US2769630A

US2769630A - Method for annealing tightly wound flat rolled metal stock

Info

Publication number: US2769630A
Application number: US417423A
Authority: US
Inventors: John D Keller
Original assignee: Individual
Current assignee: Individual
Priority date: 1954-03-19
Filing date: 1954-03-19
Publication date: 1956-11-06
Anticipated expiration: 1973-11-06

Description

Nov. 6, 1956 J. D. KELLER 2,769,530

METHOD FOR ANNEALING TIGHTLY WOUND FLAT ROLLED METAL STOCK Filed March 19, 1954 3 Sheets-Sheet 1 TEMPERATURE, F.

TIME, HOURS \g INVENTOR. wa. 1.0, Kne By J. D. KELLER Nov. 6, 1956 METHOD FOR ANNEALING TIGHTLY WOUND FLAT ROLLED METAL STOCK 3 Sheets-Sheet Filed March 19, 1954 INVENTOR.

NOV. 6, 1956 KELLER 2,769,630

METHOD FOR ANNEALING TIGHTLY WOUND FLAT ROLLED METAL STOCK Filed March 19, 1954 5 Sheets-Sheet 5 Lu v o 9 D O 8 z m 2 Q 2 u 0: Lu 0 o 5 g n: O u I: O z (D o LIJ E a 3 g 2 f I o w O E w z a z 8 :5 5 I 3: O O I c0 N 8 o o o o o o o o o o 9 co rw m r n N wasoaom Band 30 umuoncmoo -|0 maoaad NI aaruxlw svs d0 All/\LLOIICINOO 'IVWHEIHJ.

KO N

IN V EN TOR.

United States Patent METHOD FOR ANNEALING TIGHTLY WOUND FLAT ROLLED METAL STOCK John 1). Keller, Pittsburgh, Pa.

Application March 19, 1954, Serial No. 417,423

6 Claims. (Cl. 263-52) This invention relates to new and useful improvements in method of and apparatus for annealing tightly wound coils of flat rolled metal stock.

Heretofore cold rolled strip steel or other ferrous strip was annealed by placing the tightly wound coils in a boxtype annealing furnace in a non-oxidizing atmosphere in which the coil was raised to approximately the desired annealing temperature and subsequently permitted to cool. By this method the ductility and other physical properties desired in the strip for further processing and use is attained. The two greatest drawbacks of this conventional method of heat treating were the excessively long time consumed and the non-uniformity of annealing in different parts of the coils of strip. Sixty to one hundred hours or more have been required for the coils to become even approximately uniformly heated throughout and subsequently cooled, the reason for which excessive time requirement is the poor thermal conductivity of the coils because of the multitude of insulating air spaces between the convolutions of the coils. In practice the coils were stacked with their axes in vertical alignment which was an impediment to the free flow of heat into and out of the ends of the coils, approximately less than twenty percent of the heat entering and leaving the coils in that manner. By far the greater part of the heat is transferred to and into the metal radially through the coils, but because of the air spaces between the convolutions of the coils the heat conductivity in the radial direction is low since the air spaces act as a heat insulation. This not only renders the heating and cooling very slow but at the same time produces undesirably large differences of temperature within the coil, so that the strip material after processing is therefore far from uniform in quality throughout the coil.

In accordance with the present invention, the air which produces the undesirable retardation of heat flow is displaced and replaced by a gas of high thermal conductivity that is forced or injected into the spaces between the convolutions of the coils, which greatly increases the radial conductivity through the coils, improves the surface quality of the annealed metal, materially shortens the time required for heating and cooling the coils, and results in much greater uniformity of physical properties of the annealed strip material throughout all parts of the coils.

It is among the objects of this invention to subject metal strip in coil form to uniform heat treating temperature throughout the coiled strip in a minimum of time, thereby increasing the output of annealing equipment and decreasing the over-all cost of the heat treatment.

Another object of this invention is to anneal tightly woundcoils of cold rolled strip material by taking advantage of the minimum air space between the tightly wound convolutions of the coil which'has a minimum heat insulating characteristic compared with the greater insulating air space of loosely Wound coils and further minimizing the insulating or heat transfer resistance of the thin air space by substituting highly heat conducting gases for the air between the convolutions of the coil.

, 2,769,630 Fatented Nov. 1956 ice It is a further object of the invention to replace the atmospheric gases or air in such tightly Wound coils with helium gas which I have discovered to have high heat transfer properties and is favorable to maintaining the surface conditions and prevent oxidation of the cold rolled strip.

