BRUSHING PROCESS FOR CORROSION AND OXIDATION RESISTANCE
BACKGROUND OF THE INVENTION The object of the present invention is to economically produce a scale-free, smooth or directional surface finish on stainless steel coils after continuous line annealing.
Cold-rolled stainless steel sheet is used in contact with food, chemicals and pharmaceuticals because of its resistance to corrosion and oxidation. Thus the initial surface finish of the item made from stainless steel remains permanently and it is important that this finish be aesthetically pleasing, cleanable, and not detrimental to corrosion resistance. For its pleasing appearance a brushed finish, called #3 or #4 polish, is often used. This surface finish is generally produced on cold-rolled material which has been annealed and pickled by any of numerous schemes to remove oxides from the surface.
The polish finish is then achieved by abrading the surface.
This has several drawbacks. Pickling is expensive and produces large quantities of toxic effluent . The pickling acids and salts impose safety and health risks and impose environmental dangers . Maintenance of the large pickling baths requires great expenditure and the baths consume a good deal of space in a manufacturing line. Pickling removes the oxides generated in annealing and passivates the steel surface, however, it also etches and thus roughens the surface creating a roughness several times greater than the scale it removes, making subsequent polishing more difficult. Lastly, the polishing itself is
costly and negates the previous passivating effect of the pickling, making the surface less corrosion resistant.
Impregnated-abrasive brushes have become readily available only within the past 15 years employing hard materials such as SiC and A1203. New and stronger resin fibers have been recently formulated that have led to longer wear and broader acceptance and use of this abrasive technology.
Conventional stainless steel descaling practices take place after cold rolling and annealing at high temperature in an uncontrolled or controlled atmosphere. The presence of oxidizing gasses within the annealing furnace results in the formation of scales of metallic oxides of nickel, chromium, iron and other metal alloying additions. These scales adversely affect the corrosion resistance and serviceability of stainless products and must be removed after annealing. These oxides can be removed by acid pickling, electrolytic pickling or alkaline salt bath descaling. Most cost-effective pickling operations involve use of a combination of molten alkali salt bath followed by sulfuric, nitric and hydrofluoric acids .
Conventional polishing generally takes place after annealing, pickling and temper passing. Polishing may be performed in continuous coil form or in cut-to- length sheet form by abrading the surface with belts which have been coated with an abrasive mineral. The abrasion imparts a directional decorative finish which is used in many exposed applications where appearance of the stainless steel is of prime importance. This processing is costly and may result in polishing related defects during
subsequent processing operations if the abrasive residues are not completely removed after polishing. Abrasive coated belts also experience rapid wear as compared to relatively thick, impregnated-abrasive brushes. Two United States Patents disclose brushing, abrading or grinding rolled steel to remove scale. United States Patent No. 5,131,126 discloses a method for removing oxide scale from a hot-rolled stainless steel strip including the steps of applying a solution of alkaline earth metal chloride to the surface of an oxide scale layer formed on a hot-rolled stainless steel strip, allowing the solution to penetrate into the oxide layer, annealing the strip and descaling the strip with brushes. Practice of this method is not directed to the descaling of cold-rolled steel, it involves use of a chemical bath and the surface finish of the end product is not suitable for consumer use.
United States Patent No. 2,318,432 discloses removal of scale from hot-rolled steel where the ho -rolled steel sheet is first run through a scale breaker including staggered bending rollers which loosen the scale by progressively subjecting the sheet to sharp bends in opposite directions. The sheet is then brushed to remqve the scale. It is notable that the surface finish of the end product of this method is only suitable for cold rolling. To prepare a product suitable for delivery to a customer, the product must later be cold-rolled, annealed and again pickled.
A key objective of the proposed invention is to develop a commercial process wherein an attractive, scale- free, smooth or directional polished surface finish can be produced on annealed stainless steels in such a way that
pickling can be avoided entirely. A passivating step may be employed after this processing to further improve corrosion resistance, if desired. Impregnated-abrasive materials can effectively remove annealing oxides from the surface of the steel following annealing and thereby eliminate a costly, dangerous and hazardous effluent- producing pickling process.
