Ann. occup. Hyg., Vol. 46, No. 2, pp. 197-207, 2002
© 2002 British Occupational Hygiene Society
Published by Oxford University Press
Article |
Performance Testing of Three Portable, Direct-reading Dust Monitors
Health and Safety Laboratory, Broad Lane, Sheffield S3 7HQ, UK
Received 17 August 2000; ; in final form 29 November 2001;
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Three portable direct-reading dust monitors were tested in a recirculating dust tunnel and a calm air dust chamber against a range of industrial dusts with different size distributions to investigate sources of variation in their responses. Responses were found to be linear compared to reference gravimetric respirable samplers over a range of concentrations for a particular particle size distribution. Their calibration factors were dependant on particle size, particle composition and air velocity. If particle size and air velocity do not change significantly then the calibration factor can be applied to the monitor readings to give an accurate measure of dust concentration. The DataRam and HAM, factory calibrated against respirable dust concentration, were found to agree closely, whereas the Microdust gave higher readings, having been factory calibrated against total suspended particulate concentration. The calibration of the DataRam was significantly altered by either contamination of the optics with dust or by cleaning the optics. This was not observed with either the Microdust or HAM, since both monitors include a reference calibration element.
Keywords: direct-reading dust monitor; respirable dust
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Direct-reading portable dust monitors have been available commercially for many years and they offer numerous advantages over the traditional gravimetric sampler (Baron, 1994). They can provide a rapid, on-site measurement of dust concentration, are simpler and quicker to use and possibly less expensive to operate over an extended period. The ability to provide rapid measurements is particularly useful in assessing dust control measures; by pinpointing major dust producing operations, existing dust control methods could be improved or new ones implemented. Their main drawback is that their calibration can vary significantly depending on the physical properties of the dust being measured. Therefore, to obtain accurate quantitative measurements of concentration, the monitors should be calibrated using standard gravimetric methods with the dust of interest under the same working conditions.
There have been various studies on calibration factors for numerous portable direct-reading monitors. Kuusisto (1983) evaluated six monitors with respect to various commonplace industrial aerosols. Five (all photometers) had a precision of <25%. Gero and Tomb (1988) compared 46 Miniram (MIE Inc., USA) dust monitors when exposed to Arizona road dust, silica dust and limestone dust aerosols. The aerosol concentration measured by different Minirams varied by a factor of two at the extremes. Willeke and Degarmo (1988) compared the response of the Miniram operated actively at different flow rates to its response in passive mode. The concentration measured varied as a function of flow rate into the monitor, indicating an effect of flow rate on the sampling efficiency of the inlet to the sampling chamber and on the particle losses inside the chamber. Chung and Vaughan (1989) tested two respirable dust monitors with a range of dusts. The monitors showed a variable degree of agreement with a horizontal elutriator, influenced by particle size distribution and composition. Tsai et al. (1996) tested the response of three direct-reading dust monitors to different dusts having similar size distributions. They found that the Miniram was accurate within ±10–16% with regard to total dust concentration and that this was independent of dust material. Lehocky and Williams (1996) compared respirable samplers to direct-reading aerosol monitors when measuring coal dust concentrations at coal-fired power stations. Although none of the monitors tested agreed identically with each other or the respirable sampler, it was concluded that they could be used in place of traditional samplers when sampling respirable coal dust.
