Power from the exhaust pipe


Pitt researchers Shervin Sammak and Matthew Barry discuss mass transit, bus exhaust
and waste heat with Andrew Lukaszewicz,  PAT interim director of rail service delivery.

 

Early on a bright morning in Pittsburgh’s Manchester neighborhood, engine overhaul mechanic Gavin Bittner is explaining bus engines to several Pitt engineering undergraduates at a maintenance shop of the Port Authority of Allegheny County (PAT). A few yards away from the open engine compartment, Shervin Sammak and Matthew Barry discuss mass transit, bus exhaust and waste heat with Andrew Lukaszewicz,  PAT interim director of rail service delivery.

Barry, assistant professor in Mechanical Engineering and Materials Science (MEMS), and Sammak, research assistant professor in MEMS and Pitt CRC consultant, are co-PIs on a research project that aims ultimately to reduce PAT’s annual diesel fuel consumption by 800,000 gallons. They don’t propose using electricity or renewable energy. Instead, the project aims to reduce carbon-based fuel consumption by reusing a by-product of carbon-based fuel – waste heat expelled with bus exhaust fumes. A thermoelectric device inside the exhaust pipe will recover waste heat to produce electricity to power systems that now rely on electricity generated by the alternator on the diesel engine.

In the first stage of the project, Barry and Sammak are performing analysis and modeling based on a large set of historical and geographical data provided by PAT. The work is funded by the Mascaro Center for Sustainable Innovation (MCSI) in the Swanson School of Engineering.

Pitt CRC’s Sammak is enthusiastic about the project. In addition to consulting and teaching,  Pitt CRC research faculty members play vital roles  as collaborators in funded research


Shervin Sammak points out the bus exhaust on the 
roof, not the underside, of a PAT bus.

“It also aligns with our existing mission – to help researchers apply advanced computing. Pitt CRC consultants can partner as PIs in writing proposals that incorporate credible and innovative uses of computation. We know a lot about the technical aspects of advanced computing, but we can also help apply advanced computing creatively within a domain.”

For modeling a project converting waste heat to electricity, why choose buses? For many reasons, explains Barry

“Converting heat to electricity has been mostly used in space, such as on the Cassini orbiter that spent 13 years circling Saturn. Thermoelectric generators that used heat from the decay of the fuel –  plutonium dioxide – generated the electricity that powered the spacecraft.”

“We reoriented the application of waste heat from space to conservation on earth using carbon-based emissions. The best model to study is a high-mileage vehicle that operates continuously, travels at consistent speed, and produces enough exhaust to justify the investment. That vehicle is a bus. A bus is also big. It has space underneath where you can work more easily, rather than in tight spaces on a car.”

The project could eventually have a significant impact. Barry and Sammak’s proposal cites a striking figure contained in a Department of Energy report: 50 to 60 percent of the enormous amount of energy produced in the United States every year is lost as waste heat, a large portion of that coming from vehicles. According to that DOE report, waste heat from transportation represents approximately one fifth of the energy consumed in the United States.

Sammak concentrates primarily on thermodynamic modeling and analysis, working with modeling software on Pitt CRC’s Shared Memory Processing (SMP) cluster. The modeling analyzes and visualizes multiple configurations of waste heat based on 406 months of driving data from the PAT fleet and buses in other cities, including distance traveled, electrical needs, and variations in temperature and velocity.

“We’ll move from analytical modeling to developing proof of concept within a year,” Barry explains. “Then we build a prototype and begin lab testing. We hope to have a usable, economically viable device on the road in five years.”

 

 

Contact:

Brian Connelly
Pitt Center for Research Computing
bgc14@pitt.edu
412-383-0459