GB2328505A

GB2328505A - Analysis of particles flowing in a fluid

Info

Publication number: GB2328505A
Application number: GB9814896A
Authority: GB
Inventors: Heinz Umhauer
Original assignee: Palas GmbH Partikel und Lasermesstechnik
Current assignee: Palas GmbH Partikel und Lasermesstechnik
Priority date: 1997-08-20
Filing date: 1998-07-10
Publication date: 1999-02-24
Anticipated expiration: 2018-07-10
Also published as: GB9814896D0; DE19736172A1; DE19736172B4; CH693829A5; GB2328505B

Abstract

An inventive apparatus for the analysts of particles dispersed in a flowing fluid with an illuminating device, a detecting device and an evaluating device, in which the illuminating and detecting devices are so constructed that in the particle flow they define a measuring volume with a cross-section varying in the flow direction over the height thereof, and in which the evaluating device is so constructed that it only permits further evaluation of those particles, whose measured transit time exceeds a predetermined minimum signal duration, provides for the lateral surfaces of the measuring volume to have a construction changing in continuously concave manner. (Figure 3 illustrates the measurement volume).

Description

2328505 METHOD AND APPARATUS FOR ANALYZING PARTICLES The invention relates

to a method for analyzing particles dispersed in a flowing fluid. the particles being detected in an optically defined measuring volume with a differing cross-section in the flow direction over the height thereof and evaluation only takes place of those particles, whose measured transit time exceeds a predetermined minimum signal duration. The invention also relates to an apparatus for analyzing particles dispersed in a flowing fluid, having an illuminating device, a detecting device and an evaluating device, the illuminating and detecting devices being constructed in such a way that they define in the particle flow a measuring volume with a differing cross-section in the flow direction over the height thereof and in which the evaluating device is constructed in such a way that it only permits evaluation of those particles, whose measured transit time exceeds a predetermined minimum signal duration.

Scattered light particle counters are used both for quantity and size determination of aerosol particles. The definition of an adequately small measuring volume through which the particles individually move for counting and size determination is characteristic of all scattered light particle counters.

If measurement is to take place directly in the given aerosol flow (to the extent that the conditions allow this), then the measuring volume is defined with optical means. This has hitherto taken place by imaging diaphragm using two partial optical systems in a 90 arrangement (Unhauer, H.: Particle Size Distribution Analysis by Scattered Light Measurements using an Optically Defined Measuring Volume, in J. Aerosol Sci., vol. 14, No. 6, pp 765-770, 1983). The diaphragms e.g. have a rectangular aperture, so that within a certain depth of field of the imaging a parallelepipedic measuring volume is defined. This is also illuminated with one of the two partial systems and with the other the light scattered by the particles is measured. Necessarily with such a measuring volume definition, a characteristic error results from the fact that there is always a certain proportion of particles, which move through the marginal areas of the measuring volume or, in other words, are "cut off" by the edges or borders.

Apart from mathematical correction, it is possible to counteract the socalled marginal zone errors, in that apparatus and signal processing measures are taken which avoid the error from the outset, i.e. it is eliminated during the measuring process. Thus, according to the aforementioned literature reference, the error can be eliminated by a double measurement with two different detector diaphragms and comparison of the pulse heights.

A method and apparatus according to the preamble are known from the Umhauer and Berbner publication, Optical In-Situ Analysis of Particles Dispersed in Gases at Temperatures of up to 10000C, 6th European Symposium Particle Characterization, Nuremberg, Germany, March 11-23, 1995, Preprint, p 327 ff, the content of said article being fully incorporated into the subject matter of the present disclosure. By means of the known method and apparatus, it is possible to avoid measuring errors such as have hitherto occurred in the case of a purely mechanically-optically defined measuring volume, such as were previously standard, if particles flow along the marginal area of the thus defined measuring volume and are only partly illuminated by the illuminating light. They consequently scatter a smaller quantity of light onto the detector than would be the case if they were fully illuminated and therefore simulate a lower particle size than corresponds to their actual particle size. This can impair the measurement and in particular the determination of the particle size distribution. Such errors are excluded by the optically-electronically defined measuring volume according to the above document and, during evaluation, account is then only taken of particles, whose transit time exceeds a given minimum signal duration defining a measuring volume cross-section. This only permits the measurement of particle flows having a constant velocity, or the particle velocity must be continuously determined over a sufficiently long measuring time by mean value calculation and then the said minimum signal time or duration is so mathematically corrected that the measuring volume cross-section remains constant.

The problem of the invention is therefore to provide a method and an apparatus with which, whilst avoiding the aforementioned disadvantages, a correct measurement automatically takes place even with a variable particle velocity.

