Speaking Of Reliability: Friends Discussing Reliability Engineering Topics | Warranty | Plant Maintenance

ALT Samples and HALT Limits Abstract Kirk and Fred discussing two questions from a listener on HALT and ALT. Key Points Join Kirk and Fred as they discuss two common questions regarding ALT (Accelerated Life Testing) which is to quantify a component or systems time to wear out failure and when do you stop improving a design with HALT. Topics include: The number of samples needed for ALT depends on the question you are trying to answer. ALT is most relevant for mechanical systems such as gears, switches and relay’s and requires multiple samples and multiple stress levels to properly model expected lifetimes. Enough samples from different build lots need to be used to account for variation in manufacturing. To do an ALT the wear out endpoint or measurable change must be reached to make a life-use time model and this can take long test periods to complete. Few electronic assemblies and components have a life entitlement that is shorter than its technological obsolescence, and for the majority of electronics will not wear out with normal use during its technological usefulness of 10 years or more. Our second question is when do you stop changes to make a product more robust using HALT? The basic point of stopping improvement is when the product strength and capability cannot be improved with standard materials and techniques, which Gregg Hobbs Ph.D. called the “Fundamental Limit of the Technology”. Experience with HALT helps set a benchmark of what electronics are able to perform to. Different industries have different stress requirements as the down well oil drilling instruments require use to 185 degC or more. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes Click on this link to access the article from the US ARMY and CALCE titled  “Reliability Prediction – A Continued Reliance on a Misleading Approach” For more information on the newest discovery testing methodology here is a link to the book “Next Generation HALT and HASS: Robust design of Electronics and Systems” written by Kirk Gray and John Paschkewitz. The post SOR 636 ALT Samples and HALT Limits appeared first on Accendo Reliability.

Backward Blueprinting to Drive Building a Plan Abstract Philip and Fred discussing the idea of getting clear around what you want to have when working on reliability within a product, system, or program. Key Points Join Philip and Fred as they discuss the benefits of understanding the desired outcome of accomplishing a reliability plan. Topics include: A reliability plan is not a set list of specific tasks. Consider future objectives and vision for the organization – then set up tasks to get there Reliability work is like a journey with the plan providing milestones as guides Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes   The post SOR 635 Backward Blueprinting to Drive Building a Plan appeared first on Accendo Reliability.

The Need for Proper Reliability Plan Abstract Philip and Fred discussing the difference between a reliability plan and reliability testing. Key Points Join Philip and Fred as they discuss a few fundamentals around creating a reliability plan that adds value. Topics include: Who is going to use the results of each specific tasks? Start with the end in mind – focus on what you need to know to get there. The dynamic nature of planning – yes, they should change along the way. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes   The post SOR 634 The Need for Proper Reliability Plan appeared first on Accendo Reliability.

Assessing Reliability Courses Abstract Chris and Fred discuss what you need to think about when selecting your next reliability engineering course. What course do you need to do to become a better engineer (and more valuable to your organization)? Key Points Join Chris and Fred as they discuss reliability engineering courses and which ones you should be considering to further your career. Both Fred and Chris develop and deliver reliability engineering courses, but they were also once students themselves. That said, everyone has lots of different experiences and there are lot of mistakes we have all made. Where are you at? Topics include: Many reliability courses suck. We think. Have you completed a course where the instructor starts with (1) PowerPoint slides of explosions, plane crashes and other spectacular failures; (2) concluding that reliability is really, really, really important; followed by (3) a direct onslaught of statistical equations? Too many statistics. Not much more needs to be said than this. Reliability happens at the point of decision. Reliability engineering is all about making better decisions that ‘bake in’ reliability from the start. Only a few of these decisions benefit from statistical inference and data analysis. So courses that focus on statistics only are not reliability engineering courses. So by the time they were done, Chris and Fred came up with the following six considerations for your next reliability engineering course: You don’t know what you don’t know. A good course is a course that teaches you something that you don’t already know. Courses must add value. You need to learn new things that add value to whatever it is you are hoping to achieve in your career. If a course is simply about different tools but not how to select the right tool, then you are not being taught how to think about solving the problems and meeting the challenges that are specific to you. Courses must motivate – personally. Which means they must relate to your personal professional experience. If a course simply talks about organizational reputation, historical catastrophes and other really high-level concepts, then every reliability activity or skillset could be relevant. But what if your focus is reducing time to market (TTM) for a consumer electronic product? … decreasing downtime at a mineral processing plant? … or perhaps you are genuinely wanting to just dip your toes into reliability engineering? Work out your motivation first. Courses must join the dots – and not just give you the dots. The statistics you need to learn need to be relevant to solving reliability problems. Courses that simply give you the tools but don’t tell you how to put them together aren’t particularly helpful. … and beware of courses that come with software! Many courses are marketed as ‘Reliability 101’ or something similar when in fact they are a barely veiled propaganda film for the software that the so called training organization also sells. It must be well done. Is it done on an iPhone 40 meters away from the instructor in a poorly lit room? I mean come on … Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes   The post SOR 633 Assessing Reliability Courses appeared first on Accendo Reliability.

