US2821462A - Gas analyzing devices - Google Patents

Description

Jan; 28, 1958 'J. E. McEvoY GAS ANALYZING DEVICES Filed Nov. 15. 1952 BALANCEABLE NETWORK FIG. I

v INVENTOR. JAMES E. M2 EVOY A RNEY BALANCEABLE NETWORK SAMPLE 3 United States Patent 2,821,462 GAS ANALYZING DEVICES James E. McEvoy, East Cleveland, Ohio, assignor to Bailey Meter Company, a corporation ofDelaware Application November 13', 1952, Serial No. 320,323 3 Claims. (Cl. 23-255 My invention isdirected to improvements in gas analyzing devices. And more particularly, my invention is embodied in structure which. eliminates errors to which the prior art devices have been subject when employing at detecting element which is sensitive to variablesin the analyzed gas other than the one desired in measurement. The method and apparatus utilizing my improvement employs an electrically heated filamentin oxidizing combustible gases and vapors and which is electrically responsive to the heat of oxidation, or. combustion.

A recognized material limitation in the use of gas testing apparatus of the type with which I am concerned is that it is not adaptable in indicating the true extent of combustible gases and vapors .inmixtures containing varying amounts of non-combustible vapors andgases suchas water vapor, carbon dioxide and. others which have a substantially ditferent cooling efiect on a heated filament thanis exerted by air. In other words, adetecting filament. does not vary in temperature solely from. the catalytic burning. The problem of analysis in this art would be greatly simplifiednif such were the case. The filament must have direct intimatecontactwith. all of the constituents of the sample in promoting the catalytic combustion with the detected portion. Consequently, specific heat, thermal conductivity, in. short, all of' the influence of temperature variables of the sample comes into play by conducting heat away from the filament at a rate which varies with the composition of the sample. It might also be said that the resistance of the filament, and consequently, the unbalance of the including. network, has not been, in the prior art, proportionalsolely to the percentage of the detected element in the sample but to a combination of this factor and the physical properties heretofore referred to of the sample as a mixture. Although the detected portion of the gas sample itself contributes to the extraneous and additional variation of the filament temperature by reasonof itsphysical properties of specific heat and thermal conductivity, I refer to the additional variation as contributable, in a general sense, to the inert constituents, as a summary basis of reference. Thus, it is: the effect on the final indication of catalytic gas analyzers by the variable inerts that my invention eliminates.

It has been my personal conclusionthatthisdesired refinement of analysis has been recognized by many patents in the prior art in two general approaches. Obtainment. of this refinement has been by either establishing a compensating, non-catalytic filament in the electrical network in the cell of the analyzing filament, or a separate cell has been providedto allow passage, over the compensating filament, of a heated. inert gas approximating the physical characteristics of that passing,

over the analyzing filament. The disadvantages of. these schemes are obvious. The compensating, filament in the single analyzing cell would have 'to be coatedwith, or

made of, a substance whichvvould're'nder it'non-catalytic, with consequent non-parallel response with measuring filament to the physical characteristics'ofthe 'inerts; "or

it isv impractical to flow through the separate compensating cellv an inert which has variations in its physical characteristics parallel with those of the medium flowing through the analyzing cell.

Anexample. of companion cell compensation structure is. to be. found in at least the disclosure of the patents to Sullivan 2,197,370 and 2,404,993, while there are other patents such as Morgan et al. 2,211,627, 2,204,966 and2,273,981 which disclose coating or constructing the compensating filament of a substance which is noncatalytic. The patents to. Miller 2,083,521 and Sullivan 2,310,472 are of interest in their recognition of. this problem, but.these patents do not solve the problem as does the present apparatus.

The general observations in Jacobson 2,378,019 are of interest due to his observationthat when a gas sample with a low concentration of combustibles contains a considerably diiferent amount of water vapor or carbon dioxide from that in the air used for establishing the initial zerowsetting of analyzers using other means' of compensation the total heat loss due substantially to thermal conduction, convectionand direct transport by the moving gas, may be very large and may be even greater thantheheat producedby oxidation of the combustibles. Asdescribed by Jacobson, an instrument of this kind has been made useable to a practical degree by removing. moisturefrom the sample and the air used ,for zero. calibration through afdrying medium. such; as calcium chloride, activated alumina or the like, orthrough a water bubbler whose temperature is kept'cqnstant by means of a thermostat; Aside from the inconvenience to the user, thesemethocls of eliminating moisture by drying the sample are not always applicable becauseany available drying material would absorb someof the comhustiblevapors while absorbing water vapor.