According to the invention, this problem is solved with a method of the aforementioned type, which is characterized in that the lateral surfaces of 1.

the measuring volume are not straight. For the solution of the set problem, an apparatus according to the preamble provides for the lateral surfaces of the measuring volume not to be straight.

According to preferred further developments of the inventive method, the lateral surfaces are constructed in a continuously concave-varying manner and the optically defined measuring volume has hyperbolic lateral surfaces and in particular the'optically defined measuring volume has a symmetrical construction.

The said measuring volume can, according to a preferred development, be provided in that the particles are illuminated and detected by diaphragm apertures, which have continuously concave shaped boundary edges or in that the particles are illuminated and detected by diaphragm apertures having hyperbolically constructed boundary edges. In a further development, the diaphragm apertures are identically constructed in the illumination and detection path.

By means of the method according to the invention, it is possible to measure particles with a varying particle velocity, whose transit times in the measuring volume differ to a certain extent from the selected minimum signal duration. To be able to take account of larger variations in the particle velocity, according to a preferred development of the method, in the case of a varying particle velocity and therefore maximum signal duration of the signals resulting from the particles the minimum signal duration is adapted and in particular an adaptation takes place if, in the case of a velocity variation, the maximum signal duration of the signals caused by the particles has approached by 1.1 times the actual (predetermined) minimum signal duration or has moved away therefrom by 3. 8 times. According to preferred developments of the inventive method, the optically defined measuring volume has hyperbolic lateral surfaces and the optically defined measuring volume is symmetrically constructed. For creating a corresponding, optically defined measuring volume, according to a preferred development, illuminating and detecting devices are provided, the boundary edges of the diaphragm apertures are continuously concavely constructed or the boundary edges of the diaphragm apertures are constructed as hyperbolas. According to a further development, the diaphragm apertures of the detection and evaluating devices are given an identical structure.

The invention provides a primary, mechanically-optically defined measuring volume formed by illumation and detection diaphragms, inscribed in a pyramid or frustum, but which has concave or hyperbolic boundary surfaces, which are formed by concave or in particular hyperbolic edges between the base and tip of the pyramid. Thus, the optical-electronic secondary or active measuring volume of the particles admitted for evaluation and determined by the minimum signal duration remains the same, independently of what minimum measuring time is allowed. According to a preferred development, a measuring volume is formed, which is inscribed in an octahedron and having concave, particularly hyperbolic boundary surfaces in each case between the median plane and the tips of the octahedron, which are formed by diaphragms having concave or more particularly hyperbolic boundary edges.

The inventive method and apparatus can in particular be used for analyzing a s olid flowing in a gas, but also liquid particles flowing in a gas. It is possible to determine solid particles flowing in a liquid or particles of a second liquid flowing in a first liquid.

Further advantages and features of the invention can be gathered from the claims and the following description of an embodiment with reference to the attached drawings, wherein show:

Fig. 1 Fig. 2 An apparatus according to the invention for performing the method according to the invention.

A diagrammatic representation of the imaging of the inventive illuminating and detection diaphragm apertures in the particle flow.

Fig. 3 The optically defined, primary measuring volume in the particle flow in a preferred development according to the invention.

Fig. 4a The path of the hyperbolic boundary surfaces of the preferred measuring volume according to fig. 3.

Fig. 4b A plan view of the measuring volume of fig. 3 in the particle flow direction.

Fig. 5 Fig. 6 A velocity curve of the measured particles, over time, with fluctuations.

A frequency distribution of the signal durations.

Figs. 7a-c Three possible diaphragm aperture designs.

The apparatus 1 according to the invention has a tube 2, which contains a fluid flow with particles dispersed therein, such as a gas flow with solid particles. In the walls of the tube 2 are provided windows 3, 4, through which the gas flow can be illuminated and observed. An illuminating or lighting device 6 has a light source 7, preferably a white light source, a condenser system 8 and an illumination diaphragm aperture 9, together with achromatic lenses 11. An optical filter 12 is also provided. A detecting device 13 also has in the window 3 of the tube 2 a filter 14, a deviating prism 15, lenses 16, a further detection diaphragm aperture 17, a condenser 18 and a detector 19, such as in particular a photomultiplier. To the detector 19 is connected an evaluating unit 21, which in particular has a chronometer or time measuring device.

