US3457122A - Nickel alloy thermocouple - Google Patents

Description

July 22, 1969 F. s. slBLr-:Y ETAL 3,457,122

NICKEL ALLOY THERMOCOUPLE Filed NOV. 27, 1967 v(/\ 3mm: lavvmvd 'xawozmn woud :laniavdao aldnoaowum-u.

INVENTOR. FoREs S. SlaLEY a. NORMAN F. SPooNER BY @2f/nw, Md, M .E @lv-u@ ATTORNEYS United States Patent O 3,457,122 NICKEL ALLOY THERMOCOUPLE Forbes S. Sibley, Birmingham, and Norman F. Spooner, Bloomfield Township, Oakland County, Mich., assignors to Hoskins Manufacturing Company, Detroit, Mich., a corporation of Michigan Filed Nov. 27, 1967, Ser. No. 685,797 Int. Cl. H01v 1/22 U.S. Cl. 136-236 7 Claims ABSTRACT OF THE DISCLOSURE A thermocouple which substantially matches the EMF/ temperature curve of the standard Chromel/Alumel thermocouple but which has greater EMF and metallurgical stability.

The conventional Chromel/Alumel thermocouple has an electropositive leg the nominal ycomposition of which is 9.5% Cr, .4% Si, .1% Fe and 90% Ni and an electronegative leg the nominal composition of which is 95% Ni, 2.0% Mn, 1.4% Si and 1.6% Al. Chromel/Alumel thermocouples have been used for a period of over fty years for a majority of temperature measurement applications in the range of 1500 to 2000 F. The `popularity of this particular thermocouple can be attributed to its high EMF output, its relatively high resistance to oxidation, its relatively high EMF stability and relatively low cost.

However, under certain conditions Chromel/Alumel thermocouples shift rapidly in their calibration or fail prematurely. For example, under marginally oxidizing conditions, such as those often encountered in long narrow protection tubes or in reducing atmospheres, green rot corrosion occurs in a temperature range of about 15007 to 1800 F. and the EMF output-of the Chromel leg decreases substantially. In such atmospheres the chromium in the Chromel is preferentially oxidized. This is a recognized problem with Chromel/Alumel thermocouples and attempts have been made in the past to overcome it. Likewise, with Chromel/Alumel thermocouples if the wire size is relatively small, for example .020" dia., the Alumel leg oxidizes rapidly at temperatures above 1800 F. with a resultant substantial change in EMF output and premature failure of the thermocouple.

Various alloys have been proposed and developed in the past in an attempt to overcome the above-stated problems encountered with Chromel/Alumel thermocouples. While some of the prior art alloys have exhibited improved qualities in certain respects and under limited temperature ranges and in particular applications, no prior alloys have been developed which have all of the abovestated attributes of Chromel/Alumel thermocouples and which at the same time have improved resistance to oxidation and greater resistance to green rot corrosion.

The thermocouple of the present invention involves the use of two alloys which, when combined in a thermocouple, produce an EMF/temperature curve which substantially matches the standard Chromel/Alumel curve within the tolerances specified by The Instrument Society of America (ISA), as-shown in the drawings, and which at the same time exhibit a degree of corrosion resistance and EMF stability superior to the regular Chromel/ Alumel thermocouple as well as to the other prior art ICC alloys heretofore developed and proposed as substitutes for Chromel/Alumel thermocouples.

The alloy for the electropositive leg of the thermocouple of this invention has the following nominal composition: 10.3% to 11.5% Cr, 0.5% to 2.0% Fe, 0.5% max. Si, 0.5% max. Co and the balance Ni. The preferred composition of the electropositive leg is as follows: 10.5% to 11.2% Cr, 0.8% to 1.6% Fe, 0.05% to 0.4% Si, 0.0% to 0.25% Co and the balance Ni.

As is customary in melting practice, small amounts of deoxidizers are used in the melting of this alloy. These mayconsist of zirconium, magnesium, calcium and a mixture of rare earth, commonly referred to as misch metal. Other elements, such as titanium, columbium, alkaline earths, aluminum and manganese, can also be used singularly or in combination to deoxidize the alloy. The total of such elements resulting from deoxidation of the alloy should not amount to more than 0.5% and preferably should be less than 0.25 by weight.

