HOOD INVESTIGATION by HC121004052950

VIEWS: 8 PAGES: 51

									TESTING OF PROTOTYPE
ENERGY-EFFICIENT FUME HOODS

MONTANA STATE UNIVERSITY


September 12, 2001



Submitted to:
Montana State University
Bozeman, Montana


By:


                                                   Ventilation Design
Knutson Ventilation, Inc.
Minneapolis, Minnesota


Gerhard W. Knutson, Ph.D., CIH                     Industrial Hygiene
President
Knutson Ventilation, Inc.




                                 Knutson Ventilation Consulting, Inc.
                                     3404 West 60th Street, Edina, MN 55410
                                         gknutson@knutsonventilation.com
                                                 Phone (952) 928-0195
                                                    Fax (952) 920-1510
                                    CHAPTER 1

                                    SUMMARY

Part 1: Baseline Testing of the Containment Performance Based on ASHRAE 110
Procedures.

   1.     Both laboratory hoods performed adequately under normal operation.
   2.     The airflow characteristics of both hoods were not acceptable. Both hood
          demonstrated reverse flow on the work surface and a potential weakness in the
          lower portion of the hood.

Part 2: Parametric Testing of Containment Performance Based on ASHRAE 110
Procedures

   1.     The LBNL prototype hood was susceptible to imbalance of the internal supply
          air systems.
          a. At low velocity, the hood performed poorly when the supply air
              volumetric flow was high.
          b. At low velocity, the hood did not perform adequately when the supply air
              system did not operate.
   2.     The Fisher Hamilton prototype hood performed adequate at reduced face
          velocity when tested by the ASHRAE Standard 110. However, the airflow
          patterns indicated a weakness in the lower portion of the hood.

Part 3: Testing of Contaminant Performance Based on Emerging “Beyond Current State-
of-the-Art” Procedures.

   1.     Modified delivery of the tracer gas appeared to increase the challenge to the
          prototype hoods.
   2.     When the tracer gas was released close to the work surface, the performance
          of both hood deteriorated.
   3.     Because of the change in performance, further investigation into the method
          of delivering tracer gas to a laboratory hood should be investigated.

Part 4: Cooperative Effort to Improve the Prototype Hoods

   1.     LBNL Prototype Hood
          a. The screens for the supply air systems are susceptible to damage. This
             damage could affect the hood performance.
          b. Select adequate supply fans
                  i. Heavy duty for longer life
                 ii. Reduced noise levels
                iii. Provide a reproducible adjustment of the fans to ensure proper
                     function
          c. Eliminate the void space above lights



Prototype Hood Testing           September 12, 2001                             Page 1
          d. Provide a method of monitoring the supply fan performance

   2.     Fisher Hamilton Prototype Hood
          a. Revise the design of the airfoil to eliminate the reverse flow on the surface
              of the airfoil
          b. Eliminate the movable baffle
          c. Eliminate the spill tray or provide adequate means to clean the tray
          d. Modify the sash frame by recessing it into the sidewalls




Prototype Hood Testing            September 12, 2001                               Page 2
                                       CHAPTER 2

                                   INTRODUCTION

Two prototype energy-efficient laboratory hoods were installed in the Safety and Risk
Management Building at the Advanced Technology Park, Bozeman, Montana. Montana
State University (MSU) retained Knutson Ventilation, Inc. to evaluate two laboratory
hoods.




                             Laboratory
                             Exhaust Stack




                                                         Hastings
                                                         Replacement
                                                         Air Unit



             Photo 2-1: Overview of Safety and Risk Management Building

The project consisted of four parts:

Part 1: Baseline Testing of the Containment Performance Based on ASHRAE 110
Procedures.

Part 2: Parametric Testing of Containment Performance Based on ASHRAE 110
Procedures



Prototype Hood Testing             September 12, 2001                         Page 3
Part 3: Testing of Contaminant Performance Based on Emerging “Beyond Current
State-of-the-Art” Procedures.

Part 4: Cooperative Effort to Improve the Prototype Hoods

The first hood was a 4-foot [1.2-meter] bench-top laboratory hood manufactured by
Fisher Hamilton for Lawrence Berkley National Laboratory (LBNL). The LBNL hood is
a new concept hood that introduces three supply air systems as an integral part of the
hood. The prototype hood expands the concept of an auxiliary air hood by adding an
internal supply system, an exterior supply system similar to an auxiliary air system, and a
supply system where an airfoil is normally attached to a laboratory hood. The hood is
designed to operate as a constant volume (CV) hood.

The second was a 7-foot [2.1-meter] bench-top laboratory hood, with a combination sash,
designed and manufactured by Fisher Hamilton. The hood was a modified SafeAir hood
with a new design airfoil, clear sight panel, modified combination sash, and an automatic
sash-closing device that automatically reduced the sash height to a preset height. The
hood is designed to operate as either a constant volume (CV) or variable air volume
(VAV) hood.




                                              Phoenix Exhaust
                                              VAV Valves




                                                                                LBNL
                                                                                Prototype
                                          Fisher Hamilton
                                                                                Hood
                                          Prototype Hood



                Photo 2-2: Overview of exhaust system in test laboratory



Prototype Hood Testing             September 12, 2001                               Page 4
Before the testing began, MSU modified the supply and exhaust systems for the
laboratory. The exhaust system included Phoenix valves to modulate the volumetric flow
depending on the sash position. As the sash opens, the valve modulates to adjust the
exhaust flow to match a calculated volumetric flow to provide a constant face velocity
independent of the sash position.

Although the Phoenix system was designed to modulate the volumetric flow, MSU had
installed controls to override the modulation and allow the investigator to “dial in” the
desired volumetric flow. In essence, the valves became constant volume valves and
allowed the investigator to adjust the flow rates as necessary to meet the needs of the
investigation.

The supply system used the Phoenix system to maintain a specified volumetric flow off
set in the laboratory. The control system determined the total exhaust and adjusted the
supply valves to provide a volumetric offset of about 200 cfm [0.09 m3/s].




         Perforated                                       Supply Air
         Supply Duct                                      Header




                Photo 2-3: Partial view of test laboratory supply system




Prototype Hood Testing            September 12, 2001                              Page 5
Supply air was delivered through two 12-inch [300 mm] diameter perforated ducts.
Photo 2-3 shows a typical supply duct. The perforated supply duct allowed the supply
system to deliver the supply air to the laboratory in a slow controlled manner. The
resulting room air currents were generally below 30 fpm [0.15 m/s] within five feet [two
meters] of the laboratory hoods.




Prototype Hood Testing            September 12, 2001                             Page 6
                                     CHAPTER 3

                                  TEST METHODS


Test Equipment

We used a TSI Model 8455-06, Air Velocity Transducer, to collect the face velocity
readings. We used a TSI VelociCalc Model 8384A to measure the air flow through the
supply systems for the LBNL hood. Prior to conducting the tests, both instruments were
calibrated by the manufacture at their calibration laboratory.




               Photo 3-1: Face velocity measurements in the LBNL Hood

We used a Radio Shack theatrical fog generator and Regin™ smoke bottles to generate a
visible aerosol and visualize the air flow at and through the laboratory hoods.

The tracer gas tests used a Miran 205B as a detector. The detector was an infrared
spectrophotometer. At the beginning of the test period, we adjusted the detector, changed


Prototype Hood Testing            September 12, 2001                              Page 7
the wavelength to 10.7 μm, and calibrated the instrument, using methods recommended
by the instrument manufacturer. The instrument stored the data and we downloaded it to
an excel file for further analysis.

Test Procedure

Face Velocity Measurements

Using a calibrated TSI air velocity transducer, Model 8455-06, we measured the face
velocity for each hood tested. We established a grid pattern by dividing the hood face
opening into equal rectangular areas.

For the LBNL hood, we used a three by three grid, resulting in nine readings.

For the Fisher Hamilton prototype hood in the horizontal sash configuration, we used a
three by three grid (nine positions). We repeated the tests for the three horizontal
openings. In the vertical sash configuration, we used a three by six grid (eighteen
positions).

We placed the probe on a ring stand to stabilize the anemometer. We recorded the face
velocity at each point each second for thirty seconds. The data from each measurement
point were averaged, and an overall average was determined for each test condition.

