Document yv8MNDxyYdR84DJE9rLV9zpd
(a) To letamino tho magnitudo of the health hazard arising from the use of leaded gasoline for fuel in the flro unit: and secondarily,
(Is) To do terraSri tho effective length of iff of tho filter discs'used to remove the load.
77. tor.clesions
.p7
I. do -health hazard exixsts frora the breathing of load in
air w:.on toe atovo is operated in re cam or oncluauroe under norra&l
conditions of ventilation. :.rea.ll amounts of load scape through the
;jrner. a emu11 port Ion of tills Xoiud will be deposited on the surface
of cool cooking recoptaeloc while tho remaliiacr will scape into the
air.
d. There Is no danger that food prepared in receptacles
ill bo contaminated, but the unit should not be used to proper
food which cb*soa into direct cor;tact with tho flan (broiling or -
rilling).
J. vifch a rated fuel of 1.7? cc. of tetraethyl lead par
/ alien, satiisfiictory operation v;Tfch a o;inlsum of cleansing was
obtained for 111*.76 hours with a single filter unit.
III. Procedure and 7iscusslon fn order to ascertain the fate of all tho lead and thereby
tfE" 0009469
V.%;3^i'i/w:V:V'
\ \
flue gas produced and to determine Its load content. The equipment
employed for this purpose was similar to that described in a report
dated Juno 1.,
except that a larger exhaust fan was installed,
tho us of which effectively prevented buck drafts on windy days*
art orifice disc for measuring flue volumes was also installed in the
flue pip. The volumes of flu gas as asouaurod by this means .sr
employed to cheek the values obtained indirectly by analysis of the
carbon dioxide content of the flu pas. (bindings based on'the latter
values have been included in this report as in previous reports.)
'That this method was rut! factory Is des&onstrated in Table , i n
which, unions other data, tho total load content of fcha flu paa has
boon calculated from the volumes arrived at by the two methods of
measurement",.
In ''able 1 arc listed the results on the five aeries of
s
testa, D, h, >', G, 44-, .md with tho presont Coleman model, ferics G
was the ono employed for the. balance study roported in Table U. The
other data listed in Tables 2 and wore also obtained from this
sorios*
All runs except those of series cl wore made with-.fuel
containing 2,22 cc. of tetraethyl load per gallon. This type of
fuel was no longer available, however, and since it was convenient
to use a typo which was on tho v.arkot, all run In serloa 0 wore mad
0009470
lead removal was Inadequate. This inefficiency was attributed to lack of sufficient heat at the generator to cause pyrolysis of the tetraethyl load. Operation of the burner was continued, however,, until a pd-pound pressure failed to give satisfactory perforiRanco at the end of 22 hours. During this run the burner slots became clogged, with a yellowish deposit and had to bo cleaned several titles. .lien the run 'as finished, the vapor tube was almost completely filled with & grayish deposit, which bad practically closed its two .screens
cries K
;>1 of this series was carried out to dotern1no Whether the
high results obtained in series D worn duo to variability in the
porosity of the Individual filters* I n this run the burner was not
cleaned, but a clean vapor tube &r;b a flame valve were employed. The
result was in agreement with that reported for sorie D# '>2 was
made with the sane disc that had been employed, in i->l, but more heat
was applied to the generator heaters by removing all but the Inner
circle of closures in tiro heater outlets of the generator, thoreby
increasing the number of open Inga from 22 to 37 In tills tost the
h u m o r slots wore cleaned before tto unit was lighted.
nay b
aeon from Table 1 there was u alight improvement in performance.
rles {' "lie two touts in this aeries were ."ad for. the purpose of
000947,1 . ,
gradual volatilization of lead deposited In the burner following the tests In series D. hxcepfc for occasionally scraping the burner Slots,
f no attempt had been made to clean the burner itself since the start of series T.