It is still another object of this invention to provide a mixture of helium and other gases to make them more suitable for annealing purpose by reducing the explosive hazard and if hydrogen is mixed with helium any oxygen present will be absorbed to prevent oxidation of the finished surface. If any degree of oxidation is desired hydrogen is omitted and controlled amounts of oxygen may be added.

It is still a further object of the invention to positively displace atmospheric gases or air between convolutions of tightly wound strip coils with high heat conducting gases by maintaining a pressure differential between the source of supply of such gases to the coils and the exit of the gases from the coils whether the displacing gases are passed end to end through the coils or through the convolutions from the inside to the outside or from the outside to the inside of the coils.

By a tightly wound coil is meant a coil just as it comes from the coiler in which it is wound as it leaves either the last pass of the cold-reducing mill or the electroly-tic cleaning line. The amount of tension under which the strip is wound on the coiler is not the criterion, but rather the percent of void space Within the coil. Careful measurement and Weighing of numerous coils has shown that even when wound under tension so high as almost to tear the strip apart, the coil contains at least 2 /2 percent void space, or is only 97% percent solid, while at the other extreme, when wound with the lightest tension that is practicable with regard to handling the coil in subsequent processing without having the inner convolutions slip out endwise from within the outer layers, the void space still does not exceed 7 percent, or the coil is at least 93 percent solid. A tightly wound coil, therefore, in strip mill practice is one which is between 93 percent and 97 /2 percent solid. This means that, for example, if the strip is .010 inch thick, the average width of the spaces between the convolutions does not exceed $5 of an inch and for strip of this thickness the average space width is usually 7 inch.

The invention will become more apparent from a consideration of the accompanying drawings, constituting a part hereof, in which like reference characters designate like parts, and in which:

Fig. l is a time and temperature curve showing the relative temperatures of the surface and inner parts of a coil when subjected to heat treatment by the conventional method;

Fig. 2 is a vertical section, partly in elevation, through a stack of coils and enclosure showing hollow spacers or gas inlet chambers between the ends of adjacent coils of the stack adapted for injection of the high conductivity gas into the narrow spaces between the coil convolutions, and other spacers or spiders to permit egress of the displaced air; i

Fig. 3 is a plan view of an egress spacer used between certain of the coils in a stack;

Fig. 4 is an elevation view showing the piping connections and auxiliary apparatus for supplying the highconductivity gas to the distributing chambers between the coil ends;

Fig. 5 is a curve showing the relation of the thermal conductivity of hydrogen to its degree of purity, or, conversely, to its contamination or dilution with the ordinary heavy gases such as nitrogen or carbon dioxide;

Fig. 6 is a vertical sectional view showing coils stacked ventional box annealing practice.

' cover.

and arranged for the injection of high conductivity gas without the use of spacers between adjacent coil ends.

The invention will be more readily understood by reference to the temperature conditions prevailing in con- With reference to Fig. l of the drawings, assuming the point X on the curve to represent a temperature reading on the outer surface of a coil being annealed by the conventional'method in a box-annealing furnace after twelve hours time; the temperature of the median section of the coil will be approximately 600 deg. F. as indicated at Y, Fig. 1 of the drawings. If, therefore, a desired annealing temperature for processing the coil is 1100 deg. or 1150 deg F., then as shown by the bottom curve thirty hours time is required for the inner parts of the coil to reach this temperature, but the temperature of the outer layers of the coil as shown by the upper curve has been'far in excess of that required for heat treating the metal and these outer layers during a large part of the thirty hours have been exposed to overheating temperature. The curves of Fig. l have been established by actual test by placing thermocouples in the outer and intermediate sections of the coils. The great lag in temperature as indicated by the space between the upper and lower curves is due to the insulating effect of the gas or air between the coil convolutions ing gases, such as carbon monoxide, may be mixed with the helium, and for the purpose of describing the invention the symbol HC will be herein employed to designate the gas or mixture of gases having high thermal conductivity used in carrying out my process.