Smoke Visualization

The small-scale smoke visualization methods used Regin™ smoke bottles to generate a
visible aerosol within the hood. The investigators released the smoke at several locations
at each hood. In addition, we used theatrical fog, from a Radio Shake fog generator, to
visualize a large smoke challenge to the hood.

Tracer Gas Testing

The tracer gas test released a known volume of tracer gas, sulfur hexaflouride, through a
specially designed ejector within the hood. Deviating from the requirements in the
ANSI/ASHRAE Standard 110-1995, we used a tracer gas release rate of 4 lpm. We
placed a mannequin, representing a chemist or student, at the face of the hood, usually in
the center of the horizontal opening. We placed the detector probe in the “breathing
zone” of the mannequin (see photo 3-2) and the detector measured the concentration of
tracer gas.

Mannequin Position

The ANSI/ASHRAE Standard 110-1995 prescribes the position of the mannequin. The
standard assumes a “standard” five or six-foot [1.5 or 1.8 meter] bench top hood. To set
up the test conditions, the ejector (the source of tracer gas) is placed six inches [150 mm]
into the hood, representing good work practices. The mannequin is placed approximately



Prototype Hood Testing             September 12, 2001                                Page 8
in a normal working position. The mannequin is placed at the center of the hood and on
both sides, each 12 inches [300 mm] from the sidewall. Because a “standard” hood has
an airfoil which projects out past the plane of the sash, the standard positions the
mannequin near the leading edge of the airfoil. To maintain a consistent set of data, the
mannequin is placed three inches [75 mm] in front of the sash. For a combination sash,
we normally define the plane of the hood as the midpoint of the two horizontal window
tracks.




                          Tip of Detector
                          Probe




         Photo 3-2: Mannequin in center of the Fisher Hamilton prototype hood

The standard does not address a combination sash hood operating in the horizontal mode.
We tested the hood by placing the horizontal sashes in three positions: left open (all
sashes to the right), center position (sashes pushed to both sides to provide the maximum
open area in the center), and right (all sashes pushed to the left side of the hood).

Ejector Position

The airfoil design of the both low flow hoods is very different from the “standard” hood.
The LBNL prototype hand a special designed airfoil and the Fisher Hamilton prototype
has an airfoil that was “inboard” of the sash.


Prototype Hood Testing             September 12, 2001                             Page 9
After discussing the situation with MSU, we agreed to set the ejector on the work surface,
as close to the sash as the recessed work surface would allow. We modified the
ASHRAE standard by using the ejector as the datum point and measuring everything
from there. This meant that the distance from the ejector to the mannequin to the ejector
was consistent; however, for both hoods, the ejector was further into the hood and the
mannequin was closer to the sash.




Prototype Hood Testing            September 12, 2001                              Page 10
                                   CHAPTER 4

                           LBNL PROTOTYPE HOOD


                                 HOOD DESIGN


The Lawrence Berkley National Laboratory (LBNL) Hood is a prototype hood designed
with the intent of providing a low volume, high efficiency hood. Dr. Helmut Feustel,
while at LBNL, developed the initial design for the low flow hood. Subsequently,
Geoffrey Bell, Dale Sartor, and other scientists and students at LBNL developed the
concept. Under a NIST grant, Fisher Hamilton, a large laboratory hood manufacturer,
joined with LBNL to complete a prototype hood. The prototype hood has been tested by
Fisher Hamilton at their test laboratory and by LBNL at their facility in Berkeley,
California.

Hood Description




                                                             Top (Inside)
                                                             Plenum Duct




    Exhaust
    Valve




     Front Plenum
     Inlet Duct




                                                                   Bottom Plenum
                                                                   Inlet Duct




                    Photo 4-1: Overview of LBNL prototype hood



Prototype Hood Testing          September 12, 2001                           Page 11
The LBNL Hood was a prototype hood using an innovative design. The hood has a
Fisher Hamilton superstructure; however, the hood is very different from the normal
Fisher Hamilton hood. The superstructure was for a 4-foot [1.2-meter] laboratory hood
and the hood had a single vertically sliding sash. With the sash full open, the width of the
opening was 38.75 inches [980 millimeters] wide and the height, measured from the top
of the work surface to the bottom of the open sash, was 33 inches [840 millimeter]. The
available opening was 30.5 inches [770 millimeters] high, since the lower plenum
extended 2.5 inches [65 millimeters] above the work surface. The sash moved through a
small gap between the top plenum and the front plenum, both of which are described
below.

The most obvious difference between the LBNL Prototype hood and a standard
laboratory hood are the three supply plenums. Small fans, attached to the plenums, draw
room air into the plenums.


                             Perforated
                             Plate Baffle




             Recessed
             Work Surface




                                                                             Bottom
                                                                             Plenum




                  Photo 4-2: Bottom Plenum for LBNL prototype hood

Bottom Plenum

The bottom plenum is a five-inch [130 by 130 millimeter] square plenum. The plenum is
the full width of the hood. The plenum is tilted, at about a 60 degree angle, resulting in
the top of the plenum about 2.5 inches [65 millimeters] above the work surface.


Prototype Hood Testing              September 12, 2001                                Page 12
The plenum contains the screen along the long direction of the plenum. This screen adds
additional resistance to airflow and assists in providing uniform airflow. The side of the
plenum, facing the hood opening, has a second screen, also to assist in providing uniform
airflow. Finally, the plenum has a curved metal surface.

The height of the plenum is about 2.5 inches [65 millimeters] above the work surface.
This height is greater than typical. A “typical” airfoil is 1 to 1.5 inches [25 to 40
millimeters] above the work surface.

Top (Inside) Plenum

The top (inside) plenum is approximate 38 inches [970 millimeters] wide. The top is five
inches [130 millimeters] deep and the bottom is about three inches [75 millimeters] deep.
The bottom of the plenum has a fine wire screen that distributes the air inside the hood.
The screen, in cross section forms a distorted “U”. The bottom of the screen is about four
inches [100 millimeters] from the front of the front plenum panel. The top of the screen
extends about one inch [25 millimeters] from the back panel into the hood. The curvature
is not uniform with the curvature higher on the top of the hood.




                                                                     Opening Between
                                                                     Header Panel and
                                                                     Tilted Internal
                                                                     Baffle




               Top (Inside)
               Plenum Screen




      Photo 4-3: View screen for top (inside) plenum on the LBNL prototype hood



Prototype Hood Testing            September 12, 2001                              Page 13
The bottom of the screen is bout 20 inches [510 millimeters] from the top of the hood.
The top of the screen is about three inches [75 millimeters] above the bottom.

In many laboratory hoods, the momentum of the air entering the hood causes a negative
pressure in the upper chamber of the hood. The negative pressure forms a vortex inside
the upper section of the hood (usually called a “roll”) with the air flowing down the
inside of the sash. According to LBNL, the purpose of the top (inside) plenum is to
reduce the roll and eliminate the turbulence at the bottom of the sash.

Front Plenum

The front plenum is in front of the sash and discharges outside the hood. At the top, the
front plenum is four inches [100 millimeters] wide. The bottom on the plenum is two
inches [50 millimeters] wide and has a fine wire screen across the full length to assist in
distribution. The plenum is twenty inches [500 millimeters] high. The top five inches
[130 millimeters] are vertical. The bottom fifteen inches [380 millimeters] of the front
panel bends toward the hood. The bottom is flush with the front of the hood. See photo
4-1 and photo 4-4 for a view of the front plenum. When the hood is properly set up, the
front plenum has an additional panel covering the plenum.



             Front Plenum
                                                                 Exhaust
                                                                 Transition




                                                                 Top (Inside)
                                                                 Plenum Duct

                             Supply Volume
                             Controllers



                 Photo 4-4: Front plenum and top (inside) plenum duct



Prototype Hood Testing             September 12, 2001                             Page 14
As discussed below, MSU modified the hood to allow measurement of the volumetric
flow. The modification prevented the placement of the front panel.

Ventilation Measurement

Prior to the investigation, MSU modified the LBNL prototype hood to allow
measurement of the supply system volumetric flows. Specifically, MSU installed 4-inch
[100 millimeter] round PVC pipe to each of the three supply fans. Although the PVC
pipe may have modified the hood performance, we measured the flows with the pipe in
place. To ensure we would know the flow rates while testing, we left the pipe in position
during the tests.