Direct-reading personal dust monitors are extensively used as part of an exposure video visualization system (Walsh et al., 2000), also known as PIMEX (Rosén, 1993), to identify high levels of dust generated by poor work practice, in the investigation of dust control techniques and to generate hygiene training information. Field trials carried out by the Health and Safety Laboratory, UK (HSL) using the monitor with the visualization system has revealed that its response can vary significantly from one day to the next. Therefore, it was the aim of this study to investigate possible sources of monitor variability and to compare its performance with two similar monitors. The effects of dust concentration, dust composition, particle size, air velocity, monitor orientation and contamination of monitor optics on the response of these monitors were investigated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Monitors tested
MIE DataRam (DataRam)
This is a small, lightweight, personal, forward light-scattering particle detection system which relies on ambient air movement to introduce particles into the sensing chamber, which is an integral part of the instrument and is located underneath a rubber-protected hood through which the dust passes. It gives an optimum sensitivity to particles within the respirable range, determined by the variation of scattered light intensity with particle size, which the manufacturers claim give it a high correlation with standard gravimetric measurements of the respirable fraction. The monitor is auto-ranging and covers a measurement range of 0.001–400 mg/m3. An RS232 digital serial connector is included so that readings can be displayed and recorded on a computer in real time. The operating and logging parameters are programmed via the serial port by a PC using the proprietary software included. The monitor incorporates a datalogger and the logged data can be downloaded to a PC. The datalogging averaging period can be set between 1 s and 4 h. Four samples of the same monitor type were available for testing.
The monitor is periodically zeroed by placing it inside a sealable polythene bag. A hand pump with an in-line filter is used to partially inflate the bag with clean, particle-free air. The DataRam is then zeroed according to the manufacturer’s instructions. A high background reading after zeroing indicates that the optics require cleaning. Unlike the Microdust and HAM (see below), there is no physical method of adjusting the zero or checking the calibration. The DataRam is supplied factory calibrated against SAE J 726 Fine Standard test dust, which conforms to ISO 12103 Pt 1. This is largely comprised of silica and has a size distribution similar to that of Arizona road dust, another commonly used calibration dust.
Casella Microdust (Microdust)
This hand-held monitor also uses the forward light scatter detection principle to determine airborne dust concentration and is most sensitive to particles within the respirable size range. Unlike the DataRam, it uses a detachable cylindrical measurement probe, which contains the infrared light source, optics and detector. This allows readings to be taken in relatively inaccessible areas. Dust enters and leaves through a hole in the side of the probe. Although the monitor is battery powered and portable, it is too large and heavy to be considered a personal sampler. The 0–250 mg/m3 (0.01 mg/m3 resolution) version was tested in this study and included datalogging providing 32 000 data points with a logging interval adjustable from 2 to 600 s. An RS232 serial port is included for downloading data to a PC and subsequent data analysis. Unlike the DataRam, the operating parameters are programmed directly into the monitor itself and not from a PC.
The monitor is supplied factory calibrated against total suspended particulate and uses a reference calibration element to carry out a quick single point recalibration. Zeroing is carried out by attaching a hand pump and in-line filter to the probe. These are used to fill the sensing chamber with particle-free air, after which the display can be adjusted to zero.
HAM
This monitor uses the same operating principle as the previous two instruments and its response is also weighted towards particles in the 0.1–10 µm respirable size range. Like the Microdust, it has a detachable cylindrical measuring probe. It is larger and heavier than the DataRam but smaller and lighter than the Microdust. It has no logging facility, but includes a 0–2 V analogue output which can be connected to the voltage input of a datalogger. The monitor is factory calibrated using Arizona road dust and, like the Microdust, uses a calibration element as a quick single point calibration. The method of zeroing is also identical to that of the Microdust.
Dust samples
Stone dust, wood dust (sieved white pine), medium density fibreboard (MDF) dust and plain white flour were chosen as the test dusts. Realstone Ltd (UK) provided the stone dust: stone was cut using saws and the dust produced was extracted at source, filtered and collected inside a bag. The dust was slightly agglomerated due to the damp conditions and was dried inside an oven and sieved prior to use. The pine dust was bought pre-sieved (WTL International Ltd, UK) and was graded as 180 (sieve aperture 0.09 mm). The plain white flour (McDougals Supersift) was bought from a supermarket. The MDF dust was produced in the laboratory by sanding a length of MDF inside a dust tunnel (see below) using a belt sander fitted with a 60 grade sanding belt. The dust produced was captured by attaching a vacuum cleaner to the dust extraction port on the side of the sander.
