US3605484A

US3605484A - Method and apparatus to determine carbon potential in the atmosphere of treatment furnaces

Info

Publication number: US3605484A
Application number: US784496*A
Authority: US
Inventors: Joachim Wunning
Original assignee: Individual
Current assignee: Individual
Priority date: 1966-04-21
Filing date: 1968-10-03
Publication date: 1971-09-20
Anticipated expiration: 1988-09-20
Also published as: DE1673308C3; FR1526196A; GB1135306A; SE347817B; AT294462B; DE1673308B2; GB1229467A; CS151481B2; DE1673308A1; FR95622E

Abstract

AN ELECTRICAL SIGNAL REPRESENTATIVE OF THE LOGARITHM OF THE PARTIAL PRESSURE OF THE COMPONENT IN THE FURNACE IS DERIVED AND TO THIS LOGARITHMIC SIGNAL AN ELECTRICAL SIGNAL DIRECTLY RESPONSIVE TO TEMPERATURE IS ADDED TO PROVIDE FOR A TGEMPERATURE CORRECTION FACTOR, OTHER SIGNALS REPRESENTATIVE OF OTHER CORRECTION FACTORS, FOR EXAMPLE TO COMPENSATE FOR DRIFT OF MEASURING INSTRUMENTS, MAY BE ADDED, THE COMPOSITE SIGNAL IS THEN INDICATIVE OF THE CARBON POTENTIAL OF THE ATMOSPHERE WITHIN THE FURNACE.

Description

J. WUNNING 3,605 484 METHOD AND APPARATUS TO DETERMINE CARBON POTENTIAL IN THE Sept. 20, 1971 ATMOSPHERE OF TREATMENT FURNACES 2 Sheets-Sheet 1 Filed Q61.- 5, 1968 (pventor Acfi/M WOWIV/NG by 7i? p 1971 J. WUNNING 3,605,484

METHOD AND APPARATUS TO DETERMINE CARBON POTENTIAL IN THE ATMOSPHERE 0F TREATMENT FURNACES Filed Oct. 3. 1968 2 Sheets-Sheet 2 I PROPANE I COMPARATOR. MTENTIOHETER M THERMOCOUPLE I PARTIAL I 6 g c I Pgg ggg: 4 4a l 40 FURNACE E a AL flcIIr ATMOS H E I 25 I ER mg? 1 I I I I I 1 N 1s 29 n so 18 51 -52 CONTROL \V I I I4 6 FURNACE ATMOSPHERE TEST SAMPLE F 1-240 140 EaAND COMPARISON J COMPOSITION ADJUSD A! --wv wv *w M, 2?. 2 I

United States Patent M 3,605,484 METHOD AND APPARATUS TO DETERMINE CARBON POTENTIAL IN THE ATMOSPHERE OF TREATMENT FURNACES Joachim Wiinning, Bergstrasse, 7251 Warmbronn, Germany Continuation-impart of abandoned application Ser. No. 631,240, Apr. 17, 1967. This application Oct. 3, 1968, Ser. No. 784,496 Claims priority, application Germany, Apr. 21, 1966, W 41,390; Oct. 5, 1967,P 16 73 328.1

Int. Cl. G01n 7/00 U.S. CI. 73-23 14 Claims ABSTRACT OF THE DISCLOSURE An electrical signal representative of the logarithm of the partial pressure of the component in the furnace is derived and to this logarithmic signal an electrical signal directly responsive to temperature is added to provide for a temperature correction factor; other signals representative of other correction factors, for example to compensate for drift of measuring instruments, may be added; the composite signal is then indicative of the carbon potential of the atmosphere within the furnace.

PRIOR U.S. APPLICATION The present application is a continuation-in-part of U.S. Ser. No. 631,240 filed Apr. 17, 1967, now abandoned.

The present invention relates to a method, and to apparatus to determine the carbon potential of the atmosphere in furnaces for the treatment of metals, particularly ferrous metals, and more particularly relates to a method and apparatus to determine the concentration of a component of the gas Within the atmosphere, for example carbon dioxide (CO or water vapor or steam.

Furnaces for treatment of metals, particularly ferrous metals, which may be classified as annealing. furnaces although the work pieces may be heated to incandescence, contain an atmosphere which is in essential equilibrium. The carbon potential within these furnaces must be determined, and it is of particular importance to take into consideration ambient conditions existing within the furnace, such as the temperature therein, the composition of the atmosphere Within the furnace, the pressure thereof, as well as the composition of the work piece to be treated.

