CAMBRIDGE — The smoky, slightly sweet smell of mesquite filled the air outside Harvard's science and engineering buildings off Oxford Street early Monday morning. But this was no springtime diversion — instead, it was a 15-hour final exam for a class with a single, mouth-watering problem to solve: how to make the perfect brisket.
Engineers wearing custom aprons, embroidered "Harvard BBQ" in red, fiddled with sensors that reported the temperature inside their 300-pound, hourglass-shaped smoker. Students, who had been sleeping in shifts for days, monitored their smartphones for minute-by-minute updates of the readings. Tiny computer-controlled fans clicked on and off to keep the heat inside the smoker steady.
For months, this 16-person team had been designing and modeling and building the prototype for the ultimate barbecue cooker. The handcrafted Harvard smoker is their solution. Tested by countless computer simulations of virtual brisket smoking, nearly two dozen weekend smoking sessions — often in snow or sub-zero temperatures — and 220 pounds of meat, the smoker is a rigorous, data-driven tool for making a feast.
The idea for this unconventional engineering class, offered to Harvard juniors, came three years ago when engineering professor Kevin Kit Parker attended a barbecue-cooking competition in Memphis. Parker grew up in the South and has a deep appreciation for barbecue, and when he looked up from his plate that day, he saw a problem that lacked an optimal solution.
Many products have been refined by cycles of science and engineering. Barbecue, however, has been a veritable Wild West in which pit masters build mishmash setups that incorporate garbage cans, cinder blocks, a giant rotisserie. There seemed to be little in the way of deep understanding of how — or why — one smoker was better than another, Parker said.
"They are the biggest contraptions and pieces of junk you've ever seen," he said. "Everyone had their own little mojo they brought to the problem."
He thought science and smart engineering could turn the art of barbecue from something that skilled practitioners do with secretive, homespun rigs to something more accessible. So for his engineering design class back in the confines of Harvard Yard, he assigned a team of novices one of the toughest problems in the field: build a foolproof smoker that can repeatedly produce the perfect brisket, to be judged on texture, taste, and appearance.
Making a perfectly smoked piece of meat may seem to be as far as you can get from an engineering conundrum, but Parker saw it as a beast of a problem that required a deep understanding of chemistry, heat transfer, materials science, prototyping, and solving problems.
On a hike up Camelback Mountain in Arizona, he pitched the idea to executives from Williams-Sonoma, who agreed to be the client for his class. The high-end kitchen and housewares company set the general design specifications, and sent the Harvard team aprons, cooking tools, and a seemingly unending supply of brisket.
"As much as it sounds like fun, at the beginning of the semester it was truly high stress," said Michel Maalouly, a junior studying environmental engineering.
Early on, Maalouly and his classmates used a smoker already sold in stores to learn how to smoke meat, experimenting to discover what made it too tough or not smoky enough. They worked outside all winter, starting at 3 a.m. on Sundays in rotating shifts, and used temperature probes to understand how heat fluctuated in the smoker and in the meat.
In the end, the secret was to precisely control the temperature both in the smoker and in the meat over the "low and slow" smoke. They had to keep the meat below 120 degrees long enough to let the enzymes in the meat break down the collagen and make it tender; they wanted the smoker's shape to cause "cyclonic airflow," meaning the smoke would circulate down toward the brisket. While the wood would burn at 700 degrees, the meat would gradually rise over a 15-hour period to about 190 degrees.
They then built their own smoker. The class settled on a material — ceramic — and a shape that resembles a cooling tower at a power plant. The design solved one of the big problems with the commercial smoker they used, by eliminating hot spots where the meat might cook too quickly and dry out. They built an app that would allow cooks to monitor the conditions inside the smoker and share their experiences through social media.
Yinka Ogunbiyi, a junior who is studying to be a bioengineer, ran computer simulations of how smoke circulates in the smoker. But the problem solving required by the class included every aspect of product design; Ogunbiyi's research included a close study of Williams-Sonoma glassware, in an effort to make the smoker look like a product people might like to buy.
Estimates vary, but everyone in the class devoted somewhere around 40 to 50 hours a week. Maalouly learned early on that he had a severe allergic reaction to the smoke: He went to the hospital after his eyes were irritated the first time he opened the smoker. He now wears goggles and protective gear.
But how to measure success? Peyton Nesmith, the teaching fellow for the class, became certified as a judge for the Kansas City Barbeque Society, with Parker. They look for perfectly cooked brisket to take on a mahogany hue. When sliced, there should be a slightly pink section around the edge, called the smoke ring. The meat must be tender, but not falling apart. The taste, Nesmith admits, is somewhat subjective.
The Williams-Sonoma team came to Boston from San Francisco for the final project and plan to bring back the design and ideas to the company's leaders. Harvard has begun the process of patenting the smoker's design.
"They've gone from basic science to really understanding how you optimize for flavor and texture," said Patrick Connolly, chief strategy officer for Williams-Sonoma.
When asked how the brisket turned out, Connolly and a colleague both smiled.
It was a 9, they said. The highest score.