Methodology
To study the performance of the direct-reading dust monitors with changing test conditions, they need to be compared with a standard method for determining airborne dust concentration. Since the dust monitors respond primarily to dust of the respirable size fraction, they were compared with Dewell-Higgins cyclone samplers (Health and Safety Executive, 2000). It was assumed that the cyclone samplers gave negligible errors and that this was the ‘true’ measurement of concentration. This is a reasonable assumption to make at low air velocities since the cyclones are usually calibrated in calm air conditions. Sampler bias may be introduced into the results at higher air velocities.
A large recirculating dust tunnel was used to carry out most of the tests since this is capable of producing dust concentrations under steady-state conditions (Blackford and Heighington, 1986). Tests were also carried out inside a calm air dust chamber similar to that used by Blackford and Heighington (1986).
Large recirculating dust tunnel
The tunnel has been described in detail by Blackford and Heighington (1986). Since several samplers and dust monitors were exposed during a test, it was important that measurements were not influenced by concentration or velocity gradients inside the tunnel. Therefore, preliminary measurements of dust concentration and velocity uniformity were made inside the tunnel at the position where the tests were carried out.
The velocity profile was determined by traversing the tunnel cross-section using a calibrated hot wire anemometer (Dantec Electronics Ltd). Exact positioning of the probe was achieved using a grid made from 3 mm diameter galvanized steel with a cell size of 75 x 75 mm and this covered the entire cross-section of the tunnel. The anemometer probe was placed at the centre of each cell and the velocity was measured. A calibrated reference anemometer (TSI Ltd) was also placed centrally inside the tunnel just downstream of the grid so that the velocity profile could be corrected for any temporal changes in velocity, although in practice the velocity inside the tunnel was very stable. A profile was produced at two reference velocities of 0.5 and 2 m/s. The results show that the velocity profile was most uniform above a height of 
900 mm with a maximum variation of ±13% of the mean value at both velocities. All samplers were subsequently placed above this point.
The concentration profile was determined using nine Dewell-Higgins cyclone samplers positioned in the region of uniform velocity. Dust was injected into the air return region of the tunnel at a constant rate and was thoroughly mixed before it reached the measuring zone. The cyclones were loaded with glassfibre filters through which the dusty air was drawn using Rotheroe and Mitchell sampling pumps at a rate of 2.2 l/min. The concentration was calculated from the weight of dust collected by the filter, the duration of the test and the volume flow rate. Measurements were made at tunnel velocities of 0.5 and 2 m/s. The concentration profile was very uniform at both velocities, with a maximum variation of ±3% from the mean value.
Ideally, the dust concentration inside the tunnel should be constant with time. This is determined essentially by the mode in which the tunnel is operated, the constancy of the air velocity through it and the accuracy of the dust feed mechanism. The tunnel has two modes of operation. In the first mode (filtered mode) the dusty air is passed through a bank of filters before it is recirculated, which means that at the point where dust is introduced the air is always clean. If the dust feed system is constant the concentration inside the tunnel should be constant with time, assuming that the air velocity remains the same. However, the drawbacks are that large quantities of dust are used, especially at high concentrations, and the concentration is dependent on the accuracy of the feed. In the second mode (recirculated) the tunnel air bypasses the filtration system so that the dust is recirculated. If the dust feed is constant, the concentration inside the tunnel will increase with time, and so in order to obtain a constant concentration the dust feed needs to be controlled. This method uses less dust and gives better control of the dust concentration and so was used for the majority of the tests.
The dust was metered into the tunnel using a screw feed system (March Systems Ltd). This consists of a hopper into which the dust is placed. The dust settles under gravity onto a screw which rotates and feeds the dust into a venturi dust pump which is powered by compressed air and which disperses and injects the dust into the tunnel. The feed rate and hence the tunnel dust concentration can be varied by changing the size of the screw or adjusting its speed of rotation. A datalogger was used together with a hand-held aerosol monitor (HAM; Process Particulate Monitors Inc., USA) and a relay switching unit to control the dust concentration inside the tunnel. The HAM was placed inside the tunnel return line several metres downstream from the dust injection point and was used to transmit dust concentration readings back to the datalogger. The datalogger was programmed to switch a relay on or off when the HAM concentration reached a set value. By connecting the mains power of the screw feed through this relay, the dust supply was turned on or off, hence controlling the dust concentration inside the tunnel. The datalogger also recorded the measured values and displayed them in real time on a PC either as a bar graph or a graph of variation in concentration with time. Although crude, this method achieved good control.
