[Federal Register Volume 61, Number 193 (Thursday, October 3, 1996)]
[Notices]
[Pages 51708-51712]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 96-25314]


-----------------------------------------------------------------------


DEPARTMENT OF HEALTH AND HUMAN SERVICES

National Institute for Occupational Safety and Health; Draft 
Document ``Engineering Control Guidelines for Hot Mix Asphalt Pavers''

AGENCY: National Institute for Occupational Safety and Health (NIOSH), 
Centers for Disease Control

[[Page 51709]]

and Prevention (CDC), Department of Health and Human Services.

ACTION: Request for comments.

-----------------------------------------------------------------------

SUMMARY: NIOSH is seeking public comments on the draft document 
``Engineering Control Guidelines for Hot Mix Asphalt Pavers'' provided 
in this announcement.

DATES: Written comments to this notice should be submitted to Diane 
Manning, NIOSH Docket Office, 4676 Columbia Parkway, Mailstop C-34, 
Cincinnati, Ohio 45226. Comments must be received on or before November 
4, 1996.
    Comments may also be submitted by email to: dmm2@ 
NIOSDT1.em.cdc.gov. as WordPerfect 5.0, 5.1/5.2, 6.0/6.1, or ASCII 
files.

FOR FURTHER INFORMATION CONTACT: Technical information may be obtained 
from Joann Wess or Ralph Zumwalde, NIOSH, CDC, 4676 Columbia Parkway, 
M/S C-32, Cincinnati, Ohio 45226, telephone (513) 533-8319.

SUPPLEMENTARY INFORMATION: The following is the complete text of the 
draft document for public comment ``Engineering Control Guidelines for 
Hot Mix Asphalt Pavers''.

Background

    On July 8-9, 1996, NIOSH convened a public meeting in Cincinnati, 
Ohio, to discuss the scientific and technical issues relevant to the 
development of recommendations for controlling exposures to asphalt 
fume during asphalt paving operations. Representatives from labor, 
industry, and government knowledgeable of current control technologies 
for asphalt exposure met to discuss the types of remedial action (e.g., 
engineering controls, work practices) needed to reduce worker 
exposures. Participants at this meeting included representatives from 
the Asphalt Institute (AI), the Federal Highway Administration (FHWA), 
the International Union of Operating Engineers, the Laborers' Health 
and Safety Fund, the National Asphalt Pavement Association (NAPA), the 
Occupational Safety and Health Administration (OSHA), manufacturers of 
hot mix asphalt (HMA) pavers, and asphalt paving contractors.
    The participants provided detailed information on state-of-the-art 
engineering controls currently in use and discussed a draft document on 
guidelines for engineering controls in the asphalt paving industry that 
was prepared jointly by labor and industry. The draft guidelines 
provided in this announcement represent a recommendation of the 
participants to minimize asphalt fume exposures by developing and 
installing engineering controls on asphalt pavers and by providing 
training for workers.

Purpose

    The purpose of the document ``Engineering Control Guidelines for 
Hot Mix Asphalt Pavers'' is to provide information on the use of 
engineering controls for the reduction of worker exposure to asphalt 
fumes during hot mix asphalt (HMA) paving operations. NIOSH is 
soliciting public comments on the completeness and feasibility of the 
recommendations.

Document for Comment

Engineering Control Guidelines for Hot Mix Asphalt Pavers

    On July 8-9, 1996, the National Institute for Occupational Safety 
and Health (NIOSH) convened a public meeting in Cincinnati, Ohio, to 
discuss recommendations for controlling exposures to asphalt fume 
during asphalt paving operations. Participants at the meeting included 
representatives from the Asphalt Institute (AI), the Federal Highway 
Administration (FHWA), the International Union of Operating Engineers, 
the Laborers' Health and Safety Fund, the National Asphalt Pavement 
Association (NAPA), the Occupational Safety and Health Administration 
(OSHA), manufacturers of hot mix asphalt (HMA) pavers, and asphalt 
paving contractors. This meeting culminated in the development of these 
guidelines that provide information about the use of engineering 
controls (i.e., local exhaust ventilation systems) to reduce worker 
exposures to asphalt fumes during HMA paving operations.
1. New HMA Pavers
    a. Paver manufacturers should develop and install ventilation 
systems with controlled indoor-capture efficiencies of at least 80% (as 
determined by the tracer gas method described in Appendix A) on the 
following equipment:
     All new self-propelled HMA pavers weighing 16,000 pounds 
or more and manufactured after July 1, 1997.
     All new self-propelled HMA pavers weighing less than 
16,000 pounds with slat conveyors or with augers detached from the 
hoppers and manufactured after July 1, 1998.
    b. Paver manufacturers should test the ventilation systems for all 
HMA paver models and certify that the systems meet the minimum capture 
efficiency of 80% as specified in Section 1.a. To assure performance of 
the ventilation systems, manufacturers should install an indicator 
device on each HMA paver to monitor the system flow rate. Each HMA 
paver manufacturer shall provide a plate attached to the paver that 
shows:
      schematic of the ventilation system;
      acceptable operating range for the indicator device; and
      list of operator maintenance procedures.