The primary purpose of the injected high-conductivity gas mixture is to increase greatly the rate of heat conduction radially within the coils of strip, but it also has the secondary purpose of preserving orv improving the strip surface, for example by removing the rolling oil which is pressed into the pores of the steel during the cold-reducing process that precedes the annealing, which oil when vaporized by the annealing heat, if allowed to remain stagnant within the spaces between'the coil convolutions in some cases produces unsightly and hard-torernove carbon smut on the strip surfaces. Also, during the cooling part of the annealing cycle it is found impossible to prevent the entrance of some air into the space under the so-called ash can cover 11 and around the outside of the coils, by leakage through the sand seal -10 at the bottom of said cover, and in the conventional annealing process during the cooling part of the cycle this 7 oxygen-contaminated gas is drawn by the. thermal conthrough which approximately more than 96 percent of the heat which flows radially into the coil must be transferred by conduction. V

The gas between the wraps of the coils initially is air since the strip is coiled in air. On heating, part or all of the oxygen in this air may be taken up by the metal and the gas film may then consist chiefly of nitrogen. At

a some temperature, appreciable amounts of gases of uncertain composition may be given oii by thesteel. During the cooling part of the cycle which follows the heating and soaking, the contraction of the gas due to drop of temperature draws into the space between the wraps a certain amount of the protective gas surrounding the coil, whichin present conventional practice is the socalled DX or de-ox gas containing chiefly nitrogen, some carbon monoxide and about ten percent of hydrogen by volume, but which frequently has been contaminated with oxygen'by in-leakage of air into the space under cover 11 through the sand seal 10' at the bottom ofsaid air, nitrogen and DX gas does not differ .to any considerable extent from one to another and is low for all of them.

In accordance with the present invention, before and during the annealing operation the poorly conducting gases between the wraps of the coils are displaced by gases of much higher thermal conductivity, For this purpose either hydrogen or helium, or a mixture of the two, is found desirable and eflective. Hydrogen has approximately seven times and helium almost six times as great thermal conductivity as air. After heating, during the cooling part of the cycle a suflicient quantity of the high conductivity gas is injected into the space between the layers of the coils to prevent the entrance of gases of lower thermal conductivity into these spaces.

I prefer to employ hydrogen because it has the highest Thethermal conductivity of these gases, namely thermal conductivity of any known gas,' but the use of 7 pure hydrogen increases the explosion hazard. Further- 7 of strip as the high conductivity gas a mixture consisting chiefly of helium and a few percent of hydrogen, the latter less than the amount corresponding to the lower limit of explosibility in air. Other small amountsof deoxidiztraction into the spaces between the coil convolutions, causing bluing or other discoloration of the strip for a distance of about 2 inchesrinward from the strip edges. This ever-present drawback has caused strip mill operators to try all kinds of expedients in the endeavor to obviate it, without success. In my invention, by maintaining either a constant or an intermittent flow of protective gas through the spaces within 'ther'coils during both the heating and the cooling parts of the cycle, I avoid both the carbon-smut deposits and the discolored or' blue'd edges. l

oxidizing gases such as watervapor or carbon dioxide, By maintaining a flow of the, mildly oxidizing gas mixture within the coils, an oxide film of controlled thickness; is produced which is very uniform'overthe entire strip surface. At the same time, so long asthe percentage of heavy. gas added to the high-conductivity gas is not large, the resulting reduction of its thermal conductivity is not at all serious, as will be evident from the curves in Fig. '6 which refer to hydrogen.

Although the high-conductivity gases such as hydrogen and helium have from 5% to 7 times the thermal conductivity of the heavy gases such as air or the conventionatlly used de-ox gas, the overall rate, of heating and cooling of the coils is not increased in the same proportion when the internal spaces in the coils are filled with wnealing practice; which means that'so far-as rate of production only is concerned,-the radial conductivity of the coils and hence the thermalconductivity'of the gas mix- .ture which is injected into the coils need be only12 to 2% times that of air, if the same temperature differences within the coils as in the conventional annealing practice are allowed. e

For the purpose. of this invention, therefore', highconductivity gas or HC gas in defined as'a gas'imixt'ure having at least twice the thermal conductivity of air.

From Fig. 5 it will be seenthat this .would correspond On the other hand, for those classes of strip material in which it is vdesired to produce a thin but uniform oxide film on the metal surface, instead of a reducing gas such as hydrogen or carbon monoxide I add to the helium" a fairly small percentage of oxygen, oralternatively other 5 to a mixture of 33 percent or more of hydrogen with 67 percent or less of heavy gas such as nitrogen. For helium, it corresponds to about 40 percent or more of helium with 60 percent or less of heavy gases such as nitrogen.