Statistics and Reliability Abstract Chris and Fred discuss what came first … reliability or statistics? Reliability of course! There are plenty of organizations that focus on baking reliability into designs without the need for statistics. But statistics are a part of reliability engineering. So what is the relationship between the two? Listen to this podcast to learn more. Key Points Join Chris and Fred as they discuss the relationship between reliability and statistics. The reality is that engineers and designers have been designing reliable things for thousands of year – with statistics a relatively recent phenomenon. So reliability doesn’t need statistics. But if you read a textbook or complete a reliability course it feels like statistics is everything. So what is going on? Topics include: Reliability happens at the point of decision. A decision is not a statistical analysis. That statistical analysis might create information that may then go onto to help inform a decision that creates reliability. But it is not the statistical analysis that makes reliability happen. Statistics is easy to teach … sometimes. Decision-making is a reflection of culture. Which means the extent to which reliability ‘happens’ is also a reflection of culture. And culture is a challenging topic to broach in courses. But probability and statistics has textbooks and standards that can form the basis of a course. So we often have lots of reliability ‘luminaries’ simply teaching probability and statistics because they don’t have the ability to teach culture. … so there is now a ‘statistics in reliability’ industry. There are professors, reliability consultants and reliability engineers who are perceived to be valuable because they appear to have mastered probability and statistics. So when we suggest that they need to focus on good design, manufacturing and maintenance decisions instead of endless statistical analysis, those who only bring a knowledge of statistics to the table panic as they are no longer relevant. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes   The post SOR 632 Statistics and Reliability appeared first on Accendo Reliability.

What to Do to Avoid Outdated Methods Abstract This is part two of a discussion Kirk and Fred have over a recent ASQ newsletter with CRE (Certified Reliability Engineer) test sample questions at the end. Key Points Join Kirk and Fred as they discuss what to do about the perpetuation of many of the misleading and costly beliefs in the CRE book of knowledge. Topics include: When do you determine what reliability methodology is valid and useful? Predictions have been a fundamental tenet of traditional reliability engineering approaches and have little evidence of a history of correlation to actual field failure rates. See notes below for the public domain article written by the US Army and CALCE, “Reliability Prediction – A Continued Reliance on a Misleading Approach” Fred discusses specific issues with a temperature activated component, and how polymer physics was the relevant mechanism causing wear out. The fundamental way to learn how to improve reliability of components or systems is to first understand the root causes of why they actually fail in the field. Reliability in the design stage is based on good design practices, and sometimes these are overlooked and this is the purpose of testing. Unreliability of a product can be shown as a business case to justify more and better testing in the development stage. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes Click on this link to access the article from the US ARMY and CALCE titled  “Reliability Prediction – A Continued Reliance on a Misleading Approach” For more information on the newest discovery testing methodology here is a link to the book “Next Generation HALT and HASS: Robust design of Electronics and Systems” written by Kirk Gray and John Paschkewitz. The post SOR 631 What to Do to Avoid Outdated Methods appeared first on Accendo Reliability.