The LBNL hood had three rheostats to adjust the volumetric flow (see Photo 4-4). Since
the power to the hoods is related to the flow rate, MSU installed a meter to measure the
total wattage to the three fans. As we conducted the tests, we ensured the wattage
remained constant through the tests. With a constant wattage, we assumed the volumetric
flow remained constant.

Table 4-1 [Table 4-1M] and Figure 4-1 [Figure 4-1M] show the volumetric flow for the
three supply fans under four different operating conditions.



                                   TABLE 4-1
                         SUPPLY AIR VOLUMETRIC FLOW

       Watts          Top Plenum Front Plenum Bottom plenum               Total flow

       27 W           34 cfm         44 cfm         40 cfm                118 cfm
       29 W           64 cfm         61 cfm         66 cfm                191 cfm
       33 W           70 cfm         83 cfm         80 cfm                233 cfm
       42 W           70 cfm         86 cfm         89 cfm                254 cfm



                                  TABLE 4-1M
                         SUPPLY AIR VOLUMETRIC FLOW

       Watts          Top Plenum Front Plenum Bottom plenum               Total flow

       27 W    0.016 m3/s      0.021 m3/s     0.019 m3/s          0.056 m3/s
       29 W    0.030 m3/s      0.028 m3/s     0.031 m3/s          0.090 m3/s
       33 W    0.033 m3/s      0.039 m3/s     0.037 m3/s          0.110 m3/s
       42 W    0.037 m3/s      0.041 m3/s     0.042 m3/s          0.120 m3/s




Prototype Hood Testing             September 12, 2001                            Page 15
                                                                             Figure 4-1
                                                               Volumetric Flow Through Supply Fans


                                              300



                                              250
                       Volumetric Flow, cfm




                                              200


                                                                                                                        Bottom
                                              150
                                                                                                                        Front
                                                                                                                        Top
                                              100



                                              50



                                               0
                                                      27 W            29 W               33 W        42 W

                                                                      Total Fan Power, Watts




                                                                            Figure 4-1M
                                                               Volumetric Flow Through Supply Fans


                                              0.120
     Volumetric flow, cubic meters per




                                              0.100



                                              0.080
                   second




                                                                                                                        Bottom
                                              0.060
                                                                                                                        Front
                                                                                                                        Top
                                              0.040



                                              0.020



                                              0.000
                                                        27 W            29 W               33 W        42 W

                                                                        Total Fan Pow er, Watts




Prototype Hood Testing                                              September 12, 2001                        Page 16
Exhaust Plenum

The LBNL Hood has an exhaust plenum, in the back of the hood, to assist in proper
distribution. The plenum tilts slightly toward the front. At the bottom, the plenum is
about 2.25 inches [57 millimeters] deep and at the top about 3 inches [75 millimeters]
deep. The bottom portion of the baffles is perforated board. The perforations cover a
section 39 inches [990 millimeters] wide by 16 inches [400 millimeters] high. The
perforations are ¼ inch [6 millimeters] on one-inch [25 millimeter] centers (square
pattern). The only slot is at the top.

The top of the baffle extends into the duct transition. The dust transition is five inches
[130 millimeters] wide by about 34 inches [860 millimeters] wide. The baffle splits the
transition. The opening in front of the baffle is two inches [50 millimeters] wide and the
space behind the baffle is three inches [75 millimeters] wide.

The baffle does not have a bottom slot.




                                                            Sloped Front
      Top Slot                                              Baffle with Light




                  Photo 4-5: Interior baffles on LBNL prototype hood




Prototype Hood Testing            September 12, 2001                              Page 17
Tilted baffle and light

Just inside the hood, the hood has a sloped baffle plate. The leading edge is about 38
inches [970 millimeters] from the work surface and the inside edge is at the top of the
hood, 48 inches [1.2 meters] above the work surface. The sloped length of the panel is
about 16 inches [400 millimeters]. The inside edge is about 5 inches [130 millimeters]
from the back baffle and the leading edge is about 18 inches [450 millimeters] from the
back baffle. The slope is almost 45 degrees.

Between the sloped front panel and the top (inside) plenum is about 1.5 inches [40
millimeters]. The space between the sloped panel and the top of the hood is essentially a
closed space. Although the space probably did not affect the hood performance tests, the
space could allow accumulations of contaminants. This may be most significant because
the lamps are also on this space.



                              HOOD PERFORMANCE

Because the LBNL prototype hood has three supply plenums, the airflow around the face
opening is not perpendicular to the hood face. In addition, the top (inside) plenum
releases air inside the hood. Consequently, we could not directly measure the face
velocity while the supply plenums were operating. In addition, the duct system did not
allow for measurement in the duct (see photo 2-2).

We measured the total volumetric flow by shutting off the supply fans and measuring the
face velocity. We assumed the hood had no leakage and reported the results of the
average face velocity.

Face Velocity Measurements

We began the tests with a nominal 60 fpm [0.3 m/s] face velocity. When the hood
performed adequately, we lowered the face velocity to a nominal 40 fpm [0.2 m/s].
Finally, we increased the nominal face velocity to 50 fpm [0.25 m/s]. To set the
volumetric flow, we used the readout from the Phoenix valve for the LBNL hood. The
reported value is the readout and we made no effort to determine the proper altitude
correction factor for the valve.

The results of the face velocity measurements are shown in Tables 4-2, 4-3 and 4-4
[Tables 4-2M, 4-3M, 4-4M]. The left top position was A1 and the right bottom position
was C4.




Prototype Hood Testing            September 12, 2001                             Page 18
                                        Table 4-2

                            LBNL Prototype Hood
                          Face Velocity Measurements
                         Nominal 60 fpm Face Velocity
                         Phoenix Exhaust Rate 400 cfm

                      1                 2               3        4
             A      78 fpm            71 fpm          68 fpm   77 fpm
             B      61 fpm            48 fpm          47 fpm   55 fpm
             C      60 fpm            55 fpm          54 fpm   50 fpm

             Average         60 fpm
             Maximum         78 fpm            129%
             Minimum         47 fpm             78%


                                   Table 4-3
                            LBNL Prototype Hood
                          Face Velocity Measurements
                         Nominal 50 fpm Face Velocity
                         Phoenix Exhaust Rate 335 cfm

                      1                 2               3        4
             A      69 fpm            65 fpm          66 fpm   62 fpm
             B      48 fpm            48 fpm          43 fpm   43 fpm
             C      50 fpm            50 fpm          46 fpm   46 fpm

             Average         53 fpm
             Maximum         69 fpm            130%
             Minimum         43 fpm             82%

                                   Table 4-4
                            LBNL Prototype Hood
                          Face Velocity Measurements
                         Nominal 40 fpm Face Velocity
                         Phoenix Exhaust Rate 270 cfm

                    1                 2               3        4
             A      49 fpm            52 fpm          46 fpm   46 fpm
             B      48 fpm            42 fpm          39 fpm   36 fpm
             C      46 fpm            43 fpm          39 fpm   35 fpm

             Average                  43 fpm
             Maximum                  52 fpm          121 %
             Minimum                  35 fpm           81%



Prototype Hood Testing           September 12, 2001                     Page 19
                                  Table 4-2M
                            LBNL Prototype Hood
                          Face Velocity Measurements
                         Nominal 0.30 m/s Face Velocity
                         Phoenix Exhaust Rate 400 cfm

                       1            2             3          4
             A      0.40 m/s     0.36 m/s      0.35 m/s   0.39 m/s
             B      0.31 m/s     0.25 m/s      0.24 m/s   0.28 m/s
             C      0.31 m/s     0.28 m/s      0.27 m/s   0.25 m/s

             Average      0.31 m/s
             Maximum      0.39 m/s      129%
             Minimum      0.24 m/s       78%

                                  Table 4-3M
                            LBNL Prototype Hood
                          Face Velocity Measurements
                         Nominal 0.25 m/s Face Velocity
                         Phoenix Exhaust Rate 335 cfm

                       1            2             3          4
             A      0.35 m/s     0.33 m/s      0.33 m/s   0.31 m/s
             B      0.24 m/s     0.24 m/s      0.22 m/s   0.22 m/s
             C      0.25 m/s     0.25 m/s      0.24 m/s   0.23 m/s

             Average      0.27 m/s
             Maximum      0.35 m/s      130%
             Minimum      0.22 m/s       82%

                                  Table 4-4M
                            LBNL Prototype Hood
                          Face velocity Measurements
                         Nominal 0.2 m/s Face Velocity
                         Phoenix Exhaust Rate 270 cfm

                    1            2             3          4
             A      0.25 m/s     0.27 m/s      0.23 m/s   0.24 m/s
             B      0.25 m/s     0.21 m/s      0.20 m/s   0.18 m/s
             C      0.23 m/s     0.22 m/s      0.20 m/s   0.18 m/s

             Average             0.22 m/s
             Maximum             0.27 m/s      121%
             Minimum             0.18 m/s       81%



Prototype Hood Testing         September 12, 2001                    Page 20
Table 4-5 [Table 4-5M] shows the face velocity readings when the sash for the nominal
60 fpm [0.3 m/s] conditions was closed to 18 inches [45 millimeters].