German Pat. 1,071,378 (corresponding to U.S. Pat. No. 2,935,866) contains a discussion related to measuring and controlling the addition, or removal of carbon from carbon soluble work pieces, such as iron, or iron alloys, in annealing furnaces. As set forth in this patent, it is necessary when controlling the carbon content within the work piece to provide for a means to measure increase or decrease of carbon content accurately. Carbon is absorbed. or removed from the work piece, until an equilibrium condition exists between the carbon concentration within the work piece and the mixture of the atmosphere surrounding the work piece which includes carbon atoms itself. The carbon content which a sample consisting of pure iron can absorb, at a given temperature, when in equilibrium with an atmosphere within the oven in generally referred to as the carbon potential or the C-potential of the atmosphere within the furnace. Various arrangements and systems to measure this carbon potential are known, in order to adjust the atmosphere within the furnace to a specific carbon potential; reference may be made to HTM, Part A, Volume 16 (1961), Issue 1, page 7. Two methods of measuring, the direct and the indirect methods, have to be distinguished.

Direct measuring of the carbon potential is done placing 3,605,484. Patented Sept. 20, 1971 a sample in the furnace, bringing the atmosphere within the furnace into equilibrium therewith, and then determining the carbon content of the sample either, after cooling or quenching by analysis, or by determination of electrical conductivity within the furnace itself (see, for example, the aforementioned German 1,071,378, U .8. Pat. No. 2,935,866). In the indirect method, the composition of the gaseous atmosphere within the furnace itself is analyzed and the carbon potential is then calculated from known equilibrium conditions of the gaseous components within the furnace atmosphere.

The direct method of determining the carbon potential has the advantage of accuracy and simplicity; yet, it has substantial disadvantages which prevent its use in many instances. For example, the method of analysis, for example, by analyzing foils, cannot be done continuously and automatically; when introducing a separate test wire within the carbon atmosphere in order to make an electrical conductivity determination, the presence of such test wire may be objectionable and further, a high carbon potential which reaches the region of iron carbides, cannot be accurately measured. Such a high carbon potential, which is desired when forcing increased carbon content has a deleterious effect on the sensing wire and may cause suflicient damage to prevent good measurement.

The indirect method of measuring carbon potential is generally used in industry. It is known to utilize the components of the gases within the furnace atmosphere as test samples. As can be shown, it is practically only useful to use the CO CH or the H O components as a representative quantity to determine the carbon potential of the atmosphere within the furnace. The concentration of the particular component is determined. The CH component is, due to the slow change of equilibrium condition, not used much in practice, so that as measuring samples the CO and the H 0 components are usually the only ones which can effectively be utilized. Today, the 00 or H O concentration is almost exclusively determined by means of measuring the dew point, for which a lithium chloride measuring unit is used; alternatively, infrared analysis may be employed. For literature references regarding the analysis of the particular component itself, reference may be had to Chemie-Ingenieur-Technik" No. 6, 33 volume (1961), pp. 426-430. When the dew point, in C., or the partial pressure of the carbon dioxide is determined, the carbon potential is then obtained from families of curves, which relate carbon potential to the measured quantities for various temperatures of the furnace, which are the most important parameters to be determined. These diagrams assume as a precondition,

gas within the atmosphere of the furnace itself, the absolute pressure, the particular alloy composition of the material to be treated, etc. is the same as that on which the families of curves are based. This, however, is not the case in many instances. Yet, it is difficult to consider all the other factors, in actually carrying out these measuring processes, because the correction curves and diagrams necessary would unduly complicate the measuring techniques.

The indirect measuring methods, used today, require relatively complicated instrumentation and still have the disadvantage that the accuracy of measurement, as obtained in plants themselves, is frequently not suflicient to satisfy requirements. Additionally, the use of curves and diagrams to determine the carbon potential from the measuring values obtained by sensing elements is timei which is a substantial factor in the accuracy of the final carbon potential determination. This temperature determination is frequently insufficiently accurate because the temperature within the furnace is determined, in acutal practice, at only one particular point in the furnace which is usually remote from the point of removal of the gas, thus introducing errors which may reach 10 C. or more. The present invention departs from the prior method and provides for temperature compensation, the temperature being determined from the gases in the oven themselves. For accurate determination, the temperature sensing elements must also be accurate. Their calibration must be accurately known, and they should not be subject to drift; Further, thermal dynamic equilibrium within the furnace must obtain, so that temperature measurements are made under equilibrium conditions and not during deviations from an average value.

It is an object of the present invention to improve the accuracy of measuring carbon potential under continuous operating conditions of the furnace without the use tabulations involving curves and diagrams.

It is a further object of the present invention to provide a measuring method, and apparatus, in which errors arising due to lack of compensation for miscalibration of the temperature sensing elements, or due to deviation from thermal equilibrium are compensated for.

