of the effort to revise MIL-HDBK-217?
The current status as of June 14, 2011 of MIL-HDBK-217 Rev G is that the NSWC Crane working group produced a final draft but it was withheld from public review and publication was delayed due to discussions within the DoD about the future of reliability in general.
What is Physics of Failure analysis?
Physics of Failure (PoF) analysis focuses on modeling physical failure mechanisms and providing detailed information on how environmental and usage stresses lead to fatigue failure. PoF analysis presents a paradigm shift relative to reliability prediction. It changes the analysis of the system equipment from a box of parts to a box of failure mechanisms.
When would you use Physics of Failure methods instead of handbook predictions?
Physics of Failure (PoF) methods provide a detailed assessment of the environmental and usage stresses that lead to fatigue failure. PoF should be used whenever you need to understand the effects of fatigue to predict a time-to-failure (TTF), analyze the end-of-life, or understand the effects of environmental stresses such as vibration and thermal cycling.
How can I use Physics of Failure in a cost effective manner?
ANSI/VITA 51.2 section 2.1 and Figure 1 provide guidelines for selection of PoF models for different program phases. It is very difficult to apply PoF to a complex system of hardware and software. There are many failure mechanisms that invoke the limited number of failure models, and not all model parameters are available to the reliability engineer. It is subsequently the responsibility of each organization to define and document the breadth, depth, and the extent to which they apply PoF to meet their program needs and contractual requirements. PoF planning is recommended at the individual program or organizational levels.
What does VITA 51.2 provide?
ANSI/VITA 51.2 provides rules and recommendations for the application of Physics of Failure (PoF) methods to reliability prediction of electronics at the board, packaging and component levels. It is a compilation in good faith, of existing physics of failure models selected by consensus of the working group of best practices in industry.
Is the Reliability Community an industry-wide effort?
Yes, this is an industry-wide effort. The sponsors of the Reliability Community are working to encourage technology suppliers and developers to both follow the guidelines as established in the ANSI/VITA 51 specifications and documents.
Where can I get the Reliability standards?
The standards are copyrighted and distributed by VITA. They are available free to VITA members and for a fee to non-members. Documents may be purchased and downloaded from the VITA Store
Where can I go to get answers to additional questions that I might have about Reliability guidelines?
The VITA Reliability Community has established a LinkedIn
user forum. LinkedIn members can search for Reliability Community under groups. You can join existing discussions or start a new one on your specific topic.
Where can I go to learn more?
A list of articles is maintained on this website. Check there for the latest information.
How can my company get involved with the Reliability Community technical working group?
Contact VITA to become involved with either of these efforts. You must be a member of VITA to participate in working groups.
What is the basis of "Quality level of PiQ=0.1" for Rule 2.1.4-2 of the ANSI/VITA 51.1-2008 (R2013) Reliability Prediction MIL-HDBK-217 Subsidiary Specification?
The "used Pi Q" was a value used to adjust a MIL-HDBK-217FN2 prediction to match it with a failure rate derived from the field failure. It was not derived, or taken from some other source. It was a value that was found to fit the observations. The problem is that the match was not exact, and this has caused confusion in the past in trying to duplicate what was done. Generally, a conservative value was picked, and the committee tried not to deviate too far from the original MIL-HDBK-217FN2 handbook.
The values provided in ANSI/VITA 51.1 are recommendations. The user is provided with the supporting documentation and data so he or she may decide whether to use what is provided.
I am looking for some guidance with regards to inputting adjustment factors or class changes to MIL-217 components to make my prediction models more realistic compared to real life experience. Currently the predictions are far to conservative to be believable. Which of your standards would you recommend for me?
This information can be found in ANSI/VITA 51.1.
What does it mean that the VITA 51.1/.2/.3 have been “ANSI ratified”? Can I consider it as an ANSI standard (or it will become)?
VITA is accredited by ANSI meaning that VITA follows the ANSI standards development procedures and is audited to compliance to these procedures.
ANSI approval of VITA standards is intended to verify that the principles of openness and due process have been followed in the approval procedure and that a consensus of those directly and materially affected by the standards has been achieved. ANSI coordination is intended to assist the voluntary system to ensure that national standards needs are identified and met without conflict in their requirements or unnecessary duplication.
ANSI/VITA ratified standards are available from VITA.
There are many methods available for reliability prediction (e.g. 217Plus, Telcordia, UTE-C-80-811 . . .), what is the advantage of using VITA 51.1? Does the ANSI/VITA 51.1, in addition to MIL-HDBK-217F give you realistic failure rates values?
Reliability prediction accuracy depends on how closely your new device's technology and application environments align with those that the prediction methodology was based on.
All reliability prediction methodologies are based on an assessment of past device performance. Most of the data used to develop the prediction methods you mentioned were observed field or usage failures that were collected over a period of time. Some also use life testing data, which is when a device is operated at a higher temperature and/or power level to drive out failures faster than normal usage (and then the results are scaled to usage conditions based on commonly used models like the Arrhenius equation). So, whether the prediction method is a fair assessment of your current system depends on (1) whether your technology is the same or similar enough to the older technologies that were analyzed to have the same failure mechanisms, and (2) whether your application has similar characteristics as the usage environment conditions the failure rates were derived from.
So, based on consideration (1), the accuracy of a reliability prediction methodology is heavily dependent on how recently it was developed. The more recent the data, the more likely it is to be similar enough to current technologies to be useful. MIL-HDBK-217 was last revised in 1995, so it's definitely too old to be useful modeling today's electronic technologies. That said, a lot of people still use it. Customers still require it to be used. Because so many people are using MIL-HDBK-217 despite it's being so old, VITA 51.1 was developed to provide some updated models for MIL-HDBK-217. It doesn't provide any guidance for other prediction methodologies, just MIL-HDBK-217. It's not an update of MIL-HDBK-217, but a supplement to make some of the models more aligned with the current technologies.
The other prediction methods named are all newer than MIL-HDBK-217. 217Plus was last updated in 2006. Telcordia is continuously being updated and may be the most current of all of them.
Which brings me to consideration (2) (application similarity). MIL-HDBK-217 was developed by the U.S. Department of Defense based on data mostly from their products (aircraft, missiles, heavy land vehicles, etc.). 217Plus was derived from a mix of military and commercial equipment experience data. Telcordia was based almost exclusively on data from commercial telecom equipment, generally permanent installations in a climate-controlled office building. If your system is intended for any mobile usage, and especially if it's intended to be used in an airborne environment, using Telcordia is not recommended, in spite of how recently it was updated. It does not use data from airborne applications but applied a scaling factor to their predictions to approximate it.
Yes, ease of use is a definite advantage for a reliability prediction. You can quickly get an initial assessment of reliability for a new concept design. It may or may not turn out to be accurate, depending on (1) and (2), so an initial prediction should only be used until better information is available.
I’m working on a hybrid prediction using MIL-HDBK-217F Notice 2 and VITA 51.1. Vita 51.1 doesn’t specifically call out anything on hybrid parts. I was wondering if the rules for the Integrated Circuit/Microcircuit applies to the hybrid part as well?
There is no guidance for hybrids in the standard because we had no information on which to base a recommendation. You will have to use MIL-HDBK-217F Notice 2 as it is presented. Although functionally similar to integrated circuits, hybrid packaging and manufacturing processes can be very different. You may do better to look for reliability data from the manufacturer for your predictions.