In the composition of the alloy of the electropositive leg, the chromium content is maintained at a value as high as possible to minimize green rot attack and to generally promote oxidation resistance at elevated temperatures. The maximum chromium allowable in the alloy is governed by its influence on the EMF output. The iron content of the electropositive leg is 4also kept as high as possible to minimize green rot attack and, like chromium, the maximum content of the iron is governed by its influence on the EMF output of the alloy. The silicon content of the alloy forming the electropositive leg is also maintained at a value as high as possible to promote oxidation resistance. However, as is the case with chromium and iron, the maximum silicon content is governed by EMF considerations. Cobalt is present in the alloy as indicated in the above-stated compositions in trace amounts because of its presence in the raw materials used in melting of the alloy. Cobalt in larger amounts has an undesirable effect on the EMF output of the alloy but its presence in the amount of a few tenths of one percent can be tolerated. In this amount, however, it does not substantially increase the corrosion resistance of the alloy. For temperature measurement applications involving nuclear radiation cobalt in any significant amount could be objectionable.

The other alloy of this invention, utilized for the electronegative leg of the thermocouple, has a nominal composition as follows: 1.5% to 2.7% Si, 2.0% to 3.5% Cu, .4% to 1.4% Fe, 0.0% to 0.5% Co and the balance Ni. It is preferred, however, that the composition of the alloy for the' electronegative leg of the thermocouple fall within the following narrow range: 1.7% to 2.5% Si, 2.0% to 2.8% Cu, 0.5% to 1.0% Fe, 0.0% to 0.25% Co and the balance Ni.

Small amounts of misch metal and magnesium have been used as deoxidizers in melting the alloy. Other rare earths and alkaline earths may be used for` deoxidation as can be other elements such as titanium, columbium, zirconium, aluminum and manganese. The total of all such metals should not be greater than 0.5% and preferably should be less than 0.3% by weight.

In the alloy for the electronegative leg the silicon content is kept as high as possible for the same reason as in the electropositive leg, namely because of its benecial effect on oxidation resistance. The maximum allowable silicon is governed by EMF considerations. Iron and copper are added to the alloy of the electronegative leg primarily for their characteristic effects on the EMF/ temperature curve, that is, in order to conform the curve of the thermocouple to within the specied ISA tolerances for Chromel/Alumel thermocouples. As is the purposes. The recommended tolerances specified by ISA are $.09 millivolt (equivalent to about 45 F.) up to 530 F. and i%of 1% of the Fahrenheit temperature between 530 F. and 2000 F. The NBS circular referred to in Table II is published by the U.S. Department of case in the alloy for the electropositive leg, cobalt is not Commerce and is entitled Reference Tables for Thermointentionally added to the alloy; a trace amount of cobalt couples. The information concerning Chromel/Alumel is present in the raw materials. thermocouples appears at pp. through 28 of this cir- In the following table several typical compositions for cular.

TABLE 1I Departure (millivolts) from Cliromel/Alumcl Curve (Ref. June. 32 F.)

100 F 200 F. 300 F. 400 F. 500 F. 1,000 F. 1,600 F. 2,000 F.

EMF of Chromel/Alumel thermocouple (Mv.) 1 1. 52 3.82 6.09 8. 31 10.57 22. 26 36.19 44.91 ISA reeolmnended tolerances i. 09 :l: 09 :i: 09 :l:. 09 09 i. 18 i. 27 i. 33 P1 vs N1 .07 10 .11 13 .09 .05 +.04 -l-. l0 p1 vs 06 00 01 01 07 .14 20 .11 111 vs 07 .10 .01 .02 02 1s .25 .21 p1 vs 07 .10 .03 .04 .01 .20 .26 .21 p1 vs 06 10 10 .02 .05 06 -1-.04 +.15 p2 vs 06 0s 08 07 04 08 20 26 p2 vs .06 .07 +.04 07 12 .01 04 05 1 2vs. 06 07 02 04 -1-.08 .05 00 05 P2 vs 06 .07 .00 02 04 07 .10 .05 P2 vs 06 07 07 .02 .00 07 20 +.31 P3 vs .07 .11 .11 10 0s +.03 -1-.12 20 p3 vs 07 .09 01 04 +.08 06 .12 P3 vs 08 10 01 01 -1-.03 10 +.17 .11 P3 vs 0s 10 03 01 .00 .12 .18 .11 P3 vs 07 10 .10 .05 04 02 12 +.25 P4 vs 06 10 10 00 07 03 16 20 p4 vs 06 os 02 05 +.09 06 08 .01 p4 vs 07 .09 .00 02 04 10 13 .11 P4 vs 07 .09 .02 .00 01 12 .14 .11 P4 vs. .06 .00 .09 .04 .03 02 10 -.L. 25 P5 vs 05 08 06 .04 .02 +.19 +.33 +.37 P5 vs. -.05 06 +.06 +.10 +.14 -l-.10 +.00 16 P5 vs 06 07 04 07 -1-.00 06 +.04 06 P5 vs .06 .07 02 05 -1-.06 04 +.03 06 P5 vs .05 07 .05 01 02 +.1s +.33 +.42

l As published in NBS Circular 501.

