US2591760A - Gas analysis apparatus - Google Patents

Description

April 8, 1952 w. M. ZAIKOWSKY 2,591,760

GAS ANALYSIS APPARATUS Filed Nov. 22, 1944 Absorber- 1 13 a 11 13a I l 14 16 I 43 a Exbaus 46 47 fiZAQ/M/BAZ Zwzowazg I N V E NTO R.

ATTO RN EY.

Patented Apr. 8, 1 952 UNITED" ST-ATESPATENQT' OFFICE. Q

GAS ANALYSIS APPARATUS Wladimir M. Zaikowslgy, Pasadena, calm, as-

signor to Nina D. Zaikowsky Application November 22, 1944, Serial No. 564,645 7 Claims (01. ra -27 This invention .relates to the anaylsis of gaseous mixtures by thermal-conductivity measurements and is particularly useful in connection with the analysis of products of combustion of hydrocarbon fuels or coals for'the purpose of determining the ratio of fuel to air. Some difficulty has been encountered with known methods of anaylzing products of combustionby thermal-conductivity (hereafter referred to as T. C.) measurements because although oxygenof the atmospheric air used for combustion combines with the fuel to form water and carbondioxide, heretofore it has :been proposed to eliminate the water before conducting the exhaust gas to the T. C. cells. The elimination of the water from measurement is objectionable when the ratio of carbon tohydrogen in the fuel varies as it frequently does in different gasolines and other fuels, both liquid and solid. Conventional T. C. apparatus adjusted to show theoretical mixture with a fuel of one H/C ratio will give an erroneous reading with a fuel of a different H/C ratio, because the percentages of carbon dioxide in the exhaust will be different with the different fuels for said theoretical mixture. This is because conventional apparatus indicates only the amount of carbon dioxide formed in combustion, and thus the amount of oxygen consumed in the formation of water of combustion remains unaccounted for. Hereto- -fore the apparatus usually was not adapted for the measurement of the water of combustion in vapor form within the cell; besides water and carbon dioxide have opposite effects on the T. C.

of exhaust and so obscure each other. Furthermore, the water in the exhaust includes not only water of combustion but also water of atmospheric humidity, and, without special correction, the error resulting from the presence of atmospheric humidity can becomevery serious.

An object of the present invention is to determine, by T. C. measurements, changes in the total percentage of combined oxygen inthe exhaust resulting from combustion of a hydrocarbon fuel and air. 7

T. C. exhaust H/ C ratio of the fuel being burned.

;; Another object is 'to account for, in T. C.

measurements, that portion of the water in ex- ;haust gas; that is water of combustion as dis- 2 measurements, the sum quantity of two components having opposite efiects on the thermalconductivity of a gas mixture.

Another object is to reduce contamination of T. C. cells in the analysis of exhaust gas. 1

Other more specific objects and features of the invention will appear from the detailed description to follow.

In accordance with the invention, I determine the sum of two antagonistic effects produced by different components of a gaseous mixture, such as, water vapor and carbon dioxide in exhaust gas bymaking their efiects additive in the measuring circuit. One way of doing this is to supply the gas containing the two constituents to a cell of unit sensitivity in one arm of a Wheatstone bridge and simultaneously apply a dried portion of the gas to acell of substantially double sensitivity in an adjacent arm of the bridge. The efiect of carbon dioxide in the sec 0nd arm of the bridge will be substantially double that in the first arm, and there will be no effect of the water vapor in the second arm. The overall efiect of carbon dioxide and water on the bridge circuit will be substantially equivalent to the arithmetic sum of the opposite effects (on thermal-conductivity of exhaust gas) produced by the water vapor and carbon dioxide present. This effect will be substantially proportional to the total volume of oxygen consumed by the hydrocarbonfuel in the formation of the carbon dioxide and water.

It is well established that in atmospheric air, the proportion of oxygen is substantially constant, and, therefore, with apparatus operating in the manner stated above, it is possible to determine, with accuracy adequate for many practical purposes, whether air-fuel mixtures are theoretical or lean, irrespective of variations in the H/C ratio of the fuel. Thus, the above described system is applicable for measurement of the oxygen consumed in combustion of lean or bridge.

oxidizing agent such as copper oxide. Such treatment has no lasting effect on lean exhaust but converts rich exhaust into a gas composed substantially of only nitrogen, carbon dioxide and water vapor.

