1、OR-05-1 7-3 (RP-1190) Development of a New Conditioning Aerosol for Testing Electret Filters James T. Hanley Member ASHRAE ABSTRACT When ASHRAE Standard 52.2-1999 was written, Stan- dard Project Committee 52.2 recognized that the first dust loading step, also called the conditioning step, was not fu
2、lly adequate for revealing the drop in efficiency that electretfilters undergo during actual use. This ispointed out in the standards foreword: “The initial conditioning step of the dust-loading procedure described in this standard may affect the eficiency of thejlter but not as much as would be obs
3、erved in actual service.” To address this issue, and a related issue with high voltage electronic air cleaners, an ASHAE research project was undertaken, 11 90-e “Develop a New Loading Dust and Dust Loading Procedures for the ASRHAE Filter Test Stan- dards 52. I. and 52.2. ” An aerosol-based laborat
4、ory method was developed to reproduce the drop in efficiency that electret filters can undergo with use. The method exposes the filter to a high concentration of solid-phase (ie., dry) submicrometer salt aerosol particles. The approach successfully reveals the drop in efficiency for both low- and hi
5、gh-efficiency electret filters. To address differences seen in an interlaboratory comparison, additional qualification criteria were added to the condition- ing procedure; howevel; the adequacy of these additional criteria was not evaluated on the project. The conditioning procedure may also reduce
6、the measured weight gain of some filters relative to the current 52.2 procedures. A draft adden- dum to ASHME Standard 52.2 was prepared. Incorporating this new conditioning procedure into the 52.2 test standard wouldallow the standard test to better reflect the minimum efi- ciency performance of el
7、ectretfilters that may be experienced during actual use. M. Kathleen Owen INTRODUCTION Under contract with the American Society of Heating, Refiigerating and Air-conditioning Engineers (ASHRAE, Project No. 11 90-RP), RTI conducted a program to develop a new loading dust and dust loading procedures f
8、or ASHRAE filter test standards 52.1 and 52.2 (Hanley and Owen 2003). The objective was to develop a loading dust (or dusts) and procedures that better reflect actual air cleaner performance in ambient air operation. The two principal issues that drove the program were that (I) the current dust ando
9、r dust loading procedures did not adequately reproduce the reduction in filtration efficiency that electret filters undergo in actual use and (2) the current dust andor dust loading methods were incompatible with high voltage, two-stage electrostatic precipitator (ESP) air cleaners. As part of the p
10、rojects litera- ture review, the ESP issues appeared to be non-aerosol related and, thus, the technical effort focused on the electret issue. The need for a new procedure for electret filters is illus- trated in Figure 1 (Hanley et al. 1999). This figure compares the change in efficiency resulting f
11、rom ambient exposure to the change that occurs with the current dustlprocedure. In this and following graphs, the bold arrow points to the initial effi- ciency curve. With ambient exposure, the efficiency of the electret filter dropped with use; the current dust failed to show this. This behavior ha
12、d been reported earlier by Lehtimaki and Heinonen (1993) and by Hanley et al. (1 994). ISOLATION OF AEROSOL EFFECT ON ELECTRET FILTER PERFORMANCE Early in the research project, tests were performed to confirm that it was the collection of aerosol particles by the filter media that led to the drop in
13、 filtration efficiency of elec- J.T. Hanley is a senior research environmental scientist and M.K. Owen is a research chemical engineer in the Center for Aerosol Technology at RTI International, Research Triangle Park, NC. 02005 ASHRAE. 1115 , , , 0.3 m (# rn-) (# cni3) Ratio CNC CNC to OPC Concentra
14、tion 396,000 7.2 55,000 6 1,643 182 339 20,000 dP* initial in. H20 (Pa) Weight Gain MERV+ grams Filter 2 in. (50 mm) electret Lab El 11 E2 41 E3 65 A B 25 52 - 93 66 7 I 158f 12 in. (300 mm) electret A 55 99 l3 I 78 B 91 99 0.712 (192) 77 * “dp” is the differential pressure across the filter (.e., t
15、he filter pressure drop) Based on substituting new conditioning procedure into the 52.2 method. Weight of dust fed Table 5. CNC- and OPC- Measured Aerosol Concentrations and Their Ratio To investigate why the efficiency and pressure drop changes were different between the two labs, additional measur
16、ements were made of the conditioning aerosol. During operation of the conditioning aerosol generator, the concen- tration of particles was measured using both the CNC and the optical particle counter (OPC) ( 0.3 pm). The ratio of these two concentrations was then computed. Results are summa- rized i
17、n Table 5 and show that there was a large difference in the relative number of large particles between the two labs. Why the ratio of small to large particles was different between the labs is presently unknown. On the ETV program, RTI observed that using a 1% salt solution produced results similar
18、to Interteks, i.e., less than full drop and an modest pressure drop increase. While the salt solution is prepared at O. 1%, contaminates in the water used to make the mixture, old salt deposits within the Laskin generator, and contaminants within the compressed air system may effectively increase th
19、e amount of dissolved material in the water. While future interlaboratory testing will be needed to determine the exact cause or causes of the difference between RTI and Intertek, the following criteria are recommended at this time. erator should be configured to minimize drainage of salt deposits b