I have found, however, that there is a decided advantage in using helium or hydrogen which is as nearly pure as possible consistent with obtaining the desired surface-protecting or surface-modifying properties of the gas mixture as mentioned above, and preferably I use hydrogen or helium which is at least 90 percent pure, because the superfluous or superrogatory conducting value which is not required for reducing the time of the annealing cycle, shows up in a great reduction of the temperature differences within the coil and hence in much more uniform metallurgical and physical properties throughout the coils. Thus, approximately, if the thermal conductivity required on the basis of heat transfer to and from the coil exterior surface is 2 times that of air, while the thermal conductivity of slightly diluted helium gas injected into the coils is 5 times that of air, then the superfluous conductivity within the coils will be 5+2=2 /2 times, which means that if the temperature ditference between the hottest and the coldest parts of the coils at the end of the heating part of the annealing cycle is 150 deg. F., it will be reduced to approximately 150:-2 /2= only 60 deg. F. by the injection of the helium.

It is an essential part of my invention to inject the HC gas into the void spaces within the coils by the use of pressure or pressure-difference. Hitherto, the experiment had been tried of merely introducing helium into the space outside the coils and within the ash-can cover 11, apparently with the expectation that the helium would penetrate into the spaces between the wraps of the coils by natural difiusion, but no evidence of its having done so could be detected and no reduction whatsoever was efiected in the time required for annealing. In the conventional box-annealing of certain grades of stainless steel strip in coil form, a hydrogen atmosphere is regularly used around the outside of the coils to prevent oxidation, but no speeding-up of the heating and cooling has been found to result therefrom. Theoretically, if allowed time enough the hydrogen orhelium outside the coils should eventually diffuse into the internal void spaces, but because of the very small width of the spaces which ordinarily are only of the order of ih inch wide, the diffusion is so slowthat centuries would be required for the gas to penetrate. Even under the action of the considerable pressure difference or suction produced in the internal spaces by the contraction of the internal gas during the cooling part of the cycle, the external gas penetrates only about 2 inches from the ends of the coils as evidenced by the mentioned discoloration of the strip edges. 7

Natural ditfusion of the high heat conductivity gases would increase with loose winding of the coil which has been suggested where gases such as hydrogen have been used in heat treatment for other reasons than to increase thermal conductivity. However, to increase the spacing between the convolutions would defeat the advantage of high heat conductivity of the gases introduced since the resistance to heat flow through the coil increases in direct proportion to the thickness of the gas layer.

Thus if the spaces between the coil convolutions were increased by loose-windingfrom the normal 3 inch to even as little as 31000 of an inch, the advantage of the hydrogen or helium as regards heat conduction would be totally lost even if the gas penetrated perfectly. But 9 inch is still far too narrow to permit even moderately rapid penetration by natural difiusion, and if the spaces were increased to probably inch or more, the heat conduction even with hydrogen in the wide spaces would be only as fast as with the original 6 tightly wound coils with air or de-oX gas in the inch spaces as in the conventional practice.

In my invention, on the contrary, I do not depend on natural diffusion but instead I positively purge the air or de-ox gas from the internal void spaces of the coils and positively replace it with high-conductivity gas by injecting the latter under pressure, or more accurately, pressure difference. By this is meant supplying the gas to one end of each coil at a pressure which is higher than that existing at the other end of the coil. The absolute pressure is not important; all that is necessary is that the pressure difierential be sutficiently high to force the gases into and through the narrow spaces between the convolutions of the coils at a reasonably rapid rate.

With reference to Fig. 2 of the drawings, a hollow gas-distributing chamber 12 is placed on the annealing base a with its closed side down and open side facing upward. This side need not be entirely open but may have numerous perforations of size and spacing suitable for distributing the gas over the end of the coil. The chamber may be ribbed internally for strength to support the weight of the superimposed coils. 0n this spacer is placed coil 1. Suitable sealing means 15 and 16 are provided at the inside and outside of the coil to prevent lateral escape of the gas and to enable gas pressure to be maintained adjacent the bottom end of the coil. These sealing means may for example be endless asbestos rings of suitable diameters which are pressed against the top of the spacer or chamber 12 and the outside and inside rims of the bottom of the coil by the weight of the coil resting on them, but the invention is not limited to this particular means of sealing.