    c. Manufacturers should develop and implement quality control plans 
to ensure that ventilation systems on these models comply with the 
minimum capture efficiency specified in Section 1.a.
2. Existing HMA Pavers
    By July 1, 1998, paver manufacturers should make retrofit packages 
available for all self-propelled HMA pavers weighing 16,000 pounds or 
more and manufactured after July 1, 1987. These retrofit packages 
should be installed by July 1, 1999. Retrofit packages for self-
propelled pavers weighing less than 16,000 pounds with slat conveyors 
or augers detached from the hopper should be available by July 1, 1999, 
and should be installed by July 1, 2000. Manufacturers should test and 
certify that all retrofit packages installed according to the 
manufacturer's instructions meet the minimum capture efficiency of 80% 
for a specific model or equivalent class design configuration (as 
determined by the tracer gas method described in Appendix A). To assure 
system performance, manufacturers should include in the retrofit 
package an indicator device to monitor the system flow rate.
3. Inspection and Maintenance
    Owners of HMA pavers with ventilation systems should inspect and 
maintain the systems according to the manufacturer's recommendations. 
Each manufacturer should provide an operator manual containing detailed 
sketches and performance criteria for contractors to use in their 
annual assessment of ventilation systems. Annual performance 
inspections should be recorded in the operator's manual.
4. Training Program
    The National Asphalt Pavement Association (NAPA), unions, and 
equipment manufacturers should develop specific training criteria/
materials (i.e., separate document) on the operation, maintenance, and 
repair of HMA pavers.
5. Glossary of Terms
    Asphalt Paver: A self-propelled construction machine (either 
rubber-tired or crawler-mounted) specifically designed to receive, 
convey, distribute,

[[Page 51710]]

profile, and compact paving material by the free-floating screed 
method.
    Auger: A screw conveyor used to transversely distribute paving 
material ahead of the screed.
    Automatic Feeder Control: A system for automatically controlling 
the flow of paving material to the screed.
    Conveyor: A device for transferring paving material from the hopper 
to the auger.
    Conveyor Flow Gate: A device for regulating the height of paving 
material being transferred by the conveyor.
    Feeder System: The combined conveyor and auger components which 
transfer paving material from the hopper and distribute it in front of 
the screed.
    Hopper: That section of the paver which receives the paving 
material from an external source.
    Material Feed Sensor: A device used to detect a quantity of paving 
material in front of the screed.
    Operator: The person whose primary function is to control the 
paver's speed and direction.
    Screed: The device which is towed behind the tractor to strike off, 
compact, contour, and smooth the paving material.
    Screed Arm: The attachment by which the screed is connected to and 
towed by the tractor.
    Screed End Plate: A vertically adjustable plate at the outboard end 
of the screed, to retain the paving material and form the edge of the 
mat.
    Screed Extension: A fixed or adjustable attachment to the screed 
for paving at widths greater than the main screed.
    Tow Point: The point at which the screed arm is attached to the 
tractor.
    Tractor: That portion of a paver which provides propulsion and may 
also receive, convey, and distribute paving material.
    Tunnel: The passageway through which paving material moves from the 
hopper to the auger/screed.