SUBJECT MATTER OF THE INVENTION Briefly, in accordance with the present invention, gas is removed from a given location within the furnace, which given location is also utilized to measure the temperature within the furnace. Thus, the temperature determination and removal of the gas, the composition of which will be determined, is practically co-instant. The gas is analyzed and a value is determined proportional to the concentra tion of the gas component under analysis; this value is made approximately equal to the logarithm of the partial pressure of the gaseous component and this, so transformed measuring value is added to a measuring value derived from the temperature, to add a temperature correction factor thereto; from this combined added value, the carbon potential can be directly indicated. Other correction factors representative for example of composition of the atmosphere within the furnace, the absolute pressure within the furnace, the particular alloy composition of the work piece, etc. can readily be added to the measured value.

In accordance with a feature of the invention, a correction factor is obtained representative of drift of the temperature measuring element, or of lack of equilibrium of the carbon potential, by making a determination from a test sample, from time to time, and compensating, electrically, for any deviations between indicated, and actually measured values.

In accordance with another feature of the invention, an electrical signal (current, or voltage) is derived representative of the concentration of the, particular gas, which signal is made proportional to the logarithm of the partial pressure thereof. A' correction signal derived from a thermocouple is added to the signal representative of gas concentration. By making the signal representative of the carbon dioxide or vapor component of the atmosphere proportional to the logarithm of the partial pressure, correction of the final value, necessary to consider the influence of other parameters within the furnace can be easily carried out by electrically adding signals.

Correction for instrumentation drift, or lack of equilibrium is obtained by adding another electrical correction signal to the signal representative of gas concentration. This signal can be varied, from time to time, in accordance with analysis results obtained from test samples, and comparing these results with the value indicated by the instrument. The additional signal is then added in (in positive, or negative direction) in order to have the indication on the "instrument agree with actual conditions as determined from analysis of the test sample. The carbon possible to utilize a regulator which adjusts the composition of the atmosphere within the furnace. One of the advantages of the method and apparatus according to the present invention consists therein that the scale of the indicating instrument can be directly readable in percent carbon of gamma-iron so that a direct reading is obtained.

It has been found under actual operating conditions that an accuracy of i0;05% carbon can be obtained.

According to the invention, carbon potential can be determined without the use of complicated diagrams, or computer apparatus replacing. diagrams. It has been found, suprisingly, that the heat of mixing of gamma iron is equal to zero. The method and apparatus of the present invention provides for technical utilization of this discovery.

The mathematical derivation will be given. In this derivation the following definitions are used: p, (in atmospheres) is equal to partial pressure of the gas component i in the furance c (in atomic fractions) is equal to concentration of carbon in gamma-iron T K.) equals absolute temperature in the furnace 7;, is a correction coefficient for a specific alloy composition, of the work piece, and will he usually used as (in percent C.) defined as Percent 0 (pure iron sample) ,7 0

Percent C (alloy sample) The basis for the functioning of the present invention is the Gibbs-Helmholtz equation:

AG: AH-TAS wherein AG is the reaction enthalpy;

AH is the heat of reaction; and AS is the reaction entropy.

From this equation the transition of carbon from the atmosphere within the furnace in the gamma-composite crystal of the iron can be derived:

Pco

The water vapor reaction is considered in the following relationship 10g pflzFlog (1:58 o.

where the logarithm is to the base 10. I It will be noted that the above relationship consists essentially of factors which are being added, each one depending on a particular parameter existing within the furnace. The factor corrects for activity of the workpiece; How this is done will be described in detail below.

The partial pressures p or P112 respectively, are made proportional to electrical currents or potentials. Correction is then particularly simple because it can be entered by manually settable potentiometers. Any one of the correction factors has a potentiometer assigned thereto, the scale of which can be directly readable in the correction factor, so that the use of additional diagrams is unnecessary. Further, automatic compensation of at least some of the correction factors can be carried out automatically very simply, because within a certain range a shift of the zero or null point is proportional to the specific quantity to be measured, so that the correction factors to be entered can remain constant and need not be changed. I

The method according to the present invention can be carried out readily by means of an apparatus which, in accordance with another feature of the present invention, consists of a sampling assembly, introduced into the furnace, which contains a thermo-electric element within a protective cover and, an immediately adjacently arranged suction tube having at least on its interior surface a ceramic coating, from which gas is removed by suction from the furnace. Concentration of the gas so removed is determined in an analysis apparatus, which may be any form of apparatus well known in the art itself. The analysis apparatus is so arranged that it delivers a signal at least substantially proportional to the logarithm of the partial pressure of the gas component under consideration, in the analysis apparatus. Electrical signals so obtained are then added in an adding network to a potential proportional to the output of the thermo-electric elements. An indicator will then read directly in percent of carbon potential, automatically corrected at all times for temperature variations. The temperature, or the ham of which the gas analysis is done, is thus accurately determined, because temperature differences which might be caused by determining temperature at a point different from the gas removal itself is avoided. Since the determination of carbon potential is highly dependent on temperature, as indicated in the mathematical analysis above, an accurate determination of temperature 18 1mportant in order to obtain accurate measuring results.