On top of this lowest coil is placed egress spacer 13, the purpose of which is to allow the air or gas displaced vertically from the void spaces within the coils when gas pressure is applied in

chambers

12 and 14, to escape easily by flowing radialiy outward to the space between the outside of the coils and the cover 11. Spacer 13 may have the form of a spider as shown in Fig. 3. In some cases the ends of the coils have sufficient unevenness that one coil-end can be placed directly in contact with the one below it and still allow sufficient space for the displaced air to escape laterally without requiring the use of spacers such as 13.

When spacer 13 is used, a'second coil la is placed on top of 13 and above this is placed a gas-distributing chamber 14- which has openings in both its top and bottom faces and is provided with sealing means on both faces, at the inside and outside rims of the cod end. Above this is stacked a third coil in with its upper end not covered, unless the stack is to be five coils high, in which case a spacer 13 is placed on top of the third coil, a fourth coil above this, a gas-distributing chamber 14 on top of the fourth coil and the fifth coil on top of this with its top not covered by a spacer or chamber.

if, however, the coils are to be stacked four high, a spider 13 is placed on the base, the lowest coil is set on this, a gas-distributing chamber 14 with openings in both faces is placed on top of the lowest coil, a second coil on 14, a spider 13 on top of the second coil, a third coil'above this, a gas-supply chamber 14 above the third coil, and the fourth coil is set above this. i

The gas supply pipe 17 is then connected to all

chambers

12 and 14 by suitable unions, the ash-can cover 11 is placed over the stack of coils, and when all stacks of coils on the annealing base have been prepared in this way, the portable furnace, or annealing cover 2, is put in place over the stacks and the heating is started. Before or just after the start of the heating, the high-conductivity gas is admitted under pressure through pipe 17 into all gas-distributing

chambers

12 and 14 for a sufficient length of timeto purge out the internal spaces in the coils and fill them with the HC gas. Ordinarily not more than one or two minutes will be required to do this. Ihe

homo

period or else we I HC gas thereafter is ordinarily shutoff during the remainder of the heating and soaking part ofthe cycle. During the subsequent cooling part of the cycle, when there is danger that oxygen-contaminated gas from the space between the coil and the cover 11 will be drawn into the internal coil spaces by the effect of thermal contraction, either a slow seepage of the high-conductivity gas through thecoils is permitted to occur during this shots of HC gas under considerable pressure are injected at intervals. Insome cases, this slow continuous or rapid intermittent injection of HC gas is used even during the heating and soaking period.

To reduce the amount of expensive HC gas required, the space around the coils inside the cover 11 may be filled with ordinary de-ox gas admitted continuously through pipei18 as in the conventional annealing practice. A fan 7 may be used to produce circulation around the outside and through the central hole of the coils, but is much less important than in conventional annealing practice because of the high radial conductivity of the coils containing HC gas.

One preferred method of supplying HC gas to the coil stack is illustrated in Fig. 4, wherein 19 is a tank or steel bottle of helium or hydrogen as it is shipped into the plant; 20 is a similar bottle of hydrogen or carbon monoxide or oxygen or carbon dioxide as may be required to maintain or produce the kind of surface desired on the strip; 21 is' a mixing and storage tank provided with a suitable pressure-gage, a drain cock, and in the outlet pipe connected to pipe 17 leading to the gas-distributing chambers in the coil stacks a quick-acting valve 22 which may for example be a gas-tight rotary plug valve. To

' insure complete initial purging of tank 21 the latter may be filled with liquid and hydrogen or helium admitted by partially opening the valve on bottle 19 while the drain cock on 21a is opened; When all liquid has been drained out, the drain valve is. closed and the hydrogen or helium is allowed to continue to flOW'lIllIO tank 21 until the pressure therein has risen to, for example, 28 pounds per square inch; the valve on bottle 19 is closed and the valve on 20 is partially opened until the pressure in 21 has risen to, for example, 30 p. s. i., when all valves are closed. The proportion of the two gases in the mixture is indicated by the ratio of their partial absolute pressure. Instead of this simple intermittent method of.

proportioning and mixing the gases,'any of the wellknown continuous metering and used.