Jacbson goes on to classify the prior artmethods of compensation into two general methods. He firstsees thatcompensator filament as totally enclosed in a special housing or in a sealed gas-tight glass or, metaltube so .that the filament is constantly surrounded by the same air, or. any part of the sampleis prevented. from reaching it. Actually this appears closely akin to. the method I have outlined in the prior art for rendering the compensating filament non-catalytic by coating it. As Jacobson points out, this method provides very goodcompen- .sation .for voltage changes but does very little in the direction of compensating for Water vapor, carbon dioxide or other non-combustible constituents in the sample, which I. term inerts.

Secondly, Jacobson described the method of equal exposure of the detecting and compensating filaments to the sample but with the provision of acompensating filament of. relatively large size to reduce its resistance andthereby decrease its-temperature below thesurface combustion temperature. In a certain sence Jacobson treats this method as classifiable withv the'coating technique I have described. He devotes some detail to'describing the disadvantages of equal exposuretothe sample under this method of compensation, but he does; not further give classification to the category I have'recog- :nized as. parallel exposure ofthefilaments, the detecting "am concerned with providing an improvedmeans for determining the percentage of freeoxygen and/or .gaseous combustible in a flowing gas sample, such as, for exam,

pie, a sample stream of fiue gas from a combustion process or exhaust gases from an internal combustion engine. The gases with which I am concerned will generallycontain nitrogen, carbon dioxide, free oxygen, carbon monoxide, methane, or other unburned combustible. I desirably ascertain the percentage of free oxygen and, separately, the percentage of unburned combustible in the gas.

It has been explained in some detail in the patent to Johnson 2,420,430 that to test gases for combustibles, it is often necessary to provide a predetermined quantity of oxygen to insure the burning of any combustible in the gas sample. On the other hand, to test a gaseous mixture for its free oxygen constituent, it is necessary to pass over the heated catalyst filament, a mixture of a uniform rate of flow of a stream of the sample to be analyzed and a" uniform rate of flow of a stream of a gaseous fuel for combining with the free oxygen of the gas sample stream. As explained in Johnson, it, is

known that the fuel supplied to such a system may be a gas, such as hydrogen, or a vaporized liquid fuel such as methyl alcohol; Johnson discloses the use of vaporized liquid fuel for this purpose.

Other patents have dealt with this dual analysis of "a single gas sample, or separate analysis of two gas samples, by simultaneous, or serial, catalytic burning. Simultaneous provision of sources of oxygen for the comfbustible analysis and hydrogen for the oxygen analysis within the gas or gases has been contemplated in Kennedy 1,333,850 and Morgan et al. 2,273,981. Morgan et al. has even disclosed a generator with which to produce hydrogen. However, none of the prior art has contemplated plenary usage of the out-put of an electrolytic generator in a catalytic analyzer, the generator and analyzer combination providing simultaneous, and separate, analysis for the combustible and free oxygen content of gas samples.

As one of the primary objects, my invention provides a gas analyzing device capable of simultaneous and independent analysis of a single, or two independent, gas

samples for free oxygen and for combustible content.

Another primary object of my invention is to provide a gas analyzer whose record and/or indication will not reflect variations of inert constituents in the gas sample with parallel independence from separate sources of fuels for combustion with the analyzed constituent of the sample.

Another object of my invention is to provide a gas analyzer having means for generating fluids which will combine with the constituents for which the analysis is made.

Another object of my invention is to provide a 'gas analyzer which generates oxygen and hydrogen for si- 'multaneous catalytic combustion of the combustibles and invention.

Fig. 2 is an exploded view of a filament and shield used in the analyzer of Fig. 1.

.Referring to Fig. 1, I have shown therein, somewhat diagrammatically, and not to any scale, the basic units of an apparatus combination capable of analyzing a gas sample simultaneously for its combustible constituents and free oxygen. The gas sample is cleaned and prepared for analysis by an apparatus not shown here but well known in the art. What I have shownis a gas sample tube 1 through a furnace wall 2 for the initial extraction of the sample to be prepared for analysis. The

apparatus for making this extraction, as well as that for cleaning and preparing the sample for introduction into: the analyzing apparatus, is to be visualized as incorporated at some location along passage 3.