A second key objective of the present invention is to consolidate processing of cold-rolled stainless steel into a single manufacturing line. Conventional lines using a pickling step require transfer of the steel coil to a high speed temper mill where the steel is flattened; it is then leveled and cut to length and polished as individual sheets or polished in coil and then cut into sheets. By eliminating the pickling step and replacing it with a brushing/polishing step, the high speed tempering step may be avoided and the tension-leveling step may be inserted after annealing but before brushing to ensure a superior finish upon brushing. Elimination of the high speed tempering step allows consolidation of the cold-rolled stainless steel production process into one continuous line, saving energy and production costs. The cut-to- length facility can then be in line with the annealing and brushing operation. SUMMARY OF THE INVENTION
In accordance with the present invention, a method is described for descaling and polishing flat cold- rolled metal strip which has been previously annealed, comprising the steps of annealing the strip and abrasively removing the oxide using thick, mineral-impregnated brushes or materials. The surface may be subsequently passivated
to restore full corrosion resistance. By eliminating the pickling step, a huge cost savings is recognized by eliminating the need for acid and alkali salt baths. This invention results in fewer environmental and safety risks usually associated with the use of pickling baths. The method includes steps of : a. annealing cold-rolled stainless steel; b. tension leveling the annealed cold-rolled stainless steel; c. wet brushing the steel to remove oxidation and to impart a desired finish to the steel; d. passivating the steel; e. slitting the steel to trim its edges; and f . shearing sections of the steel to a desired length, wherein the steps are performed in a single processing line. The present invention is also directed to a manufacturing line for carrying out the steps of the method described above.
In accordance with a further embodiment of the present invention, a method is presented for improving productivity of an existing conventional mill for producing cold-rolled stainless steel, having one or more pickling tanks wherein the mill may be converted to include an abrasive brushing station, eliminating the pickling steps. The mill may be further converted to include a tension leveler, a slitter and a shear in the same line as the annealing furnace and the abrasive rollers.
BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 shows a schematic diagram of a conventional processing of cold-rolled stainless steel;
Fig. 2 shows a schematic of the present invention;
Fig. 3 shows a graph of the surface roughness of the trial samples after abrading both annealed and 2D finished cold-rolled stainless steel; and
Fig. 4 shows a graph of the gloss measurements of the trial samples after abrading both annealed and 2D finished cold-rolled stainless steel.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
In the practice of the invention after hot rolling, annealing, pickling and cold rolling, the stainless steel is annealed in a continuous annealing line at 1500 to 2200°F. These lines typically subject the strip to several treatments after annealing to remove the oxide generated in the annealing furnace. Traditional methods for removing the oxide are by pickling, or treatment with strong acids, such as hydrofluoric or sulfuric acid, to remove the oxides. The pickling step also removes impurities, such as sulfides, that can result in pitting of the stainless steel. A conventional pickling process is shown in Fig. 1. In Fig. 1, a coil 10 of steel is unrolled and a steel strip is consecutively passed through an annealing furnace 18, a molten salt bath 20, a first rinse 21, a first strong acid bath 22, a second rinse 23, a second strong acid bath 24, a third rinse 25, a weak acid bath 26, a fourth rinse 27, a first temper pass 28, an abrasive belt polisher 29 and a second temper pass 30. In the present invention, the pickling treatment or treatments (A) are replaced by an abrading treatment as shown in Fig. 2.
As shown in Fig. 2, annealed and unpickled surfaces may be cleaned and/or conditioned using an impregnated-abrasive brush 40 on a continuous annealing or processing line 10 by abrading the sheet/strip surface. The present invention differs from the past conventional practice in that both the scale removal and the application of a directional decorative finish may be imparted to the stainless steel surface after the continuous line annealing operation but without the use of costly pickling solutions and/or abrasives after pickling.