                                    Table 4-5
                              LBNL Prototype Hood
                           Face velocity Measurements
                   Nominal 60 fpm Face Velocity: Sash Half Open
                          Phoenix Exhaust Rate 400 cfm

                      1            2               3            4
               A      117 fpm      113 fpm         113 fpm      106 fpm
               B      110 fpm      111 fpm         105 fpm      106 fpm

               Average      110 fpm
               Maximum      117 fpm                106 %
               Minimum      106 fpm                 95%


                                    Table 4-5M
                              LBNL Prototype Hood
                           Face velocity Measurements
                   Nominal 0.3 m/s Face Velocity: Sash Half Open
                         Phoenix Exhaust Rate 0.01 m3/s

                      1            2               3            4

               A      0.59 m/s     0.57 m/s        0.57 m/s     0.56 m/s
               B      0.56 m/s     0.56 m/s        0.53 m/s     0.54 n/s

               Average      0.56 m/s
               Maximum      0.59 m/s       106 %
               Minimum      0.53 m/s        95%



Smoke Visualization

Work surface

Smoke released along the work surface was lazy and demonstrated some reverse flow
(see Photo 4-6). Smoke released in front airfoil has a slight tendency to move forward
toward the bottom supply.




Prototype Hood Testing           September 12, 2001                           Page 21
                                                          Smoke Showing
                                                          Reverse Flow at
                                                          Work Surface




        Photo 4-6: Smoke released along work surface in LBNL prototype hood

Top of hood

Smoke release in the top cavity of the hood, between the rear baffle, the bottom of the
open sash, and the sloped light panel, was drawn into the large slot on the top of the
hood. The airflow patterns confirmed the higher velocities at the top of the hood
opening.

Face Capture

   1.     Smoke released outside the hood, without the supply fan operating, moved
          toward the opening and was captured by the hood.
   2.     With the supply fans operating at the design point, and the total exhaust
          volume reduced, the hood did not capture smoke released at the face of the
          hood.
   3.     With the supply fans on and with sufficient exhaust volumetric flow, smoke
          released outside sash opening was captured. However, the down flow from
          the front plenum pushed the smoke downward before it was captured.
   4.     By adjusting supply flow, the hood developed a quiescent spot in the center of
          the face opening. A slight increase in supply air, or decrease in exhaust,
          caused the smoke to escape capture from the hood.


Prototype Hood Testing           September 12, 2001                             Page 22
Supply air

Smoke released into the bottom supply appears to have a higher velocity on the fan side
(right side) of the hood. The smoke does not move all the way to the end of the plenum.

Smoke released from the theatrical fog generator into the bottom plenum remained on the
work surface for several minutes. The smoke appeared to enter more on the fan side of
the hood; however, it never got to the back of the work surface.




                  Photo 4-7: Smoke released from the bottom plenum

Smoke released in the front air intake appeared to be uniformly distributed across the
front of the hood. However, it may have had slightly less airflow on the sides.

Smoke release in the top (inside) air supply tended to rise within the hood, toward the
exhaust slot.

Smoke released outside the hood, three inches [75 millimeters] above the front plenum,
rose with room air currents. The angle also had an effect on the smoke patterns.




Prototype Hood Testing           September 12, 2001                            Page 23
Smoke released above the top (inside) plenum tended to raise then follow the tilted
baffle.

Smoke released in the front plenum fell across the face of the hood and the exhaust drew
the smoke into the hood within the top eight inches.

Tracer Gas Tests

60 fpm [0.3 m/s] Tests

Based on recommendations provided by Geoffrey Bell of LBNL, we adjusted the supply
fans to provide about 270 cfm [0.13 m3/s]. The split between the three plenums was as
shown in Table 4.1 and Figure 4.1.

With the LBNL prototype hood set to provide a nominal face velocity of 60 fpm [0.2
m/s], we measured the tracer gas levels. We released tracer gas, sulfur hexafluoride, at 4
lpm and monitored in the breathing zone of the mannequin. Because chemists come is all
sizes, we used three mannequin heights. The ASHRAE standard required a mannequin
with a breathing zone 26 inches [650 millimeters] above the work surface. At that height,
the mannequin’s head hit the front supply plenum. The situation would have been worse
if the front panel were attached to the hood.

Several investigators are currently using a mannequin with a breathing zone 18 inches
[450 millimeters] above the work surface. Consequently, we used that height. Finally,
we chose the highest mannequin height that would allow the mannequin to stand under
the front plenum, or 22 inches [550 millimeters] high. The results of the tests are
summarized in Table 4-5.



                                TABLE 4-5
                            TRACER GAS TESTS
                             SASH FULL OPEN
                   NOMINAL FACE VELOCITY 60 FPM [0.3 M/S]
                    SUPPLY SYSTEM AT 400 CFM [0.19 M3/S]

       Mannequin Height                Left           Center         Right

       26 inches [650 mm]            < 10 ppb       < 10 ppb       < 10 ppm
       22 inches [550 mm]            < 10 ppb       < 10 ppb       < 10 ppm
       18 inches [450 mm]            < 10 ppb       < 10 ppb       < 10 ppm


Without changing the exhaust of supply flows, we lowered the sash to 18 inches [450
millimeters] above the work surface and repeated the tracer gas tests with the mannequin




Prototype Hood Testing            September 12, 2001                              Page 24
height 18 inches [450 millimeters] above the work surface. The results of all three
positions and the sash movement test were all below the limit of detection, 10 ppb.

Because the LBNL hood performed well with all three of the mannequin heights and the
low front panel made the 26-inch [650 millimeters] mannequin’s head touch the panel,
we conducted most of the remaining tests with a 22-inch 9550 millimeters] mannequin.

40 fpm [0.2 m/s] Tests

We reduced the volumetric flow through the LBNL hood to provide a nominal face
velocity of 40 fpm [0.02 m/s] and conducted tests with a 22-inch [550 millimeter]
mannequin. Because of the significant reduction in volumetric flow, we tested the hood
in the center position at several different supply air setting. When the controls were set at
a minimum, the meter indicated 27 Watts. We tested at five different supply airflows
(see figure 4-1 for the flow rates) and with no supply air. We also tested with the
mannequin at 18 inches [450 millimeters]. The results are shown in Table 4-6 [Table
4.6M].



                                         TABLE 4-6

                              TRACER GAS TESTS
                               SASH FULL OPEN
                         NOMINAL FACE VELOCITY 40 FPM
                           SUPPLY SYSTEM AT 270 CFM

Mannequin         Supply Air                    Left           Center          Right
Height         Power       Flow

22 inches      0 Watts         0 cfm                          110 ppb
22 inches      27 Watts        120 cfm        < 10 ppb        < 10 ppb    < 10 ppb
22 inches      29 Watts        190 cfm                        < 10 ppb
22 inches      33 Watts        230 cfm                        4,000 ppb
22 inches      43 Watts        250 cfm                        >10,000 ppb

18 inches      0 Watts         0 cfm                          2,800 ppm
18 inches      27 Watts        120 cfm        < 10 ppb        < 10 ppb       < 10 ppb
18 inches      29 Watts        190 cfm                        100 ppb


The test data clearly shows loss of control when the supply fans are off or when the
supply fans provide too much supply air.




