Procedure: The following describes the step-by-step SOP for weighing filters using the OSHA Automated Weighing System. The filter media is 37-mm diameter low-ash polyvinyl chloride (PVC). These preweighed filters are used by the field compliance officers for the collection of respirable and nonrespirable particulate material. These PVC filters in their "flying saucer" type cassette are exclusively recommended for respirable sampling. An abstract of a memorandum that describes their use in the field is attached. (Appendix A) The sealed cassettes with the preweighed and numbered filters are shipped from the SLTC.

1. Only authorized laboratory personnel are allowed in the weighing area. Lock the weighing room after business hours and when not in use. Challenge any non-laboratory worker entering the weighing area without an escort.
   
   
2. The automatic weighing program which utilizes robotics is described in Appendix B. Appendix B outlines the sequence of events the robot performs to weigh the filters. Additionally, the software commands that control the robot are listed. Analysts should refer to Appendix B for details of the robot’s operation.
   
   
3. The filters are weighed in a laboratory environmental controlled room with the temperature being controlled to 21 ± 1ºC (70 ± 2ºF) and relative humidity to 35 ± 5%. Deviations from these parameters must be addressed before the weighing of filters. Consult your supervisor or arrange for repair services to correct the problem.
   
   
4. Strictly comply with good laboratory practices with no smoking or eating in the weighing room or laboratory. Before handling filters or gravimetric samples, work areas should be cleaned with damp towel and adsorbent pads placed on the bench top. Handle all filters with tweezers or a gloved hand. Preweighed filters are not handled with bare hands.
   
   
5. The PVC filter assemblies are purchased from MSA, Inc. and are packaged in a plastic sleeve containing 40 assemblies. The filter assemblies are removed from the sleeves and a bar code is affixed to the aluminum cone. The assemblies are place on a 75-position weighing rack. The filters will remain on this rack for the entire weighing process. Desiccate all filters in a vacuum oven before pre-weighing. The initial desiccation is under a vacuum (-15 inHg) at 60ºC using a dry air bleed for a total of 36 hours. Subsequent desiccations are under a vacuum (-15 inHg) at 40ºC using a dry air bleed for 15 hours. A minimum of two desiccations is required. All assemblies must have two consecutive weights that are ±10 µg of each other.
   
   
6. Change desiccant well before it is depleted (e.g., completely changes color — blue desiccant turns pink). Depending on the local relative humidity, this might be required several times a week or only weekly. A filter is used to clean air metered into the vacuum oven when releasing the vacuum. Change this filter as needed or annually. Release the vacuum to the oven in a controlled and gradual manner.
   
   
7. The pre-weighed assemblies are returned to the plastic sleeves with blue filter spacers separating each assembly and shipped to MSA, Inc. MSA places each PVC filter assembly in a polystyrene cassette. Each cassette is sealed with shrink-wrap tape and flow tested. The cassettes are then returned to the SLTC. Six cassettes are selected at random from the lot of returned cassettes to be analyzed. After going through the desiccation procedure, the difference in pre-weights and post-weights is checked for outliers.
   
   
8. Use the following protocol when receiving and logging in gravimetric filter cassette samples. Use the exclusive bar code number affixed to the filter cassette as the sample submission number. Replace another sample submission number with the bar code number by lining it out and initialing it. Record the changes on the OSHA 91A and include a note on the OSHA 91B analysis report form. Use the following statement in the note. "When requesting gravimetric analysis, please use the affixed filter bar code number as the sample submission number." If the bar code number does not appear on the OSHA 91A, include in a note that "xxx" submission number corresponds to "yyy" bar code. The set of laboratory numbers is attached to the sample cassette for gravimetric analysis. Stamp the numbers on the OSHA 91A as the samples are originally logged in. After the gravimetric analysis has been completed, return the filters in the labeled petri dishes to the proper drawer in the sample room for the additional analysis.
   
   
9. After verifying and recording the condition of seals and number agreement, remove filters from their cassettes and place on a 75-position weighing rack. Desiccate all filters (except oil mist, C_rO₃ or samples with the potential to lose volatile organic compounds) in a vacuum oven before pre- and post- weighing. The initial desiccation is under a vacuum (-15 inHg) at 40ºC using a dry air bleed for a total of 36 hours. Subsequent desiccations are under a vacuum (-15 inHg) at 40ºC using a dry air bleed for 15 hours. A minimum of two desiccations is required. Change desiccant well before it is depleted (e.g., completely changes color — blue desiccant turns pink). Depending on the local relative humidity, this might be required several times a week or only weekly. A filter is used to clean air metered into the vacuum oven when releasing the vacuum. Change this filter as needed or annually. Release the vacuum to the oven in a controlled and gradual manner.
   