In exhaust treated with hot copper oxide, the sum percentage of carbon dioxide and water will increase when the mixture becomes rich (because more carbon and hydrogen "will be oxidized, whereas the amount of nitrogen willremain substantially the same) and the sum percentage of carbon dioxide and water will decrease when the mixture becomes lean (because the-amount of nitrogen Will remain substantially the same and there will also be free oxygen-present,whereas there will be less carbon and hydrogen to be oxidized). Thus, through the use of treatment of exhaust gas by a non-gaseous oxidizing agent, I extend the described method of T. C. analysis for the indication of both rich and lean air-fuel mixtures.

The methodso far described is :imperfect. in that any Water present in ltheair of combustion as humidity is indicated as Water of combustion. To eliminate errors due toatmosphericlhumidity, I employ the following procedure.

I supply exhaust gas to a. firstcell of unitsensitivity in one arm of the bridge at a pressure .low enough-to maintain alLthe Water-originalIy present in'the gas in vapor phase, and supply atmospheric air at the same low pressure to 'a cell of unit sensitivity in an opposite arm :of the The efiect of atmospheric humidity on the second cell neutralizes the effect due to atmospheric humidity in the first cell-with sufficient accuracy for many practical purposes.

I also provide additional-cells of double sensitivity in both arms of the bridge in series with the first-mentioned cells. The double sensitivity cell in the first armis supplied with-atmospheric air saturated with'water vapor. The'double sensitivity cell in the second arm is "supplied with exhaust gas at atmospheric pressure-so that some of the water in the exhaust gas condensesand the efiect of the water in'the exhaust gas-on'the cell is that of Water vapor of saturation, which effect is cancelled by the water-saturated-air in the double sensitivity cell in the first arm. The double sensitivity cell in the-second arm'isfitherefore, really responsive only to the carbon dioxide in the exhaust, and since the-cel1 is of 'double sensitivity, it cancels the effect of carbon'dioxide in the first low pressure-cell. The overall result is that the balance of the bridge "is-determined by the arithmetical sum'of the TIC. efiects of the carbon dioxide and water=of-combustion in the-exhaust gas.

A fullunderstanding of the invention maybe had from the following detaileddescription with reference to the'drawin'g, in which:

Fig. 1 is a schematic diagram of one system in accordance with theinventiony-and Fig.2 is a schematic diagram of a'modification of the system shown inFig. 1.

Referring first to Fig. 1, there is disclosed a bridge circuit having two T. C. filaments -'-I'Iland II arranged in series inone-arm and two C. filaments I2 and I3 arranged in series in the opposite arm. The bridge is completed by fixed resistance arms I4 "and I5 and'azresistor lfi having an adjustable tap thereon, the latter convstituting. one end of one diagonal of the bridge.

A meter I1 is connected :between the ''movable :contact and the juncture ofjthe .two:arms containing-the filaments. A battery i8 is connected in series with the variable resistor 28 in the other diagonal of the bridge.

As shown in Fig. 1, exhaust gas is supplied in series to all of the filaments. Thus it enters through a conduit I9, passes through a constriction 29 and into a treating cell 20 containing heated copper oxide or other oxidizing agent for thepurpose of completely burning all fuel and preventing the entry of raw fuel intoany of the TIC. cells. Heat'maybe provided'in any known way but an electric resistor 60 energized by a "battery 6I is illustrated. From the treating cell "20'the'gas flows through a conduit 2| past a cell containing filament I0, thence through a drying cell 22 which abstracts moisture therefrom. Thereafter it flows through a conduit 23, past cells containing filaments I2 and I3 and through a treating cell 24 which absorbs carbon dioxide 'from the gas. Thereafter it flows through a conduit 25, past a cell containing filament II to an exhaust pump 2I whichis of suficient capacity to :maintain a relatively low pressure on the :downstream side of .the restriction 2 9. The-pressureshould be lgept low enough to maintain all of the waterpresent in the exhaust :gas in the vapor phase.

'Each cell containing one of the filaments I0 to 1'3, inc'lusivegis connected to the adjacent'portion of the conduit, preferably bya diffusion passage, so-that interchangebf'gas will take place between the conduit and the cells Without objectionable currents inthe -cells.