several heats of the alloys for the electropositive (desig- Table II indicates that most of the combinations listed nated P) and electronegative (designated N) legs of the produce thermocouples which fall within the yISA tolerthermocouple are set forth: ance up to 500 F. and most combinations produce thermocouples which fall within the ISA tolerances be- TABLE L CHEMICAL COMPOSITION, PERCENT n l 7 N tween 500 F. and 2000 F. Those combmations which Cr Fe s1 C0 C r l do not fall within the ISA tolerances throughout the full o rr range of temperatures do, however, match the standard ig S23? 00 Chromel/Alumel curve within t.24 millivolt (equiv- .19 0.23 alent to *10 F.) up to 1000 F. and within 1% of the gi; Fahrenheit temperature from l000 to 2000 F. The ISA .79 tolerance ranges for Chromel/Alumel thermocouples are :gg 1:81 I shown graphically in FIGURE l and the EMF/tempera- Jg 60 ture curve of selected ones of the combinations set forth 7 in Table II are also shown in FIGURE l.

i In order to compare the stability of the thermocouple The compositions set forth above indlcate that the new formed with the new alloys with the stability of the alloy for the electropositive leg of the thermocouple dltregular Chmmel/Alumel thermoeouple at different temfers from ,regular Chromel primarily in its hlgher chromi- G5 perature ranges, samples of both types of thermocouples um content and its substantially higher iron content. The were tested in protection tubes in gas-tired ingot and bar composition of the alloy for the electronegative leg of the furnaces, In these tests 8 gauge wires were inserted in thermocouple differs from regular Alumel in that it does protection tubes open at the cold end, and exposed for not contain aluminum or manganese and contains a Subperiods of two to four weeks at temperatures from l900 stantial amount of copper and a higher silicon content. to 2300 F. in an ingot furnace and from 1700o to 2300 In Table II below are set forth the EMF versus tern- F. in a bar furnace. The EMF changes for both sets of perature values for all twenty-five combinations of the thermocouples at 2000 F. are tabulated in Table III. ve electropositive (P) elements and of the ve electro- In each test both thermocouples were exposed in the same negative (N) elements listed in Table I. The ISA tolerprotection tube and the EMF readings taken in a calibraances expressed in millivolts are included for comparison tion furnace at 2000 F. after the test.

TABLE IIL-COMPARATIVE STABILITY OF NEW ALLOY THERMOCOUPLE AND CHRO- MEL/ALUMEL THERMOCOUPLE (EACH LINE REPRESENTS SEPARATE TEST) [EMF change at 2,000 F.]

As indicated in Table I\II, appreciable changes in calibration occurred in the regular Chromel/Alumel samples tested in the ingot heating furnace. The condition in the protection tubes was such as to cause preferential oxidation of the Chromel leg of the thermocouple. No such drastic changes occurred in connection with the thermocouples formed with the new alloys. The test results set forth in Table III show that the new thermocouple has substantially greater stability than the regular Chromel/ Alumel thermocouple in this typical industrial application. These tests also indicate that the thermocouples of the new alloys show a positive shift in calibration While most of the regular Chromel/Alumel thermocouples shifted negatively. In other words, the effect of the oxidizing test environment on the new alloys is to cause the thermocouple `to indicate a furnace temperature slightly higher than the actual temperature Whereas the eifect of that environment on the regular Chromel/Alumel thermocouple causes the thermocouple to indicate a temperature substantially less than the actual furnace temperature. Obviously a positive shift in EMF is preferred to a negative shift as a fail safe feature.

In order to compare the new alloy for the electropositive leg with a regular Chromel element from the standpoint of their abilities to resist green rot corrosion, 14 gauge wires of regular Chromel and of the new electropositive alloy were subjected to alternately oxidizing and reducing conditions (air and cracked ammonia) at l600 F. for 350 hours. For the next 60 hours the wires were subjected at 1600" F. to an endothermc gas commonly used in heat treat furnaces. At the end of 410 hours the temperature was raised to 1800 F. and the test continued for an additional 1008 hours in an atmosphere of wet cracked ammonia. EMF readings against a stable standard were taken every 24 hours. 6

The change in the EMF for the two sets of wires is shown in Table IV.

TABLE IV.-COMPARATIVE STABILITY OF REGULAR CHROMEL AND NEW ELECTROPOSITIVE ALLOY UNDER OXIDIZING AND REDUCING CONDITIONS [Cumulative change in EMF (mv)] Iss The results tabulated in Table IV indicate that the samples of the new alloy for the electropositive leg of the thermocouple showed considerably less EMF change than regular Chromel. As was true of the test tabulated in Table III, the EMF of the Chromel alloy shifted negatively and the EMF of the new alloy shifted positively. After completion of the test, magnetic comparator readings were taken along the lengths of the two sets of wires. In the case of the regular Chromel wire, at certain locations the wire became somewhat magnetic whereas in the case of the wire formed of the new electropositive alloy the wire remained non-magnetic along its entire length. Photomicrographs taken of cross sections of the regular Chromel Wire showed that intergranular oxidation occurred. Similar photomicrographs of the new alloy wire showed no evidence of intergranular oxidation.

