Page 156 - Air Pollution Control Engineering

03_chap_wang.qxd 05/05/2004 12:48 pm Page 135
Cyclones 135

Table 14
Multiple Regression Results for ψ
50
Parameter Coefficient Standard deviation
exp (F ) 0.0414
0
F −0.713 0.008
1
F −0.172 0.033
2
2
R = 0.981; GSD = 1.12
Source: (Adapted from Lidén and Gudmundsson (49), with permission from Taylor and Francis;
http://www.tandf.co.uk/journals.)

Table 15
Parameters a , a , and a for Eq. (68)
1 2 3
Cyclone a a a
1 2 3
Stairmand 0 4.848 0.627
SRI 0 8.196 −3.239

4.4. The “Cyclone Family” Approach
Kenny and Gussman (53) recently presented a detailed experimental study based
on three cyclone “families” (whose relative dimensions are in fixed proportions to the
body diameter) named Extra Sharp Cut Cyclone (ESCC), Sharp Cut Cyclone (SCC),
and Gussman Kenny (GK). The authors refer to the ESCC cyclone as designed to give
a very sharp penetration curve, intended for ambient air monitoring. The base of the
cyclone is cylindrical rather than conical. The SCC family is based on the design of
the SRI–III cyclone (54) and, according to Kenny and Gussman, provides both a
sharp cut and stable performance under loading. The GK cyclone has a wider pene-
tration curve and was designed for use in the workplace and indoor air monitoring.
The detailed geometrical characteristics of the families are not clearly stated in the
article. However, the authors give indications that led to the dimensions listed in
Table 16, which may contain deviations.
The experimental work, described in detail by Maynard and Kenny (55), included
the testing of 36 diverse cyclone combinations, resulting from modular combinations
of cyclone parts, all with a body diameter D of 17.5 mm. At the first stage, the
c
cyclone tested comprised more than the three families. In a second phase, the experi-
ments were focused on the three families: seven ESCC, seven SCC, and eight GK
cyclones were tested. The test aerosol was made of poly-dispersed glass microspheres,
3
with nominal diameters up to 25 µm and density of 2450 kg/m . The size distribution
was measured with an aerodynamic particle sizer (APS 3310 or APS 3320, manufac-
tured by TSI Inc.). Penetration curves were determined, as a function of particle size
and gas flow rates. Each test was characterized by the cut diameter D (the diameter
50
with 50% collection efficiency) and by the sharpness of the penetration curve, given by