Prototype Hood Testing              September 12, 2001                              Page 25
                                  TABLE 4-6M
                              TRACER GAS TESTS
                                SASH FULL OPEN
                         NOMINAL FACE VELOCITY 0.2 M/S
                           SUPPLY SYSTEM AT 0.13 M3/S

Mannequin        Supply Air                Left         Center        Right
Height        Power       Volume

550 mm        0 Watts        0 m3/s                    110 ppb
550 mm        27 Watts       0.056 m3/s   < 10 ppb     < 10 ppb    < 10 ppb
550 mm        29 Watts       0.090 m3/s                < 10 ppb
550 mm        33 Watts       0.11 m3/s                 4,000 ppb
550 mm        43 Watts       0.12 m3/s                 >10,000 ppb

450 mm        0 Watts        0 m3/s                    2,800 ppm
450 mm        27 Watts       0.056 m3/s   < 10 ppb     < 10 ppb      < 10 ppb
450 mm        29 Watts       0.090 m3/s                100 ppb


50 fpm [0.25 m/s] Test

We increased the volumetric flow to provide a nominal face velocity of 50 fpm [0.25
m/s] and repeated the tests. Table 4-7 [Table 4-7M] shows the test results.



                                   TABLE 4-7
                              TRACER GAS TESTS
                               SASH FULL OPEN
                         NOMINAL FACE VELOCITY 50 FPM
                           SUPPLY SYSTEM AT 330 CFM

Mannequin        Supply Air                Left         Center        Right
Height        Power       Flow

22 inches     0 Watts        0 cfm                       10 ppb
22 inches     27 Watts       120 cfm      < 10 ppb     < 10 ppb      < 10 ppb
22 inches     29 Watts       190 cfm                   < 10 ppb
22 inches     33 Watts       230 cfm                   < 10 ppb
22 inches     43 Watts       250 cfm                     10 ppb

18 inches     27 Watts       120 cfm      < 10 ppb     < 10 ppb      < 10 ppb




Prototype Hood Testing           September 12, 2001                        Page 26
                                TABLE 4-7M
                            TRACER GAS TESTS
                              SASH FULL OPEN
                       NOMINAL FACE VELOCITY 0.25 M/S
                         SUPPLY SYSTEM AT 0.13 M3/S

Mannequin         Supply Air                   Left           Center          Right
Height         Power       Volume

550 mm         0 Watts        0 m3/s                           10 ppb
550 mm         27 Watts       0.056 m3/s      < 10 ppb       < 10 ppb       < 10 ppb
550 mm         29 Watts       0.090 m3/s                     < 10 ppb
550 mm         33 Watts       0.11 m3/s                      < 10 ppb
550 mm         43 Watts       0.12 m3/s                         10 ppb

450 mm         27 Watts       0.056 m3/s      < 10 ppb       < 10 ppb       < 10 ppb


                                      DISCUSSION

Supply Air Balance

Although the hood opening was small, at low velocities the supply air is clearly critical to
the hood performance. The front plenum pushed any contaminate away from the
breathing zone of the operator almost providing a clean air island around the operator’s
head. However, when the balance is not perfect, the hood performs poorly.

Noise

Although we did not conduct a noise survey, the noise from the supply fans was
sufficiently loud to be annoying. Before the hood can become marketable, the noise
levels need to be addressed.

Space Above Lamps

The gap between the inclined baffle in the front of the hood and the hood super structure
does not appear to have any function. Smoke released into the space tends to stay in the
space.

Front Supply Plenum

To measure the airflow through the front supply plenum, we took off the front panel.
This resulted in a small angle, which projected out 1.25 inches. Although smoke patters
did not appear to show any effect, the angle is a departure from good aerodynamics and
may have affected the hood performance.




Prototype Hood Testing             September 12, 2001                              Page 27
Screen

The screen for the top and front supplies was distorted. Under normal laboratory use, the
screen will only be worse. Although the distortions may not have affected the hood
performance, the manufacturing methods need to be changed before the

Supply Fans

The supply fans are inexpensive plastic fans. They are not designed to operate in a
laboratory on a continuous basis.

Hood Monitor

Good laboratory design requires that each laboratory hood to have a monitor that allows
the chemist or student to monitor the hood periodically. The delicate nature of the
balance between the supply and exhaust systems indicate that the supply systems and the
exhaust should be monitored.




Prototype Hood Testing            September 12, 2001                             Page 28
                                     CHAPTER 5

                            FISHER HAMILTON HOOD


                                    HOOD DESIGN

The Fisher Hamilton hood was a prototype of the SafeAir Concept hood. The hood was a
7-foot [2.1 meter] bench-top laboratory hood with a combination sash. The combination
sash could operate as a vertical sash or as a horizontal sash.




                  Photo 5-1: Fisher Hamilton prototype Concept hood

Photo 5-1 shows an overview of the Fisher Hamilton prototype Concept hood. The
vertical sash is half closed and the horizontal windows are closed. The most unusual
characteristic of the hood is a partially self-closing sash. If the sash were raised above
eighteen inches [450 millimeters], the sash automatically closed to about eighteen inches
[450 millimeters] above the Work surface.



Prototype Hood Testing            September 12, 2001                              Page 29
According to Jon Zboralski, representing Fisher Hamilton, the design intent was to
operate the hood with a face velocity of 100 fpm [0.5 m/s] with the horizontal windows
open. With a constant velocity hood, the volumetric flow would provide a nominal face
velocity of 80 fpm [0.4 m/s] at automatically reduced opening and a nominal face
velocity of 60 fpm [0.3 m/s] with the sash held in place at the sash latch.

Sash

The combination sash has four horizontal windows in two tracks. The horizontal
windows are top hung. Two panels are 15 inches [380 millimeters] wide by 38 inches
[950 millimeters] high and two panels are 22 inches [550 millimeters] wide by 38 inches
[950 millimeters] high. The maximum opening is 34 inches [860 millimeters] wide. The
smaller sash allows the chemist to use the sash panel as a splash shield. The panels are
placed in the sash so that the chemist can always have a sash shield when operating the
hood in a horizontal mode.




           Photo 5-2: View of sash stop on Fisher Hamilton Prototype hood




Prototype Hood Testing            September 12, 2001                            Page 30
Vertical Operation

The sash moved freely vertically. However, when the sash was above 18 inches [450
millimeters] open, the sash would not remain in place and slowly closed to about the
eighteen-inch [450 millimeter] height. The position where the sash stopped depended on
the height where it was released. The higher the sash when released, the lower the sash
when it came to rest. However, the position was always no higher than eighteen inches
[450 millimeters].

Since the chemist could desire to have the vertical sash full open, the hood has a sash stop
to hold the vertical sash at 27 inches [690 millimeters].

Horizontal Operation

With horizontal operation, the sashes can be moved independently. However, the
maximum opening is 34 inches [860 millimeters] wide by 28.25 inches [720 millimeters]
high.




               Sight Panel




                                                                     Work Opening




                 Horizontal Sashes




                       Photo 5-5: Horizontal sash with sight panel



Prototype Hood Testing               September 12, 2001                             Page 31
Sight Panel

The sash has a glass sight panel near the top to allow the operator unobstructed view into
the top of the hood without increasing the volumetric flow. The sight panel reduces the
maxim open area, in the horizontal mode of operation, to 28 inches [710 millimeters]
high.

The top of the sight panel extends to the bottom of the metal panel that reduces the
bypass opening. The glass is not sealed and has a small gap if about 3/16-th inch [5
millimeters]. The bypass opening is about 2.5 inches [65 millimeters].


Baffle and Plenum

The prototype hood has the standard SafeAire Baffle, consisting of three panels. The top
and bottom are fixed. The middle panel pivots at the bottom modifying the size of the
slots. The operator can move the baffle by a lever arm in the right side wall of the hood.




                          Photo 5-5: Baffles in closed position



Prototype Hood Testing             September 12, 2001                             Page 32
The center panel pivots on its bottom edge. The top of the bottom panel extends about an
inch above the bottom of the middle panel. Consequently, the slot is horizontal. The
open distance is about one inch [25 millimeters] and is independent of the baffle position.
The top of the center panel is about one inch [25 millimeters] above (and behind) the
bottom of the top panel. When the panel is in the closed position, the center panel leans
against the top panel and there is essentially no opening. In the “normal” position, the
baffle has a one-inch [25 millimeter] horizontal slot. In the “open” position, the baffle
has a two-inch [50 millimeter] horizontal slot.