Fig. 2 diagrammatically shows the imaging of the diaphragms 9, 17 in the particle flow with the flow dirction S. From it can be gathered the shape of the diaphragm apertures 9, 17 of this embodiment. The diaphragm apertures 9, 17 have continuous, concave and in particular hyperbolically widening edges 22. As can be gathered from figs. 3 and 4b, the two diaphragms 9, 17 form a primary, optical measuring volume, which is inscribed in a frustum oriented parallel to the flow direction S and whose base is perpendicular to the flow direction S, the boundary surfaces 15 of the thus formed. primary measuring volume pass in a continuously concave manner within the frustum and the boundary.edges 22 are set back in concave manner with respect to the tl'- - 6 edges of the surrounding frustum. The edges 22 are in particular constructed as hyperbolas and the boundary surfaces are hyperbolic.

Thus for the path h of the lateral surfaces 15 of the optically defined measuring volume V m, as a function of the radial spacing s thereof or the cross-section A. in the case of a corresponding height: h - 1/(s) 2 1/AM, and consequently if h is the distance covered belonging to the minimum signal duration-(: hA, constant = As a result of the aforementioned, inventive choice of the geometrical design of the optically defined measuring volume V m with hyperbolic boundary surfaces or the diaphragm apertures forming said measuring volume with hyperbolic boundary edges, on giving a fixed minimum signal duration Y, the volume flow m within the thus formed actual measuring volume always remains constant, even with a variation of the particle velocity. Under these conditions, the count rate Z = N/t m is alone the measure for the c. oncentration C N = Z/ M Consequently the velocity can change at random within certain limits. For the volume flow applies:

2 2 v4(s) = 4 hI-C (s) ' 2 in which 4(s) = A. the (square) maximum measuring volume cross-section, whose actual value can change, h is the actual, external measuring volume height belonging to s and -t' is the fixed, predetermined minimum signal duration. The minimum signal duration 'C means that all particles, whose flight and therefore signal time is smaller than ^C are discarded.

The two diaphragm images perpendicular to one another span the measuring volume (fig. 3). The hyperbolic boundary surfaces commence at h min and end at h max (predetermined by the diaphragm design). Corresponding, the variable measuring volume cross-section has an absolute maximum at h im of 4s 2 max and a minimum at h max of 4s 2 min The surface normal to S 2 is the flow direction. Starting from a particular, given velocity v, a particle flight time or signal duration -C is established, in such a way that h. < h.9 h min max in which h - v.^V W being roughly in the centre between h min and h max).

The square surface 4(s) 2 - AM belonging to h is the maximum crosssection of the actual measuring volume, through which move all the particles with T>T and consequently h < h 'and with respect to which (or T) the already mentioned particle selection takes place. Thus, with T, with respect to the particular velocity v, the actual, maximum, lateral extension of the measuring volume Is fixed.

The following applies for (a) 2 6 2 V H = v. 4(s) 4(5) 2 v v tonst. h If the velocity changes (for a fixed T) there are changes both to (s) 2 and also h, in such a way that the product (0) 2. h - 1 V 4 remains constant. Here V is designated the base volume and its value is known. The parallelepipedic base volume V always maintains the same value, if h changes for a fixed -C due to velocity changes, but not the complete effective measuring volume V m, which is larger. Its value is obtained from the sum of the parallelepipedic base volume V and the residual volume over 2 the square surface 4(s) defined by the hyperbolic lateral surfaces and extending up to the height h max. The measuring volume V m becomes e.g. smaller, if h increases with increasing velocity, because the residual volume then always becomes smaller. As T is always predetermined. the volume flow V is also known. Thus, the count rate Z, which is measured with such a measuring volume definition, is the measure for the particle concentration CN, independent of velocity changes. If the count rate changes, this is due to a concentration change and not a velocity change. The velocity may change with random frequency. It need not be known, if it is ensured that the velocity changes do not exceed certain limits.

If a particle flows through the primary measuring volume, it is only detected, if its minimum transit time through the measuring volume, i.e. the time in which it reflects light on the detector, t is at least as long as the predetermined minimum signal duration t (corresponding to lateral extension s). The particle 2 in fig. 4b with is just detected. Thus, the evaluating device has a timer through which particles with a shorter transit time than At., such as particle 3, are excluded from further evaluation. Thus, within the mechanically-optically determined, primary measuring volume, as described hereinbefore, there is an optoelectronic definition of a secondary, active measuring volume, which has a parallelepipedic construction, the boundary planes being perpendicular and parallel to the particle flow direction S.

Independently of the particle precision, the invention eliminates marginal effects, such as occur with purely mechanically-optically formed measuring volumes, in that particles flowing along the edge of such a volume and which are therefore only partly illuminated and consequently simulate a smaller particle in the detector than corresponds to their individual size.