The abrasive brush 40 roller or wheel contains hard, mineral abrasives which are impregnated within the brush fibers. This yields a very abrasive yet flexible contact patch with the steel surface. The roughness, reflectivity and surface finish are controlled by the brush arrangement, roll speed, contact pressure and the size of the abrasive particles in the brush 40. Preferably, the relative surface speed of the brush 40 to the steel surface is 1400 to 4000 feet per minute, most preferably, 2400 feet per minute. The pressure, or power consumption of the brush 40, not including drive train losses, is preferably .25 to 2.5 hp per inch of brush width, most preferably .66 hp per inch. This abrasive brush 40, wheel or belt has enough flexibility to contour with the steel surface and is sufficiently abrasive to remove any oxide scale layers or provide a distinctive, directional polish-type finish, depending on the grit of the abrasive employed. Impregnated-abrasive brushes, readily available only within the past several years, now offer significant cost savings over standard abrasive belts which are coated with an abrasive mineral. One such brush is a Scotch-Brite® #36
grit A1203 9A Extra Coarse Wheel, produced by 3M of Minnesota.
Since the impregnated-abrasive brush has mineral contained within the fibers, a more consistent finish is generated during the surface conditioning action. As the impregnated brush wears, "new" abrasive is continuously being exposed to the steel surface which improves product consistency and promotes much longer wear life compared to standard belts coated with an abrasive. The life of such rolls is quite long, and the operation of the continuous line thus need not be regularly interrupted to replace the rolls.
The brushes 40 and/or surface of the steel strip 12 are preferably wet during the abrasion process. The wetting may be accomplished through either spraying with a wetting solution or submerging the brushes and steel during the abrasion process.
Through abrasive action of the impregnated- abrasive brush, the surface of pickled materials can be conditioned to match the polishing finish produced using conventional coated abrasive belts, i.e., #3 and #4 polished finishes. The surface roughness and gloss produced by varying the number of passes of the abrasive brushes is set forth in Table 1, below, presented in conjunction with Example 2. Surface roughness is a very important indicator of the quality of finish of the steel . As a point of reference, experimental data shows that a #3 finish has an average surface roughness (Ra) of 22 micro inches with a range of 8-38 micro inches. The Kool Line® finish was shown to have an average surface roughness of 22 with a range of 17-32 micro inches. As shown below, surface
roughness readings of the abraded annealed, but not pickled steel strip were comparable to these readings for steel prepared by standard finishing processes.
The ease with which the strip may be polished and the uniformity of the polish is enhanced by the strip's flatness. To that end a tension-leveling device 42, preferably but not necessarily inserted after the anneal furnace 18 step and before the abrasive brushes 40, can provide excellent flatness for uniform brush contact. This also obviates the need for subsequent temper mill flattening.
The strip may then be conveniently passed through a passivating bath 44 and a subsequent passivating wash 45 which restores full corrosion resistance to the abraded surface. After the abrasion step, although the finish of the steel will appear bright, impurities will remain. Retention of these impurities, such as sulfides, in the surface of the steel will often result in pitting. "Passivating" the strip with an acid, such as nitric acid, will remove impurities. After passivation, the coil of strip may then be discharged from the continuous line.
As shown in Fig. 2, the steps of slitting the strip and shearing the strip to length can also be performed on the same processing line as the annealing, brushing and passivating steps. The strip can be passivated and subsequently slit in a slitter 46 to a desired width and then cut to length in a shear 48. The slitting step does not necessarily have to follow the passivating step although it is preferable that it does so. In conventional methods, annealing and pickling, temper mill flat passing, polishing and slitting and
re-temper passing are separate steps that are carried out at different speeds. These steps are usually followed by the low speed steps of roller leveling and cutting the coil to length. The present invention removes the need for separate lines having different processing speeds, allowing production of a finished product, finished and cut to size, from a single line. The combination of all steps into one line removes the need for multiple coilings and uncoilings of the strip and transport of the strip to other lines, resulting in lower production costs and energy usage, reduced floor space and less opportunity for damage of the sheets .