                           Photo 5-6: Baffles in Open position

The bottom slot is fixed. It is about one inch [25 millimeters] above the work surface.

The top slot is 0.5 inch [13 millimeters], but does not extend across the center of the
hood. A molded plastic piece covers the center 32 inches [810 millimeters].

The baffle panels do not extend all the way to the sides. The panels have a ¼- inch [6
millimeter] slot on both sides. The baffles are free to move side to side. The panels
could touch on one side with the other side having a ½ inch [13 millimeter] slot.



Prototype Hood Testing             September 12, 2001                             Page 33
The rear plenum is two inches deep [50 millimeters] at the bottom slot, about 4 inches
[100 millimeters] deep at the middle slot and six inches [150 millimeters] deep at the top
slot. The slots run the full height of the hood, about 48 inches [1.2 meters].

Airfoil

The Fisher Hamilton hood has an airfoil that is flush with the top of the workbench.

The airfoil consists of three major parts:

   1)      In front of the airfoil is a 1-1/4 inch [30 millimeter] diameter “Belly Bar”,
           extending about 3.5 inches [90 millimeters] in front of the sash.
   2)      The top of the airfoil is a round leading edge, about 1 inch [25 millimeters] in
           diameter, with a flat plate that extends about 2.5 inches [65 millimeters] into
           the hood.
   3)      A small trough below the flat plate.




                                                       Top of Airfoil




                                                                           Gap between
                                                                           Airfoil and
                                                                           Work Surface




          Belly Bar



                      Photo 5-7: Airfoil with the sash raised vertically




Prototype Hood Testing              September 12, 2001                             Page 34
                                                               Tray under the
                                                               Flat portion of
                                                               Airfoil




         Sash Frame




      Photo 5-8: Airfoil with vertical sash down and airfoil rotated to show trough


                              HOOD PERFORMANCE

Before we began to test the Fisher Hamilton hood, Dr. Alan George stated that we should
assume the hood the Fisher Hamilton hood would operate as a VAV hood. In the recent
ventilation remodeling of Laboratory 110, MSU equipped the room with a Phoenix VAV
system. In addition, MSU installed a controller on each hood to allow the investigator to
“dial in” the desired volumetric flow.

Although the intent of MSU is to operate the hood as a VAV hood, we tested the hood as
if it were a constant volume hood. We dialed in the required volumetric flow and
maintained the flow throughout the tests. Since the VAV system works on the Phoenix
sash position indicator, if the sash did not move the volumetric flow would be constant
and the hood would operate as a constant volume hood. However, we modified the
volumetric flow rate when we change the sash position.

Since the baffle can be modified by the user, we discussed the baffle position with MSU.
Changing the baffle would add an additional variable to the tests. Consequently, we



Prototype Hood Testing            September 12, 2001                             Page 35
selected a single baffle position. For all the testing, the baffle was in the normal or center
position.




                 Photo 5-8: Baffle adjustment leaver in Normal position


Face Velocity Measurements

We began the tests with a nominal 60 fpm [0.3 m/s] face velocity. When the hood
performed adequately, we lowered the face velocity to a nominal 40 fpm [0.2 m/s].
Finally, we increased the nominal face velocity to 50 fpm [0.25 m/s]. To set the
volumetric flow, we used the readout from the Phoenix valve for the Fisher Hamilton
prototype hood. The reported value is the valve readout. We made no effort to determine
the proper altitude correction factor for the valve.

The results of the face velocity measurements are shown in Tables 5-1 through 5-6
[Tables 5-1M through 5-6M]. The left top row was labeled A and the bottom row was
labeled C. The first column was on the left.




Prototype Hood Testing              September 12, 2001                               Page 36
                                        Table 5-1
                            Fisher Hamilton Prototype Hood
                              Face Velocity Measurements
                                   Vertical Sash Open
                             Nominal 60 fpm Face Velocity


       1               2             3                 4          5                  6

A      61              64            64                66         63                 67
B      67              63            63                63         60                 58
C      62              59            70                68         61                 25

              Average                61 fpm
              Maximum                70 fpm
              Minimum                25 fpm

Note: The lower left (position C6) reading does not appear to be correct.



                                       Table 5-2
                            Fisher Hamilton Prototype Hood
                              Face Velocity Measurements
                                 Horizontal Sash Open
                             Nominal 60 fpm Face Velocity


       Left                          Center                       Right
       1      2        3             4      5          6          7     8            9

A      60     55       57            74       69       67         58        58       58
B      61     56       60            69       64       70         56        55       53
C      64     64       68            76       75       82         72        70       56

Average       61 fpm                          72 fpm                        60 fpm
Maximum       68 fpm                          82 fpm                        72 fpm
Minimum       55 fpm                          64 fpm                        53 fpm

Note: The volumetric flow was not changed for the center position measurements.




Prototype Hood Testing            September 12, 2001                             Page 37
                                       Table 5-3
                           Fisher Hamilton Prototype Hood
                             Face Velocity Measurements
                                  Vertical Sash Open
                            Nominal 50 fpm Face Velocity

      1               2            3                 4      5                6
A     54              60           58                57     58               67
B     47              48           47                48     51               53
C     40              42           50                52     46               53

             Average               52 fpm
             Maximum               67 fpm
             Minimum               40 fpm

                                      Table 5-4
                           Fisher Hamilton Prototype Hood
                             Face Velocity Measurements
                                Horizontal Sash Open
                            Nominal 50 fpm Face Velocity

      Left   315 cfm               Center 285 cfm           Right   315 cfm
      1      2      3              4      5      6          7       8      9
A     39     40     45             48     48     41         47      46     44
B     38     44     51             44     43     48         44      46     43
C     41     57     59             50     56     50         55      46     46

Average      46 fpm                         48 fpm                  46 fpm
Maximum      59 fpm                         56 fpm                  55 fpm
Minimum      38 fpm                         41 fpm                  43 fpm

                                       Table 5-5
                           Fisher Hamilton Prototype Hood
                             Face Velocity Measurements
                                  Vertical Sash Open
                            Nominal 40 fpm Face Velocity

      1               2            3                 4      5                6
A     51              47           52                46     48               50
B     44              41           45                45     45               47
C     29              41           35                31     37               50

             Average               44 fpm
             Maximum               52 fpm
             Minimum               29 fpm




Prototype Hood Testing           September 12, 2001                       Page 38
                                         Table 5-6
                              Fisher Hamilton Prototype Hood
                                Face Velocity Measurements
                                   Horizontal Sash Open
                               Nominal 40 fpm Face Velocity

       Left                           Center                      Right
       1      2        3              4      5        6           7         8        9
A      42     39       43             42     40       37          38        36       42
B      35     39       43             39     39       39          41        39       44
C      37     41       54             40     38       40          46        41       47

Average       41 fpm                         39 fpm                         41 fpm
Maximum       54 fpm                         42 fpm                         47 fpm
Minimum       35 fpm                         37 fpm                         36 fpm




                                        Table 5-1M
                              Fisher Hamilton Prototype Hood
                                Face Velocity Measurements
                                     Vertical Sash Open
                               Nominal 0.3 m/s Face Velocity

       1               2              3               4           5                  6
A      0.31            0.33           0.33            0.33        0.32               0.34
B      0.32            0.32           0.32            0.32        0.31               0.29
C      0.31            0.30           0.36            0.34        0.31               0.11

              Average                 0.31 m/s
              Maximum                 0.36 m/s
              Minimum                 0.11 m/s

Note: The lower left (position C6) reading does not appear to be correct.