For purging the coils, the quick-acting valve 22 is suddenly opened wide for a period of usually one to two chamber used isthe bottom chamber 12, and the coils are stacked on top of each other without any spiders 12 or chambers 14, preferably with sealing means 15 and 16 between adjacent coil ends at the inner and outer rims of the coils, but in some cases without any sealing means and with adjacentcoil-ends in direct Contact. Considerably higher gas pressure must then be used for purging in a given time, and if seals are not used between the coil-ends the Waste of the expensive, HC gas is likelyrto be considerable and the purging especially of the top coils much less efiective. V V V Referring to Fig. 2, it will be evident that alternatively the flow of HC gas into and through the spaces between the coil wraps can be produced by introducing the HC gas through pipe 18 into the space between the'ash-can cover 11 and the outside of the coils at a pressure only slightly higher than atmospheric, and producing within chambers 12 and 14 a suction, by connecting the outer end of pipe 17 to a suction pump of any of the wellknown types located outside the furnace. The pressure difference between the approximately atmospheric pressure in spiders 13 or at the'top of the uppermost coil and the suction .within

chambers

12 and 14 causes the HC gas to flow downward in the narrow internal spaces within the bottom and top coilsand upward within the middle coil shown in Fig. 2. Because of the danger of mixing devices may be minutes, depending on the volume of tank 21 and the pressure of the gas in it, in relation to the number and size of coils to be purged.

For maintaining the presence of the HC'gas within the coils, valve 22 is preferably opened Wide at intervals for short periods, rather than kept partly open all the time. The reason for this is that because the coils are wound with considerable tension which produces radial pressure of each wrap on the adjacent ones, when removed from the coiler mandrel this pressure often causes localized buckling of some of the layers near the inside of the 'h ollow-centered coils and opens some triangular-shaped longitudinal spaces in the coil which often are about /2 inch Wide; and at low-velocity flow of the .HC gas too much of the gas would flow through these localized large spaces and too little through the distributed narrow spaces between the convolutions. The velocity in the narrow spaces varies directly as the pressure diflerence but in the large spaces only as the square root of the pressure difference, hence the'relative proportion of gas undesirably discharged through the large spaces can be re-" duced' by having the gas flow at high velocity for a short time rather than slowly for a long time.

In a modification of the method, the only gas-supply method is not preferred.

drawing air into the space under the cover 11, this alternative methodvis not preferred- When it is used, cover 11 must be completely free of leaks and a liquid seal instead of a sand seal 10 should beused at its base.

In still another modification as shown in Fig. 6, no spacers or chambers are used between the coil ends, but the coils are stacked one on top of another and on top of the uppermost coil is placed a solid cover 23 which preferably has a seal ring 24 between it and the top of that coil. This cover plate may be held tightly against the coil end by along bolt 25 passing down through the hole in the middle of the COll'lQ thebase plate on which the lowest 'coil rests. HC gas isadmitted under pressure through pipe 17 into the closed hollow center of the coil stack, and flows spirally outward throughthe narrow spaces within the coils, between their convolutions. For several reasons, this modification is not preferred. Because of the much longer flow path for the spirally outward instead of vertical flow of the gas, either much higher gas pressure must be used or a muchlonger time. taken for purging out the initially present air and filling the interior void spaces with the HC gas. The'waste of gas due to leakage radially outward across the ends of the coils may be considerable, unless a soft packing or sealing means is used between the coil ends. Furthermore, since the spaces between the coil wraps are usually somewhat narrower near the middle of the coil height than nearthe ends, the purging of the middle'parts of the coils may be less etfective.

To reduce or eliminate the leakage of gas outward 2 gas at a pressure slightly higher; than atmospheric under the cover 11 around the outside of the coils through pipe .18, and producing the required pressure difierence by connecting pipe 17 to a suction pump outside the furnace,"

creating a suction within/the hollow/"center of the coil stack. The HC gas then flows spirally inward between the wraps of the coils. For the mentioned reasons this "In caseswhere it is considered permissible to sacrifice the outermost convolution of the coil by allowing this outer layer to become blued or oxidized, the ash can covers 11 can be dispensed with and no de-ox gas used for protecting the strip but only the HC gas injected within the coils. This has the advantage that the heat transfer from the radiant tubes 4 or other heating elements to the outer surface of the coils can be greatly speeded up, first because one radiation jump is eliminated and second because the radiating temperature is not limited by what the metal of an inner cover can Withstand. Much greater advantage can then be taken of the five-to-sevenfold conductivity of nearly pure helium or hydrogen, and at the same time the high maintenance cost of the inner covers is eliminated.