The cleaned sample flow in passage 3 is led to the base central housing structure 4 where the catalytic combustion takes place. This housing 4 forms an area into which is flowed, not only the sample of gas to be analyzed, but the substances to be mixed with the sample prior to catalytic combustion. The electrical networks which are generally responsive to the results of the catalytic burning are shown as dual units separately responsive to the two analyses and indicating, separately, the combustible constituent of the sample and the free oxygen constituent. The housing structure 4 itself has a desirable function other than mere support of various element combinations. To perform its function most efiiciently this housing is preferably of a metallic material which gives a'large heat storage for insulation of the responsive elements'of the analyzer from ambient tem- Structure for maintaining the temperature'of this housing 4 at a desirable level above ambient tmperature will be described subsequently.

Specifically, the metallic housing 4 has three different streams of fluids flowing into it. It is necessary for the successful operation of catalytic analyzers of this type that these fluid supplies to the analyzing cells in the housing 4 be controlled as to their flow rates. Consequently, the sample in conduit 3 is held to a constant pressure by means of regulator 5 prior to introduction into the housing 4.- This regulator 5 may be mounted directly upon hous ing 4, and it may be quite similar to the regulators of the patent to Johnson 2,43 8,973.

Tracing the path of the sample in conduit 3 on through the regulator 5 and into the housing 4, it may be seen how the sample is split into two branches for simultaneous introduction into two separate analyzing cells.

In'eachbranch is established an orifice for precise determination of the rates of flow required for proper cata lytic action in the analyzing cells. Orifice 6 meters its portion of the sample into the cell analyzing the gaseous sample for its combustible content. Orifice 7 performs a similar'function on its portion of the gaseous sample which will be analyzed in the second cell for the free oxygen content.

Prior to" either sample portion being introduced into the respective analyzing cell, each portion is mixed with an" element which will promote combustion of the constituent to be detected on the detecting filament in the analyzing cells. Considering, first, the analysis made in cell 8, theprovisions are to be noted for supplying oxygen with the sample which is metered through orifice 6. Generally, in the prior art, a source of compressed air has been disclosed for this purpose, or a tank of oxygen has been controlled through various valves to mix its contents with' the 'gas sample prior to catalytic combustion. None of these provisions of the prior art are necessary with my invention because electrolytic generator 9, as an integrated portion of my analyzer, forms a convenient, simple and eflicient source of oxygen for mixing with the gas sample.

It has been found quite feasbile to design the size of generator 9 to supply the maximum quantity of oxygen needed to mix'with the sample portion analyzed in cell 8. No valves or flow controllers between generator 9 and the analyzing cells are necessary as control of the basic output of the generator is a function of the electric power applied toits eelctrodes. Therefore, pipes 10 and 11, conducting oxygen and hydrogen from generator section 9 into housing 4, may be quite short; in fact, the

generating section may be attached directly to the housing 4, considering only the connections necessary to gain access to the generator for maintenance and addition of water and electrolyte.

Considering the other product of the generator 9 which a portiontif the gas sample "is analyzed'for its free oxygen content "as'it was "analyzed for its combustible content in cell 8. The gas sample is mixed with hydrogen which combines with the'free oxygen content resistance of the detecting filament in indication of the tree oxygen content of the gaseous sample portion.

It is, of course, to be appreciated that with initial consideration given the size of the electrical signal needed for actuation of a balanceable electrical network, thesize of the filaments, their cells and the rates of'flow'required for the sample portions and the-oxygen and hydrogen must be carefully considered in a unitary design. The more flexible and'logical point of-adjustment of the various variables is in the size of the orifices 6 and 7 for the sample of gas. These orificestructures may be made up as easily inserted fixtures with various, specified sizes of passages for the gas sample. A-set oforifice=sizes can be predetermined which will establish ranges of-actuationfor the balanceable networks to'span'the wide quantitative variation, in the samples, of the analyzed constituents.