   
10. Use an electronic balance capable of weighing to the nearest 0.001 mg. Record the test weights, temperature and humidity in the appropriate notebook daily before using the balance to weigh filters. After weighing the test weights, press SELECT 1 twice on the robot’s balance. The balance is automatically zeroed prior to each weighing. The balance should be checked by a qualified service technician every twelve months or when performance is not acceptable.
    
    
11. Check the calibration of the balance annually with National Institute of Standard and Technology Class S-1.1, ASTM Class 1 weights or equivalent. Use one of the NIST-traceable weights stationed at the balance. Record its weight into a logbook. One or more other weights — 1- to 1000-mg weights, which are also NIST-traceable — are used to check the balance and the measured values are documented in the annual balance audit.
    
    
12. Weigh the filters using the protocols described in Appendix C. Equilibrate filters in the environmental controlled room for six hours prior to each weighing. Use extreme care to avoid spilling loose dust from the cassette assembly.
    
    
13. Weigh all filters at least twice.
     
    1. Differences greater than ±10 µg in the two weights require a third desiccation. If the filter has gained significant weight (e.g., >0.2 mg), it must be reweighed until the weights agree within five percent of the total weight.
        
    2. A weight of <66 µg can be reported as "None Detected, Reporting Limit = 66 micrograms."
14. Analyze (weigh) a set of quality control samples with each analytical set. Quality control samples should be treated in the exact same manner as the compliance samples. Report and obtain quality control results. Acceptable QC results must be within the control limits established before the compliance samples can be reported. Complete a QC Corrective Action Sheet for any outliers and take appropriate steps to remedy the problem.
    
    
15. A detailed description of how to perform the calculations using the GRAV (Excel) software is given in Appendix C. Calculations are based on taking the difference between a pre- and post- weights in milligrams. Subtract a positive weight for the blank from the net weight. Then divide the results by an air volume in cubic meters to obtain a particulate concentration in air. DO NOT add negative weights of the blank to the gravimetric weight. These calculation conventions are included in the GRAV program.
    
    
16. Report the results of the gravimetric analysis using the following IMIS (analyte) codes. Also report any detected blank values. Additional codes can be found in Appendix A. Report gravimetric results using the conventions below and with these two IMIS codes:
     
    G302 Sample Weight and blanks reported in mg (Y)
    
    G301 Gravimetric Analysis reported in mg/m³ (M)
    Report additional results based on the gravimetric determination using the appropriate IMIS codes listed and report in mg/m³ (M):

IMIS	Analyte	P104	Portland Cement
9130	Particles Not Otherwise Regulated (Respirable Fraction <1% Silica)	W103	Wood Dust (General)
		9210	Hardwood Dust
9135	Particles Not Otherwise Regulated (Total Dust)	9211	Softwood Dust
		C120	C120 Coal Dust (>5% Silica)
5010	Oil Mist	9040	9040 Coal Dust (<5% Silica)
1300	Glass Fibrous or Dust	A100	Aluminum Dust

Report additional results based on the determination using other analytical techniques (e.g., XRD, UV-Vis spectrophotometer, AAS). The IMIS codes listed are for the more frequently requested additional analysis:
 

IMIS	Analyte	S103	Nonrespirable Quartz
9010	Respirable Quartz	Z102	Zinc Oxide

Care should be taken to report appropriate IMIS codes based on whether the sample is respirable or nonrespirable. Size determines respirable vs. nonrespirable particles. Respirable dust particles are those that are small enough to enter the gas exchange area of the human lung. For OSHA compliance purposes, these respirable particles are collected using a 10-mm nylon cyclone where the mean size collected is approximately 3.5 µm. For practical purposes in reporting gravimetric results, when the sample has been collected using a cyclone at a sampling rate of 1.7 L/min, the sample is respirable. Respirable PEL standards often include a limitation that the dust does not contain silica in certain percentage. Consult with your team leader as whether to report this respirable PEL without a silica analysis. Below are current conventions for reporting gravimetric samples using the LISA database. These instructions were drafted to help acting team leader’s understand the appropriate reporting conventions. Our new Oracle database (LISA) should fine-tuned in the future to eliminate many of the current shortcomings.

ALL GRAVIMETRIC SAMPLES REQUIRING SUBSEQUENT CRYSTALLINE SILICA ANALYSIS
 
1. All samples should be Ok'd, Dated and Released when the gravimetric analysis is reviewed. The samples should be released in the LISA database. The samples and paperwork should be transferred for crystalline silica analysis immediately after the completion of the supervisor review.
   