Prototype Hood Testing              September 12, 2001                           Page 39
                                     Table 5-2M
                           Fisher Hamilton Prototype Hood
                             Face Velocity Measurements
                                Horizontal Sash Open
                            Nominal 0.3 m/s Face Velocity

      Left                        Center                         Right
      1      2      3             4      5       6               7     8      9
A     0.31 0.28 0.29              0.38 0.35 0.34                 0.30 0.29 0.29
B     0.31 0.28 0.30              0.36 0.33 0.36                 0.28 0.28 0.27
C     0.32 0.33 0.35              0.39 0.38 0.42                 0.37 0.36 0.29
Average      0.31 m/s                    0.36 m/s                      0.30 m/s
Maximum      0.35 m/s                    0.42 m/s                      0.37 m/s
Minimum      0.28 m/s                    0.33 m/s                      0.27 m/s
Note: The volumetric flow was not changed for the center position measurements.

                                       Table 5-3
                           Fisher Hamilton Prototype Hood
                             Face Velocity Measurements
                                  Vertical Sash Open
                            Nominal 50 fpm Face Velocity

      1             2              3             4           5             6
A     0.27          0.30           0.29          0.29        0.29          0.34
B     0.24          0.25           0.24          0.25        0.26          0.27
C     0.20          0.22           0.25          0.26        0.23          0.27

             Average               0.26 m/s
             Maximum               0.34 m/s
             Minimum               0.20 m/s

                                      Table 5-4
                           Fisher Hamilton Prototype Hood
                             Face Velocity Measurements
                                Horizontal Sash Open
                            Nominal 0.25 m/s Face Velocity

      Left   315 cfm               Center 285 cfm            Right   315 cfm
      1      2      3              4      5      6           7       8      9
A     0.20   0.20 0.23             0.24 0.24 0.21            0.24    0.23 0.23
B     0.19   0.23 0.30             0.22 0.22 0.24            0.23    0.23 0.22
C     0.21   0.20 0.30             0.30 0.28 0.25            0.28    0.23 0.24

Average      0.23 m/s                     0.25 m/s                   0.24 m/s
Maximum      0.30 m/s                     0.30 m/s                   0.28 m/s
Minimum      0.19 m/s                     0.21 m/s                   0.22 m/s



Prototype Hood Testing           September 12, 2001                        Page 40
                                         Table 5-5M
                               Fisher Hamilton Prototype Hood
                                 Face Velocity Measurements
                                      Vertical Sash Open
                                Nominal 0.2 m/s Face Velocity

       1                2              3            4             5              6
A      0.26             0.24           0.26         0.23          0.24           0.25
B      0.22             0.21           0.23         0.23          0.23           0.24
C      0.15             0.21           0.18         0.16          0.19           0.25

                 Average               0.22 m/s
                 Maximum               0.26 m/s
                 Minimum               0.15 m/s


                                         Table 5-6M
                               Fisher Hamilton Prototype Hood
                                 Face Velocity Measurements
                                    Horizontal Sash Open
                                Nominal 0.2 m/s Face Velocity

       Left                            Center                     Right
       1         2      3              4      5     6             7       8      9
A      0.21      0.20   0.22           0.21 0.20    0.19          0.19    0.19   0.21
B      0.18      0.20   0.22           0.20 0.20    0.20          0.21    0.20   0.22
C      0.19      0.21   0.27           0.20 0.19    0.20          0.23    0.21   0.24

Average          0.21 m/s                     0.20 m/s                    0.21 m/s
Maximum          0.27 m/s                     0.21 m/s                    0.24 m/s
Minimum          0.18 m/s                     0.19 m/s                    0.29 m/s



Smoke Visualization

We tested the hood with smoke while the horizontal sash was open. The smoke patterns
were similar for each face velocity. However, with the lower face velocities the motion
was not as distinct.

Top of Airfoil

Smoke released on top of the airfoil, with or without the mannequin, was lazy (see Photo
5-9) on the airfoil. It tended to drift up past the mannequin toward the breathing zone.



Prototype Hood Testing               September 12, 2001                          Page 41
                          Photo 5-9: Reverse flow on the airfoil


Under the Airfoil

Smoke released under the airfoil curls back above the airfoil. Although the air is “clean”
air, it shows a lack of aerodynamics.

Smoke released between the belly bar and the airfoil tended to rise and run up the front of
the mannequin. At times, this flow induced reverse flow at the airfoil and entrained some
smoke at the airfoil.

Work Surface

Smoke on the work surface was lazy, especially at low volumetric flows. It tended to lay
on the work surface, sometimes with a slightly forward motion, then migrate toward the
rear slot.

When the hood baffle control was in the “Open” position (the middle slot was open
cutting off flow to the bottom), the air was lazier and had frequent reverse flow. Smoke


Prototype Hood Testing             September 12, 2001                             Page 42
on the work surface rose toward the center slot rather than moving toward the bottom
slot.

With the hood in the “closed” position (the center baffle was closed), the hood
demonstrated an improved sweeping action at the work surface.

Hood Opening

Smoke entered the hood is a smooth controlled manner.

Sash frame

The sash frame extends into the workspace by about 1.25 inches [30 millimeters]. Smoke
released at the side between the airfoil and the sash frame moved toward the sidewall,
then forward to the sash frame. Once it approached the sash frame, the smoke rose about
half way up with some eddying forward on the inside lip of the frame. Although the
smoke moved forward, it did not pass beyond the plane of the sash.

Behind the sight panel

Smoke released behind the sight panel tended to rise into the upper portion of the hood.
As it rose, it entered air entering into the flow of bypass air entering the hood.

Behind the vertical sash

Smoke release behind the open vertical sash tended to fall toward the bottom of the sash.
Occasionally, as the smoke approached the bottom of the sash, it eddied out to, and
possibly beyond, the plane of the sash.

Tracer Gas Tests

For each face velocity, we used the Phoenix controller to set the volumetric flow and
conducted a face velocity. The setting on the Phoenix control was noted. When we
wanted to return to the same face velocity, we set the controller and assumed the face
velocity was the same as previously measured.

With the Fisher Hamilton prototype hood set to provide a nominal face velocity of 60
fpm [0.3 m/s], we measured the tracer gas levels. We released tracer gas, sulfur
hexafluoride, at 4 lpm and monitored in the breathing zone of the mannequin. The
ASHRAE standard required a mannequin height of 26 inches [650 millimeters] above the
work surface. Because we conducted the tests on the LBNL prototype hood with the
mannequin’s breathing zone 22 inches [550 millimeters] above the work surface, we used
the 22-inch [550 millimeter] mannequin for the Fisher Hamilton prototype hood.




Prototype Hood Testing            September 12, 2001                             Page 43
                                  TABLE 5-7
                              TRACER GAS TESTS
                     NOMINAL FACE VELOCITY 60 FPM [0.3 M/S]

Sash Position          Face Velocity       Left           Center        Right

Vertical               67 fpm [0.34 m/s]     < 10          < 10           < 10
Vertical               61 fpm [0.31 m/s]     < 10          < 10           < 10
Vertical half open                           < 10          < 10           < 10

Horizontal             61 fpm [0.31 m/s]     < 10                         < 10
                       72 fpm [0.37 m/s]                   < 10

Note: The center position test in the horizontal position did not modify the volumetric
      flow resulting in a higher face velocity.




                              TABLE 5-8
                          TRACER GAS TESTS
                 NOMINAL FACE VELOCITY 50 FPM [0.25 M/S]

Sash Position          Face Velocity       Left           Center        Right

Vertical               52 fpm [0.26 m/s]     < 10          < 10           < 10

Horizontal             46 fpm [0.23 m/s]     < 10                         < 10
                       49 fpm [0.25 m/s]                   < 10



                                  TABLE 5-9
                              TRACER GAS TESTS
                     NOMINAL FACE VELOCITY 40 FPM [0.2 M/S]

Sash Position          Face Velocity       Left           Center        Right

Vertical               44 fpm [0.22 m/s]     < 10          < 10           < 10

Horizontal             42 fpm [0.21 m/s]     < 10                         < 10
                       39 fpm [0.20 m/s]                   < 10




Prototype Hood Testing            September 12, 2001                             Page 44
                                     DISCUSSION

Airfoil Design

The airfoil is very flat. It is almost an extension of the work surface. The chemist will
have a tendency to place apparatus and containers on the work surface. With the
extended work surface, the potential source of contamination will not be as well
controlled as if the work were conducted inside the hood. Good work practices
recommend that the chemist conduct all work at least six inches into the hood.