The output of the electrolytic generator 9 is fixed by -the size of the electrodes, the electric power applied to them and the degree to which the electrodes are immersed in their'eelctrolyte. 'It is generally anticipated that an excess of oxygen and hydrogenneededfor catalytic combustion of theanalyzed constituents 'Wlll be supplied the cells 8'and 12. Thisoutput is basically set by the size 'of the generator 'sectionand the size of'conduits 1'0 and "'11 are'fixedfor themaximum output expected of generator 9 so excessive back-pressure will not build upin the housing. -As the output of thegenerator will fluctuate with variations in the voltage of the electric power applied,-a constant-voltage transformeris considered necessary under normal conditions of operation. 'I have indicated,- diagrammatically, at 13 both a transformer and a system of rectification for the A.-C. applied to the electrodes of thegenerator.

The electrodes of the genera-tor are shown immersed 'below the level of the electrolyte. It is commonpractice to give 'the innerelectrodecorrugations in order that its area will compare-with" that of the outer electrode. Both electrode 14 and "electrode 15 are supported by sturdy metallic posts which also serve as conductorsfor the curreritfrom 1'3.

The-combination of the generator has a particular 'fea- However, generators of the takes place to a detrimental extent. 'To retain the advantage of the asbestos cloth in its ability to allow free passage' of 'the ions in'the electrolyte between electrodes,

while eliminating the intermingling, I have provided a metallic shield as a substitute for an upper portion of the asbestos barrier. Thearrangement of the metallic shield 17 and asbestos'shield 16 can be clearly seen in Fig. 1. The metallic shield 17 is-atta-chedto the top of the genera- -tor, extending down, between the electrodes, to the ex ipected minimum level of-the electrolyte. The asbestos shield 16 iswired, or otherwisesolidly fixed, to the lower end :of the shield -17 and may extend down to the bottom of the generator, certainly far enough to prevent the gas discharging*from-theelectrodes from crossing to the wrong side of the barrier.

Thus, as the level'of'the electrolyte lowers, the metallic shield .17 continues to insure ,positive separation of "the accumulated, .generated "gas prior to usageby"the analyzer section. The shield retention within this constant heat source.

"17 is fixed at a length 'whichw'ill allow the electrolyte usage to occur over a substantial, predetermined period without sinking to the lower asbestos shield '16. This arrangement is unique in generators of this size'andin generators'having a constantdemand upon them.

It is-also to be noted, at this point, that the "location of the orifices and the various passages for the sample, oxygen and hydrogen within the metallic housing 4 provides-aconvenient structure for maintaining the tempera ture of the various fluids at a constant level as they are metered. The-housing 4 not only heats the gaseous fluids to a common, consistent temperature, but maintains the orifice structures themselves unchanging by reason of their To maintain housing 4 at a desired temperature level, heater 18 is conveniently recessed into the housing and is placed under control of thermostat 19 by means of relay 20. And it also might be noted at this time that the temperature level atwhich housing -4' is maintained by heater 18 insures that water formed during the catalytic combustion will not collectin the cells and vary the temperature of the filaments by reason of its heat absorbing properties. As amatter of fact, should now begin to strongly impress any one versed in this art that the structure of my analyzer gives adequate provisions for safeguarding the fact that the variation in temperature between the measuring and compensating filaments of the analyzing cells will be solely a function of the degree of catalytic combustion on the measuring filament.

And finally, it should be obvious that if each set of two filaments in cells 8 and 12 are established, essentially, as legs in a'Wheatstone bridge included in balanceable networks 21 and 22, the restoration of balance to these networks will give the desired indications of combustibles and free oxygen present in the gaseous sample extracted through conduit 3.

The advantages of the analyzer over the one disclosed in Johnson 2,420,430 may be readily apparent from one standpoint it it is observed that the arrangement of Johnsons Fig. 3 is not necessary with the present invention. As the present analysis is made in parallel, and independently, in the present invention, the compensation of the oxygen analysis for air added need not be made. However, it may be desirable, in some instances, to utilize a combination of the two analyses. This is simply provided for by a single balanceable network instead of 21 and 22 or by the mechanical linkage I have shown as actuated by the two balancing motors. Essentially link 23 is simultaneously positioned by the motors of networks 21 and 22. The resultant of these positionings determines the vertical position of link 24 which actuates pilot valve 25 and for indicator 26. The output of valve 25 can be utilized to control, desirably, from the two variables, combustibles and oxygen. It the single balanceable network is desired, the arrangement could be similar to that of Fig. 13 in Johnson 2,420,430.