20、ack into the Laskin reservoir. The compressed air system for the conditioning aerosol generator should be equipped with adequate water and oil traps and a high-efficiency particle filter to remove contaminants from the supply air. An additional qualification test should be established based on the r
21、atio of small (measured by CNC) and large (measured as 0.3 pm) particles. Based on the data available, a ratio 20,000 is recommended. DISCUSSION ON INCORPORATING THE NEW CONDITIONING PROCEDURE INTO 52.2 The Laskin nozzle conditioning process provides a means of accelerating the drop in efficiency th
22、at electret filters may undergo in real-life applications. Incorporating this procedure into ASHRAE 52.2 (and 52.1) will require committee discus- sion and consensus on several issues as well as addressing public review comments. Within the context of that future committee discussion, the following
23、questions, comments, and recommendations are provided. 1. What is the endpoint for the conditioning? It is recommended that the desired endpoint be to reason- ably reproduce the minimum efficiency observed in real- life applications. This does not necessarily mean nuliifi- ing the effect of all elec
24、trostatic charge in the filter (such The Laskin generator should be thoroughly rinsed with distilled or filtered deionized water prior to each test to minimize the accumulation of internal salt deposits. The outlet aerosol flow tube attached to the Laskin gen- ASHRAE Transactions: Symposia 1123 as i
25、s achieved with isopropanol dipping methods), although that does appear to be the endpoint for currently marketed electret filters run under ambient exposure conditions. Thus, taking current electret filters to their minimum is a reasonable approximation of the minimum efficiency that may be experie
26、nced during their actual use. This may or may not hold true for future electret filters. The Laskin nozzle conditioning process allows for less-than-total nullification of the electrostatic charge should that be needed in the future. Should the method allow the conditioning duration to be based on “
27、CT” (the concentration of the aerosol concentration X time (duration) of conditioning)? Yes. Running to a target CT makes sense and is consistent with the underlying theory that particle deposition on fibers reduces the efficacy of the fiber charge to enhance filtration. The rate at which these part
28、icles are delivered to the filter is not expected to be a factor. However, a maxi- mum in-duct concentration of 1.0 x lo6 cmJ is recom- mended to prevent excessive coagulation of the conditioning aerosol. This will allow laboratories some flexibility in designing the capacity of their conditioning a
29、erosol generation system. Should there be an upper limit on conditioning? Yes. Based on the ambient and laboratory exposure results for the 12 in. (300 mm) electret filter, a condition- ing CT of 3.1 x lo8 particles min represents approx- imately three months of full-time use. Taking one year of ful
30、l-time use as a maximum filter service life, it is recom- mended that a maximum level of conditioning be speci- 2. 3. There is also the risk of “overexposing” a short service- life filter to the extent that the efficiency passes though the minimum before conditioning is completed. Should conditionin
31、g be performed incrementally until a minimum efficiency is reached? Yes. To address the concerns raised in question 4 (unnec- essarily long conditioning durations and the risk of filter “overexposure”), it is recommended that conditioning be performed in incremental steps and stopped when two consec
32、utive steps show no further significant drop in effi- ciency. A minimum incremental CT of 3.2 x 1 O7 particles cm-3 min is recommended to ensure that a filter is not underconditioned. Should the conditioning apply to all filters or only to electret filters? It is recommended that the conditioning pr
33、ocedure be applied to all filters. The test lab cannot be certain a filter contains no electrostatic charge that contributes to the filters efficiency. Should the method require measurement of the concen- tration of the challenge aerosol with a CNC or simply rely on the Laskin generator operating co
34、nditions (e.g., air pressure and the number of nozzles operation)? The output of the Laskin nozzle generator cannot be guar- anteed. Nozzles can clog, hoses can become discon- nected, air pressures can drift, fluid levels can drop, air leaks can develop, etc. It is therefore recommended that the use
35、 of a calibrated CNC be required to monitor the challenge aerosol concentration during conditioning. 5. 6. 7. fied as a CT of 1.2 x lo9 particles min (i.e.lfour times the three-month CT). Establishing an upper limit on 8. Should the method require measurement of the she distribution of the condition
36、ing aerosol? conditioning serves two purposes: (a) it provides a stop- ping point in the event that the efficiency of the filter continues to drop and (b) it provides a conditioning endpoint that would represent less than complete nullifi- cation of the electrostatic charge in the event such “long-
37、lasting” electret filters become available. It is anticipated that this maximum CT limit will not be needed for currentlv marketed filters. 4. Should the conditioning CT (concentration x time) be the same for ail filters or should filters with relatively short service lives, typical for 1 and 2 in.