In my copending patent application serial N 0. 477,675 "filed March 1, 1943, 1 have described and broadly claimed: systemsinvolvi'ng such supplyof exhaust gas at low pressure to the test I cells and systems employing such difiusion passages leading to the test cells.

in the system :of Fig. 1, variations in the resistance of the filament III are determined by changes in the T. Cnof the gas in the-cell, Which in turn -is chiefly determined by changes in the :carbon dioxideand water vapor present in the exhaust. Changes in the resistance of filaments and I3 are'determinedchieflyby-changes in the amount of carbon dioxide present the exhaust. Changes in the-resistance of 'filament I I are-minor and are determined by changes in all constituents other than water and carbon .dioxide.

.Let it be assumed that in the absence ef-any cooling effect due toTf-C, the resistances-of all the filaments "would be equal and the bridge balanced. Then in operation the balance tends 'tobe upset 'because of the 'difierent -T. Cfs of the difFerentgases'in the -differentcells; thus:

-Rmis-a function of AB+C R12 is a function of .AB

Ria'isa function of -A-B Ru-is a functionof A Where:

.50 istheiit. Clef-water vapordioxide to filament The net effect on the balance of the bridge is obtained by adding the resistance changes in one arm of the bridge and subtracting those in an adjacentarm since an increase fresistance in one side deflects the meter in the same direction as does a decrease in the other arm. Thus:

Hence, the net effect of the system is to produce an indication proportional to the arithmetic sum of the separate efiects of water and carbon dioxide. -One molecule of oxygen (02) combines with carbon to produce one molecule of carbon dioxide (CO2) and combines with hydrogen to produce two molecules of water (21-120). However, the efiect of a small change in the percentage of carbon dioxide on the thermal-conductivity of exhaust is substantially equal in magnitudeto the effect produced by twice the change in the percentage of water vapor. Thus in Gas Analysis by Measurement/of Thermal- Conductivity by H. A. Daynes, on page 119 in Table 1'7, it is stated that the relative deflections per 1% change of carbon dioxide and water vapor, respectively, are -0.l32 and +0.067 where the deflection produced by 1% change of hydrogen content in air is taken as equal to +1.000.

As a result, the system of Fig. 1, when once balanced for theoretical mixture of one hydrocarbon fuel and air, will also be approximately balanced for theoretical ,mixture of any other hydrocarbon fuel and air because substantially the 'same effect is produced on the system by the carbon dioxide resulting from combination of one molecule of oxygen with carbon as by the water vapor resulting from combination of one molecule of oxygen with hydrogen. It is to be understood that it is not essential or even preferable to conduct a single stream of gas past all the cells. The gas can be divided into parallel streams and treated separately, it being merely necessary to deliver untreated gas to filament l0, dried gas to filaments l2 and I3, and gas dried and treated to remove carbon It may be desirable to employ parallel flow in some instances to prevent lag in the arrival of gas to some cells with respect to other cells, or to prevent different pressures in different cells. 7

Obviously, a single cell having a filament of double length can be-substituted for the two cells l2 and I3.

The present invention utilizes the fact that the effect of water vapor is opposite to that of carbon dioxide in T. C'. apparatus. whereas all previous attempts have been in the direction of eliminating either the carbon dioxide or the water so that one would not obscure the effect of the other.

The system of Fig. 1 has the defect that it does not distinguish between water of combustion and water that was present as humidity in the air of combustion. Hence it would correctly indicate theoretical mixture only'so long as the atmospheric humidity did not change. V

Fig. 2 shows a system that is not only balanced for theoretical mixture of fuels having different H/C ratios but also compensates for atmospheric humidity, and thus the balance of the bridge is not affected by changes in atmospheric humidity. This system employs the same general bridge arrangement as that of Figlubut has two '1. C.

. filaments 30 and 32 ofequal sensitivity in opticcontaining filament 32.

site arms of the bridge and two T. C. filaments 3| and 33 of double sensitivity (1. e., filaments twice as long) in opposite arms, the filament 3| being in the same arm with filament 30, and filament 33 being in the same arm with filament 32. Exhaust gas to be tested enters through a conduit 43 and passes through a cell 2011 corresponding to the cell 20 of Fig. 1 which contains copper oxide or other oxidizing agent for preventing any unburned fuel from reaching the cells. The gas leaves the treating cell 20a through. a conduit 45, which conducts the main stream of the gas pastthe filament 3| and is transferred to the cell containing the filament by diffusion. A portion of the gas from conduit 45 is drawn through a restriction 46 into a conduit 41 which is connected by a diffusion passage to the cell The gas is drawn through the conduit 41 and maintained at a suitable low pressure at which all water remains in the vapor phase, by a pump 54 and a constant pressure valve 49. The pump and valve also maintain a flow of atmospheric air at low pressure through! a restriction 50 into a conduit 5| connected by a diffusion passage to the cell containing filament 30. The last filament 33 is positioned in a cell containing air and connected to a closed chamber 52 containing a body of water 53 so that the filament 33 is at all times exposed to air saturated with water vapor.