Introducing a sampling assembly within the furnace enables construction of a simple unitary test transducer, which does not require substantial modifications of, or within the furnace itself.

FIG. 1 is a partially vertically sectional view of a test assembly in combination with the apparatus, shown partly in schematic form;

FIG. 2 is an electrical circuit diagram, in schematic form, indicating measuring of water vapor concentration; and

FIG. 3 is a modified circuit diagram, particularly suited for measuring carbon dioxide concentration within the atmosphere in the furnace, in schematic form.

Referring now to the drawings and particularly to FIG. 1:

The furnace, not forming part of the invention, is not illustrated; its wall 2 is formed with an opening into which a sampling and test assembly generally shown at 1 can be inserted. The test assembly 1 includes a gas suction tube 3, 4, formed of two telescoping tubes; the outer one, tube 3, may consist of chromium nickel steel, and the inner one, tube 4, may be of ceramic material in order to prevent changes'in the gas being removed which might arise if metallic surfaces, at the temperature of the oven, have catalytic effects thereon. The gas inlet is shown at 5. Immediately adjacent the gas inlet 5, and within the common sampling assembly 1 and secure to a common sampling assembly head 11, is a double-walled protective tube 7 containing a thermo element, the active point 6 of which is located adjacent the suction inlet 5 of gas tubes 3, 4. Since both tubes 3, 4 and the tube 7,

protecting the thermo element, extend within the furnace immediately adjacent each other, the temperature at the active point 6 of the thermo element correspond, in actual operating conditions, exactly to the temperature of the gas being removed at point 5, so that the temperature of the gas as actually removed is determined. The sampling assembly head 11 contains a third tube 8, extending within the interior of the furnace, the interior of which communicates with the furnace interior by means of perforations 9 formed in the Wall therein. Tube 8 is adapted to hold work samples, shown at 10, which can be introduced into the furnace to initially adjust, and calibrate the instrument, as will be discussed in detail below. The sampling head 11 is formed with a removable cover 20 and each one of the tubes extending within the furnace itself is covered by a plug 21.

Gas removed from the interior of ceramic tube 4, and entering through inlet 5 is first led to a gas filter .12, arranged immediately within the sampling head 11. It is then taken off by means of an outlet pipe 13, and sucked by means of a pump 22 to an analysis apparatus 14. Since filter 12 is arranged immediately within the sampling head 11, which is itself heated by contact with the metallic tubes 4-, 7, 8, condensation of gas Within the filter 12 is avoided yet it is unnecessary to add additional separate heating means for the filter. I I

The analysis apparatus 14 which will be discussed in detail below, supplies an electrical voltage (or current) to operate indicator 15 (FIG. 1). This indicator can be directly set to indicate percent of carbon potential. It is arranged in a cabinet 23 and contains the circuits further illustrated in FIGS. 2 or 3. A value representative of temperature is applied thereto over conductors 140, connected to the active point 6 of the thermo electric element.

Potentiometers

16, 17 and 18, are manually settable in accordance with directly indicating scales 24 likewise arranged on cabinet 23 in order to compensate for parameters or ambient conditions existing in the oven, such as the particular composition of the atmosphere in the oven, pressure within the oven, or composition or particular alloy of the work piece within the furnace.

Referring now to FIG. 2, the partial H O pressure is determined by a lithium chloride humidity-measuring element and, in this case, forming the analysis instrument 14. A humidity sensing resistance 14a is included in one branch of a bridge network 25, one diagonal connection of which is fed by a rectifier 26 supplied by potential from one winding of a transformer 50. The structure of the rectifier itself need not further be explained and it may be an ordinary kind of a bridge rectifier; it is indicated schematically only in FIG. 2. The other diagonal portion of the bridge 25 is connected in series to a potential which is determined by the temperature, as obtained from thermo element 6. The conductors from thermo element 6 are connected across a variable resistance 27 which is again fed by potential from a rectifier 28 connected to the transformer 50. Thus the sum of the potentials obtained from instrument 14, in essence from the humidity sensitive element 14a and the potential determined by the thermo electric element 6 are added. To this added signal, a group of three further signals are added which consist of potentials obtained across the diagonals of three