The samplers were arranged inside the tunnel as shown in Fig. 1. A total of three cyclone samplers was considered adequate because of the uniformity of dust concentration within the sampling region. These were placed close to the monitors. Four samples of the DataRam were tested but not always at the same time, since some were returned for recalibration during the tests owing to severe contamination of the optics and some were in use on other projects. DataRam 1 was chosen as the reference monitor and was present throughout the tests. At the start of testing all of the DataRams, with the exception of DataRam 3, had been newly calibrated by the manufacturer. Each DataRam was fixed upright inside the tunnel with the display facing the airflow. This meant that the dust-laden air would have to hit the DataRam and be directed upwards before passing into the hood containing the sampling chamber of the instrument. The probes for the Microdust and HAM dust monitors were fixed upright with the inlet hole facing the airflow. Each of the cyclones was loaded with 25 mm glassfibre filters which were conditioned and weighed before and after exposure. Air was drawn through these using small self-regulating sampling pumps (model L2 SF; Rotheroe and Mitchell Ltd).
|
A direct-reading aerosol sizer (API; TSI Ltd), placed
2 m upstream from the samplers, was used to measure the size distribution of the airborne dust (mean mass aerodynamic diameter). A dust diluter accessory was used so that the sensor did not become overloaded at high dust concentrations. The density of the dust is required for the API: the density of the stone and flour dust was measured using a density bottle; measurement of the pine dust and MDF dust density was not possible with this technique since this involves immersing the dust in liquid which would cause it to swell or float. Although not ideal, measurements were therefore used from previous work on the same materials (Thorpe and Brown, 1995), based on the physical dimensions and weight of the unsanded materials. The densities were as follows: stone dust, 2.37 g/cm3; flour, 1.39 g/cm3; pine dust, 0.403 g/cm3; MDF dust, 0.578 g/cm3. Several measurements of particle size were made throughout a test and statistics taken. The principal advantage of using the API or any other real-time instrument for measuring particle size is the speed with which results can be obtained. In addition, temporal variations in particle size can also be investigated. However, the instrument has drawbacks, one being the high acceleration forces which act on the particles causing them to reach sonic velocities as they pass through the nozzle. Such non-Stokesian conditions can affect the size response as a function of particle density and shape. Also, the high forces can de-agglomerate particles causing them to appear smaller than they were in the tunnel. These effects should be considered when analysing the API data. Each sample of the DataRam was zeroed prior to each test and programmed to take a measurement of concentration every 5 s. The Microdust and HAM were calibrated, zeroed and adjusted before each test and programmed to record a measurement every 5 s.
Tests were carried out mostly at 0.5 m/s with the tunnel operating in recirculating mode, although some measurements were made with stone dust and pine dust at a tunnel velocity of 2 m/s. The stone dust and flour were also generated with the tunnel in filtered mode in order to increase the particle size. After several tests some samples of the DataRam gave a high background reading when zeroed. When this occurred, the optics were cleaned using canned air as described in the monitor’s operating manual. This often resulted in a significant change in the instrument’s calibration and is discussed later.
Evidence gained from previous on-site measurements revealed that exposure of the DataRam to very high dust concentrations, even for a short duration (several minutes), can contaminate the optics and affect its calibration. To investigate this effect, one sample each of the DataRam and Microdust were tested several times (for a few minutes per test) inside the tunnel at dust concentrations beyond the upper detection limit of the monitors. In between these tests the monitor response was checked at lower concentrations.