We claim:

1. A thermocouple comprising an electropositive element composed of an alloy consisting essentially of 10.3

.to 11.5% Cr, 0.5% to 2.0% Fe, 0.5% max. Si, 0.5%

max. Co and the balance Ni connected to an electronegative element composed of an alloy consisting essentially of 1.5% to 2.7% Si, 2.0% to 3.5% Cu, .4% to 1.4% Fe, 0.0% to 0.5% Co and the balance Ni.

2. A thermocouple comprising an electropositive element composed of an alloy consisting essentially of 10.3% to 11.15% Cr, 0.5% to 2.0% Fe, 0.5% max. Si, 0.5% max. Co, not more than 0.5% deoxidizing elements and the balance Ni connected to an electronegative element composed of an alloy consisting essentially of 1.5% to 2.7% Si, 2.0% to 3.5% Cu, .4% to 1.4% Fe, 0,0% to 0.5 Co, not more than 0.5 deoxidizing elements and the balance Ni.

3. A thermocouple comprising an electropositive ele- Iment composed of an alloy consisting essentially of 10.5% to 11.2% Cr, 0.8% to 1.6% Fe, 0.05% to 0.4% Si, 0.0% to 0.25% Co and the balance Ni connected to an electronegative element composed of an alloy consisting essentially of 1.7% to 2.5% Si, 2.0% to 2.8% Cu, 0.5% to 1.0% Fe, 0.0% to 0.25% Co and the balance Ni.

4. A thermocouple comprising an electropositive element composed of an alloy consisting essentially of 10.5% to 11.2% Cr, 0.8% to 1.6% Fe, `0.05% to 0.4% Si, 0.0% to 0.25% Co, less than 0.25% deoxidizing elements and the balance Ni connected to an electronegative element composed of an alloy consisting essentially of 1.7% to 2.5% Si, 2.0% to 2.8% Cu, 0.5% to 1.0% Fe, 0.0% to 0.25% Co, less than 0.3% deoxidzing elements and the balance Ni.

5. A thermocouple comprising an electropositive element composed of an alloy consisting essentially of 10.5% to 11.2% Cr, 0.8% to 1.6% Fe, 0.05% to 0.4% Si, 0.0% to 0.25% Co and the balance Ni connected to an electronegative element composed of an alloy consisting essentially of 1.7% t0 2.5% Si, 2.0% to 2.8% Cu, 0.5% to 1.0% Fe, 0.0% to 0.25% Co and the balance Ni, said thermocouple producing an EMF/temperature curve from 32 to 2000 F. which lies within 110 F. up to 1000 F. and within i1% of the Fahrenheit temperature from 1000 to 2000 F. of the standard ChromeL/Alumel EMF/temperature curve as shown in the drawing.

6. A thermocouple comprising an electropositive element composed of an alloy consisting essentially of 11% Cr, 1.2% Fe, .215% Si, less than .25% C0 and the balance Ni connected to an electronegative element composed essentially of 2.2% Si, 2.5% Cu, .8% Fe, less than .25% Co and the balance Ni, said thermocouple producing an EMF/temperature curve from 32 to 2000 F. which lies Within i10 F. up to 1000 F. and within 11% of the Fahrenheit temperature from 1000" to 2000 F. of the standard Chromel/Alumel EMF/temperature curve as shown in the drawing.

7. A thermocouple comprising an electropositive element composed of an alloy consisting essentially of 11% Cr, 1.2% Fe, .25% Si, less than .25% Co and the balance Ni connected to an electronegative element composed essentially of 2.2% Si, 2.5 Cu, .8% Fe, less than .25% Co and the balance Ni, said thermocouple producing an EMF/temperature curve from 32 to 2000 8 F. which lies within 1.09 mv. up to 530 F. and i3/4 of 1% of the Fahrenheit temperature between 530 F. and 2000 F. of the standard Chromel/Alumel EMF/ temperature curve as shown in the drawing.

References Cited UNITED STATES PATENTS OTHER REFERENCES McElroy, D. L.: ORNL-2467, U.S.A.E.C., received Sci. Lib., Mar. 10, 19,58, pp. i, 33 and 34.

Reference Data for Radio Engineers, 1943, pp. 23 and 24.

WINSTON A. DOUGLAS, Primary Examiner A. BEKELMAN, Assistant Examiner U.S. C1. X.R.

Images