Appendix A

Laboratory/Factory Test Procedure

    Engineering controls (i.e., ventilation systems) for HMA pavers 
will be evaluated in a laboratory setting (i.e., manufacturers' plant, 
shop, or warehouse) in which ventilation control efficiency will be 
measured using smoke and tracer gas tests. The smoke test will be used 
as a qualitative test to visualize airflow patterns around the paver 
and ventilation system, and to ensure appropriate testing conditions 
for conducting the tracer gas tests. The tracer gas test will be used 
to quantitatively measure the volumetric airflow rate and capture 
efficiency of the ventilation system.
    Ventilation systems will be evaluated in a large bay area at the 
manufacturing plant or testing shop. The paver will be parked with the 
screed and rear half of the tractor positioned in the bay area 
(referred to as the testing area) and with the front half of the 
tractor and engine exhaust ducts positioned outside the building. An 
overhead garage door or other barrier can be used to separate the two 
areas. A garage door can be lowered to rest on top of the tractor, and 
the remaining doorway openings around the tractor can be sealed to 
isolate the paver's front and rear halves. The screed will rest on the 
ground with edger plates extended one foot on each side of the screed. 
The flow gates at the back of the hopper should be closed as far as 
possible and the remaining tunnel opening should be blocked off. During 
the performance evaluations the idle speed for the paver, which can 
affect the exhaust rate of the ventilation system, will be set near the 
typical revolutions per minute (rpm) that are maintained during normal 
paving operations.
Safety
    Following are safety precautions for each test:

     Handle smoke generating equipment that can be hot with 
appropriate caution.
     Make sure that the smoke generators do not set off fire 
sprinklers or create a false alarm.
     Avoid direct inhalation of smoke from the smoke generators 
because the smoke may act as an irritant.
     Transport, handle, and store all compressed gas cylinders 
in accordance with the safety recommendations of the Compressed Gas 
Association.
     Store the compressed cylinder outdoors or in a well-
ventilated area.
     Stand back and let the tank pressure come to equilibrium 
with the ambient environment if a regulator malfunctions or some other 
major accidental release occurs.
Smoke Test
    A smoke generator is used to produce theatrical smoke as a 
surrogate contaminant. The smoke is released through a perforated 
distribution tube traversing the width of the auger area between the 
tractor and the screed and supported above the ground under the augers. 
The smoke test helps to identify failures in the integrity of the 
barrier separating the front and rear portions of the paver. After 
sealing leaks within this barrier, smoke is again released to verify 
the integrity of the barrier system, to identify airflow patterns 
within the test area, and to visually observe the ventilation system's 
performance.
    The sequence of a typical smoke test is outlined below:

     Position paving equipment within door opening and lower 
the overhead door.
     Seal the remaining door opening around the tractor.
     Place the smoke distribution tube directly underneath the 
auger.
     Connect the smoke generator to the distribution tube (PVC 
pipe, 2-inch diameter, 10 feet long, capped on one end, \1/4\-inch 
diameter holes every 6 inches on-center).
     Activate video camera if a record is desired.
     Activate the ventilation system and the smoke generator.
     Inspect the separating barrier for integrity failures and 
correct as required.
     Inspect the ventilation system for unintended leaks.
     De-activate the ventilation system for comparison 
purposes.
     De-activate the smoke generator and wait for smoke levels 
to subside.
     Disassemble test equipment.
Tracer Gas
    The tracer gas test is designed to: (1) calculate the total 
volumetric exhaust flow of the prototype design; and (2) evaluate the 
effectiveness in capturing and controlling a surrogate contaminant 
under the ``controlled'' indoor conditions. Sulfur hexafluoride 
(SF6) will be used as the surrogate contaminant. A real-time 
SF6 detector should be calibrated in the laboratory prior to the 
test. There are several methods for calibrating the SF6 detector. 
The least labor-intensive method requires the use of multiple 
compressed gas cylinders with known concentrations of SF6. The 
SF6 concentrations should include at least four concentrations 
ranging from zero to 50 ppm SF6 in nitrogen. An industrial hygiene 
sampling bag such as a 12-liter Milar bag can be filled from 
each cylinder, then the bag can be hooked to the detector, and the 
response of the detector can be recorded for each concentration.
    Another method for calibrating the SF6 detector requires the 
use of two compressed gas cylinders, one with pure nitrogen and another 
with 50 ppm (or higher concentration) SF6 in nitrogen. Four 
different concentrations of SF6 are made by mixing different 
volumes of fluid from the two cylinders into sampling bags. The bags 
are mixed, hooked to the detector, and the response of the detector is 
recorded for each concentration.