Furthermore, in annealing furnaces heated with gas fuel which form the majority, and also in those heated by oil firing, by the use of my method radiant tubes as well as inner covers can be eliminated and the flame fired directly in the space between the refractory walls 2 of the furnace and the outside of the stacked coils. Direct firing has hitherto been unsatisfactory because a slowburning lazy flame had to be used in order to avoid damaging the inner covers 11, which resulted in overheating the top part of the coil stack while the bottom parts remained too cold. By my method, with the HC gas injected into the coils, the heat absorption by them is so fast that high-temperature highly radiant flames can be fired directly around the coils, not only resulting in uniform heating from bottom to top of the stack but still further speeding up the heating rate. Only the outermost convolution of the coil is oxidized by the flame, the inner Wraps or convolutions being protected by the HC gas in the spaces between them. By this modification of the method the high maintenance cost of radiant tubes is eliminated.

In the conventional method of annealing, the cooling part of the cycle ordinarily takes twice as long as the heating and soaking part, chiefly because the heat which is to be dissipated must first be radiated from the outside of the coils to the inner covers 11, and since radiation is proportional to the diiference of the fourth powers of the absolute temperatures, during cooling the rate of heat radiation is inordinately slow. By my invention, with the inner covers dispensed with and the coils filled with HC gas, during the cooling part of the cycle the furnace 12 is lifted oif and cool air is blown directly against the outside of the coils themselves. With only one instead of two radiation jumps (because no inner cover is present) the rate of heat radiation is doubled, and with rapid conduction of heat from the interior of the coils to their outer surface and'the additional rapid extraction of heat by blowing air or even by spraying water against the outer surface, the cooling time is very greatly reduced, and the output of each annealing base is correspondingly greatly increased.

Although several embodiments of the invention have been herein illustrated and described, it will be evident to those skilled in the art that various modifications may be made in the details of construction and operation without departing from the principles herein set forth.

I claim:

1. The process of annealing sheet metal strip, said process including the steps of assembling the strip by spirally winding the same under tension to form a compact coil between 93 and 97 /2% solid, charging a gas of at least twice the thermal conductivity of air between the convolutions of the wound strip under pressure suflicient to displace the atmosphere normally present, and subjecting the charged coil to annealing temperature for a period to uniformly anneal said strip.

2. The process of annealing sheet metal strip, said process including the steps of assembling the strip in spirally wound coil form with from to 7 inch spacing between the convolutions of the strip, charging a gas of at least twice the thermal conductivity of air between the convolutions of the wound strip under pressure suflicient to displace the atmosphere normally present, and subjecting the charged coil to annealing temperature for a period to uniformly anneal said strip.

3. The process of annealing sheet metal strip, said process including the steps of assembling the strip by spirally winding the same under tension to form a coil with 2 /2 to 7% void space between the convolutions throughout the coil, charging a gas of at least twice the thermal conductivity of air between the convolutions of the wound strip under pressure suificient to displace the atmosphere normally present, and subjecting the charged coil to annealing temperature for a period to uniformly anneal said strip.

4. The process of annealing sheet metal strip, said process including the steps of assembling the strip by spirally winding the same under tension to form a compact coil between 93 and 97 /2% solid, charging a gas of at least twice the thermal conductivity of air between the convolutions of the wound strip under pressure suflicient to displace the atmosphere normally present, and subjecting the charged coil to annealing temperature for a period to uniformly anneal said strip, while maintaining the gas between the convolutions of the coil under pressure greater than the pressure of the atmosphere outside the coil.

5. The process of annealing sheet metal strip, said process including the steps of assembling the strip by spirally winding the same under tension to form a compact coil between 93 and 97 /2% solid, charging a gas of at least twice the thermal conductivity of air between the convolutions of the wound strip under pressure suflicient to displace the atmosphere normally present, and subjecting the charged coil to annealing temperature for a period to uniformly anneal said strip while maintaining a pressure produced flow of said charged gas between the convolutions of the coil during at least part of the annealing cycle.

6. The process of annealing sheet metal strip, said process including the steps of assembling the strip by spirally winding the same under tension to form a compact coil between 93 and 97 /2% solid, charging a gas of at least twice the conductivity of air and of a character selected to produce a desired surface condition of the metal strip in the presence of heat between said convolutions under pressure suflicient to displace the atmosphere normally present and subjecting the charged coil to annealing temperature for a period to uniformly anneal said strip.

References Cited in the file of this patent UNITED STATES PATENTS