I NVESTIGATIONS ON P ARTICLE D YNAMICS宝宝怎么添加辅食
IN A P LATE TYPE E LECTROSTATIC P RECIPITATOR USING
D OUBLE-P ULS
E H OLOGRAPHY
Hans-Joachim S CHMID and Heinz U MHAUER
Institut für Mechanische Verfahrenstechnik und Mechanik
Universität Karlsruhe (TH), D-76128 Karlsruhe, Germany
E-Mail: ha-jo.schmid@ciw.uni-karlsruhe.de
Abstract
The paration of particles from gas within electrostatic precipitators is mainly depend-ing on the flow conditions. The flow field is governed by the mutual interaction of the electric field with the gas ions and the turbulent channel flow. Especially for wire-plate electrostatic precipitators this caus co
ndary flows, well known as the 'ionic wind'. The influence of the ionic wind on the precipitation of particles has been discusd widely in the past.
Double-pul holography was applied to investigate the particle motion within a labora-tory scaled plate-type electrostatic precipitator with round wires as discharge electrodes. This experimental method allows to determine size, location and velocity of all particles within a specified volume at one distinct moment. Subquently frequency distributions as well as spatial distributions of the particle collective within the precipitation zone are evaluated.
A quantitative analysis of the migration velocities in the near-wall region and a compari-son with theoretical migration velocities, leads to the conclusion that the ionic wind has an order of magnitude about some 10 cm/s.
1 Introduction
identifying
The motion and precipitation of particles in the duct of an electrostatic precipitator depends mainly on the electric field strength, the accumulated particle charge and the gas flow. Many theoretical investigations (e.g. Bernstein and Crowe 1979, Yamamoto and Velkoff 1981, Shaugnessy et al. 1985, Liang and Lin 1994) as well as experimental investigations (e.g. Yabe et al. 1978, Leonard et a
l. 1983, Larn 1986, Riehle and Löffler 1993b, Miller and Schwab 1996) has shown the prence of a significant interaction between the ionic current emanating from the discharge electrode and the turbulent flow field of the neutral gas molecules. Opposite of the discharge electrodes a condary flow towards the collecting electrode is formed, becau at the locations there are the highest ion concen-trations. On the other hand the law of continuity requires a reversal flow at symmetry lay-ers between two electrodes. This should result in large eddies at the walls, as schemati-cally shown in fig. 1. Furthermore a supplemental turbulence production by the ionic
wind has to be suppod Therefore it is not obvious, whether the ionic wind ad-vances or deteriorates particle precipita-tion process.
The aim of the investigations prented in this paper is to examine the effect of condary flows and to discuss their influence on particle precipitation.
2 Experimental Approach 2.1 Double-pul Holography
As experimental approach puld-lar holography has been chon. It is a short-time photographical imaging technique bad on the formation of a spatial image. If particle concentrations are not too high, an 'in-line' arrangement has proven very suitable for the investigation of the dynamic behaviour of the disperd pha (Schäfer and Umhauer
1987, Schäfer 1986). Par-ticle concentrations about 1000 1/cm 3 have lead to optimum results. In this arrangement a parallel co-herent beam emitted by a puld ruby lar, ex-panded to a diameter of 80mm travers the region to be investigated and forms the hologram on the holo-gram plate by superposi-tion with the diffracted light (fig. 2a). In this way a comparatively large vol-ume (up to 350 cm 3) may be imaged. The application of two short puls allows to record all particles within that volume twice.
The systematic evaluation of a hologram (fig. 2b) is done mi-automatically. This leads to the knowledge of the 2x3 coordinates and the size of each individual particle within a specified volume with a very high reliability but is also time-consuming. Becau the in-terval between two puls is known (in the investigations between 40 and 100 µ
s), the
Fig. 1:Sketch of the electric field lines in a
wire-duct precipitator and of the result-ing field lines.
Fig. 2:Schematic reprentation of the arrangement for
holographic recording of a particle enmble (a)
and evaluation of holograms (b).
velocity as well as more complex quantities such as kinetic energy or momentum of indi-vidual particles can be calculated.
