Volume 48, Issue 9- September 2013
What if it was possible to calculate the exact windspeeds that occurred during Hurricane Katrina, take that info back to a lab, and apply those storm conditions to your products to see how they perform? This isn’t an ASTM test that would replay hurricane speeds for a specified time. This is a “real storm,” that can replicate a hurricane (or tornado) in its entirety in both duration and strength—a test that could introduce real-world conditions including installation errors and manufacturer defects. What if someone else paid for the bulk of the research—to the tune of $4 million in fact?
This is exactly the scenario playing out at the University of Florida with the help of one Michigan entrance product manufacturer. Forrest Masters, associate professor of civil and coastal engineering, gets dozens of requests per year to perform research—most of which he turns away. So what made Henry Upjohn, CEO at Special-Lite Inc., based in Decatur, Mich., stand out when he approached Masters approximately four years ago? To hear Masters tell it, it was Upjohn’s brilliant engineering mind (For more on Upjohn, see page 92).
Brilliant mind aside, the fact that Upjohn was willing to foot three quarters of the projected $4 million dollar bill, was also enticing. And he’s not keeping the research for himself. Building product companies, including doors and windows, even his competitors, are welcome to test their products at the university after Upjohn completes certain phases of his work.
Simulating a Storm
The hurricane simulator is 24 feet wide by 18 feet tall, and can simulate 450 psf loads on vertical systems and 230-mph winds over horizontal/sloped systems. It is powered by a 14-foot fan (for a video tour go to usglassmag.com and click on the studio).
“We don’t rely on over simplistic tests,” says Masters. “We go for full, natural fury.”
When the test is performed, as I learned first-hand, when you close your eyes, you will hear a whooshing sound—it will feel like a real hurricane taking place outside your doors. The test makers can even measure strains, displacements, air leakage and the forces required to hold the test subject in place.
“We build a computer model of what we are testing then we tweak it to ensure that the computer gives us the same results as the experiment,” says Masters. “Once we have a high degree of confidence in the model, we can introduce design improvements or installation errors to see how performance is affected. This approach obviates the need to run lots of full-scale experimental tests, which are expensive and time-consuming to set up.”
Masters, also a licensed structural engineer, teams up with students at the university both inside the lab for computer modeling and outside the lab to chase storms. Florida, with the highest number of hurricanes taking place each year, makes it the ideal stomping ground for this innovation. Masters is the ideal person to spearhead it.
“I love what I do and enjoy working on different problems that most people don’t want to touch,” he says. “I am not scared of trying new things. Ingenuity always wins.”
That, and a whole lot of science.
“When Henry first visited me he told me the science comes first,” says Masters. “He insisted that we work at the leading edge—taking risks and transforming the industry. Our concept is much bigger than the scope of one company. It’s about American competiveness. It’s about producing the best technology we can that is still economical.”
But the road to get here was not cheap.
“The research was expensive. Most people aren’t prepared for that level of effort and cost,” says Masters. “It got a lot bigger along the way. His [Upjohn’s] expectations for the final outcome grew.”
Were either of these two masterminds ever wary of their path? Did they ever think about scaling back?
“I felt good about this from the beginning,” says Masters. “Our team is knowledgeable and passionate and we think things through. What made everything come together was Special-Lite. I knew this would work. They are looking to be leaders in the field.”
An Unlikely Pair
This is just the type of thing that turns Upjohn’s inquisitive and mechanical mind into overdrive.
“This insurance friend felt one of the issues was a result of the current way hurricane doors and exterior building products are evaluated,” says Upjohn. “If you are going to change the test procedures what do you change them to? I had met Forrest through this friend and he had said what we really needed was a way to evaluate products based on a hurricane experience. The idea was to build a simulator in which we could put large pieces of building products—even the side of a house—inside and simulate a storm.”
The goal is to then have the industry build standards based on this research.
“In an actual hurricane the wind is coming and going and the pressure is changing,” explains Upjohn. “You can put a door in a mild situation but if you test it, and have it shake for a few hours as it would during a hurricane, then it will fail. The idea was to build something that could evaluate products in these real-world conditions.”
Special-Lite has produced impact products sold in hurricane zones for 10 years, and much of this is in school applications. The company was looking to expand, as it was running out of growth potential in schools. In 2012, it bought Universal Pultrusion, a manufacturer of fiberglass doors and frames that opened up a new world of opportunities for Special-Lite. This is one place where some of this data will be used to further develop products.
Generating the Data
“This will take a lot of playing back storms [in numerous simulations] to get the research we want,” says Upjohn. “It will be interesting to see what we will get out of it.”
He is more than anxious to get started, particularly in testing some of the company’s new composite products.
“One of the nice things about composites is you can get engineering properties you can’t get otherwise and you can get them for less money. We have been working on different composite sections to understand how they degrade and perform under different types of loading and we already understood how you transfer those loads in our flush door,” says Upjohn.
He says the company will start taking many of Special-Lite’s new designs and run them through the simulator to see the failure mechanisms. Then we will build or enhance our FEA model to further refine our approach.
“Every once in a while something comes around and you think it will fail in a certain way and it turns out differently,” says Upjohn. “If you are going to make cost-competitive products you have to evaluate different ways of doing things so you come up with the most efficient use of the products. For us it will be a really good development tool.”
Upjohn set an official cut-off date where development of the simulator stops and research begins—that was July 1. Again, Masters and Upjohn are in alignment with respect to this issue as well. Masters cautions companies that R & D efforts don’t have to go on indefinitely. “You want to get to the end point and that’s the next generation product,” he says.
“We are heavy planners—90 percent planning. Then we execute. People are usually blindsided by how much research comes out in a quick timeframe.”
Once this abundance of research is produced there are a few things that both are in agreement will and will not happen.
“We think our findings will ultimately affect test labs,” says Masters. “But that’s not our role, though we would be open to working with them.”
Both agree that this research could influence upcoming building codes. “There is nothing in the current regulations about maximum door deflections,” for example, says Upjohn. “That will have to be in the codes at some point.”