Tories 0 .'oforo series 1 to* to wore star todf the burner, vapor tub,
and M a s e valve were cleaned as thoroughly a possible by scraping away ull deposited scale, following this by treatment of tho contam inated surfaces ith hot dilute nitric acid and a h o t 20' solution of arawnla! acetate, "'he cabinet, hood, and piping equipment were also dismantled and the surfaces were thoroughly cleaned first by scraping with a wiro brush and then by wiping down the surfaces with clean rags. For this tost, as we have stated above, the only fuel available was that containing 1.?? cc. of tetraethyl lead per gallon, .a in the previous tests, each period of operation extended over 5*5 hours, the unit being allowed to cool to room temperature before the noxt test was started. In order to operate the unit as consistently as possible, the flue aa temperature was observed by means of a pyrometer and the tomperature v.aa kept practically constant by %anIpulating the air pressure and flame valve* The temperature control was chosen so as to give a constant consumption of 0.L0 gallon of fool per hour. The unit operated satisfactorily for a period of 1ll.. hours- at air
K o o o m m
request was received that the test be continued until a pressure of 50 pounds was,required Tor operation. (The unit had still been operating at pressures below ^0 pounds*) The cabinet and hood wore therefore reassembled and the period of the test was extended to 11?;.76 hours, "his. waa done with the uncloaned burner and disc, the only clean parts installed being, a vapor tube ami flarr.o valve* at ..the cifici of this period, the stove and hood equipment were Again dismantled and thoroughly cleaned, "'ho burner, vapor tut-ss, and flame valve were cleaned as before at the start of the test, while the hood, cabinet, and piping were wiped down with strips of filter paper* --.11 deposits and solutions wore rosorvoei for loud analyses, fho rvsul La of which are recorded In Table h* Two sets of values' ;*r given to enable comparison of the adequacy of tho methods employed in determining the volume of flu vases* In this table correction ia m a o for the lead present in the ns*!! quantity of fuel drained at the start of each run after th enex'ator had attained proper operating tempera ture. correction of the amount of lead In the flue gas had also been made for the amount deposited on the walls of th cabinet, hood, and piping up to the locus of the precipitator tube*
In Table 2 arc listed the values for lead found in tho flu gas for the individual periods of operation. These values have not been corrected for the amount of lead lost an deposit on the
K 0009473
filter by the lead salts, Longer operation caused a slight increase In the lead -passed by the burner, 'This slight decrease in efficiency say be due to a number of causes: 1, The gradual filling-up of tho lower portion of the generator, plus
the heat, way cause tho filter case gradually to expand, until it practically seats itself over the generator heater outlets. This causes a drop in tho heat output at tho heater outlets, and Impairs the pyrolytic efficiency of the generator. This condition can ho remedied by installing a pin over each heater outlet to keep the generator from seating itself over tho heater outlets,. 2 , Tho lead deposited on the lower aurfacc of the generator acta as a heat insulator and thus prevents the attainment.of the necessary pyrolytic temperature. This night be restedSod by Increasing the si of tho generator unit or by providing the generator with a ' baffle plate which would keep too deposit off the lower wall. The use of the latter would also tend to prevent wetting of the - filter disc by the boiling gasoline. It my very well be that the latter condition causes the passage of a email amount of liquid fuel through the filter, thus tending to Increase slightly the quantity of load passed Into the air*
in Table 3 are recorded the average figures for the passage of load into the flue gas during successive -periods of u 5*5 hour run. The verugoa for comparable periods in 32 completed 3.5 hour runs were employed in compiling those data, Tho table also records the
accepted .maximum safe limit of 1*5 mg*, per 10 cubic we tera Is also
indicated Thus it may be seen that whenever ventilation la necessary
(hen the unit is operated in enclosed spaces) considerably more ven
tilation will be required during the first hour than during any sub
sequent period*
In Interpreting the dilution data, It should be borne in T-
wind that a forced draft was employed in these tost and teat under
normal operating conditions thrro will .be a smaller volume of flue gas
with a consequently higher lend coneantratlon Ventilation under
these conditions
i. be avers*! tines that Iruilcatod in tho table*
bun J
'
Pollowing tho tost in serios 0, a J*5 hour tost was wad
with the thoroughly cleaned unit and a new Johns-baaovllle filter.
