Click & Close Ads
Click & Close Ads
PITMASTERS, THIS WILL be hard to swallow: Some of the best smoked brisket comes out of Cambridge, Massachusetts. And it practically cooks itself.
The secret is in the smoker. And the secret to the smoker is everything a group of 16 Harvard engineering students could do to turn the finicky art of smoking meat into a controlled, semiautomated science.
In prototype form, the smoker looks like a combination of a giant pepper mill, a tandoori oven, and V.I.N.CENT from The Black Hole. It weighs 300 pounds. It has a refueling chute built into the side of it. And it uses a proportional-integral-derivative controller, a Raspberry Pi, and fans to regulate its own temperature, automatically producing an ideal slow-and-low burn.Why would a group of Harvard students build this? Because they had to. Engineering Problem Solving and Design (ES96) is a required class for engineering majors at Harvard—most of them, anyway.
“Mechanical engineers, it’s not a required class for them,” says bioengineering major Jordan DeGraaf. “There are no mechanical engineers who take this class. They just run away.”
ES96’s projects and professors vary each semester, but the concept stays consistent. Classes need to work as a team—across disciplines such as biology, electrical engineering, and environmental engineering—to design, build, and test a solution to a real-world problem.
Each class is one of a kind, because those problems are all over the map. Past projects included color-changing camouflage, weather-balloon systems, and an app meant to curb gang violence. Whatever the project happens to be, the team needs to finish in less than four months.
If you’d spent any time with Kevin “Kit” Parker, you wouldn’t be surprised his class cooked this thing up.
An Officer and a Professor
Sitting in his office at Harvard’s School of Applied Engineering and Science, the 6′ 5″ Parker sips from an equally giant jar of sun tea. It’s less than an hour before his students, still smoking briskets outside, present their work to fellow students, faculty, chefs, reporters, and venture capitalists, along with execs from the project’s target client, Williams-Sonoma.
If Parker is nervous, it doesn’t show. His perma-grin transmits confidence. He’s excited. He’s surrounded by artifacts of his life: Little jars filled with artificial heart valves produced by modded cotton candy machines, “organs on chips” for studying concussions, and his 6-year-old daughter’s pink Flower Princess bicycle.
The Harvard BBQ team spent 880 man-hours cooking 220 pounds of brisket in their quest to develop the ultimate smoker.
In a booming, Tennessee-flavored voice, he talks about how proud he is of his students. When he’s not teaching classes or researching heart and brain ailments, Parker is regularly serving in Afghanistan. A lieutenant colonel in the US Army Reserve, he has shuttled between Cambridge and Kabul for the past 13 years.
“I know this 18- to 30-year-old demographic,” Parker says. “I’ve been in the lab with them, I’ve been in the battlefield with them, I’ve pulled all-nighters with them, I’ve stood in the desert and cried with them when one of them got killed. Motivating them sometimes requires a pat on the back—and sometimes a boot.”
This time, it required 220 pounds of brisket. That’s how much meat his class prepared between late January and early May using their device and the Big Green Egg, a popular backyard smoker with a cult-like following, as a theoretical competitor. They also smoked hundreds of pounds of virtual brisket, using modeling software to test everything before firing up the coals.
Click & Close Ads
Click & Close Ads
Brisket Breakdown
Smoking brisket is no brisk task. It takes about 15 hours to smoke a 10-pound slab, and this smoker doesn’t change that. Important things happen in stages over that time. During the first few hours, when the meat is cool and moist, smoky flavors infiltrate the flesh. Charcoal heats up chunks of wood, which causes lignin in the wood to denature. These compounds impart smoky flavor to the meat.
The Science of Smoke Indirect heat cooks the brisket “slow and low,” giving enzymes the optimal amount of time to turn tough collagen into melt-in-your-mouth gelatin. Wood adds flavor during the first few hours: Its lignins denature as it burns, releasing tasty smoky compounds. Too much heat can leave you with shoe leather, so controlling temperature is paramount. Flare-ups from drippings can cause an uneven cook; a water pan helps snag and neutralize falling fat.