It will be apparent from an inspection of Fig. 2 that filament 30 is exposed to atmospheric air at low pressure and filament 32 is exposed to the exhaust gas at low pressure, whereas filament 3| is exposed to exhaust gas at substantially atmospheric pressure and filament 33 is exposed to air saturated with water vapor. The conduit 45 supplying exhaust gas at atmospheric pressure to the filament 3| is of such length that the gas therein has opportunity to cool to atmospheric temperature so that some of the water vapor therein condenses and the gas to which filament 3| is exposed is saturated with water vapor substantially at the same temperature as the air filling the cell 33.

In the operation of the system of Fig. 2, the resistances of thediiferent filaments are affected by the T. C.s of the gases to which they are exposed in the following manner:

R 0 is a function of Ch R32 is a function of B+Cc+Cn R31 is a function of 2(-B+C) Ra: is a function of 2(03) Where:

R30 is the change in resistance of filament 30 due to T. C.

R32 is the change in resistance of filament 32 due to T. C. R31 is the change in resistance of filament 3| due to T. C. R33 is the change in resistance of filament 33 due to T. C.

. B is the T. C. of carbon dioxide Ce is the T. C. of water vapor of combustion Ch is the T. C. of water vapor of atmospheric humidity C5 "is the T. C. of water vapor saturation changes in resistance due to .T. C. and reversing the signs -..of the componentsin the right half .ofthebridge, we, have It will be observed .that the defiection is, therefore, a .function of the concentration of carbon dioxide plus that portion of the water vapor .concentration that is 'dueto water of combustion.

ter effective to show the difference between the T. C. of the carbon dioxide of combustion and the water of combustion. The resistance of filament 3| .is a function of the carbon dioxide concentration in the exhaust and the water of saturation. Since filament 33 is exposed to water vapor-of saturation, it compensates for the effect of the water on filament 3|, leaving the latter effective only to show the concentration of carbon dioxide. Since filament 3| is twice the length of filament 32, it neutralizes the effect of carbon dioxide on the filament 32 and produces an effect responsive to the carbon dioxide, which effect is additive with respect to the effect on filament 32 of the water vapor of combustion.

Aside from other considerations, the system of Fig. '2 has the practical advantage over that of Fig. 1 of eliminating the need of desiccants for drying the gas and reagents for removing the carbon dioxide.

It is to be understood that equations have been used for explaining the operation because they afford a simple and understandable method of describing 'what takes place, and not to indicate or infer great precision of results. oftentimes, the thermal-conductivities of gaseous mixtures do not follow the simple rule of proportions.

Thus, the correction for atmospheric humidity will be more accurate when the volume of water resulting from combustion is small'than when it is large. ACCOlfdiIlgtO the book of H. A. Daynes previously referred to, page 112, the increase in thermal-conductivity resulting from 1% increase in Water vapor decreasesias the total water vapor concentration increases up to about 20%, and becomes negative thereafter. My own measurements indicate that the increase in the T. C. of exhaust gas due to atmospheric humidity is greater than the increase in the T. C. of the atmospheric air due to that humidity. For this reason, if the atmospheric air contains say 2% water vapor and the exhaust contains water of combustion, the compensation for atmospheric humidity of the system of Fig. 2 will be'moreaccurate than one might expect it to be from the graphs shown in the above quoted reference for moist air, because the presence of carbon dioxide in exhaust increases the incrementof T. C. corresponding to unit change in water vapor concentration.

An example will illustrate how important it is to compensate, even imperfectly, for atmospheric humidity: According to Daynes, atmospheric humidity equivalent .to -.2% of .water :vapor by vol- .ume changes the T. :C. nearly .as much. as .1% of .carbomdioxide would. Inasmuch as the-exhaust of a leanmixture willcontainnot morethan 8% carbon dioxideby v lume,.-fai1ure to correct for theatmospheric ihumiclity can introduce an error of aboutl2% in case the fuel contains no hydrogen. On theother hand, if the fuel contains one gram mole of hydrogen for-each gram atom of carbon, the exhaust will contain approximately 8% carbon dioxide and 8% water vapor of combustion, by volume. "Since the water .vapor has approximatelyhalf the effect on T. C. and acts oppositely, themagnitude of the efiect .on T. Cris substantially that of-4% dry carbon dioxideor 8% of water vapor alone. In such .-a case, failureto correctfor 2% water vapor due to atmospheric humidity can introduce an error 013.25%.