[[Page 51711]]

    The sampling bags should be clearly marked with the appropriate 
concentration of SF6 that each contains. Bags can be reused; 
however, they should be emptied prior to reuse and they should only be 
filled with approximately the same concentration of SF6. A bag 
used to hold 50 ppm SF6 in a previous test should not be used to 
hold the 2 ppm SF6 sample in the next test because of the 
possibility of residual gas causing an incorrect calibration point.
    Using either calibration method or an equivalent method, a 
calibration curve, not necessarily a straight line, can then be 
calculated to fit the data and convert the detector's response to an 
actual SF6 concentration.
    To increase the likelihood of independence for each SF6 
concentration reading, program the SF6 detector to a minimum 
sampling interval of 30 seconds. Larger intervals may be required based 
on the model of SF6 detector and the experimental setup.
    100% Capture (to quantify exhaust volume): A known volumetric flow 
rate (0.90 liters per minute) of SF6 is released into the 
ventilation system. The release point must be upstream of the 
ventilation system's fan and downstream of the ventilation system's 
hood to ensure 100% capture of the released gas. The supply tank of 
pure SF6 is connected to the release point via a pressure 
regulator, flow controller, and \1/4\-inch tubing.
    A \1/4\-inch diameter hole is placed in the ventilation system's 
exhaust duct half way between the fan and the outlet of the exhaust 
dust. A 12-inch long and \1/4\-inch outside diameter stainless steel 
tube (sampling probe) is inserted into this exhaust-duct hole 
perpendicular to the exhaust air flow. The sampling probe should be 
sealed at the end and have several \1/8\-inch diameter holes, one inch 
on-center along one side. The number of holes depends on the diameter 
of the exhaust duct. An 8-inch exhaust duct would require use of a 
sampling probe with six \1/8\-inch holes. These holes should be 
positioned perpendicular to the exhaust air flow and must all be inside 
the duct when sampling. The tubing connecting the sampling probe to the 
detector should be airtight to ensure that the sample is pulled from 
within the exhaust duct and not from the surrounding area. The exhaust 
volume is then calculated using the following equation:

where
    Q(exh)=volume of air exhausted through the ventilation system (lpm 
or cfm) (To convert from liters per minute (lpm) to cubic feet per 
minute (cfm), divide lpm by 28.3.)
    Q(SF6)=volume of SF6 (lpm or cfm) introduced into the 
system
    C*(SF6)=Concentration of SF6 (parts per million) detected 
in exhaust and the * indicates 100% capture of the released SF6.

    If there is more than one ventilation system exhaust duct, then the 
above procedure should be repeated for each. Sufficient time should be 
allowed between tests for the background readings to drop to below 0.2 
ppm SF6. Background readings must be subtracted from the detector 
response before calculating the exhaust volume.
    To quantify capture efficiency, SF6 is released through a 
distribution plenum located under the augers between the tractor and 
the screed. A discharge hose feeds pure SF6 at a flow rate of 0.90 
lpm from the pressure regulator, through a mass flow controller 
(precision rotameter), and into the distribution plenum. Accuracy of 
the flow controller will greatly affect the accuracy of the test and 
should be #3% or better. The plenum is ten feet long and is designed to 
release the SF6 evenly throughout its length. The same multi-port 
sampling wand, sampling location, and detector, as used in the 100% 
capture test, is also used in this test.
    At least five consecutive measurements will be taken and an average 
value will be calculated. If the SF6 volumetric flow rate is the 
same for both the 100% capture test and capture efficiency test, then 
the capture efficiency is calculated using the following equation:

where
    =capture efficiency
    C(SF6)=Concentration of SF6 (parts per million) detected 
in exhaust
    C*(SF6)=Concentration of SF6 from 100% capture test