Calm air dust chamber
A series of tests was also carried out using a conveniently smaller laboratory bench test chamber, similar to that used by the manufacturers during monitor calibration. The experimental set-up was similar to that used by Chung and Vaughan (1989). It consisted of two enamelled steel and glass chambers of dimensions 1 x 1 x 1 m stacked one on top of the other (Timart Precision, Borehamwood, UK). Laboratory air was drawn into the chamber through a HEPA filter at the top, after which the aerosol was introduced and agitated to disperse it. An aluminium honeycomb sheet was placed at the top to reduce turbulence as the air was drawn into the sampling chamber, with another one at the bottom. There was another HEPA filter at the bottom of the chamber to filter the dusty air before it was vented to the atmosphere. Air flow was provided by the laboratory extraction system and was adjusted using the main laboratory control or using butterfly valves located either side of the bottom filter.
Each chamber housed a circular mesh carousel onto which samplers were placed. They rotated at one revolution every 5.5 min with a reciprocating action. Samplers placed on the same circumference on the carousel should be exposed to the same overall concentration of dust and local airflow conditions. In these tests, the samplers were placed on the lower chamber carousel, equally spaced and on the same circumference. One cyclone sampler was attached close to the inlet on each sample of the DataRam and Microdust so that they were exposed to the same dust concentrations. The test duration depended on the aerosol concentration and was such that the weight of dust collected on the filters was always >500 µg. The dust monitors were programmed to record a reading every 5 s.
For low dust concentrations, a fluidized bed aerosol generator (TSI Inc, USA, supplied by BIRAL, UK) was used to feed dust into the upper chamber. This produced an aerosol which was highly dispersed and de-agglomerated before it entered the chamber. For higher concentrations, the fluidized bed was replaced with the screw feed/venturi system used in the tunnel tests. In both instances the aerosol was passed though a 85Kr radioactive neutralizer prior to entering the chamber. This reduced the electrostatic charge on the aerosol, thereby minimizing losses inside the chamber caused by electrostatic attraction between the dust particles and the chamber walls. The extract flow was adjusted to give a velocity through the test chamber of a few cm/s. The exact velocity profile inside the chamber was unknown, but was probably not critical since the samplers were placed on the carousel.
The samplers were tested using ISO 12103 Pt 1 A2 fine dust, also described as SAE fine dust (Particle Technology Ltd, UK) which, according to their respective operating manuals, is also used by the manufacturers for factory calibration of their instruments.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Recirculating dust tunnel
Calibration factors
The three cyclone sampler concentrations agreed within 5% of each other and so averages of these were used for comparison with the direct-reading dust monitors.
The responses of three samples of the DataRam, Microdust and HAM to dusts of different size distribution and composition with change in respirable dust concentration are shown in Fig. 2a–e. The solid black line represents a 1:1 relationship. Mean calibration factors (slope of the line) for the different monitors and dust types are shown in Table 1. The coefficient of variation (CV) is the ratio of standard deviation and mean value, expressed as a percentage; therefore, the lower the CV of the mean calibration factor, the more linear the response of the monitor. All the monitors show good linearity when measuring respirable stone dust, pine dust and MDF dust (Fig. 2a–d). However, there is much more scatter in the flour results (Fig. 2e).
|
|
Figure 2a shows that the DataRam and HAM agreed closely with the measurements of respirable stone dust with the tunnel operating in recirculating mode. This is probably because the size and density of the airborne stone dust is similar to SAE fine dust and Arizona road dust used by the manufacturers in the calibration of the DataRam and HAM, respectively. The average measured mass mean diameter of the stone dust is 4.03 µm and the average mass mean of the SAE calibration dust is quoted as 2–3 µm. Likewise, the densities of these dusts are similar: 2.37 g/cm3 for stone dust and 2.6–2.65 g/cm3 for the calibration dust. Increasing the mean particle size of the stone dust aerosol from 4.03 to 6.43 µm by operating the tunnel in filtered mode caused the DataRam and HAM to overestimate the respirable dust concentration by 10–20%.