It was not previously recognized that a cold- rolled surface covered with a hard, brittle oxide scale could be polished, but experimentation showed that the thinness and the brittleness of the scale made it surprisingly less difficult to remove and that the underlying metal surface was extremely smooth, having not been etched by acid pickling. The present invention is also directed to increasing the productivity of an existing line used for the preparation of cold-rolled stainless steel. As shown in Fig. 1, one or more pickling steps (A) are used to prepare cold-rolled stainless steel according to conventional processes. These pickling steps are eliminated through practice of the present invention. An existing mill having a line for manufacturing cold-rolled stainless steel may be modified by eliminating the pickling tank(s) (A) and installing an abrasive brush 40 to replace the pickling tanks (A) . The brush 40 is preferably installed into an existing pickling tank along with a water
source for wetting the brushes. Further modifications to an existing conventional line preferably include installing a tension leveler 42 before the brushes 40, installing a slitter 46 after the annealing furnace 18 and installing a shear 48 at the end of the processing line.
EXAMPLE 1 A small-scale pilot trial was conducted on narrow 304 2D finished strips in light of promising preliminary results achieved using alumina/Sic brushes on the same substrate on a 52" wide line. On this small processing line, brush loading, abrasive wheel types and brush rotational speeds could be readily manipulated and controlled. A trial, "best practice", using an alumina impregnated wheel, #36 grit, and .25 Hp/inch of width brush loading, yielded an acceptable polished finish on top of the pickled 2D surface in just one pass. Based on these favorable findings, it was decided to apply this processing to 304 annealed but not pickled material, thereby eliminating the costly and hazardous pickling processing altogether. This trial is presented below in Example 2.
EXAMPLE 2 The original trials involved abrasive polishing of a 304 strip which had a 2D surface finish. This strip was abraded with various mineral wheels and produced good results in terms of surface roughness characteristics as compared to standard Kool Line® (employing hydrogen annealing, no pickling and temper rolling with an embossed set of work rolls) and polished finishes. Based on the success of the trial outlined in Example 1 above, it was decided to abrade an annealed but not pickled 304 strip.
The pilot line includes a single head abrasive wheel. Power was supplied by a vari ble-speed 15 HP motor. All line conditions remained constant during the trial and matched the "best practice" as shown in Example 1. The alumina wheel and all equipment were unchanged from the original trials. A brush oscillator was used at all times. All surface polishing was conducted wet using 60°F taD water.
All line conditions are presented below and remained constant during the trial in order to judσe the consistency of this process. The only variables introduced were the number of abrasive passes and the use of a 7A.VF, Very-Fine, wheel to smooth the surface after a rough wheel had been used. The line speed was maintained at 35 fum and the motor output was held at 60% of maximum load. If parasitic losses are 30% of load, then the net Hp aoolied to the pilot strip would be 30% of 15 Hp or 4.5 HD. Over an 8" wide mult of steel, the applied Hp/inch would be approximately 0.56. Brush speed was held at 1000 mm which yields a contact speed of approximately 3,142 fpm for a 12" diameter wheel. During the trial, the surface finish was visually evaluated. Surface roughness and gloss, at an angle of 60°, were recorded in the transverse rolling direction after a representative section had been conditioned by stopping the line. Hand samples were also obtained from the coil ends for additional testing and evaluation.
Two of the three 8.00" wide mults of 304 annealed but not pickled material from the coil were processed. Two different abrasive wheels were employed and up to three passes were applied to the surface. A total of 12 separate processing measurements were taken during the trial, shown in Table 1.