    If the SF6 volumetric flow rate is not the same for both the 
100% capture test and the capture efficiency test, then the capture 
efficiency is calculated using the following:

where C(SF6) and Q(SF6) refer to the values obtained during 
the capture efficiency test and Q(exh) was calculated from the 100% 
capture test.
    A total of four pairs of the 100% capture tests and capture 
efficiency tests will be performed with the ventilation system's 
overall capture efficiency determined from the average of all four 
trials.
    Between each test (after a pair of 100% capture test and capture 
efficiency test), the paver should be shut down and background SF6 
measurements should be monitored to determine if any SF6 had 
accumulated in the test area. If SF6 has accumulated (>2.0 ppm), 
the integrity of the barrier system should be checked and the test area 
should be well ventilated before proceeding. Sufficient time should be 
allowed between tests for the background readings to drop to below 0.2 
ppm of SF6. Background readings must be subtracted from the 
detector response before calculations are made.
    The sequence for a typical test run is outlined below:
     Position paving equipment and seal openings as outlined 
above.
     Calibrate (outdoors) flow meters at approximately 0.9 lpm 
of SF6.
     Drill an access hole in the ventilation system's exhaust 
duct for insertion of the detector's sampling probe and position the 
sampling probe into the exhaust duct.
     With the ventilation system activated, begin monitoring 
for SF6 to determine background interference levels.
     While maintaining the SF6 tanks outdoors or in a 
well-ventilated area, run the discharge tubing from the mass flow meter 
to well within the exhaust hood to create 100% capture conditions.
     Initiate flow of SF6 through the flow meter and allow 
it to reach steady-state (should take only a minute).
     Continue monitoring until 5 readings are recorded.
     Deactivate the flow of SF6.
     Remove the discharge tubing to an outdoor location.
     End the 100% capture test. (Leave the tractor engine 
running.)
     Initiate monitoring to establish background interference 
until levels drop to <0.2 ppm.
     Locate an SF6 distribution plenum under the auger 
area and connect the discharge tubing of the flow meter.
     Initiate SF6 flow through the mass flow meters and 
monitor until approximate steady-state conditions appear (about one 
minute) and take at least 5 readings.
     Discontinue SF6 flow and quickly remove the 
distribution plenum and discharge tubing from the auger area and remove 
to an outside location.
     Continue monitoring to determine the general area 
concentration of SF6 which escaped into the test area.
     Discontinue monitoring when concentration decay is 
complete.
     Turn off the ventilation system and paver engine; 
calculate the capture efficiency.
     Repeat four times.

Example Test Run and Calculations

    The paver was positioned and smoke was used to visually test the 
system.

[[Page 51712]]

Smoke was seen coming in the top of the overhead door. The opening in 
the overhead door was sealed and the smoke test revealed no other 
problems.
    For simplicity of example, the SF6 detector was calibrated and 
adjusted to read directly SF6 in ppms. The SF6 flow meter was 
calibrated using a bubble meter.

------------------------------------------------------------------------
                                                              Flow rate,
                         Trial No.                               lpm    
------------------------------------------------------------------------
1..........................................................        0.903
2..........................................................        0.908
3..........................................................        0.899
4..........................................................        0.900
------------------------------------------------------------------------

    The mean flow rate was ((0.903 + 0.908 + 0.899 + 0.900)/ 4)) 0.903 
liters per minute (lpm).
    The sampling probe was placed in the exhaust duct of the 
ventilation system and background samples were registered by the 
detector. The tubing (pure SF6 outlet) from the flow meter was 
placed through the hood and into the duct of the ventilation system 
(upstream of the fan). Readings were as follows:

------------------------------------------------------------------------
                                                              Detector  
                     Task                      Reading No.  reading, ppm
                                                               of SF 6  
------------------------------------------------------------------------
Background...................................            1        0.0051
                                                         2        0.0062
                                                         3        0.0048
                                                         4        0.0050
                                                         5        0.0066
                                                         6        0.0062
                                                         7        0.0058
Start SF6....................................            8        6.3   
                                                         9       22.0   
                                                        10       21.8   
                                                        11       21.9   
                                                        12       21.7   
                                                        13       21.8   
End..........................................           14       21.9   
------------------------------------------------------------------------

    At least five consecutive measurements are needed; in this case, 
the last six data points were used. The eighth reading (6.3 ppm) does 
not reflect steady-state and was not used in determining the average. 
The mean concentration of SF6 is 21.85 ppm (the average of those 
six points). The mean background value is 0.0057 ppm. These values were 
used to calculate the volumetric flow rate from Equation 1.

Q(exh)+0.903 / 28.3 / (21.85-0.0057) * 106 = 1460 cfm.

