An experimental investigation involving the design, fabrication, and testing of an ambient sampling inlet and two additional Stokes-scaled inlets is presented here. Testing of each inlet was conducted at wind speeds of 2, 8, and 24 km/h (0.55, 2.22, and 6.67 m/s), and characterized for particle sizes between 5 and 20 õm AD. The base-line ambient sampling inlet, which operates at 100 L/min, was developed to interface with a Circumferential Slot Virtual Impactor aerosol concentrator. The inlet displays wind-speed independent characteristics with a penetration above 90% for a nominal particle size of 10 õm AD for all wind speeds. Particles up to 11.5 õm AD are sampled through this inlet with a penetration above 80% at all wind speeds. In an effort to test the validity of Stokes scaling to assist in the design of inlets, two additional inlets were designed to accommodate design flow rates of 400 L/min and 800 L/min, with the 100 L/min unit as the base inlet. Scaling was achieved by applying a Stokes scaling factor to selective parameters, such as inlet aspiration gap, annular gap, window height, and the rise which is the vertical distance extending from the lower flange to the base of the window. The scaled inlets display wind independent penetration characteristics close to 95% for a nominal particle size of 10 õm AD. The scaled inlets also have the ability to sample particles up to a size of 13 õm AD with a penetration in excess of 80% at all wind speeds. Observations from the plots of penetration against the Stokes number based on the free stream velocity suggest that it is insufficient to use only Stokes-scaling for inlet design. A modified velocity ratio defined for omnidirectional inlets was incorporated into a summary of results obtained for all combinations of BSI units and wind speeds. Also, a correlation equation based on the Stokes number and a modified velocity ratio was developed as a model for predicting performance among the BSI family of inlets. This correlation used in unison with Stokes-scaling provides promise for predicting performance and improving the overall design process of inlets.
Identifer | oai:union.ndltd.org:TEXASAandM/oai:repository.tamu.edu:1969.1/5736 |
Date | 17 September 2007 |
Creators | Nene, Rohit Ravindra |
Contributors | McFarland, Andrew R., Haglund, John S., Hassan, Yassin A. |
Publisher | Texas A&M University |
Source Sets | Texas A and M University |
Language | en_US |
Detected Language | English |
Type | Electronic Thesis, text |
Format | 2902326 bytes, electronic, application/pdf, born digital |
Page generated in 0.0018 seconds