   
2. The gravimetric samples associated with crystalline silica samples should be faxed when the crystalline silica samples are released.
ALL OTHER GRAVIMETRIC SAMPLES
 
3. All other samples should be Ok'd, Dated and Released when the gravimetric analysis is reviewed.
   
   
4. If additional analyses are requested (i.e., ICP, AAS), please make sure that a note is added stating that "The samples have been resubmitted for ‘X’ analysis. Results will be reported separately."
   
   
5. If the gravimetric results yield a result less than the PEL for the additional analysis requested, report the additional analysis at the same time the gravimetric samples are released.
   
   
6. These reports should be completed by the analyst who performed the gravimetric analysis. Each IMIS analyte must be reported using the gravimetric results and a "<" symbol. A note should be included on the FM91B stating that: "The gravimetric analysis yielded an air concentration that was less than the PEL. The samples will not be analyzed unless special circumstances exist and the CSHO calls the SLTC."
   
   
7. If the additional analyte was for mist and a bulk was submitted, it should be also reported at this time with the following note: "The gravimetric analysis for the air samples yielded an air concentration that was less than the PEL. The bulk sample will not be analyzed."
   
   
8. If it appears that the CSHO only used the preweighed filters because they were convenient, and a weight was not required, please include a note with the gravimetric analysis that "Preweighed filter should only be used when a gravimetric analysis is required."
17. Confidence limits for the gravimetric analysis are defined by the reporting of a sampling and analytical error (SAE). The sampling or pump error is assumed to be ±5% and the analytical error is derived by analyzing single blind Quality Control samples. A group of 100 data points is necessary to establish the analytical error. The current SAE for gravimetric analysis is 0.083 (as of 3/24/03) and covers the range of 1.0 – 5.0 mg. This gravimetric SAE will be reported for respirable and nonrespirable PNOR (IMIS 9130 and 9135) and other results based solely on a gravimetric determination. The quartz results (IMIS 9010) have a separate SAE, and this number should be referenced for that analysis.

Inlet                   Outlet	New Stainless support (drawing)

MSA cyclone holder with cassette	Bendix holder with cassette

Table 1 PARTIAL LIST OF SUBSTANCES FOR AUTOWEIGHING SUBMISSION
Analyte	IMIS	mg/m³
-Alumina
Total dust	0160	15
Aluminum Metal (as AI)
Total Dust	A100	15
Respirable fraction	A110	5
Ammonium Sulfamate
Total dust	0185	15
Respirable fraction	A111	5
Barium Sulfate	B101	15
Total dust	B104	5
Bismuth Telluride, undoped
Total dust	0370	15
Respirable Fraction	B110	5
Boron Oxide
Total dust	0380	15
Calcium Carbonate
Total dust	0505	5
Respirable fraction	C130	5
Calcium Hydroxide
Total dust		15
Respirable fraction		5
Calcium Oxide	0520	5
Calcium Silicate
Total dust	C112	15
Respirable fraction	C122	5
Calcium Sulfate
Total dust	C104	15
Respirable fraction	C123	5
Carbon Black	0527	3.5
Cellulose
Total dust	0575	15
Respirable fraction	C124	5
Coal dust (<5% Sio2)
Respirable fraction	9040
Coal dust (>5% SiO2)
Respirable fraction	C120
Cotton dust (The SLTC will supply special preweighed cotton dust filters and cassettes on request.)
Dicyclopentadienyl Iron
Total dust	0904	15
Respirable fraction	D100	5
Emery
Total dust	1016	15
Respirable fraction	E102	5
Grain dust
(oat, wheat, barley)	G109	10
Glycerin (mist)
Total dust	1363	15
Respirable fraction	G115	5
Graphite, synthetic
Total dust	1366	15
Respirable fraction	G100	5
Gypsum
Total dust	1367	15
Respirable fraction	G101	5
Kaolin
Total dust	1568	15
Respirable fraction	K100	5
Limestone
Total dust	1593	15
Respirable fraction	L100	5
Magnesite
Total dust	M113	15
Respirable fraction	1615	5
Magnesium Oxide fume
Total particulate	1610	15
Marble
Total dust	1626	15
Respirable fraction	M114	5
Oil Mist(Mineral)	5010	5
Particulates not otherwise regulated (PNOR)
Total dust	9135	15
Respirable fraction	9130	5
Pentaerythritol
Total dust	1987	15
Respirable fraction	P157	5
Perlite
Total dust	2035	15
Respirable fraction	P101	5
Plaster of Paris
Total dust	2127	15
Respirable fraction	P102	5
Portland cement
Total dust	0557	15
Respirable fraction	P104	5
Rouge
Total dust	2229	15
Respirable fraction	R102	5
Silica, amorphous, precipitated and gel	9050
diacotomaceous earth, containing <1% crystalline silica	S112
Silica, crystalline
Cristobalite,
Respirable dust	9015
Quartz
Respirable dust	9010
Tripoli, (as quartz)
Respirable dust	S114
Tridymite,
Respirable dust	9017
Silica, fused
Respirable dust	9013
Silicates (<1% crystalline silica)
Mica,
Respirable dust	9075
Soapstone,
Total dust	9025
Respirable dust	S121
Silicon
Total dust	2235	15
Talc (containing no asbestos)
Respirable dust	9030