    The average background value, 0.0057 ppm, was subtracted from the 
average 100% capture value, 21.85 ppm. In this case, the background 
value was negligible.
    The same flow meter and SF6 flow rate were used for the 
capture efficiency test. The tubing was removed from the ventilation 
system hood and connected to the 10-foot distribution plenum. Readings 
were as follows:

------------------------------------------------------------------------
                                                              Detector  
                     Task                      Reading No.  reading, ppm
                                                                 SF6    
------------------------------------------------------------------------
Background...................................            1         0.092
                                                         2         0.084
                                                         3         0.078
Start SF 6...................................            4        28.1  
                                                         5        18.8  
                                                         6        19.6  
                                                         7        19.7  
                                                         8        20.9  
                                                         9        17.3  
                                                        10        19.4  
                                                        11        18.9  
                                                        12        19.6  
------------------------------------------------------------------------

    At least five consecutive measurements are needed; in this case, 
the last eight will be used. The fourth reading (28.1 ppm) was high; in 
this case it reflects the flow controller overshooting the set point 
during the startup of SF6 flow, and this point is not used in 
determining the average. The mean concentration of SF6 is 19.28 
ppm; the average background concentration was 0.0847 ppm.
    Because we used the same SF6 flow rate in both the exhaust 
volume test and the capture efficiency test, the calculations are 
simplified. From Equation 2, the capture efficiency is (19.28-.0847) / 
(21.85-0.0057)* 100 = 87.9%.
    This procedure was done four times with the following results:

------------------------------------------------------------------------
                                        100%      Capture      Capture  
             Trial No.                capture,  efficiency,  efficiency,
                                      ppm SF 6    ppm SF 6        %     
------------------------------------------------------------------------
1..................................      21.84       19.20         87.9 
2..................................      21.67       19.95         92.1 
3..................................      21.74       18.10         83.3 
4..................................      21.93       19.01         86.7 
------------------------------------------------------------------------

Statistics

    Calculate the overall average of the means:

m = (87.9 + 92.1 + 83.3 + 86.7) / 4 = 87.5%

    Calculate the estimated standard deviation:

s={((87.9-87.5)2 + (92.1-87.5)2 + (83.3-87.5)2 + (86.7-87.5)2) / 
(4-1)}0.5
    ={ (0.16 + 21.16 + 17.64 + 0.64) / 3}0.5 = 3.63

    If the number of trials, n, is different from 4, then (n-1) is used 
in the denominator of this calculation and the numerator is the sum of 
all n squared differences, rather than just 4. Choose the number t 
(from the Student's t-distribution table at the 95th percentile) from 
the following table, based on the value of n:

t: 6.31 (n=2) 2.92 (n=3) 2.35 (n=4)
    2.13 (n=5) 2.02 (n=6) 1.94 (n=7)
    1.90 (n=8) 1.86 (n=9) 1.83 (n=10)

    Calculate a test statistic (T):

T=m-t*s / n0.5

    For this example: T = 87.5-2.35 * 3.63 / 40.5 = 83.2.
    If T > 80.0, then decide (with 95% confidence) that efficiency is 
greater than 80%. In this example, we are 95% confident that the 
efficiency is greater than 80%.
    If T  80.0, then the conclusion that the efficiency is 
greater than 80% cannot be made from these data.

Equipment

Smoke Test
Smoke generator
2 inch x 10 foot Schedule-40 PVC perforated distribution pipe
Tracer Gas Tests
Compressed cylinder of 99.98% SF6 with regulator
Flow controller such as a precision rotameter
\1/8\-inch ID x 20-foot Teflon tubing and snap valves for SF6 
distribution
Primary Flow Calibrator
\1/2\-inch ID x 10-foot Copper tubing with \1/32\-inch holes every 12 
inches SF6 distribution plenum
Gas monitor calibrated for SF6
Calibration gases, nitrogen and at least one SF6 concentration in 
nitrogen
12-liter Mylar gas sampling bags
Ventilation System Evaluation
Air Velocity Meter
Micro manometer w/Pitot Tube

    Dated: September 27, 1996.
Linda Rosenstock,
Director, National Institute for Occupational Safety and Health Centers 
for Disease Control and Prevention (CDC).
[FR Doc. 96-25314 Filed 10-2-96; 8:45 am]
BILLING CODE 4163-19-P