MDF dust, with a mass mean particle size of 10.4 µm, was overestimated by the DataRam by a factor of 1.5. Pine wood dust, with a mass mean diameter of 16.1 µm, was overestimated by the DataRam and HAM by a factor of 2–2.5. The DataRam also overestimated the respirable concentration of flour dust, but there was much more scatter in the results, as shown in Fig. 2e and indicated by a high CV in the average calibration factor results shown in Table 1. This is a result of the large variation in particle size of the flour dust aerosol, which generally increased as the concentration increased, probably because the dust was not dispersed as well at the higher concentrations. The variation of the calibration factor (ordinate) with particle size for flour dust is shown in Fig. 3 and it is clear that there is an increase in calibration factor with particle size.
|
Table 1 also shows that there is less variation in particle size for stone dust, pine dust and MDF dust, indicated by a lower coefficient of variation in mean particle size. The particle size dependency of the direct-reading monitors can be explained by comparing the particle size-dependent light scattering characteristics of the dust monitor with standard sampling conventions. The standard BS EN 481 (British Standard Institution, 1993) shows that when the dust particle size exceeds
12 µm, respirable samplers (cyclones) cut-off and stop sampling, if they correctly follow the convention. However, light scattering monitors such as the DataRam, Microdust and HAM continue to detect particles up to 30 µm. Therefore, as particle size increases, light scattering monitors increasingly overestimate the true respirable dust concentration.
The Microdust consistently read higher than the DataRam for all dust types, overestimating the measurements of respirable stone dust by a factor of 2.8, MDF dust by 5 and pine dust by 6.8. This is because the DataRam was calibrated against the respirable concentration while the Microdust was calibrated against the concentration of total suspended particulate, which approximates to the inhalable concentration.
Air velocity
Changing the air velocity through the tunnel from 0.5 to 2.0 m/s had little or no effect on the ratio of monitor response to cyclone concentration for the DataRam when tested with stone dust. However, when tested with pine dust, the ratio of monitor response to cyclone concentration for both the DataRam and Microdust decreased as the velocity increased, as shown in Table 2. The cause of this is not clear, although it may be that since the monitors rely on air movement to introduce the aerosol into the sensing zone, the higher inertial properties of the larger pine dust particles at the higher velocity may have caused a greater proportion of them to impact out before entering the sensing zone. Also, the performance of the cyclone sampler at the higher tunnel velocity may give rise to a bias in the results which may also depend on particle size.
|
Very high dust concentrations
Anomalous readings obtained with the DataRam have been excluded from the tables of results and the graphs. These readings usually occurred after the DataRam had been subjected to very high concentrations of dust or after the optics had been cleaned in the laboratory. During the tests, DataRam 4 was used for field work and was exposed to concentrations of dust which were beyond the instrument’s limit of detection. On return there was a high background reading when zeroed, an indication that the optics had become contaminated. When re-tested inside the tunnel using stone dust, the calibration factor had increased by
25%. Tests carried out inside the tunnel at very high dust concentrations gave a similar result. After exposing the Microdust and DataRam 4 to five consecutive short bursts of very high stone dust concentrations and one high concentration of pine dust, the DataRam gave a high background reading when zeroed. When tested at a lower concentration of stone dust, the calibration factor was unchanged at a concentration of 8 mg/m3, but had increased by 90% at a concentration of 1 mg/m3. Whenever the DataRam gave a high background reading, the optics were cleaned according to the operating manual, using moisture-free canned air. This was directed close to the surface of the optics using a flexible nozzle. After cleaning the DataRam always indicated that it was now operating correctly. However, the DataRam’s calibration factor could be as much as 65% lower than the original value. The reason for this is not clear, although it is of some concern.