1. Double click on the GRAV icon. The GRAV icon is usually located on your desktop on computers using WindowsNT with the Excel client program installed. Also, the GRAV program can be reached with the icon entitled with one of these terms "AA_Weighing" on some computers.
   
   
2. Filter Data. Purpose: Transfers filter weighing data files from the ALPHA to the NTSERVER to enable calculations.
    
   1. Select Weighing Samples button. From the Main Menu of the GRAV program select Filter Data from Rack.
      
      
   2. A dialog box will appear titled "Form 1". The default drive that should appear is entitled \\ROBOT\WEIGHING or \\USD4\WEIGHING. If not, reselect it by clicking on it from the list of choices provided and shown by clicking on the down arrow.
      
      
   3. Select the BIN subdirectory on the \\ROBOT\WEIGHING or \\USD4\WEIGHING drive, then click on one of your raw data files (e.g., Apr27135.1). Click OK.
      
      
   4. Select Used from the dialog box. New is for filters that have been preweighed but not used in the field. Once the data transfer is complete, the Main Menu will reappear.
      
      
   5. If you have more than one data file, repeat steps a-d.
3. From the Main Menu select Sample Range and then New Sample Range in the Project dialog box, up to 100 samples. Type in your sample ranges using common conventions (e.g., R74001–23, R74102, R74111–2, etc.). Enter QC sample ranges with the prefix of Q (e.g., Q45600–3). After completing the range enters click OK. Purpose: Identification of sample sets for weighing calculations.
   
   
4. Weighing Calculations. Purpose: Calculates the weight of material on the filter.
    
   1. Select (double click) Weighing Calculations from the Main Menu, then click on Select.
      
      
   2. The dialog box shown lists the sample number, new weights, and used weights for one sample in the sets identified. Select the new and used weights that you want to use in the sample weight calculations. The program defaults to the lowest weights unless you change them. Click OK to accept your choices. Select or confirm a choice on the subsequent samples.
      
      
   3. When you get to the QC’s a second dialog box will appear titled "Bar." Enter the bar code (e.g., I490) for the QC sample identified in the Select dialog box, then click OK.
5. Generate Report. Purpose: To print a copy of the new and used weights for each sample and QC with the designation of the chosen weights.
    
   1. After the last sample is displayed in 4c above, click on the Report button. A window opens with a text file that lists all weights with your selection for each filter chosen in step 4b.
      
      
   2. Edit the document file using hard returns, when necessary, to make it easier to be read when printed. Change the printer settings in Notepad to have the left and right margins to be 0.25 inches. Print the file to a networked printer (e.g., \\NTSERVER\XRAYHP4). Close Notepad and return to the Select dialog box.
6. Sample calculations. Purpose: Calculates the air concentration for all samples using the sample weight and sample air volume.
    
   1. Click the Excel button in the Select dialog box.
      
      
   2. Two workbooks will open up in the Excel program. Select the one titled "MAINMENU.XLS." Click Here to Run the Weighing Program begins the procedure to perform the calculation. A third workbook will open up titled "WEIGHT.XLS" and seen as a tab at the bottom of the screen.
      
      
   3. Make sure air volumes are correct and fill in "AIR" for QC samples and "ABLNK" for QC blanks. Click Here to Start will open a workbook that contains all the sample information along with the individual chosen weights listed and a net weight calculation.
      
      
   4. Highlight (click and drag the mouse pointer) an individual sample set including the blank from this workbook’s column A. Select Range, then Click Here for Blank Corrections! to perform the air concentration calculation on that set. Repeat for each set of samples.
      
      
   5. To Print and Exit, Click Here! button gives a printout of the air concentration in milligrams per cubic meter along with the net weight found on each filter in milligrams.
      
      
   6. Close Excel.
7. After exiting, the Main Menu will reappear. Select Load Resources to Lisa. Purpose: Transfers the data (both weight and air concentration) to Lisa for inclusion in the sample results record.