A final feature of this invention is disclosed in connection with the flow valve, or regulator 5, similar to that of Johnson 2,438,973. It is quite possible that regulators of this nature may be required on the other fluid supplies introduced into analyzer 4 as representative of any consurner, or utilizer, of fluids under pressure.

It must next be noted that regulator 5 has at .least one vertically reciprocating part, which assumes various heights, depending on the variations of pressure in the source of fluid supplied. If an interruption should occur in the fluid supplied, the reciprocating part, or member 27, would drop, by gravity, to the bottom of its bore, as it is called in the Johnson patent. Normally, the fluid flows into the bore below the conical head of the member 27 and the resulting pressure built up in the borerraises the member 27 until a cross-slot is uncovered in varying degree to allow a flow of the fluid to atmosphere. A variable portion of the fluid is wasted. in.order.that,a constant predetermined pressure will be maintained in A 7 the bore depending upon the weight of the member 27 and cross-sectional area of the conical head. I regard the waste passage formed by the bore slot and head of member 27 as essentially a variable orifice regulating the pressure of the fluid, the member 27, consequently,

, assuming various positions within a predetermined range.

With the foregoing structure available in regulator 5, and the equivalent of member 27, may be found in other regulators, I suddenly perceived that the motion provided could be utilized to provide a signal or control action if the fluid supply should depart from its expected range. However, there is only a very small amount of power available in these members represented by 27. Actuation of a circuit switch by member 27 is not a straightforward problem. Eventually I arrived at the structure disclosed for accomplishing the required function. I attached a small magnet 28 to the elongated end of the member 27 and arranged a metallic disc 29 on the end of actuating arm 30 of switch 31 at a point where the magnet would attract the disc a sufiicient distance to actuate the switch when the fluid sample pressure dropped to a predetermined minimum.

It is, of course, possible that other types of regulators can employ the principle of my invention and other fluids than in the present disclosure could be so monitored. It is also possible to have a switch 31 in circuits other than the simple annunciator circuit of light 32 specifically disclosed. Also it is simple to arrange the magnet 28 and disc 29 so that actuation of switch 31 will occur at the other extreme of its range. Once switch 31 is actuated, it can be included in a control circuit to send a control valve, or other member, to a safe extreme of its range of operation. Other results may be required of my actuating mechanism and it is flexible enough to be readily adapted to many circumstances which can be envisioned.

Fig. 2 is ofiered as a disclosure of the essential structure of the filament supports and novel elements associated therewith to be found in both cell 8 and cell 12. In each cell, insulating base 40 supports the terminals and filaments 41 and 42. The filament 42 is the measuring one while 41 represents the comparison filament. A single shield, or shell, fits over both of these filaments to prevent direct impingement on the filaments of the mixture introduced into cells 8 and 12. Shield-shell 43 fits over the filaments and attaches snugly to its base 40.

The structure of shield-shell 43 is unique in arrangement. It is to be noted that a partition 44 isolates filaments 41 and 42 from each other. With this internal isolation of filaments provided, a fine metallic mesh screen 45 is carried across one-half of the end of shield 43 and a plate, bearing an aperture 46 therethrough, is carried across the other half of the end of 43. With this struc ture of the cells complete, the sample gas does not flow over the filaments as in so many prior art devices but is introduced along the outside of the walls of shield 43, and this tangential introduction swirls and thoroughly mixes the constituents of the streams coming into the cells. Actually, that portion of the mixture which reaches the filaments is accomplished by the phenomenon of diffusion, common to gaseous fluids.

With the structure defined above, catalytic combustion on filament 41 is prevented from being identical to the catalytic combustion occurring on filament 42. This desired result is obtained very satisfactorily by greatly limiting the rate of diffusion to the compensating filament 41 by means of the small aperture 46 in shield 43. The rate of diffusion to the compensating filament 41 is reduced so drastically by aperture 46 that insignificant catalytic combustion occurs on the surface of this filament while at the same time the rate of difiusion maintains a passage of the mixture over the compensating filament 41 which is proportional in rate to that over measuring filament 42. Whereas, with this arrangement, all the physical properties of the gaseous sample and air or oxygen or hydrogen as well as the products of catalytic combustion have a cooling effect on measuring filament 42, simultaneous with its temperature elevation due to catalytic combustion, the same effect is maintained proportionately upon compensating filament 41. With proportionality in effects on the opposed sensitive bridge elements of networks 21 and 22, with the exception of the raised resistance of the measuring filament 42, due to the catalytic combustion thereon of the analyzed constituent, the resulting unbalance of the networks 21 nd 22 truly reflect the percentage by volume of these constituents existing in the analyzed gaseous sample.