further bridge networks

29, 3d, 31. The particular value of each one of the potentials is determined by potentionmeters 16' 17, 18 contained in one branch of the bridge network. Each of the bridge networks are supplied by

separate rectifiers

33, 34, 35, connected across the three diagonals of the

bridges

29, 30, 31. The sum of all these potentials, that is the sum from bridge 25, the potential corresponding to the thermo electric element 6 and from

bridges

29, 30, 31 then applied over an amplifier 32 to an indicator 15 which is directly readable in percent of carbon potential. The amplifier 32 can also supply the controller in order to maintain the indicated percent carbon within specific limits by regulating input of air or other gases to the furnace as will appear in more detail in connection with FIG. 3.

The circuit according to FIG. 2 is arranged to permit the addition of fixed values to the output obtained from analysis instrument 14 and the thermo electric element 6. The analysis instrument 14 is so arranged that the output obtained therefrom corresponds to the relationship indicated in Equation 2 above. That is, an output potential proportional to log PHZO. Detailed consideration of Equation 2 shows that, when the average operating conditions are approximately within a medium range, which is usually the case in actual practice, that the other factors within Equation 2 can be so transformed so that constant correction factors can be added therein. These correction factors are obtained by proper adjustment of

potentionmeters

16, 17 and 18 in accordance With scales 24; it is of course also possible to continuously vary the setting of

potentiometers

16, 17 and 18 depending upon the output of other sensing elements or transducers which influence the setting of the potentiometers in accordance with variation of ambient conditions, such as for example variation of ambient pressure within the furnace.

The composition of the work piece has to be compensated for; ordinarily it is customary to define the carbon potential with respect to pure iron. In ordinarily used nonalloy steel, the effects of deviations due to activity of the material are within the limit of tolerance of the equipment. Yet, the addition of alloying materials even in only little alloyed steels have an influence which can no longer be ignored. The activity of carbon in the gamma-crystal can be sufiicient so that the carbon content may change up to 20 percent from the desired value, if the alloy content is not considered. Alloying elements which form carbides have an even greater effect.

The influence of alloying elements on the activity of the carbon is generally considered in thermodynamics by means of a correction co-efiicient. There are numerous publications disclosing experimental results and considering mathematical relationships determined by the influence of various alloying elements (see, for example, Arch. for Eisenhuttenwesen, Vol. 35 (1964), pp. 999-1007 Correction for the activity constants in the

above Equations

1 and 2 appears as the additive constant. When measuring the carbon potential, such correction can be done, like the other corrections, by shifting the zero or null point; in practical effect, and as illustrated in FIG. 2, this is readily accomplished by adding a correction potential.

It would be possible to compensate for the influence of several alloying elements in any one steel composition separately, that is, to provide a separate potentiometer for each alloying element. Since, however, the mutual influence of the various alloying elements must also be considered, it is usually easier in actual practice to determine a correction factor for any particular type of steel directly and then to adjust a corresponding potentiometer in accordance with a predetermined scale setting.

As has been noted, in practical effect it is useful to utilize a correction factor for the carbon potential itself, rather than a coefficient which compensates for the activity within the carbon crystal. In accordance with the present invention, this coefiicient for correction for the carbon potential can readily be determined. Referring now to FIG. 1, the third tube 8 includes a test. sample 10 consisting of pure, that is, unalloyed iron; further, the test sample consisting of the work piece is inserted therein. Both test samples are exposed to the temperature and atmosphere of the furnace for a sufficient time in order to obtain equilibrium. Thereafter, the test samples are removed from tube 8, quenched and analyzed for carbon content. The coelficient to be determined then is proportional to the quotient of the carbon content of the unalloyed, that is, pure sample (which is the same as that of the carbon potential for pure iron, of course) and of the alloyed work piece or sample. Mathematically, the relationship is percent C (pure iron sample) percent 0 (alloy sample) which, of course, corresponds to the definition given above.

Since this coefficient is, in part, dependent on carbon concentration, it is recommended to utilize a standard carbon addition, such as for example 1% C and to make the determination of the coeificient at least approximately in the region of such a value.

The scale of the correction potentiometer, which is intended to correct for the composition of the alloy, may contain not only the coefficient above-referred to but may further contain indications showing the particular types of steel to be treated. This greatly simplifies setting by even unskilled personnel.

If a carbon nitrating process is used, then the influence of the nitrogen absorbed by the iron and corresponding to the addition of ammonia can readily be compensated for.