In this cane fool containing 2 ,o2 co* of tetraethyl lead por gallon
was used and tho conditions of operation wore the sets ua In the 0
series* Those data offer additional proof that the poor performance
in tho case of orlos D, . and P was due to Insufficient heat at tho
generator heaters arid to volatilisation of lead which fiad deposited
in the burner buring tests in serios D* On the basis of the results
of series -J {'"able 2.) and tost J, it m y to predicted that pox'fornanco
would improve with length of operation of the same disc and that the
turner would be as effective with fuel of higher load content as with
fer-r lower*
K 0009475
least after ovary ? hour of operation* After a tost of 21 hours
duration, it as necessary to Insert a clean vapor tube and flat
valve.
tn series f- and P , no cleanings were required and. operati on
were satisfactory#
tn serios <l9 operation was very satisfactory* >arnr slots
\ vapor txi.be and flame valve were changed after
nspoetIon showed t h & t t i l vapor tubo 5vaa still
'liter screens.
/< * a
is time,$ > -
*#.< #
for reasons dascri bod
disc was removed
the unit. ..hen the reques 4K.t
n^ 1 1nue th o t e s t t h e CI <? Q was replaced but could
,n place because o f ga lllnb; of the bolt provided
for this purpose. On this account, a new bolt had to be ;-mde.
In series no operational difficulties were encountered
except at the start when the now disc was inserted. This did not
.fit properly Into Its copper seat and u entail leak developed. On
inspection, it was seen that the seating b*ad occurred ufc an angle
which gouged out sufficient matrial to cause the leak, a now copper
seat was inserted and the teat was satisfactorily concluded. In
order to avoid this difficulty in the field, it might bo advisable
to incorporate 5.nto the burner casting & series of pins so placed
that;'the filter disc would just fit inside fchm. This would provent
000947G
Reported by* f. Cholak
f\. >.ok/ary u* Younker * Tu^^oy
pproved t . Kobert A. cedo, director
l;ate: Kay 2h, 19^3
H g 000947T
0008478
"alio 1
l-saur, of ocsulKa
ill VCflfc5l
^,-Te$t Apit-S\V S* ^Series ;r?1 8-1
e-2
p-l r-2
ah-feries<
1
;ur&tion of Tost
hours }'*'
5-5
5-5 3-5,
5.5
Oallons Fuel t-sad io.Ii5 1*73
1.59
1.73 1.61
*7-9
l.liii
Gallons per Four
3.1+75 0.1+9
0.1*53
0.1*95 0.1*7
o.ij17
0.1*12
bead in Fuel Used
Grar.s
Lead in Flue has
Gr&na
31.5
7.5*
5-3 1.02
8-73
0.67
5.16 l.iin
o./U c.629
vF).57(a) !i.55U !
1;.23
0.23?
------ 1
% of Total rb In
Flue das
Eemarks
25.8 19.2 18.2 15.35
Only outer circle of heater outlets of fSee+arator used Uev filter, cleaned vapor tube s.nd i' lm n valve but burner not cleaned, beater outlets sum as above. 3*uso filter as K-I, burner slots cleaned, and all but inner circle of heater out lets of generator opened. Openings increased fro* 2
.to 57. ..... ........ -- rr dame as 2-2 except now fil ter disc containing now Johna-aansvi11 filter*
I2 .u5 5.07(a) 6.rr
%&&& as above but Lorloo'
filter used.
burner, vapor tube, and ,
flam valve cleaned as well
as hood, piping, and eabi-
not. Fro ah' filter* bal
ance study.
iquip.TiQnt cleaned follow
ing balance study, new
;
filter disc. .
:
This series of tests with 1*77 cc. Fb(Ft)4/alien fuel. All otrier runs with 2 .b2 cc. ffciFt/i/galloo fuel*;
i a ) Those values are not corrected for lead In fuel brained from lighting valve at start of each 3*5 hour
run or for load recovered la hood, cabinet, and piping* Only total load recovered in flu gas as
:
calculated fro electrolytic precipitator tul-o analyse lias been used*
'
** i, A-C .-D a-- s .j--p i-O A-H A--7 ' ?-- '. A-L J--P A-- A 0-0 A-P i-i A-0 Ar -U ,:-V
A-X -Y - > ** - '**AA ;-m fi i-CC '-- 1):> Tl*?