The unitary construction and assembly of the base 40, filaments 41, 42, and shell 43, of Fig. 2 provides the possibility of pro-assembly and calibration, for ready removal and replacement, without change in the individual or overall characteristics of the system. Furthermore, this unitary assembly of measuring and comparison filaments, insures an unchanging relation of heating, ambient effects, and the like.

What I claim as new and desire to secure by Letters Patent of the United States, is:

1. A catalytic gas analyzing cell structure removable from and insertable into a gas sample receiving chamber in a heated block member including, a non-metallic insulating head portion supporting two similar filaments and their terminals, the filaments supported in spaced parallelism, a metallic shell having an open end closely engaging the head portion, surrounding the filaments, and having a central dividing partition extending between the filaments to form two chambers, one around one filament and the other around the other filament, and end closure structure for the other end of said shell having gas difiusing openings of different degree into the two chambers from said gas receiving chamber.

2. A continuous gas analyzing apparatus, including a heated metal block, a combustion chamber in the block, a heated catalytic filament mounted in the combustion chamber, a mixing conduit in the block and connected to the combustion chamber, a gas pressure regulator having a pressure responsive valve member freely movable vertically relative an orifice provided in the regulator to maintain the gas pressure on one side of the orifice constant, a source of gas to be analyzed, a conduit connecting said source of gas to the regulator, another conduit connecting the regulator and the said mixing conduit, a source of gas to be combined with a constituent of the analyzed gas, a conduit connecting the source of gas to be combined with a constituent in the analyzed gas with the mixing conduit, a balanceable electric network including the heated catalytic filament, a switch actuated by the movable valve member of the regulator at a predetermined point in the range of regulating movement, and an electric circuit controlled by the switch and adapted to signal a predetermined position of said movable valve member.

3. The apparatus of claim 2 wherein the movable valve member of the regulator has a magnet at one end and is positioned for operation of the switch.

References Cited in the file of this patent UNITED STATES PATENTS 1,356,598 Cahill Oct. 26, 1920 1,529,470 Dowd Mar. 10, 1925 1,770,059 Barber July 8, 1930 1,918,702 Hebler et al. July 18, 1933 1,988,841 Hayward et a1 Jan. 22, 1935 2,273,981 Morgan et al Feb. 24, 1942 2,335,032 Sullivan Nov. 23, 1943 2,378,019 Jacobson June 12, 1945 2,404,993 Sullivan July 30, 1946 2,420,430 Johnson May 13, 1947 2,438,973 Johnson Apr. 6, 1948 2,488,758 Binford Nov. 22, 1949 2,652,315 McEvoy Sept. 15, 1953

Claims (1)

1. A CATALYTIC GAS ANALYZING CELL STRUCTURE REMOVABLE FROM AND INSERTABLE INTO A GAS SAMPLE RECEIVING CHAMBER IN A HEATED BLOCK MEMBER INCLUDING , A NON-METALLIC INSULATING HEAD PORTION SUPPORTING TWO SIMILAR FILAMENTS AND THEIR TERMINALS, THE FILAMENTS SUPPORTED IN SPACED PARALLELISM, A METALLIC SHELL HAVING AN OPEN END CLOSELY ENGAGING THE HEAD PORTION, SURROUNDING THE FILAMENTS, AND HAVING A CENTRAL DIVIDING PARTITION EXTENDING BETWEEN THE FILAMENTS TO FORM TWO CHAMBERS, ONE AROUND ONE FILAMENT AND THE OTHER AROUND THE OTHER FILAMENT, AND END CLOSURE STRUCTURE FOR THE OTHER END OF SAID SHELL HAVING GAS DIFFUSING OPENINGS OF DIFFERENT DEGREE INTO THE TWO CHAMBERS FROM SAID GAS RECEIVING CHAMBER.

Images