:It will be apparent that when the sensitivities of;resistors'3l and 33 aresimilar to the sensitivitiesnof *resistorsgifi; and 32, the apparatus illustrated in Fig. .2 becomes suitable for the measurement of the totalwater resulting from complete combustionof the fuel. Thus, this apparatus .provides for the indication of the air-fuel ratio supplied to a combustion device utilizing definite hydrocarbon fuels just as such air-fuel ratio isdeterminable from the measurement of the carbon dioxidea'lone. Therefore, the invention is also useful for-determining air-fuel ratio by measuring the percentage of the-Water of com- 'ciistionalone.

'In Fig. '1 the resistors 10, 'H, l2 and i3 are shownyplaced within cells lla,.lla, 12a and 13a, respectively. The arrangement is such that the walls of each related vpair of cells (cells Illa and 12a constituting one pair and cells Ila and i3a constituting the other pair) or all of said cells are kept at equal temperatures by forming them ,in a metal block or blocks. In Fig. l the dis- .tances between the cells [to and l2a and between cells Ma and -l3a are shown disproportionately large for the sake of clearness of illustration, but in actual practice the cells are preferably placed asclose to each other as is prac- =ticable.

Similarly, in Fig. 2, the filaments 30, 3|, 32 and .33 are-positioned within cells 30a, 3|a, 32a and.33a,.respectively, and the walls of the cells ,are:preferably=-kept at the same temperature for respective pairs or for all of said cells .byforming them .in a metal block.

It isaunderstood that .in thesystem'of Fig. 1, the arrangement is such that on the up-stream side of i'therestrictors :25 the gas remains at a temperature. sufficiently high to maintain all Water in the vapor phase at atmospheric pressure, .but before the exhaust gas reaches any cell, it is miteferablycooledto the substantially lower temperature ofthewalls ofthe'cells (which may be kept at atmospheric temperature). *This insures that no water is condensed from the exhaust on the up-stream side of the restrictor.

Itis also to be understood that in the system of Fig. 2, the temperatures along the gas conduits, including portions 43'and 45, are such that no condensation-of-water or unburned fuel occurs on the up-stream side of the restrictor 36, so that the filament 32 is exposed to exhaust containing all the carbon dioxide and all the water'that'resulted from the combustion of the fuel.

It'is alsoto be understood that the reaction chamber 20 in Fig. l or-Zfla-in Fig. 2 can bepositioned on the upstream side of the restrictor, and this arrangement may be advantageous in that it permits utilizationof the cooling resulting from the expansion of the gas across the restrictor 29 for dissipation of the heat acquired by the gas in the reaction chamber.

Where in the claims a hydrocarbon fuel is referred to, the, expression is intended to include fuels which may contain free carbon and/or free hydrogen in addition to hydrocarbon compounds.

Various departures from: the exact system shown and described in detail can be made without departing from the invention, which isto be limited only to the extent set forth in the appended claims.

I claim:

1. Apparatus of the type described, comprising: first and second thermal-conductivity cells of equal sensitivity and third and fourth thermalconductivity cells having sensitivities approximately twice that of the first cell; means for supplying exhaust gas to be analyzed to said second cell at a sub-atmospheric pressure low enough to maintain all water in said exhaust gas in the vapor phase; means for supplying atmospheric air to said first cell at said sub-atmospheric pressure; means for supplying exhaustgas to said third cell at such pressure that water saturation results; means for supplying air saturated with water vapor to said fourth cell; and bridge means for comparing the combined resistance of the first and third cells against the combined resistance of the second and fourth cells, whereby said first cell cancels the effect of atmospheric humidity on said second cell, said fourth cell cancels the effect of water vapor in said third cell, and said third cell cancels the effect of carbon dioxide in the second cell and produces an equal effect in additive relation to the effect of water vapor in said second cell.