Referring now to FIG. 3; this figure, basically corresponds to the circuit of FIG. 2. For simplicity, a single bridge network, that is bridge 29, has been illustrated in order to consider the correction factors previously discussed. Of course, in FIG. 3, as well as in FIG. 2, various serially connected bridge circuits can be used. A further bridge network 37 is added to the bridge networks 29, in similar manner. A source of electrical potential 137 is connected across one of the diagonal cross con nections of bridge 37, the other diagonal cross connection being placed in series with bridge circuit 29 and the connection to element 32; A variable potentiometer 38 is adjusted in order to introduce a correction potential, the value of which is determined by deviation between indicated value, that is value indicated by measuring instrument 15 and actual value as determined from analysis of a sample of iron strip, or foil introduced into the furnace.

The additional compensation compensates for aging and drift of the thermal elements providing for temperature measurement, as well as for errors which arise due to lack of thermo dynamic equilibrium within the furnace. The accuracy of measurement obtained is thus, practically, limited only by the possible errors in analysis of the iron sample strip. It has been found in actual practice that adjustment of potentiometer 38, to compensate for variations in thermo dynamic equilibrium, and in calibration of the temperature measuring elements is only necessary once or twice weekly. Changes in calibration of the measuring instrument, and changes in thermo dynamic equilibrium within the furnace occur only slowly.

The anaylsis instrument 14, instead of containing a humidity sensitive resistance, may be made to be sensitive to the partial pressure of the carbon dioxide component. The analysis instrument 14 then contains an infra-red analysis device 240, operating in accordance with infra-red absorption and delivering a potential which is dependent on the partial pressure of the carbon dioxide component introduced therein. By suitable choice of the length of the measuring chamber within unit 240, the output is made approximately proportional to the logarithm of the partial pressure of the carbon dioxide, that is, mathematically to log p The relationship of the output will then be in accordance with Formula 1 above.

Although the output should relate only to a particular component within the gaseous mixtures introduced thereto, the composition of the gas, in the aggregate, has a substantial influence on the result of the output from analysis apparatus 240.

A so-called cross-sensitivity exists which in the past 7 (percent 0) I has been compensated by special arrangements within the instrument itself, which introduced complications. For example, the particular gas to be considered can be absorbed specially; or, the undesired gases can be introduced into a special comparison or test vessel for sampling; or additional and known gases can be introduced or used for comparison. In accordance with the present invention, such undesired cross-sensitivity can be used directly for correction and calibration of the measuring instrument. The cross-sensitivity for the partial pressure of carbon monoxide is suppressed, that is compensated only insofar that it is just barely considered in accordance with the Equation 1 above in connection with the correction factor 2 log p Since carbon dioxide as Well as carbon monoxide are measured as partial pressures, due to the dependence of infra-red absorption on their density, and the absolute pressure within the furnace and thus Within the measuring chamber is approximately constant over long periods of time, correction for pressure is not usually necessary.

In actual operation it has been shown that an error tolerance of i0.05% C can be readily achieved and frequently improved on. Automatic control of maintenance of carbon potential can be obtained easily. The indicator 15 can be formed with maximum and minimum contacts, activated by the indicator pointer. Alternatively,a continously effective comparator can also be utilized. A manually settable potentiometer 52, is adjusted for a certain carbon potential value; it controls a potential which is compared in comparator 51 with the ouput from amplifier 32., also applied to indicator 15. Comparator 51 then controls a valving arrangement, consisting of

valves

53, 54 which, respectively, introduce propane or air into the atmosphere of the furnace. For example, if the carbon potential value to too low, propane is added by opening valve 53; if the carbon potential is too high, a small quantity of air is added by opening of valve 54.

The determination as to whether to measure the partial pressure of water vapor, or carbon dioxide depends on the general ambient conditions of each application. Measuring the water vapor pressure has the advantage that the measuring instrument 14 is comparatively inexpensive, because an infrared analysis device need not be used. Measurements of water vapor pressure usually are based on measurements of steam pressure, so that the result obtained is usually generally substantially proportional to the Water vapor partial pressure (p Transducers to determine the dew point, or humdity, such as lithium chloride are comparatively simple and cheap, particularly in comparison with analysis apparatus for carbon dioxide. One of the disadvantages of measuring the water vapor partial pressure is, that the vapor may condense within the ducting 13 (FIG. 1) when the dew point of the sampling gas exceeds the ambient temperature. It may thus be necessary to utilize heated ducts; arranging the filter 12 in the sampling assembly head 11, in accordance with the present invention and as shown in FIG. 1 avoids the necessity of heating the filter itself and avoids the major diificulties with condensation. When utilizing a lithium chloride element, the transformation temperature t of lithium chloride is measured directly, which is directly proportional to the water vapor partial pressure p In order to compensate variations of carbon monoxide and hydrogen components within the atmosphere of the oven, when utilizing the partial water vapor pressure as a base, a pair of potentiometers can be adjusted and utilized to introduce correction factors: one for variations of ambient atmospheric pressure, and the other for variations of the product of the volumetric portions of carbon monoxide and hydrogen. Since the atmospheric pressure variations are usually slow, correction twice daily in accordance with barometric pressure is sufficient. The scale on the potentiometer can thus be set to indicate from 700 to 780 torr. Calibration is in accordance with the Equation 2 above.