Why Do We Use Outdated Methods Abstract This is the first part of two podcast that Kirk and Fred have on a recent ASQ Newsletter with sample CRE (Certified Reliability Engineer) test questions at the end and how outdated and for Kirk how it is perpetuating irrelevant and misleading reliability paradigms Key Points Join Kirk and Fred as they discuss the CRE training process and testing and Fred’s long involvement in the CRE world. Topics include: Fred talks about the his reasons and process of getting the CQE (Certified Quality Engineer) and teaching the courses that will allow you to pass the CRE test, which was out of date 20 years ago and his efforts to change the CRE body of knowledge to make it more current and relevant. The large amount of CRE information and direction Fred believes may be misleading but some information is still beneficial. Kirk discusses how he only found out what the CRE world was all about long after he had been a repair technician in a calibration lab and a field service engineer in capital semiconductor manufacturing systems,  had met and worked with Gregg Hobbs, PhD., creator of the methods of HALT and HASS at the Storage Technology Company in 1989. Models of wear out and material loss in mechanical systems such as tires or  brakes can be modeled and those models are relevant, but the many mechanisms in silicon, solder, or PCB assemblies have more life entitlement than will ever be needed. Technology obsolescence comes long before wear out in most electronic systems. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes Click on this link to access the article from the US ARMY and CALCE titled  “Reliability Prediction – A Continued Reliance on a Misleading Approach” For more information on the newest discovery testing methodology here is a link to the book “Next Generation HALT and HASS: Robust design of Electronics and Systems” written by Kirk Gray and John Paschkewitz.   The post SOR 630 Why Do We Use Outdated Methods appeared first on Accendo Reliability.

The Problem with Textbooks Abstract Chris and Fred discuss the apparent differences between what you find in textbooks and what you are supposed to do when it comes to good reliability engineering. Key Points Join Chris and Fred as they discuss reliability engineering textbooks and how they relate to good reliability engineering practices. Full disclosure … Chris has written a reliability engineering textbook. But perhaps unsurprisingly, Fred and Chris tend to really agree on this – reliability textbooks are not the answer to every reliability engineering problem. Topics include: Textbooks tend to focus on a few statistical concepts. They might reference lots of different distributions in appendices or addenda, but textbooks by and large give you equations that focus on the exponential and perhaps normal distributions. So why is this? Many textbooks first started coming out before we all had laptops and calculators. And the exponential and normal distributions are easier to use when you have to solve problems ‘by hand.’ But many textbooks have failed to take the next steps to accommodate what even our most basic spreadsheets are able to do these days. And another problem … is that some people NEED these old approaches for them to remain relevant. Some professionals out there realistically aren’t comfortable with some of the newer approaches to reliability engineering practice. And because their old approaches are in the textbooks … they protect their turf! So we don’t move on. … and then there are Bodies of Knowledge (BOK). You know, the documents that different societies need you to study to pass their exam (which you pay for). And the good thing about the old school approaches to reliability engineering? You can do it all by hand … for that exam. So we again reinforce outdated approaches to reliability engineering. And if you don’t want to think about reliability, then old approaches are great. Because they are so oversimplified, they can be reduced to really simple numbers and concepts. Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes   The post SOR 629 The Problem with Textbooks appeared first on Accendo Reliability.

HALT & Activation Energy Abstract Carl and Fred discussing a listener-submitted topic about highly accelerated life testing. Key Points Join Carl and Fred as they discuss the difference between ALT and HALT, what models can be used, and the limitations. Topics include: HALT is intended to find product weaknesses, not a life test HALT will help you know the failure mechanism, which is key to improving the product reliability FMMEA: Failure Mode, Mechanisms, and Effects Analysis Always ask: what are the assumptions and limitations that go into a particular model before using it Standards often don’t include limitations Not feasible to write a generic test plan for temperature and humidity that applies to all technologies Start with the failure mechanism; if don’t know the failure mechanism you can use HALT to learn the mechanism of failure Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes   The post SOR 628 HALT & Activation Energy appeared first on Accendo Reliability.

Selecting Tools for a Plan Abstract Carl and Fred discussing how to select the tools that will be used in reliability plans. Key Points Join Carl and Fred as they discuss what makes for good reliability plans. Topics include: Wrong way: selecting the tools that you have always used, merely because they have been used before Right way: begin with reliability vision and gap assessment Good reliability plans require good preparation Wrong way: reliability plan is merely a list of tests Ask who needs the results of tests, and what decision are they going to make Right way: ask what we are trying to achieve? A test plan is not a reliability plan 6-step process to achieve high reliably Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches. Download Audio  RSS Show Notes   The post SOR 627 Selecting Tools for a Plan appeared first on Accendo Reliability.