38、(25 and 50 mm) filters, be conditioned to a lesser degree than filters with service lives of one year or longer, typical of 12 in. (300 mm) filters? As was seen in Figures 9 and 1 O, the 2 in. (50 mm) electret filter required only about one-tenth the amount of condi- tioning of the 12 in. (300 mm) d
39、eep electret filter. From a practical point of view, running all filters to a maximum fixed exposure will either take an excessive amount of testing time or require excess output capacity for the conditioning aerosol generation system. While this would be useful to confirm that the proper size distr
40、ibution is achieved, it is not recommended as a requirement. The use of proper water, salt, and air purity should be sufficient. The cost of the instrumentation is substantial and it is oriented to laboratory research. However, if interlaboratory tests show discrepancies, this would be a useful diag
41、nostic tool. Are additional specifications and qualification tests needed? Yes. It is recommended that the following specifications and qualification tests be implemented if this procedure is adopted: 9. Uniformity of the conditioning aerosol (CV 20,000) Water purity (distilled or filtered deionized
42、) Maximum in-duct conditioning aerosol concentra- tion: io6 Salt purity (reagent grade) 1124 ASHRAE Transactions: Symposia 10. How should the possible effect of the new conditioning procedure on weight gain be addressed? If it is desired to keep the weight gains close to those obtained with the curr
43、ent conditioning procedure (ASHRAE 1999), one approach would be to perform the new conditioning procedure and efficiency measure- ments on one filter with a second filter used for weight gain determination that would not include the new condi- tioning procedure. DRAFT ADDENDUM FOR THE 52.2 STANDARD
44、A draft addendum to ASHRAE 52.2 that incorporates the new conditioning procedures was prepared as part of the project objectives. The draft addendum includes the follow- ing: incremental conditioning of the filter until no further decrease in efficiency is observed, generation of the conditioning ae
45、rosol with Laskin noz- zles using 0.1% KCl solution (I gram KC1 per liter water), conditioning of the filter based on CT, a minimum incremental conditioning CT step of 3.2 x io7 particles cm-3 min, a maximum cumulative conditioning CT of 1.2 I O9 par- ticles cm-3 min, which allows stopping condition
46、ing even if the filter is still dropping in efficiency, requires use of a CNC to measure the concentration of the conditioning aerosol, establishes qualification test criteria for uniformity of the conditioning aerosol across the test duct, and specifies the purity of water and potassium chloride to
47、 be used. stage electrostatic precipitator (ESP) air cleaners. The main effort of the project focused on the electret issue. An aerosol-based laboratory method for accelerating the drop-in efficiency that electret filters may undergo in real-life applications was developed. The method exposes the fi
48、lter to a high concentration of solid-phase (i.e., dry) submicrometer salt aerosol particles. The method successfully reveals the drop in efficiency for both low- and high-efficiency electret filters over the full 0.3-10 pm size range (when the filter has a low mechanical efficiency). To address dif
49、ferences seen in an interlaboratory comparison, additional qualification criteria were added to the conditioning procedure. The conditioning procedure may also have reduced the measured weight gain of some filters relative to the current 52.2 procedures. A draft addendum to ASHRAE Standard 52.2 was prepared. Incorporation of this conditioning procedure into the 52.2 and 52.1 test standards will allow the standard tests to better reflect the minimum efficiency performance of electret filters that may be experienced during actual use. REFERENCES ASHRAE. 1999. ANSUASHRAE Standard 52.