Whereas in the described embodiment, the hyperbolic boundary surfaces of the diaphragm apertures 9, 17 face the incoming particle flow S (figs. 2 and 3) and form an entrance boundary and the planar base surface of the measuring volume forms an exit boundary (fig. 7a), there can also be a flow against this measuring volume in the reverse direction (fig. 7b). It is also possible to provide hyperbolic boundary surfaces on the entrance and exit sides, so that the measuring volume would be inscribed in an octahedron. A corresponding diaphragm is shown in fig. 7c.

However, the compensation of velocity changes by changing (s) 2 only occurs within certain limits. A velocity curve with fluctuations and transit time behaviour is shown in fig. 5.

If the velocity change is too great, h leaves the range between h min and h max and the conditions for the operation of the method are no longer satisfied. In this case, t must be correspondingly defined again, so that h min < h < h max is again fulfilled.

To be able to take account of these facts, it is necessary to detect the excessive velocity changes, i.e. also the velocity must be sufficiently accurately checked if excessive changes are to be expected. As the signal time --C is permanently checked for each particle, within a certain measuring time t h it is also possible to give the velocity, also in the case of the measuring volume defined according to fig. 3. As a result of the design of this measuring volume (in the case of regular operation and v = constant). for the particle flight time and therefore signal duration -C, there is a frequency distribution Q 0 (T), as is qualitatively represented in fig. 6.

The proportion QO (-rlzax) of signals with a length of. ma results from those particles moving through the cross-section s 2 min Theyxall have the same signal duration ^C, because here the measuring volume is defined max and its height has everywhere the same value h-h mam (h max and h min representing system constants and are not parameters).

For checking the velocity limits, within which for a specific ^r the method is effective, It is merely necessary to roughly detect the upper limit of the distribution at -Cmax and to this contributes the fraction r1. max in its characteristic form. There is no need to representatively determine the entire distribution Q,(Z), which once again means that the measuring time t m can be relatively short. For the velocity limits apply:

v min h min /T and v max " h max /er From this it follows that with retained parameter -C and rising velocity _Cmax.>-T must remain, whereas with falling velocity there must be T min 4 T and therefore _r max < h max. -C h min because h must not become larger than h max and not smaller than h,,,,. On passing below or above these limits, as stated, the parameter ^r must be refixed in accordance with the changed situation (takes place automatically b 1 y computer program). If the signal duration --rmax in fig. 6 become ever smaller with rising velocity and approach ever more the value T, then-t must be reduced, etc.

The clear detectability of the fraction of the maximum signal durations (fig. 6) is a great advantage with the measuring volume definition procedure according to the invention. The maximum signal durations come from all those particles which move through the planar cover surface of the measur ing volume V m Thus critical velocity changes can be detected relatively early.

When reference is made to velocity changes, a distinction must be made between slow, continuous changes (drifts) and rapid variations about a mean value (fluctuations). A combination of both is also possible.

In the case of a high fluctuation frequency, i.e. a frequency greater than the reciprocal value of the necessary measuring time t m (this case being allowed), the distribution Q 0 (C) at the upper end will not have an abrupt boundary or limit, but will instead rise continuously to the value 1 (or 100%) within a more or less wide range. Here again, by checking (averaging) the -values in the upper range of the distribution it is possible to relatively rapidly establish whether the aforementioned conditions are respected for the correct operation of the scattered light particle counter. In all cases there is a quasi-on line control.

If the expected, average velocity v 0 v 0 = 1 mls = 1 Am/ps h 0 = 5. 10-5 m - 50 m T - 5. 10- = 50 PS (5 0) 2 = 10-8 m 2 pm a square of 100)M X 100 J91n 4 2 The quantities designated 0 represent the average operating point, which in the case of a variable particle velocity can move "upwards" and "downwards" in fig. 1 (fig. 3).

It follows for the base volume V 2 V - 4.(s 0). h 0 m 2.10 A2 m 3 = 2. 10 6 FM 3 and for the volume flow = VMIT = 4.10- 8 m 3 /S = 4. 10- 5 l/S Another possibility for giving the adaptation limits consists of indicating the margin for the velocity changes. Readily implementable and also realistic is e.g. to allow fluctuations of velocity within a range of 0.5 v 0 ', v 0 < 2v 0 This merely constitutes an example and somewhat modified values could be used without casting doubts on the functionality of the method. Then, for the measuring volume design the following apply:

h min vmin' -c 2.5.10- 5 m = 2 5 jam and h max = V max-c 1.10 m -4 = 100 pm as well as 4.s 2 V/h min max 8.10- 8 m 2 - 8.10 4 p 2 and 4.9 2 min " V1h max = 2.10-8 m 2 = 2. 10 4 PM 2 The cross-section of such a measuring volume is shown in fig. 7a. The physical diaphragms creating this measuring volume by imaging would have the same appearance.