When the humidity of the ambient air varies substantially, then further corrections can be entered by one or 10 additional potentiometer bridge networks as discussed previously in connection with FIG. 2. Such a correction may be necessary if pure propane gas is used. The product vol. percent CO vol. percent H results in a value of 760 when pure propane and absolutely dry air is used; pure propane itself will yield a value of 840. These values change by about 40 units for each percentage of humidity within the air in the furnace. Thus, compensation for ambient humidity of air introduced may be necessary.

I claim: 1. Apparatus for determining the carbon potential of the atmosphere in a furnace, comprising:

means continuously determining the temperature of the atmosphere Within the furnace at a given point and deriving a corresponding electrical temperature signal; means removing an atmosphere sample from immediately adjacent said given point at which said temperature is determined; means analyzing the concentration of the CO component of the atmosphere and continuously generating an electrical-gas-characteristic signal representative of a defined function of the partial pressure of the CO component, said defined function being:

wherein:

c (in atomic fractions) is equal to concentration of carbon in gamma-iron; p, in atmospheres, is equal to partial pressure of the gas com-ponent in the atmosphere; T in degrees K. equals absolute temperature in the furnace; and 7 is a correction coefiicient for a specific alloy composition of a work piece in the furnace; means continuously electrically adding said temperature signal and said gas characteristic signal; means indicating a characteristic of said combined,

added signal to, continuously display the carbon potential directly; means intermittently generating an electrical error signal representative of the error value between the carbon potential of a standard material sample, as derived from an intermittently analyzed sample and the carbon potential as indicated; and means adding said electrical error signal representative to said error value to said added temperature and gas characteristics signal to reduce any error value to zero.

2. Apparatus as claimed in claim 1 including signal generating means connected to said adding means to add further electrical signals to said added temperature and gas-characteristics signals, said further electrical signals being representative of additional correction factors obtained during operation of said furnace.

3. Apparatus according to claim 1, wherein said electrical error signal is generated intermittently in about weekly intervals and added in about weekly intervals to said added temperature and gas characteristics signals.

4. Apparatus for determining the carbon potential of the atmosphere in a furnace comprising:

means continuously determining the temperature of the atmosphere Within the furnace at a given point and deriving a corresponding electrical temperature sig nal;

means removing an atmosphere sample from immediately adjacent said given point at which said temperature is determined;

means analyzing the concentration of the water vapor component of said atmosphere and continuously generating an electrical gas-characteristic signal representative of a defined function of the partial pressure 1 1 of the water vapor component, said defined function being:

wherein:

c (in atomic fractions) is equal to concentration of carbon in gamma-iron; p, in atmospheres, is equal to partial pressure of the gas component in the atmosphere; T in degrees K. equals absolute temperature in the furnace; and is a correction coeflicient for a specific alloy composition of a Work piece in the furnace; means continuously electrically adding said temperature signal and said gas characteristic signal; means indicating a characteristic of said combined,

added signal to continuously display the carbon potential directly;

means intermittently generating an electrical error signal representative of the error value between the carbon potential of a standard material sample as derived from an intermittently analyzed sample and the carbon potential as indicated; and

means adding said electrical error signal representative of said error value to said added temperature and gas characteristics signal to reduce any error value to zero.

5. Apparatus as claimed in claim 4 including signal generating means connected to said adding means to add further electrical signals to said added temperature and gas-characteristics signals, said further electrical signals being representative of additional correction factors obtained during operation of said furnace.