;-oo A-AH -JJ
3*5 3*5 5*5 3*5 3q 3*
3*5 3*3 3*5 3*57 3*5
/*rJ/;oJ 3*3 3*5 3*5 >5 *?^*****5t5s" 5*5 3.*5**
-37/* 5fV* J3?**5r 3*5 ;-o 3o 3*5 3.6 n ri
7.0
10.5 4.0
17.5 20.9 4*5 y .0
31.5 35.0
36.5? fi2.07
t -57 !,3.16 52.66
56.16 Yr,07*LLP 65.16 60.06 70.16 ? A .6O 77.16 ;.0,66 O't, .l-` 37.66
91.16 93.66-
yfi.l 101.66
105.16 103,66 112.26 II6.76
5.90 7.05
5.65
Lyo
'
k .h 5.33 5.3S
M3 1*45 lt.50
h*4 3*y?
A .17 6.20
li..22
3.63 5.36
3.59 n!..9-si3.
2 .50
5 *05
j.05
5.05
f5/T*<3&7{
5.33
6..B
7.10
6.30
7.03
' 3 *60
AVg, 5<07
20-22 lb. ' 20-23 lb. .20-23 lb.
23-30 lb. 25-23 lb.
25-35 lb. 26-55 lb. 30-55 lb.
50--56 lb. 26-55 lb.
2-3J4. lb.
26-3I4 lb. 23-56 lb.
28-58 lb.
26-35 lb. 6-38 lb. 26-57 lb. 26-/<0 lb.
30-1*1 lb. 30-62 lb.
35-3 lb. 4'-i.* lb. . 30-39 lb. 50-fa lb.
33-4 lb.
? H 4 lb.
lb.
*6 - 4 lb. 36-67 lb. 35-50 lb. 56-5 lb.
-iiO-50 lb.
At fchle point it fm crlglnal'ly deolded to end test* flood and pipng or cloar.et, but test vas continue to '}0 lb. presauro or, roquest of th :f;'ice of tbo uurtarsater tonerai.
# 0009479
K F OO034SO
#
:7M ir ntervul Inntos
15
15 JO
60
60
30
Tblo }
Data. for bead Distribution in bine bun for u 3.5 bour Verst of' Series 0 (Values based or Averages for 3d Complete 3*5 2our buna)
rb in blue Gas
U 5*73 12.62
17.13 27.82 28.18 11.67
t of Total ?b in Fine Gas
32.95 9.C9 -
. 12.31-
19.76
17.If2 . ip*^4lj-
Mg. Pb in d u e bus per Hour
75.68 27*1+2 ilii* 23.36
t . Pb/lG (.U . M O .J&8
17.81 hJ
3*26'
9. 9 5 ' 3.62
5.18
3.08
Dilution Of Flue Cm bequired to Reduce
Pb Concentration to 1.3 ic. Fb/10 Cu. U
11.6 v
5.2 (
2.I7J
6*6
2.12 2.06
ai o 138.75 (O.I3B8 grass)
100.00
t v .-i
fei
{ V?>
idA-verage amount of lead In fu el consumed por-p.b ncur test c. 7p5 ^rosta
_
f&vera^o per cent of total lead in flue as * 5*7
5.
'Average amount of flue gas per hour * z6Q 0 cu. ft. 75*8 cubic meters
k /t 000548
ralanco
Gallons of fuel used
M1 Load In fuel used
,-j.y
Load in fuel drained from ,: starting valve and not used
if?.$ 6-9.57 &
0.it2 g.
f Corrected lead i n fuel used
Lead found in flue ga {precipitator samples)
Lead found in cabinet, hood, and flu piping
Total Load in Flue Has
Load in vapor tub and flasse valve
^Juead in burner part
Lead In filter disc and housing
Lead unaccounted for
wr .Tp^tal load recovered W:
4
0: % A -- based on amount of flue gas as calculated
B -- hasad on amount of flue gas as calculated
i.
"i;lo 1:
t t t l y o r . i G V l o o Jl
9* 15 t*
h +Js '
A
5*11 t* oJil t* 1.77 6* 7 2 .10
.'5 .73# 9 .U5.2 19a > .90
1.7 6 s* . 1.96,1
I9 .15 37.59 E*
10 0 .00^ 9&.QU
fros C., armlyis fro orifici readings
5*^5 s* !*1 s* 1.77 * 72.10 a>* l.'s2 &
6.11, 9.t5, i. 98; 80.90;= I.53;
39 .15 S * ? .75 t
10 0 .0 0 9 9a .1)2^