Belly Bar

The airfoil, on the hood, is inboard of the sash. Since a chemist tends to position
himself/herself at the opening of the hood defined by the sash, the chemist would lean
into the hood. The hood has a 1-1/4 inch bar extending five inches in front of the airfoil.

Movable Baffles

The hood has an adjustable baffle. The tests were conducted with the baffle in the
normal position. Since the baffle can be adjusted easily, the chemists should be
instructed in the use f the baffle. Although many hood manufacturers advertise that the
hoods can be adjusted for heaver and lighter than air solvents, most applications in
chemical laboratories do not have this potential. The low solvent releases result in a
concentration usually less than a few hundred ppm. At these concentrations, the
difference in density is not significant.

Auto Sash

The automatic sash closure is a mixed blessing. Although automatically lowering the
sash, with a VAV system, allow for some energy conservation, the inability of operating
with the sash open above 18 inches [450 millimeters] could be a disadvantage in using
the hood.

Spill tray

The small spill tray under the airfoil has the potential of accumulating spills. The tray
could become a secondary source of volatile contamination. A more serious problem
could develop is incompatible chemicals are spilled into the tray.

Sash Frame

The sash frame was oversized. The frame extended into the hood opening and caused
turbulence as the air enters the hood. The turbulence resulted in reverse flow. Although
the reverse flow did not appear to break the plane of the hood, the turbulence could have
been removed by recessing the frame into the sidewalls.




Prototype Hood Testing             September 12, 2001                             Page 45
Differences from Concept Hood

The prototype hood installed at MSU has been modified by Fisher Hamilton. According
to Jon Zboralski, the new Concept SafeAir hood includes the following modifications:

   1)     The rear plenum is deeper (2.75 inches [70 millimeters] vs. 2 inches [50
          millimeters]).
   2)     The bottom slot is larger (2.75 inches [70 millimeters] compared to the
          prototype that had a 1 inch [25 millimeters] slot).
   3)     The side slots are “minimized.”
   4)     Top baffle plate moved forward several inches.
   5)     The top slot now extends the full width of the hood.
   6)     The new hood has two fixed baffle panels rather than the three in the
          prototype
   7)     The baffles are fixed.
   8)     The airfoil and sash were modified to allow plugs to enter.




Prototype Hood Testing          September 12, 2001                          Page 46
                                     CHAPTER 6

                                  SPECIAL TESTS


The ASHRAE ejector releases tracer gas through a screen approximately 14 inches [350
millimeters] above the work surface. In addition, the tracer gas has a vertical momentum.
Consequently, the challenge to the hood is in the upper region of the hood.




             Ejector Bonnet




                              Figure 6.1: ASHRAE ejector

On the other hand, both prototype hoods demonstrated slow removal of smoke from the
work surface with occasional reverse flow. In addition, both hoods drew a large
percentage of the exhaust through the top of the hood. Consequently, both hoods were
strongest in the upper regions of the hood.

In an effort to challenge the weaker portion of the hood, we modified the tracer gas
delivery in three ways.



Prototype Hood Testing            September 12, 2001                             Page 47
Inverted Ejector

We turned the ejector upside down and ran the tests in the Fisher Hamilton prototype
hood. We began the tests with the face velocity at a nominal 40 fpm [0.2 m/s] and
increased to 50 fpm [0.25 m/s] and 60 fpm [0.3 m/s]. All the tests were in the center
position. Table 5-1 shows the test results.



                                                TABLE 6-1
                                    FISHER HAMILTON PROTOTYPE HOOD
                                            TRACER GAS TESTS
                                            INVERTED EJECTOR

Sash Position                            Face Velocity               Left                Center                 Right

Horizontal                               42 fpm [0.21 m/s]                               1200 ppb
Horizontal                               46 fpm [0.23 m/s]                               2500 ppb
Horizontal                               62 fpm [0.31 m/s]                               1800 ppb


Figure 6-1 shows the concentration over time. The high peaks and low valleys are typical
of significant spillage from a laboratory hood.



                                                    Figure 6-1
                                            Fisher Hamilton Prototype
                                                 Inverted Ejector
   Concentration, ppb




                        10000
                         8000                                                                                       40 fpm
                         6000                                                                                       50 fpm
                         4000                                                                                       60 fpm
                         2000                                                                                       Reference
                            0
                                                         116
                                                               139
                                                                     162
                                                                            185
                                                                                  208
                                                                                        231
                                                                                              254
                                                                                                    277
                                                                                                          300
                                    24
                                          47
                                               70
                                                    93
                                1




                                                         Time, seconds




Prototype Hood Testing                               September 12, 2001                                                 Page 48
Some of the test results were above the calibration of the instrument. In previous effort,
we demonstrated the meter is nearly linear for a significant range. When the instrument
looses linearity, the slope of the curve increases so that the readings on the instrument are
lower than true concentrations. Consequently, we extrapolated the calibration curve and
estimated the concentrations. For readings above 2000 ppb (2 ppm), the extrapolation is
at best precarious.

Hitchings Ejector

The ASHRAE committee, SPC 110, reviewing the current standard has discussed
alternate approaches to the ejector design. Dale Hitchings has suggested a modified
ejector that consisted of four 12-inch [300 millimeter] long porous tubes used as air
bubblers for aquariums. He constructed an ejector, as seen in photo 5-5.

When assembling the Hitchings ejector, we observed a very high background. We
assumed we spilled tracer gas into the laboratory while moving the vertical array around.
During the calibration of the flow rate, the room concentrations increased significantly.

While we conducted the first test, the concentration increased slowly toward the end.
After we turned off the gas, the concentration continued to increase, about three fold.
The high level was through out the room, not just at the hood.

Upon investigation, the wind blew from the exhaust stack toward the Hastings air-
handling unit. Consequently, we considered reentry a possibility. However, the design
of the stack and the location of the air-handling unit should not have been a problem.

We repeated the tests. When we observed elevated tracer gas results, we took the
detector outside abut could not detect any tracer gas at the intake to the Hastings unit.
When we returned to the laboratory, the concentration in the laboratory was still
elevated. Further investigations showed low tracer gas levels in the corridor outside the
laboratory indicating that the problem was not reentry.

Because the tracer gas levels were high at the floor, we followed the high concentrations
to the cabinet under the hood. When we placed the probe inside the cabinet, the
concentration exceeded 66,000 ppb (66 ppm)!

The route of traverse is clear. With the low ejection rate, the pure tracer gas, and the
discharge directly on the work surface, the heavy tracer gas “fell” through the crack and
filled the cabinet. Through diffusion, the tracer gas entered the room with the effect
similar to reentry.

Due to the problems associated with the spillage through the crack, and the length of the
time to remove the tracer gas from the laboratory, we discontinued the tests of the Fisher
Hamilton hood with the Hitching ejector.




Prototype Hood Testing              September 12, 2001                              Page 49
We performed preliminary tests of the LBNL hood with the Hitchings ejector. We tested
the hood with the mannequin at the center of the hood. Tables 6-3 and 6-4 show the test
results.



                                    TABLE 6-2
                             LBNL PROTOTYPE HOOD
                            USING HITCHINGS EJECTOR
                         22-INCH MANNEQUIN AT CENTER

       Face Velocity                 Supply air                           Tracer Gas

       60 fpm [0.31 m/s]             0 cfm          [0 m3/s]              3700 ppb
       60 fpm [0.31 m/s]             118 cfm        [0.056 m3/s]           170 ppb
       60 fpm [0.31 m/s]             254 cfm        [0.12 m3/s]            < 10 ppb


                                    TABLE 6-3
                             LBNL PROTOTYPE HOOD
                            USING HITCHINGS EJECTOR
                         22-INCH MANNEQUIN AT CENTER


Face Velocity                Supply air                    Tracer Gas

60 fpm [0.31 m/s]      118 cfm       [0.056 m3/s]          4700 ppb
53 fpm [0.27m/s]       118 cfm       [0.056 m3/s]           855 ppb
44 fpm [0.22 m/s]      118 cfm       [0.056 m3/s]           170 ppb


Conclusion

The preliminary tests results clearly indicate that the LBNL prototype hood, and possibly
the Fisher Hamilton prototype hood, performance depends, in part, on the location of the
tracer gas release. Further investigation of this phenomenon is warranted.




Prototype Hood Testing            September 12, 2001                             Page 50

								
To top