The novel method in particular uses hyperbolic boundary surfaces. However, different arrangements are conceivable. In fig. 7a the hyperbolic boundary surfaces face the incoming particle flow and form the entrance boundary. The base surface of the measuring volume forms the exit boundary (fig. 7a). However, there could be a flow against such a measuring volume in the reverse direction (fig. 7b) and the method criteria are still fulfilled. A third possibility would involve having hyperbolic boundary surfaces both on the entrance and exit sides of the measuring volume (fig. 7c).

AS to which of these arrangements is most favourable will inter alia be decided by the associated signal characteristics, such as the slope steepness in conjunction with signal triggering. On said signal characteristics is e.g. dependent the sharp definition with which the criterion Tx-C, i.e. the lateral measuring volume extension, is controlled. A part is also played by other standpoints, e.g. the coincidence error problem.

Hitherto the method has been considered under the condition that all the particles (independently of their size) move through the measuring volume with the same velocity. A velocity change should always affect all the part icles in the same way. This prerequisite is realistic for the particle size range x 4 10 pm, a range which represents the actual domains of such scattered light particle counters. However, further extending considerations reveal that the present method can also function if the particles of the collective to be investigated do not all move with the same velocity through the measuring volume, i.e. if there is a specific frequency distribution of the particle velocities. A correlation may or may not exist between the size x and the velocity v of the particles (twodimensional frequency distributions of particle size and velocity).

Claims

1. Method for the analysis of particles dispersed in a flowing fluid, the particles being detected in an optically defined measuring volume with a differing cross-section in the flow direction over the height thereof and only particles, whose measured transit time exceeds a predetermined minimum signal duration are accepted for evaluation, characterized in that the lateral surfaces of the measuring volume are not straight.

2. Method according to claim 2, characterized in that the lateral surfaces are constructed so as to continuously concavely change.

3. Method according to claim 1, characterized in that the optically defined measuring volume has hyperbolic lateral surfaces.

4. Method according to claim 1 or 2, characterized in that the optically defined measuring volume is symmetrically constructed.

5. Method according to one of the claims 1 to 4, characterized in that the particles are illuminated and detected by means of diaphragm apertures. which have continuously concavely constructed boundary edges.

6. Method according to one of the claims 2 to 5, characterized in that the particles are illuminated and detected by diaphragm apertures, which have hyperbolically constructed boundary edges.

7. Method according to claim 5 or 6, characterized in that the edges of the diaphragm apertures are symmetrically constructed.

8. Method according to one of the claims 5 to 7, characterized in that the diaphragm apertures in the illuminating and detection paths are identically constructed.

9. Method according to one of the preceding claims, characterized in that with a changing particle velocity and therefore maximum signal duration of the signals brought about by the particles, the minimum signal duration is adapted.

10. Method according to claim 9, characterized in that an adaptation takes place if, in the case of a velocity change, the maximum signal duration of the signals brought about by the particles approach to within 1.1 times the actual (predetermined) minimum signal duration or move away therefrom to within 3.8 times.

11. Apparatus for the analysis of particles dispersed in a flowing fluid, with a detecting device, an illuminating device and an evaluating device, the illuminating and detecting devices being constructed in such a way that in the particle flow, they define a measuring volume with a differing crosssection in the flow direction over the height thereof and in which the evaluating device is so constructed that it only permits further evaluation of those particles, whose measured transit time exceeds a predetermined minimum signal duration, characterized in that the lateral surfaces of the measuring volume are not straight.

12. Apparatus according to claim 11, characterized in that the lateral surfaces of the measuring volume are constructed so as to continuously concavely change.

13. Apparatus according to claim 11, characterized in that the optically defined measuring volume has hyperbolic lateral surfaces.

14. Apparatus according to one of the claims 11 to 13, characterized in that the optically defined measuring volume is constructed symmetrically.

15. Apparatus according to one of the claims 11 to 14, characterized in that the boundary edges (22) of the diaphragm apertures (9, 17) are continuously concavely constructed.

16. Apparatus according to one of the claims 13 to 15, characterized in that the boundary edges (22) of the diaphragm apertures (9, 17) are constructed as hyperbolas.

17. Apparatus according to claim 15 or 16, characterized in that the boundary edges (22) of the diaphragm apertures (9, 17) are symmetrically constructed.

18. Apparatus according to one of the claims 15 to 17, characterized in that the diaphragm apertures (9, 17) of the detection and evaluating devices (6, 13) have an identical construction.

11