From the quantities, frequency distributions as well as profiles of concentration or par-ticle velocities across the gap may be derived. Furthermore it is possible to examine the spatial distribution of the disperd pha in the flow (Neumann and Umhauer 1991). Be-cau pul holography is a concentration related measuring technique, the results are al-ways bad on spatial averaging (Raasch and Umhauer 1977)
2.2 Laboratory-scale Precipitator
The experimental tup for recording the holograms is shown in fig. 3.
A single vertical duct with a cross
ction of 100 x 100 mm 2 is ud as
model of an electrostatic precipita-tor. The precipitation zone with a length of 250 mm and 5 round wires with a diameter of 0.3 mm as dis-charge electrodes is following an inlet zone with a length of 500 mm.The dimensions of the laboratory-scale precipitator allow the detection of a comparatively large part of the
channel at once. It is obvious that
the dimensions are really small compared to industrial devices, but Riehle and Löffler (1992 and 1993a)
have shown that, provided there is
只要你说你爱我2geometrical similarity, also experi-ments in smaller devices can be
transferred if at the same time elec-trical similarity is prerved.
The insulating walls, fixing the dis-charge electrodes, are broken to hold the lens at one side and the
hologram plate at the opposite side.
The particle are fed into the channel
by a sieve stimulated ultrasonically.
嘿朱迪歌词The mesh size was chon to be only a little larger than the particle
size to assure perfect dispersion. To inhibit deposition of dust on the lens and the holo-gram plate the particles are fed only in a slit, located in the centre of the duct and perpen-dicular to the collecting electrodes. Besides this assures, that the particles are in a flow re-gion, which is not disturbed by wall irregularities, necessary to optically acces the pre-cipitation zone. The particles which are leaving the electrostatic precipitator are collected in a cyclone which is positioned downstream, ahead of a suction fan.
Fig. 3:Schematic view of the experimental tup for holographic recording
The optical arrangemant allows to record holographically in x-direction the whole region between the dis-charge electrodes and one collecting electrode at once. In flow- (y-) di-rection maximally a region longing
from the inlet to the third discharge electrode is accessible (fig. 5).
Most of the data prented in this
paper is gained by evaluation of very flat volumes (referred to as
'evaluated layers') situated perpen-dicular to the mean flow direction山顶洞人英文
(compare with fig. 4). The first layer is at the position of the first
discharge electrode (y=25mm), the
cond layer is situated between the first and cond discharge electrode
(y=50mm).人教版初一英语教材
As disperd pha a narrow fraction of glass beads with a mean diameter of 20 µm is ud. In this size range a very good holographic detection is assured. Although particles of the sizes are easily precipitated, the evaluation of the data shows very interesting de-tails of the fluid flow.
3 Experimental Results
As experimental parameters fluid velocities ranging from 0.5 to 2.0 m/s and applied volt-ages varying from corona ont voltage up to 30 kV, which corresponds with a mean electric field strength of 6.0 kV/cm were chon.
The data gained by the evaluation of the
holograms may be analyd in many different ways. A vector plot of a parti-cle collective is shown in fig. 5. This figure reprents the projection of the particle enmble in z-direction, each arrow indicating one particle, with the foot of the arrow reprenting the posi-tion and the arrow itlf the particle ve-locity. This gives a very good qualita-tive impression of the particle motion.It should be especially noticed, that even particles very clo to the wall can
be imaged. Furthermore momentary
local fluctuations of the particle con-centration, which may be entailed by small irregularities of the particle feed,can easily be detected. This is a specific advantage of this whole-field method,
ndaFig. 4:Holographically accessible volume and evaluated volumes for data prented here.
202530x-position / mm
y -p o s i t i o n / m m
Fig. 5:Vector plot of particle velocities. Pro-jection in z-direction.
which is hard to attain by other experimental techniques.
A more quantitative description is possible with profiles of the migration velocities across the gap from the middle of the channel (x=0mm) to the wall (x=50mm) as shown in fig.6. Hereby a local averaging of this velocity component over small ctors has been con-
ducted. The error bars indicate the standard deviation within each averaging ctor.
