Does Color Temperature matter when designing a fixture for a hazardous location?
The Half Life of Knowledge + the Role of Proactive Engineering in Successful Product Certification
We all know that technological innovation is exploding today. But what we may not think too much about is the flip side of this. While scientific knowledge and technology increases exponentially, knowledge is also becoming obsolete at a rate faster than ever before. In fact, Harvard mathematician Samuel Arbesman argues in his book The Half-Life of Facts: Why Everything we Know Has an Expiration Date, that knowledge has a half life. Arbesman’s research shows that the time for the field of medicine to double its knowledge is 87 years; mathematics, 63 years; chemistry 35 years and; genetics, 32 years. But this doubling also implies a half life of knowledge. Arbesman cites the example of medical researchers investigating hepatitis and cirrhosis. When a panel of experts were given research papers in the field that were 50 years old and asked to determine how much was still true, they found that half of this knowledge was outdated in 45 years. The field of scientometrics deals with the measurement and analysis of science, technology and innovation and what it shows is that much of what we know today will soon be incorrect. This half life of facts has far reaching ramifications. For instance, writing in IEEE Spectrum, Robert N Charette asserts that an engineer’s half life today is estimated at 10 to 12 years or less (Charette, 2013).
Another area where the explosion / half life of knowledge has implications is in the certification industry. Third party standards testing and certification organizations exist to ensure the safety of the public. In North America, the largest and most well known of these are UL and CSA and in Europe, the CE. Other certifications such as NSF for food, water and consumer products apply internationally. On top of these are countless smaller regional standards testing organizations catering to specialized industries. Manufacturers pay fees that can be as high as tens of thousands of dollars per test for products like large engines or gas fired generators, and it can take many months or even years of non-destructive or destructive testing before a product is certified. Collectively, certification and standards agencies hold the responsibility for ensuring the safety of the millions of products sold to consumers around the globe. The real test of a product though, is not during certification, but afterwards when certified products are used in the real world in different application scenarios. Often real world usage leads to new scenarios not encountered in the lab environment. When this happens, and certification bodies are alerted, it often results in updates to the standards.
The rapid pace of technological change results in a continuous stream of new product designs, and new safety standards to deal with them. The most famous law of technological change is Moore’s law, which states that the number of transistors in integrated circuits doubles every 18 months. It first appeared in an article inElectronicsmagazine in 1965 (Moore, April 19, 1965).
Recent research comparing a variety of hypothesis of technological progress from Moore (Moore’s law), Wright, Goddard, Sinclair and Nordhaus demonstrates exponential increase in production over 62 different diverse technologies (Nagy, Farmer, Bui, & Trancik, 2013). But as our technology has grown exponentially, our ability to model its long term impacts has only grown arithmetically (Tenner, 1997). This widening gap between exponential and arithmetic growth has serious consequences for society. Those aspects of new technology that fall outside our ability to model can and often lead to unintended consequences. As futurists do their job predicting the bright future of robotics, self-driving cars, artificially intelligent systems, nanotechnology, or CRISPR-engineered genetic life forms, we all know that such technology always comes with surprises. It’s the job of the rest of society to try to mitigate their serious real world impacts. Standards organizations are part of the safety network to whom this task is delegated.
By their very nature, standards agencies face challenges with new technologies; they cannot remain static, yet there is always a lag from when new technology appears to when standards are mature to certify products as safe. Certification standards are forced to constantly evolve because technology is constantly evolving. If you are on the bleeding edge of technology and have incorporated the latest whizbang scientific discovery into your product design, you just may find yourself in a fuzzy gray zone, a position of being ahead of knowledge that certification labs possess. This is when the half life of knowledge has a concrete business impact. A standard that has been good for older technologies may suddenly find itself too restrictive for new technology.
In hazardous location areas, the maximum surface temperature of the electrical apparatus must stay below the ignition temperature of the potentially explosive atmosphere it is operating within. Each explosive substance (dust or gas) can be assigned within a corresponding Temperature Class. Testing and certification agencies test and assign a maximum temperature that a device’s external enclosure can reach, called a T-code (Wikipedia, 2018).
The higher the T-Code number that a Device Under Test (DUT) can achieve, the lower its maximum device surface temperature and the more environments it can operate in. The purchaser and installer must compare the auto ignition temperature of the specified hazard with the maximum surface temperature of the equipment installed. The auto ignition temperature must be higher than the equipment T-Code. So the object is to try to design the product for the highest T-Code number possible. Because our luminaires offer a wide spectrum of temperature colors and lumens, they employ a large range of different components, resulting in many surface temperature profiles. It was not economically viable to test such a large number of products.
So what do you do when your latest design causes you to bump up against the limits of certification testing? As we learned, sometimes, we have to have confidence in our own abilities. Surprisingly, agencies are responsive to your expertise, if you can back up your claims with solid empirical and verifiable evidence. In fact, industry can play an important role in helping certification agencies refine and fine tune their blanket standards. They err on the side of caution, but sometimes that can impact economic feasibility, or can result in a certification that can severely limit the products markets. To avoid such outcomes, you can play a proactive role in helping certification agencies by educating them.
The luminaire family we were testing had 14 mainline LED bins (Furness, 2008), and 2 different LED families that we wanted to be able to sell, along with at least one alternate part number for each, and 4 different power levels. This gave us a total of (14 x 2 x 2) x 4 = 56 x 4 = 224 different LED bins that had to be tested. At an average of $5K USD for each test, and a test time of 1 to 3 days per test, for a small company, this was not economically feasible. Based on arguments of LED device physics pertaining to LED flux bins, CRI and CCT, along with LED binning tests conducted inhouse, we were able to drop the testing from 224 tests to 8 tests, a substantial reduction. We wrote a report with our findings and submitted to the testing agency. Subsequently, the agencyagreed with our findings and granted a project specific variance in the test procedure, making the design economically feasible to test and certify.
This experience of working with the certification agency taught us that by playing a proactive role, clients can educate them on knowledge gaps, which can lead to significant improvements of the certification process for our products, leading to a win-win situation. We were pleasantly surprised to learn the flexibility that agencies have in accommodating a client’s specific needs. Sometimes agencies are flexible enough to work with your inhouse engineering expertise to interpret the standard in light of new, compelling scientific and / or engineering results that you present to them, and make exceptions to the rules. The key is to play a proactive role. Do the research and be prepared to back up your claim with strong evidence. Filling the knowledge gap in fast moving technological sectors can pay big dividends.