Long-term exposure to dust
The effects of long-term exposure to lower dust concentrations on contamination of the DataRam optics (assumed to be when the monitor gave a high background reading) was investigated. This was achieved by analysing the running average respirable and inhalable concentrations, calculated from all the cyclone sampler measurements made during this study and the cumulative exposure time. It was found that newly calibrated DataRams 1 and 2 were exposed to approximately the same concentration of dust and for a similar period before their optics became contaminated (65–76 h at 5.3–5.7 mg/m3 respirable). After cleaning, the optics of DataRam 2 became recontaminated after only 16 h exposure at 3.4 mg/m3 respirable. This indicates that cleaning the optics with compressed air is only partly effective. DataRam 3, whose exposure history was unknown at the start of testing, required cleaning after only 20 h at 5.8 mg/m3 respirable. It should be pointed out, however, that these figures are only rough guides, since samples of the DataRam were exposed to various dust types.
Throughout the tests, the Microdust and HAM did not exhibit the same calibration problems as the DataRam. The optics still became contaminated, indicated by an inability to zero the monitor, but blowing clean air into the probe using the supplied hand pump or in some instances using canned air always returned the zero. Also, the use of a reference calibration element and adjustment of monitor gain meant that calibration was always maintained at the factory set value.
Calm air exposure chamber
The concentrations inside the calm air dust chamber measured by the four cyclone samplers were all within 3% of the mean for each test, which shows that the cyclone samplers and dust monitors placed on the same circumference of the carousel were exposed to the same concentration of dust. Figure 4 shows the effects of DataRam orientation on measurements of temporal change in dust concentration and measurement of the calibration factor. All samples of the DataRam were exposed at the same time and so any differences in response were due to the sampling characteristics of the monitor alone. Placing the DataRam on its side results in the most stable measurement of dust concentration and the highest calibration factor. This indicates that the DataRam sampled more effectively and resulted in smaller errors when placed on its side inside the enclosure. Therefore, all further measurements were made with the DataRam in this position. The observed orientation dependence of the DataRam would probably also apply to the Microdust and HAM monitors. This is because all three instruments rely completely on localized air movement to introduce dust into their sensing zone. Since air movement through the calm air chamber is so low, positioning of the inlet becomes critical.
|
Table 3 shows that the response of all the dust monitors to SAE Fine Standard calibration dust was linear over the range of concentrations measured and a linear fit to the data gave a correlation coefficient close to unity in each case. The slopes of the graphs were close to unity for each DataRam (when placed on their side on the turntable), which is encouraging since they had been calibrated at the factory using the same dust. The response of the Microdust was about twice that of the DataRam, which was a little lower than, but otherwise consistent with, the measurements made inside the dust tunnel. These results confirm the differences in calibration for the Microdust, compared to the DataRam and HAM, observed using the large, recirculating tunnel.
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The response of each monitor was found to be linear with respirable dust concentration when the size distribution of the test dust remained constant, which was largely the case with stone dust, pine wood dust and MDF dust. However, monitor response was found to increase as the mass mean particle size of the airborne dust increased. This was most apparent with flour dust, where particle size tended to increase with increasing concentration. Closest agreement between the DataRam and HAM and respirable dust concentration was for stone dust, which is most similar to the calibration dust (SAE Fine Standard).
Tests carried out inside the dust tunnel and the calm air dust box showed that the Microdust consistently read higher than the DataRam and HAM dust monitors by a factor of 2–3 for all dust types, including SAE Fine Standard calibration dust. This is because the monitor was calibrated against total suspended particulate and not the respirable concentration.
The effect of air velocity on the response of the dust monitors was found to depend on the type of dust being measured. Increasing the velocity had little effect for stone dust, but when tested with pine dust the calibration factor of the DataRam and Microdust decreased as the tunnel velocity increased, even though the particle size remained unchanged.
Calibration of the DataRam was found to be susceptible to contamination and subsequent cleaning of the optics. Contamination of the optics with dust often resulted in an increase in monitor response, whereas cleaning the optics using compressed air often resulted in a decrease in monitor response. Since the DataRam has no method for checking the calibration, there is no way of knowing to what extent the calibration has been affected. This was not the case with the Microdust and HAM monitors, since they both use a reference calibration element to check and adjust the calibration prior to making a measurement. Examination of all the results of this study has revealed that the DataRam can be used for 70 h at an average respirable dust concentration of 5.5 mg/m3 before the optics become contaminated with dust. After cleaning with compressed air the optics became recontaminated more quickly, indicating that this method of cleaning is not completely effective.
