Development of Ultra-High Speed Detectors to Study the Physics of Turbulence

Cornell University, Ithaca NY

Investigators

Abstract

Particle tracking in turbulent flows requires very high temporal (approx. 100,000Hz) and spatial resolution imaging detection systems (512 x 512 pixels). Such a system generates data at very high rates, which need to be transferred, stored, and analyzed. In addition three-dimensional tracking of many particles requires the simultaneous stereoscopic imaging by four cameras. Traditional detector technology cannot handle the resulting onslaught of data (> 26 gigabytes of data per second per camera) that would need to be streamed into memory and stored for further processing. Here, we will develop the next generation, ultra-high speed, high-resolution particle tracking instrumentation for following the motion of approximately 300 minute tracer particles in three dimensions with a temporal resolution of 100,000Hz. This new system will be two orders of magnitude faster than any existing conventional imaging system of comparable spatial resolution. The instrument will be based on stereoscopic imaging with four ultra-high-speed, high-resolution pixel array detectors (PAD) that will be developed at Cornell. Each PAD has 1024x512 "intelligent" pixels. The development of the uniquely designed PAD detectors, the data acquisition system, and the data analysis software will result in a technological breakthrough for the Lagrangian analysis of high Reynolds number turbulent flows. Fully developed turbulence is ubiquitous. Virtually all engineering and naturally occurring flows (e.g., atmospheric, oceanographic, and astrophysical flows ) involve high Reynolds number turbulence. The transport and mixing properties of turbulence ultimately impact our daily lives. In systems from the mixing and chemical reactions in a turbulent burner or internal combustion engine to the transport of pollutants (or bioagents) in the atmosphere, or even the "simple" mixing of milk into a coffee cup, the understanding of turbulence is essential. It is clear that many aspects of turbulence are best studied by following the motion of fluid particles. The difficulty of tracking particles in high Reynolds number turbulence has thus far made it impossible to test many long-standing predictions. This project represents a major step to confront this issue. The detection and analysis system to be developed will simultaneously follow 300 particles at a rate of 100,000 pictures per second. (A standard TV camera takes 30 pictures per second.) This high speed will be achieved by using pixel array technology in which each pixel has local intelligence. Data of this kind will enable theoretical advances in a number of important applications such as scalar mixing and cloud formation. Moreover, the technology, once developed, is likely to impact broader fields than can not be envisioned at this time. For example, tracking of particles is at the heart of a number of environmental problems (e.g., contaminant or bioagent dispersion). A successful PAD fabrication will likely result in more sensor chips than needed for this study. Many of these chips would be made available to the larger community.