6. Apparatus to determine the carbon potential in the atmosphere of a furnace comprising means to sense the condition of the atmosphere in the furnace, said means being located within said furnace and including a thermocouple (6, 7) to determine the temperature within a point in the furnace and supplying a temperature signal;

a suction tube and means to suck gas from the furnace through the tube, said tube having a suction inlet located immediately adjacent the position of the thermocouple within the furnace;

a gas analysis apparatus (14) connected to said suction tube to analyze gas sucked therefrom, said apparatus including means to sense the partial pressure of the component of the gas sucked through the tube and providing an electrical gascharacteristic signal;

a function generator having said gas characteristic signal and said temperature signal applied thereto and providing a transformed gas-characteristic output signal, for the gas being C wherein c (in atomic fractions) is equal to concentration of carbon in gamma-iron; v p, in atmospheres, is equal to partial pressure of the gas component in the atmosphere; T in K. equals absolute temperature in the furnace; and is a correction coeflicient for a specific alloy compositon of a Work piece in the furnace; manually settable electrical potential generating means (29, 30, 31, 37) generating adjustment signals to represent'fixed factors and to represent the error value between the carbon potential of a standard material sample, as derived from an intermittently analyzed sample and the carbon potential as indicated; an adding circuit connected to and electrically adding said transformed gas-characteristic signahsaid' tem- 12 perature signal and at least one of said adjustment signals and providing an indicating signal; and an indicating means (15) connected to said adding circuit and having said indicated signal applied thereto and being directly readable in percent of carbon potential.

7. Apparatus according to claim 6 wherein said sensin means includes a gas filter (12) located immediately adjacent the inlet and of said suction tube.

8. Apparatus according to claim 6, wherein said sensing means includes a sensing assembly introduced into said furnace; and said sensing assembly includes a test sample tube to hold a standard sample, said tube having perforations to establish communication between the interior of the test sample tube and said furnace to expose said test sample to the atmosphere inside the furnace.

9. The apparatus as claimed in claim 6, wherein said manually settable potential generating means includes sources of potential and manually settable Potentiometers (16, 17, 18, 38).

10. The apparatus as claimed in claim 6, wherein said manually settable potential generating means includes means generating a fixed potential (33, 34, 35, 37); seriesconnected bridge circuits (29, 30, 31, 37) having one diagonal connection connected to said sources of fixed potential and the other diagonal connections connected in series; and manually settable potentiometers (16, 17, 18, 38) in a branch of each of said bridge circuits.

11. The apparatus as claimed in claim 6, wherein said means to sense the partial CO pressure includes an infrared spectrum-absorption apparatus.

12. The apparatus as claimed in claim 6, including control means connected to said indicating means, valve means regulating the composition of the atmosphere within said furnace; said control means controlling the valve means to keep the carbon potential Within the furnace at a desired point by addition of air or gas capable of changing the carbon content of the atmosphere within the furnace.

' 13. Apparatus to determine the carbon potential in the atmosphere of a furnace comprising means to sense the condition of the atmosphere in the furnace, said means being located within said furnace and including a thermocouple (6, 7) to determine the temperature within a point in the furnace and supplying a temperature signal;

a gas analysis apparatus (14) connected to said suction tube to analyze gas sucked therefrom, said apparatus including means to sense the partial pressure of the component of the gas sucked through the tube and providing an electrical gas-characteristic signal;

a function generator having said gas characteristic signal and said temperature signal applied thereto and providing a transformed gas-characteristic output signal, for the gas being H O component:

wherein c (in atomic fractions) is equal to concentration of carbon in gamma-iron; p, in atmospheres, is equal to partial pressure of the gas component in the atmosphere; T in K. equals absolute temperature in the furnace; and 'y is a correction coefficient for a specific alloy composition of a work piece in the furnace; manually settable electrical potential generating means (29,-30, 31, 37) generating adjustment signals to represent fixed factors and to represent the error value between the carbon potential of a standard material sample, as derived fronr-and intermittent- 13 ly analyzed sample and the carbon potential as indicated; an adding circuit connected to and electrically adding said transformed gas-characteristic signal, said temperature signal and at least one of said adjustment 5 signals and providing an indicating signal; and

an indicating means (15) connected to said adding circuit and having said indicated signal applied thereto and being directly readable in percent of carbon potential.

14. The apparatus as claimed in claim 13, wherein said means to sense the partial H O pressure is a lithium chloride humidity indicator and includes a bridge network (25) having one branch thereof formed by a humidity sensitive resistance (14a); said adding circuit including connection means interconnecting one diagonal connection of said bridge circuit with said thermocouple.

References Cited UNITED STATES PATENTS 2,935,866 5/1960 Schmidt 7327 2,949,765 8/ 1960 Thayer 7327 2,648,976 8/1953 Bur 73--23 3,128,323 4/1964 Davis 148-165X 3,070,990 1/1963 Krinov 73-25 3,237,928 3/1966 Warman 14816.5X 10 3,329,495 7/1967 Ohta 73-23X OTHER REFERENCES Steel Carburization and Decarburization-J. K. Stanley, The Iron Age, J an. 21, 1943, pp. 31-35.

15 RICHARD c. QUEISSER, Primary Examiner C. E. SNEE III, Assistant Examiner