Acknowledgements—We would like to thank Dr M. Piney and Mr A. Griffin of the Field Operations Directorate (FOD) of the Health and Safety Executive (HSE) and Mr P. Roberts of HSL for their contributions.
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Baron PA. (1994) Direct-reading instruments for aerosols, a review. Analyst; 119: 35–40.[Medline]
Blackford DB, Heighington K. (1986) The design of an aerosol test tunnel for occupational hygiene investigations. Atmos Environ; 20: 1605–13.
British Standards Institution. (1993) Workplace atmospheres—size fraction definitions for measurement of airborne particles, BS EN 481. London: British Standards Institution.
Chung KYK, Vaughan NP. (1989) Comparative laboratory trials of two portable direct-reading dust monitors. Ann Occup Hyg; 33: 591–606.
Gero AJ, Tomb TF. (1988) Miniram calibration differences. Appl Ind Hyg; 3: 110–4.
Health and Safety Executive. (2000) Methods for the determination of hazardous substances. General methods for sampling and gravimetric analysis of respirable and inhalable dust, MDHS 14/3. Sudbury, UK: HSE Books.
Kuusisto P. (1983) Evaluation of the direct reading instruments for the measurement of aerosols. Am Ind Hyg Assoc J; 44: 863–74.
Lehocky H, Williams PL. (1996) Comparison of respirable samplers to direct-reading real-time aerosol monitors for measuring coal dust. Am Ind Hyg Assoc J; 57: 1013–8.
Rosén G. (1993) PIMEX®: combined use of air sampling instruments and video filming: experience and results during six years of use. Appl Occup Environ Hyg; 8: 344–7.
Thorpe A, Brown RC. (1995) Factors influencing the production of dust during the hand sanding of wood. Am Ind Hyg Assoc J; 56: 236–42.[Web of Science]
Tsai CJ, Shih TS, Lin JD. (1996) Laboratory testing of three direct reading dust monitors. Am Ind Hyg Assoc J; 57: 557–63.
Walsh PT, Clark RDR, Flaherty S, Gentry SJ. (2000) Computer-aided video exposure monitoring. Appl Occup Environ Hyg; 15: 48–56.[Medline]
Willeke K, Degarmo SJ. (1988) Passive versus active aerosol monitoring. Appl Ind Hyg; 3: 263–6.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Meijster, E. Tielemans, J. Schinkel, and D. Heederik Evaluation of Peak Exposures in the Dutch Flour Processing Industry: Implications for Intervention Strategies Ann. Hyg., October 1, 2008; 52(7): 587 - 596. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Schneider and K. A. Jensen Combined Single-Drop and Rotating Drum Dustiness Test of Fine to Nanosize Powders Using a Small Drum Ann. Hyg., January 1, 2008; 52(1): 23 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Thorpe and P. T. Walsh Comparison of Portable, Real-Time Dust Monitors Sampling Actively, with Size-Selective Adaptors, and Passively Ann. Hyg., November 1, 2007; 51(8): 679 - 691. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. ROSEN, I-M. ANDERSSON, P. T. WALSH, R. D. R. CLARK, A. SAAMANEN, K. HEINONEN, H. RIIPINEN, and R. PAAKKONEN A Review of Video Exposure Monitoring as an Occupational Hygiene Tool Ann. Hyg., April 1, 2005; 49(3): 201 - 217. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. EDMAN, H. LOFSTEDT, P. BERG, K. ERIKSSON, S. AXELSSON, I. BRYNGELSSON, and C. FEDELI Exposure Assessment to {alpha}- and {beta}-Pinene, {Delta}3-Carene and Wood Dust in Industrial Production of Wood Pellets Ann. Hyg., April 1, 2003; 47(3): 219 - 226. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||