2. Thermalconductivity apparatus comprising: a bridge circuit including in one arm a first cell; means to supply to said first cell exhaust gas under low pressure at which water in the gas remains in the vapor phase; a second cell of equal sensitvity to said first cell and in opposing rela- 10 water vapor; and means for neutralizing the effect on said bridge circuit of water vapor in said second cell, whereby said bridgecircuit is rendered additively responsive to water and carbon dioxide in saidfexhaust'gas, said last mentioned means comprising a third cell of double sensitivity in said one arm 'of the bridge, said third cell containing atmospheric air saturated with water vapor; and means for maintaining the temperature of gas in said second and third cells the same.

i 4. In apparatus for analyzing exhaust gas resulting from combustion of a hydrocarbon fuel with air, the combination which comprises four thermal conductivity cells each including a resistor mounted therein, each of said resistors having an electrical conductivity varying with the composition of the gas supplied to the cell in which it is mounted, the first and second thermal conductivity cells being of one sensitivity and th ethird and fourth conductivity cells being of twice that sensitivity, ,means for supplying a portion of said exhaust gas to said first cell at a pressure low enough to maintain all water in said exhaust gas in the vapor phase; means for supplying air at the said low pressure to said second 0511; means for supplying a portion of said exhaust gas to said third cell at a pressure high enough to produce water vapor saturation therein; and means for supplying air saturated with water vapor to said fourth cell;

tion thereto in said bridge, and means for supplying it with atmospheric air at said low pressure, whereby said second cell neutralizes the effect on said bridge of atmospheric humidity in said first cell; a third cell of approximately double the sensitivity of said second cell and in aiding relation therewith, and means for supplying exhaust gas to said third cell at a pressure such that water saturation exists; therein; and a fourth cell of. equal sensitivity to and in opposing relation thereto said third cell, and means for supplying atmospheric air saturated with water vapor to said fourth cell; whereby the effect of said third cell neutralizes the effect of carbon dioxide in said first cell and substitutes the same effect 'in additive relation to the effect of water vapor in said first cell, and the effect of said fourth cell neutralizes the effect of water vapor in said third cell.

3. In thermalconductivity apparatus comprising a bridge circuit, the improvement which comprises: a first cell of unit sensitivity mounted in one arm of said bridge circuit; means for supplying to said first cell exhaust gas under low pressure at which water in the gas remains in vapor phase; a second cell of double sensitivity in another arm of the bridge; meansfor supplying said second cell with exhaust gas at a high pressure at which the gas is saturated with and means for measuring the difference between the sum of the resistances of the first and fourth cells and the sum of the resistances of the second and third cells.

5. In apparatus as defined in claim 4 means for treating said exhaust gas with a solid oxidizing agent to convert all carbon and hydrogen therein to carbon dioxide and water vapor prior to supplying said portions of exhaust gas to said first and second cells.

6. In apparatus for analyzing exhaust gas resulting from combustion of a, hydrocarbon fuel with air, the combination which comprises two pairs of conductivity cells, each including a resistor mounted therein, each of said resistors having an electrical conductivity varying with the composition of the gas supplied to the cell in which it is mounted; means for supplying said mixture and a reference gas to the respective cells of one pair of cells at a relatively low pressure at which a condensible component in said exhaust gas remains in vapor phase; means for supplying said mixture and said reference gas to the respective cells of the other pair of said cells at a relatively high pressure; means for maintaining the vapor pressure of said condensible component equal in the cells of said second pair; and means interconnecting said resistors for measuring a combination of the resultant effects of said gases on the resistors in said cells.

7. In thermal conductivity apparatus for determining the composition of exhaust gas that is at a high temperature and at a relatively high pressure and that contains moisture in, such an amount that the moisture would' condense at such pressure at atmospheric temperature, the improvement which comprises a bridge circuit including first and second cells in adjacent arms thereof, said cells being exposed to atmospheric temperature, means for supplying such exhaust gas from a source to said first cell, said means including means for reducing the pressure of WLADIMIR M. ZAIKOWSKY,

REFERENCES. CITED The; following references;v are: of: record. in. the file: oft'this; patent:

12 UNITED; SEIATES PATENTS Number 10 Number:

Name- Date 7 Rodhe Mar. 17; 1925 Harrison Oct. 2'7, 1931 Schneider Apr. 26, 1932 Hebler- 'etall July 18; 1933 Scheffler' Aug. 31, 1943 FOREIGN PATENTSv Country Date Germany" li'eb. 20, 1926 GreatrBritainiuuueOct; 31, 1938 mimnce May 16, 1938

Images