Fig. 6:Particle migration velocity profiles for different flow velocities and different lay-ers. U = -25 kV.
溃退
It is well known, that the component of the electric field strength perpendicular to the wall in the first layer (y=25mm) pass through a minimum at about x=15mm (e.g. Leutert 1971, Miller et al. 1994). The measured profiles of the migration velocities also show a minimum at almost the same location as the electric field strength.
Furthermore the influence of particle charging kinetics can be obrved. The migration ve-locities for a fluid velocity of 2m/s are significantly lower than for 1m/s, which corres-ponds with decreasing residence time and hence particle charge for increasing mean flow velocity.
A comparison of the migration velocities in the near-wall re-gion for different fluid veloci-ties and locations yields an in-teresting insight in the EHD-flow. Nearby the wall the elec-tric field strength is virtually independent from the position in streamwi direction. If one assumes that the particles have
take down
been expod to a homogene-ous electric field E (with E as
the field strength at the wall)and the particle charge calcu-lated according the field charging theory (Pauthenier
and Moreau-Hanot 1932), the
stationary theoretical drift ve-25 mm 40 mm 25 mm 50 mm 25 mm 50 mm w x / (m / s )
v y
= 0,9 m/s v y = 1,25 m/s v y = 2,0 m/s
streamwi velocity:y-position:Fig. 7:Comparison between measured and calculated migration velocities near the wall.
locity may easily be calculated. The measured migration velocities will be compared with the theoretical values, calculated as outlined.
This is shown in fig. 7 for an applied voltage of -25kV and different mean fluid velocities and locations. In layer one, opposite of the discharge electrode, the migration velocities are always significantly higher than theoretically predicted. On the other hand at layer two, between the 1st and 2nd discharge electrode, measured and calculated values corres-pond very well. This may readily be explained by condary flows: Whereas the calcu-lated drift -velocity is relative to the fluid flow, the measurements yield the migration -ve-locity towards the wall, which includes condary flows. Therefore the quantity denoted ∆v in fig. 7, which is computed according to
∆=−()−−()==v w w w w meas theor y mm meas theor y mm 2550may be ud as a measure for the strength of the ionic wind. Although ∆v scatters signifi-cantly for the three different fluid velocities, it can be stated, that the ionic wind, at least for round discharge wires, is about veral 10 cm/s.
4 Summary and Conclusions
金泡泡The prented examples show, that the method of puld lar-holography is well suited to study the particle motion and especially the characteristics of condary flows. The very detailed description of
the disperd pha, namely the knowledge of location, size and velocity of each individual particle, yielded by a systematic evaluation of parts of the hologram may be prented in different ways.
The visualisation of the particle motion by means of vector plots allows a good impres-sion of the motion of the particles and of local variations in particle concentration. Profiles of the migration velocities correspond well with the expected cour, derived from the known cour of the electric field strength.
The evaluation of the near-wall region shows the influence of condary flows. It could be shown that the ionic wind has an order of magnitude of about some 10 cm/s. This will surely have a great impact on the motion of small particles in the submicron range which have naturally a smaller migration velocity. Therefore at least opposite of the discharge electrodes the motion of small particles is mainly influenced by condary flows in addi-tion to the direct particle drift due to electrostatic forces.
Although the evaluation is time consuming, only few holograms taken until now lead to very detailed informations. Additional holograms, taken at identical conditions should be recorded, to further increa statistical reliability, especially for evaluating velocity fluc-tuations within small ctors. Furt
hermore investigations of the strongly three-dimen-sional condary flow, emanating from barbed wire electrodes shall be done.
Acknowledgements
This work was supported by the 'Deutsche Forschungsgemeinschaft (DFG)'(Lo 142/23-1).
References
Bernstein, S.; Crowe C.T. 1979: Interaction Between Electrostatics and Fluid Dynamics
in Electrostatic Precipitators. 2nd Symposium on the Transfer and Utilization of Particulate Control Technology, July 23 to 27, 1979, Denver, Colorado.