Table 1: 304 Annealed but not Pickled, Coil 7820609 B, .035" x 8.00"
Coil No. Grit/ Brush HP/I Ra Rz Rrnax Peak Line
Passes Mineral rpm μ in Speed, fpm
B5 1 #36A 1000 0.56 20 177 237 419 35
B5 1 #36A 1000 0.56 20 175 242 306 35
B5 2 #36A/#36A 1000 0.56 27 216 243 469 35
B5 2 #36A/#36A 1000 0.56 29 197 276 263 35 ω c
DO B5 2 #36A/#36A 1000 0.56 25 213 259 450 35
B5 3 #36A/#36a/#36A 1000 0.56 18 152 183 456 35 m ω B5 3 #36A/#36A/#36A 1000 0.56 24 210 262 481 35
_z m
H B5 #36A/#36A/#36A 1000 0.56 28 218 302 375 35
B5 3 #36A/#36A/#36A 1000 0.56 20 191 225 444 35
N> σ> Bl 1 #36A 1000 0.56 26 212 278 350 35
Bl 1 #36A 1000 0.56 29 221 256 375 35
Bl 2 #36A/7AVF 1000 0.56 9 86 137 331 35
Note; XA" represents A1203
From Table 1, the average surface roughness values were determined for the various surface polishing practices employed. These average values are plotted in Fig. 3 along with reported surface roughness averages obtained from a first pilot trial. Typical roughness values for Kool Line® (KL) (Ra = 22) and for #3 polished material (#3) (Ra = 22) are also presented in Fig. 3 for purposes of comparison. As shown in the figure, the transverse roughnesses for the annealed but not pickled substrate yielded average Ra numbers which were slightly higher than those for the first trial using the same wheel, #36A, but with the 2D substrate as the starting material. The roughness of the annealed but not pickled material did not appear to change appreciably after the second and third passes with the coarse #36A wheel as shown in Fig. 3. The roughness of the second trial coil (Bl) , prepared with the #36A wheel, decreased from over 20 micro inches to less than 10 micro inches af er conditioning with the very fine 7AVF wheel .
Gloss values during the trial were obtained using 3M's BYK Gardner Glossmeter and are illustrated in Fig. 4. Gloss values for the annealed but not pickled starting substrate were only 19 in the transverse direction before surface abrading. The gloss increased to 65 after one pass with a coarse #36A wheel and then to 110 when a coarse #36A wheel was followed by a fine 7AVF wheel, as shown in Fig. 4. A sample of 304 2D material was also measured for comparative purposes. The 2D finish had a high gloss reading of 138 in the as-received condition and the gloss decreased to 62 following one pass with a #36A coarse wheel.
Visual observations were made during the trial . With the exception of the first 500 feet of the first coil, which had wavy edges, all materials received uniform coverage using the coarse #36A wheel. This same #36A wheel was used in the first pilot trials and showed no signs of wear or degradation over five months even though it had been used often during this time for other projects. The #36A Scotch-Brite® wheel did not show any evidence of "clogging" after the trial was completed. No build-up of residue was observed in the pilot line rinse area. The Scotch-Brite® wheel only removes the oxide scale and very little, if any, metal is removed during processing. Any of the oxides present on the annealed but not pickled surface would be washed away with the rinse water. Thus, the #36A wheel produced a uniform and visually clean surface in just one pass when applied over the 304 annealed but not pickled substrate. Surface Ra values were in the 20 to 30 micro inch range using the #36A wheel over the 304 annealed but not pickled substrate, comparable to Kool Line® and #3 polish finishes. Roughness values can be reduced using the Scotch-Brite® process by applying smoother grit finish wheels to the annealed but not pickled substrate.