The Science of Smoke
Indirect heat cooks the brisket “slow and low,” giving enzymes the optimal amount of time to turn tough collagen into melt-in-your-mouth gelatin. Wood adds flavor during the first few hours: Its lignins denature as it burns, releasing tasty smoky compounds. Too much heat can leave you with shoe leather, so controlling temperature is paramount. Flare-ups from drippings can cause an uneven cook; a water pan helps snag and neutralize falling fat.
But that’s just the flavor-infusion process. The bulk of the cook influences texture and taste. Slow, steady heat breaks down collagen into gelatin, making the meat tender. Controlling heat and oxygen is crucial: Too much heat or wood—or too little oxygen or time—and the meat can get tough and taste awful.
Controlling those variables is a science that exists in the minds of experienced pitmasters, barbecue jedi who instinctively know how much to nudge a vent, what wood to use, and when to refuel in order to shape the profile of the meat. It’s science disguised as sorcery. “A pitmaster will know exactly how much linear distance he has to adjust the vent in order to get the correct temperature,” says junior bioengineering major Paul Kaczor.
The thought of a group of Ivy Leaguers—some of whom had never tasted brisket before—automating the process may send blue smoke billowing out of a pitmaster’s ears. But this smoker isn’t meant for pitmasters. It’s meant for the rest of us. The smoking process is filled with contradictory advice, requires hours of babysitting, and is intimidating for novices.
“Old-school pitmasters are like, ‘I cook mine in a garbage can,’ and there’s a point of pride in that,” Parker says. “A lot of the cutting edge is when you take an art form and drag it back onto scientific turf and turn it into an algorithm. I don’t think we’ve diluted the artistic component with this. I think we’ve mitigated the risk of the artistic component so that other people can participate.”
Meat Market
There’s always a real-world client for the ES96 class projects, an organization that sets parameters for what’s needed in their market. Williams-Sonoma played that role for this smoker. The gourmet cookware retailer wanted an easy-to-use unit that would look fine on an outdoor patio, had software controls, offered predictable results, and smoked the competition. In exchange, Williams-Sonoma supplied the meat.
The team had other key obstacles. They could use only brisket, which is notoriously hard to cook: It’s a lousy cut of meat unless it’s perfectly smoked, braised, or cured. There’s also a drought-induced brisket shortage going on right now—although getting meat on demand from Williams-Sonoma solved that issue.
“We gave them 220 pounds of brisket during a world brisket shortage, which I think we contributed to,” says Pat Connolly, executive vice president and chief strategy and business development officer at Williams-Sonoma.
Parker had his own goals, too. The smoker would have to meet the Kansas City Barbeque Society’s official contest rules. Other types of smokers allow precise temperature control, using gas or electricity as a heat source. But those smokers generally produce less-flavorful meat, and they also don’t hew to KCBS rules. This smoker would need to use charcoal and wood: Superior for flavor, but harder to control.
The team also limited themselves to using mesquite, which can be overpowering in terms of flavor compared to hickory or oak. But mesquite contains four times as much lignin as hickory, meaning more potential for complex flavor if they could expertly harness the smoke.
“We cooked all these briskets in silica, computer simulations, before we did it in our smoker,” Parker says. “We had all this data and knew exactly how it was going to go down. The control loops are so tight out there. We tamed this beast. We control the fire, and we use the fire to control the chemistry inside the meat.”
Smoke Machine
That control starts with the smoker’s shape. It’s inspired by the hyperboloid cooling towers in nuclear power plants: The tapered middle forces heat and smoke upward like a cyclone to spiral evenly at the grill surface. According to the class, this creates a uniform temperature around the meat, cooking it evenly. Another bonus: The twisty turbulence of the smoke helps the meat absorb a smokier flavor.
Fuel Basket
The basket makes adding and removing fuel easier, and allows for more uniform temperature control. No hot or cold spots!
The ceramic body is optimal for retaining and radiating heat. According to the class’s research, ceramic has lower heat-conductivity properties and higher heat capacitance than cast iron or stainless steel.
Then we get to the automated adjustments. The system takes temperature readings from inside the smoker, compares them to the ideal temperature for that moment in the smoking process, and uses integrated fans to adjust the heat as needed. “It’s not so simple as what’s the temperature now, turn the fans on,” explains electrical engineering student William Jameson. “It’s more about what you’re approaching in terms of temperature. It doesn’t overshoot and get too hot.”
The team also hooked its system and fan to the Big Green Egg to compare results. After all, there are add-on fan-based temperature-control accessories like PitmasterIQ and BBQ Guru out there already. The Harvard tech worked well with the Egg, but with their hyperboloid meat bot, it produced perfect results. The Egg has a hot spot over its vent, contributing to an uneven cook. Plus, refueling the Egg required removing the meat and grill rack, a flaw the team tackled with a side chute.
And finally, ease of use. The accompanying app gives live readings of the temperature inside the smoker and the meat, and it coaches the user through the process. Recipes and tips are planned for the app—they’ll dynamically adjust based on the weight and type of meat—and the team plans to expand beyond brisket-based algorithms.
Click & Close Ads
Click & Close Ads
Snow Job
Nature didn’t exactly cooperate with the project. The team had to contend with the most brutal winter in Boston history, as about 100 inches of snow fell during the semester. Much of the testing had to be done in inclement weather—with a start time of 3 am.
“We ran experiments every Saturday regardless of the weather,” says junior Salathiel Ntakirutimana, an electrical engineering major. “It snowed heavily for two or three of those smokes. It rained at least four times, and it hailed at least once.”
The weather was beautiful at the smoker’s unveiling, a sunny 80-degree day in May. Based on taste tests from the crowd, the smoker’s forecast was just as bright. The brisket was fall-apart tender. You’d pick up a piece of the flat, and gravity would pull it apart. The flavor was delicious and smoky. Adding sauce seemed blasphemous.
“I’ve tried barbecue cooked over mesquite, and sometimes it’s like it’s been cooked over tires,” says Peter Degnan, director of product development at Williams-Sonoma. “It gets so overly smoked. This was very nice. It was pleasant.”
The Next Course
Even though the class is over and the brisket passed important taste tests, there’s still work to do in order to bring the smoker to market. Three students—DeGraaf, Kaczor, and Yinka Ogunbiyi—are working to make it more versatile and address Williams-Sonoma’s wish list. Because the prototype was hand-made in a ceramics lab, the shape was lumpier, heavier, and smaller than the class’s original design. Over the summer at Harvard Innovation Labs, the trio hopes to make the rig lighter for easier shipping and to widen the cooking surface to accommodate more kinds of meat.
Over the course of the engineering class's semester of outdoor smoker testing, 100-plus inches of snow fell in Boston—its worst winter ever.
“Thanksgiving is Williams-Sonoma’s biggest selling time, and they want to confirm that this can also cook turkey and do rotisserie chicken,” Ogunbiyi, a bioengineering major, says. “That’s our main focus in terms of redesigning it.”
Click & Close Ads
Click & Close Ads
The team may also perform a few aesthetic nips and tucks to make the smoker a bit more Williams-Sonoma-friendly.
“A lot of Williams-Sonoma’s existing products are stainless steel,” DeGraaf says. “The ceramics that we’re using are better for the actual cooking process—it allows you to have a much more even temperature. So any stainless steel that we add will just be for aesthetics.”
Even if Williams-Sonoma doesn’t bite, the smoker may still end up in stores. The three students are launching a Kickstarter campaign to fund a professional prototype, and they’re starting their own company. Parker and teaching assistant Peyton Nesmith will serve as advisers, and other students in the class are also offering input as needed. They’re focusing on the smoker for now, but they also hope to create more smart cooking appliances.
“I think some of the other students were most interested in the engineering element, while we were interested in both the engineering and the business,” Ogunbiyi says. “We wanted to pursue that and see if we could commercialize the barbecue.”
As for old-school pitmasters who want to test their skills against this bold new smoker, they may get their chance. Once the next-generation prototype is ready, the smaller team intends to compete in contests from Texas to Kansas City to Memphis. “We’re fully competition-legal,” Kaczor says. “And I had a bag of BBQ Lay’s chips today, so I can definitely eat more barbecue.”Click & Close Ads
Click & Close Ads
Sunday, July 5, 2015
