Studying Engines in Action
Hank Hogan
Although it was enough for a famous “little engine” to think it could, more than wishful thinking is needed to move things along in the real world. Motors produce movement by using an energy source, which for internal combustion engines is a chemical reaction. For researchers, monitoring the chemical reactions that take place in a combustion chamber as an engine burns fuel presents a challenge.
To handle such measurements, Christopher L. Hagen, at the time a graduate researcher in mechanical engineering at the University of Wisconsin-Madison, and Scott T. Sanders, an assistant professor of mechanical engineering at the university, developed a hyperspectral sensing system based on infrared absorption spectroscopy. Hagen is now with Chevron Energy Technology Co. in Richmond, Calif.
This setup was used to investigate homogeneous charge compression ignition combustion, which could lower emission and increase efficiency of engines. It consists of a quartz tungsten halogen lamp, a 550-μm multimode fiber (MMF1), lenses (L1, L2 and L3), a metal single-cylinder homogeneous charge compression ignition (HCCI) research engine with sapphire windows, 62.5-μm multimode fiber (MMF2), and an extended indium gallium arsenide (ex-InGaAs) camera-coupled grating spectrometer.
The researchers developed the system to help their investigation of homogeneous charge compression ignition combustion engines. In these engines, a roughly uniform mixture of fuel and air ignites because of compression. This technology can lower emissions and increase efficiency as compared with conventional spark-ignited gasoline or diesel engines. Consequently, the engines have been investigated extensively for the next generation of automobiles.
Homogeneous charge compression ignition combustion does not use a spark plug or fuel injection control. Instead, the homogeneous mixture of fuel and air ignites in many places simultaneously, with the start of ignition dependent on chemistry. To understand this process, accurate probing of conditions inside the chamber is required, which can be challenging because of the heat, pressure and transitory chemical species present in the chamber. Moreover, the measurements must be acquired without disturbing anything.
The researchers, therefore, employed absorption spectroscopy, a technique that integrates every substance’s spectra along the line of sight. This averaging type of measurement works because the mixture is uniform.
In their setup, the scientists used an Acton spectrometer and a Princeton Instruments InGaAs camera with a spectral response from 1.0 to 2.2 μm. The instrument acquired 900 spectra per second over the 1600- to 1850-nm range, with a resolution of 0.75 nm. They used a quartz tungsten halogen lamp from Newport Corp. of Irvine, Calif., to illuminate the combustion chamber through sapphire windows as the fuel isooctane burned in the engine, and they measured the result. Hagen noted that the measurement involved some unknowns.
“We did not know the absorption strength of isooctane vapor at elevated temperature or pressures for our chosen wavelength range prior to the experiment,” he said.
They examined the absorption of water and isooctane in part because of their spectral separation. They averaged readings over many engine cycles to reduce noise and used postprocessing of the data, resulting in clear measurements of water vapor, fuel density and temperature. Such information is a step toward characterization of the chemistry in a homogeneous charge compression ignition combustion engine. The work will be published in the
Journal of Near Infrared Spectroscopy.
Hagen said that some further improvements of the instrumentation depend upon the development of enabling technologies. For example, a bright, broad and stable supercontinuum source would help increase the signal-to-noise ratio. A faster camera also would help.
He noted that the ideal instrument would record time-resolved temperature, pressure and concentrations of all species of interest with high spatial resolution and without the need to average. That, unfortunately, cannot be done yet, but — like the little engine that could — the investigators are not giving up.
The technique could have other uses besides monitoring engines. It could be used in applications ranging from combustion analysis to medical imaging to military sensors, Hagen said.
Contact: Christopher L. Hagen, Chevron Energy Technology Co., Richmond, Calif.;
e-mail: clhagenco@yahoo.com.
LATEST NEWS