EXAMPLE 3 A 3M Scotch-Brite® wheel was dressed down to
9.75" diameter from the original 12" diameter as manufactured by 3M and was mounted onto a belt grinder. Five coils of 304 grade stainless steel, 50" wide, were cold-rolled to 0.059" (0.062" nom. ) , then processed through a cold anneal and pickle line while bypassing the entire pickling section. A trial coil (#7996245) was then placed
on the belt grinder to determine if the wheel could uniformly remove the annealed unpickled substrate. A list of processing variables, including line speed, wheel speed, belt grinder spindle pressure (measured in amps) , pre- contact roll wrap angle post-contact roll wrap angle, entry reel tension, surface roughness and, most importantly, visual appearance is shown in Table 2.
The trial began by determining the minimum power requirements needed to drive the wheel and belt grinder's spindle without contacting the surface of the coil, at both operating speeds of 900 rpm and 1,800 rpm (Table 2, set up 1 and 2) . The contact roll position was then placed at the two o'clock position, or 1.75" offset towards the entry end of the unit . The belt grinder processing variables as listed in Table 3 were initially arranged as noted in setup 3 by the belt grinder operator. During the first pass the wheel pressure and contact roll wrap angles were adjusted as documented in setups 3 and 12. At the conclusion of setup 12, it was visually observed that all of the unpickled substrate had been removed, and the coil exhibited a uniform surface appearance except for some heavy chatter. At this point setup 13 was used to process the coil and the surface roughness was checked. The Ra was 23 micro inches, the same as the average for #3 Polish. To try to eliminate the chatter problem, setups 14 through 25 were tried, but these set ups increased the chatter. Setup 13 was determined to be the best for removing the annealed unpickled substrate while producing a uniform finish. The belt grinder was then changed back to setup 13 and the remainder of the coil was processed. Tests were cut and an
additional surface roughness test was taken (Ra 22 micro inches) . Since the surface roughness and appearance were virtually the same as with setup 13 , the process was proven to be capable of being duplicated on a consistent basis. Setup 13 was then used to apply a second pass to both sides of the coil . Results of this process are shown in Table 3 setup 27. Surface roughness results were taken and shown to be Ra 16 to 21. The most important aspect of adding a second pass to the coil was that the chatter was dramatically reduced. To try to further reduce the surface chatter a third pass was applied to just one side of the coil using the same processing parameters as originally used in setup 13. Tests were taken after the third pass . Surface roughness tests showed the coil to have an Ra of 12 to 15 micro inches . Tests received from the head end of the second pass still exhibited some remnant chatter, but tests from the tail end of the third pass had no chatter.
Wear rate of the 3M Scotch-Brite® wheel was also measured. The wheel which was originally used on the belt grinder at 9.75" diameter had been reduced to 9.63" while processing the coil #7996245. During this time frame it was calculated that 12,000 linear feet was passed through the belt grinder. The total wheel loss was calculated to have been 91.3 in3. This measurement was then translated into wear life per linear foot of processed coil.
Total wear volume expected for a new 12" 3M wheel, assuming the scrap size is 7", would be 3,730 in3. Total wear usage of 3M Scotch-Brite® wheel in terms of total line as footage: 3,730 in3/0.007306 in3/ft = 510,539 ft. The above information was then applied to the average processing rates for a cold anneal and pickle unit to achieve an approximate cost per ton in incorporating this type of system into a cold anneal and pickle line. The costs involved in using the brushes (wheels) of the present invention were low and the wear rate of the brushes is long enough for substantially continuous operation of the line.
As can be seen from the preceding Examples abrading annealed, but not pickled cold-rolled stainless steel yields steel with a finish comparable to finishes commonly available on the market. This is shown primarily by the comparable surface roughness (Ra) achieved in the inventive process. The substitution of abrasive brushes provides a safe, clean and economical alternative to standard pickling processes.
Table #2 - Parameters evaluated and modified to produce a uni form directional finish comparable to #3 Polish
ω c
CD to
H
H C H m to z VD m m
H c m r- r σ>
to c
CD to
m o to x m m
30 c r- m t
The above invention has been described with reference to the preferred embodiment. Obvious modifications and alterations will occur to others upon reading and understanding the preceding detailed description. It is intended that the invention be construed as including all such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof .