Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.


  1. Advanced Patent Search
Publication numberUS3688571 A
Publication typeGrant
Publication dateSep 5, 1972
Filing dateDec 11, 1969
Priority dateDec 11, 1969
Publication numberUS 3688571 A, US 3688571A, US-A-3688571, US3688571 A, US3688571A
InventorsAtkins Anthony G, Estes Bay E, Renner Richard L
Original AssigneeUnited States Steel Corp
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Apparatus for determining flatness deviation in sheet or strip
US 3688571 A
Continuous monitoring of the shape of cold rolled strip is provided by passing the strip, under tension, over a roll. A line of sensors on the roll surface measures the longitudinal stress distribution across the width of the strip indicating the location, magnitude and type of flatness defect which will appear when the tension is removed.
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

United States Patent Atkins et al.

[451 Sept. 5, 1972 [54] APPARATUS FOR DETERMINING FLATNESS DEVIATION IN SHEET OR STRIP legheny County, both of Pa.

[73] Assignee: United States Steel Corporation,

[22] Filed: Dec. 11, 1969 [21] Appl. No.: 884,087

3,324,695 6/ 1967 Sivilotti ..72/12 3,334,508 8/ 1967 Martin ..73/144 3,442,104 1969 Misaka et al. ..73/ 144 3,481,194 12/ 1969 Sivilotti ..73/144 3,499,306 3/1970 Pearson ..73/159 3,413,846 12/1968 Flinth ..73/141 3,526,l 14 9/1970 Wistreich ..72/9 3,554,025 1/ 1971 Andeisson et al. ..73/ 144 Primary ExaminerRichard C. Queisser Assistant Examiner-Herbert Goldstein Attorney-Rea C. Helm [57] ABSTRACT (g1 Continuous monitoring of the Shape of cold rolled strip is p g the p under tension of over a roll. A line of sensors on the roll Surface sures the longitudinal stress distribution across the [56] References cued width of the strip indicating the location, magnitude UNITED STATES PATENTS and type of flatness defect which will appear when the tension is removed. 2,674,127 4/1954 Garrett et a1. ..73/159 2,809,519 /1970 Kaestner ..73/159 3 Claims, 10 Drawing Figures III l8 1 1a 34 32 H I /6-- l6 2e 2;, I 20 @222; 3a 9 H To To EZZ/a; Display Un/f Z PATENTEDSEP 5:912

SHEET 1 OF 3 DISPLAY UNIT FIG 2 m. mum I4 m? mam 8 0 TE m w 4 mm n TY .L MN D w m m m A m a m m w w 8 m o n x rm wnw 1 F m, m M0 6 0 2 v a T Attorney PATENTED E 5 SHEET 2 OF 3 M in I/T/T/T fT/T/T fW/Tn EDGE T E IVS/0N QMRTER POINT T E IVS/0N CENTER TE IVS/0N QUARTER POINT T E IVS/0N E 06E TE IVS/ON FIG. 3



' & RICHARD L. RENNER A g M PATENTEDsEP 51912 3.688.571 sum sur 3 FIG. 5A H6. 55



Attorney APPARATUS FOR DETERMINING FLATNESS DEVIATION IN SHEET OR STRIP This invention relates to method and apparatus for determining the deviation from flatness in a sheet or strip and more particularly for continuously determining deviations from flatness in cold rolled strip steel as it is being rolled under tension and passing over a bridle roll under tension to a coiler.

When a steel strip is subjected to non-uniform deformation conditions in the roll bite internal stresses are created within the strip as longer over-rolled material attempts to line up with shorter under-rolled portions. In heavier gage strip, the stresses may not be of sufficient magnitude to buckle the strip but in the lighter gages of tinplate and foil it is more likely the stresses will relieve themselves by buckling the strip on exit from the mill. The extend of the defect depends on the thickness of the strip and the amount of non-uniform deformation.

Surface flatness of strip can be determined by mechanical measurements but this is most difficult with thin material, particularly if the material is under tension. In principal, specular reflection could be used to measure shape but the interpretation of results would be complicated and of questionable reliability. Visual observation of strip flatness is unreliable and almost impossible on a high speed production mill with the strip under high tension and masked by flow of lubricant and coolant. Thus, there is no method that we are aware of to assess the quality of flatness in a strip during rolling.

According to our invention, a line of load cells is mounted inside the bridle roll to measure the strip tension across the width of the strip. The distribution of the tension across the width of the strip then becomes a measure of flatness as to magnitude, location and type of deviation from flatness.

It is, therefore, an object of our invention to provide a method for monitoring the flatness of strip as it passes over a roll under tension.

Another object is to provide a method for determining the magnitude of flatness and deviation from flatness of strip as it passes over a roll under tension.

Still another object is to provide apparatus for continuously determining the flatness and deviation from flatness of strip as it passes over a roll under tension.

A still further object is to provide apparatus for determining the location, magnitude and type of deviation from flatness of strip as it passes over a roll under tension.

A still further object is to provide such apparatus for determining flatness regardless of the level of tension in the strip.

These and other objects will become more apparent after referring to the following specification and drawings, in which:

FIG. 1 is an isometric diagrammatic view of cold rolling with the bridle roll of our invention;

FIG. 2 is a schematic sectional view through the center of the bridle rolling together with connections to slip rings, timer and display unit of our invention;

FIG. 3 is a partial sectional view along line III-III of FIG. 2 showing the load cell and transfer pin arrangement;

FIG. 4 is a chart showing a typical signal display for a strip with center buckles on a five-pen strip chart recorder;

FIGS. 5A and 5B illustrate a strip with loose edges and the longitudinal stress distribution;

FIGS. 6A and 6B illustrate a strip with tight edges and the longitudinal stress distribution;

FIG. 7 illustrates a multi-point chart recorder readout of loose edges; and

FIG. 8 illustrates an oscilloscope readout of tight edges.

Referring now to the drawings, and particularly to FIG. 1, reference numeral 2 indicates an uncoiler from which a strip S is unwound. The strip S passes over and around entry bridle rolls 4 through work rolls 6, over and around exit bridle roll 8, over and around bridle roll 10 and is coiled on recoiler 12. This is a typical arrangement of a cold roll steel strip mill in which the strip S is under tension between the bridle rolls.

Referring now to FIG. 2, tension measuring exit bridle roll 8 is generally hollow and has five compression type load cells 14 mounted inside roll 8. The load cells may be Model 3108- manufactured by Lebow Associates, Inc., Oak Park, Mich. Other types of transducers may be used in place of the load cells. A transfer pin 16 is associated with each load cell 14 and extends from the load cell 14 through a hole 18 in the shell of roll 8, protruding beyond the surface of roll 8 very slightly to contact the underside of a thin rubber sleeve 20 covering roll 8. Strip S will just contact the projection in sleeve 20 caused by transfer pin 16 as shown in FIG. 3. The load cells 14 and transfer pins 16 are arranged so that holes 18 are in a straight line across the surface of the roll 8 parallel to its axis.

Roll 8 has a shaft 22 with a timer assembly 24 on one end and a slip ring assembly 26 on the other end. Timer assembly 24 has a set of contacts 28 which are closed by a cam 30 when pins 16 arrive at point A where strip S makes contact with roll 8 until pins 16 arrive at point B where strip S leaves contact with roll 8 as shown in FIG. 3.

Slip ring assembly 26 has a pair of input rings 32 with input connections to each cell 14 but for the purposes of illustration, a connection to only one load cell 14 is shown. A pair of output slip rings 34 is connected to each load cell 14 but for the purposes of illustration, the rings and connections to only one load cell 14 is shown. A pair of slip ring brushes 36 connects power to load cells 14 from a suitable power source 38 through connections 40. Another pair of slip ring brushes 42 connects the output from load cells 14 to a display unit 44 through connections 46. Contacts 28 are connected to display unit 44 through connections 48. Display unit 44 is a five-pen strip chart recorder such as manufactured by Brush Instrument Division of Clevite Corporation, Cleveland, Ohio, which records the signals from load cell 14 during the time when contacts 28 are closed. Journal boxes 50 of shaft 22 have load cells 52 connected to display unit 44 by connections 54.

Under optimal conditions, internal stresses are evenly distributed throughout the strip and the normal reaction tensile loading force of strip S on roll 8 is carried evenly distributed across the face of roll 8. The strip will remain flat when tension is removed. If the strip is rolled so as to develop uneven internal stresses, such as loose or over-rolled sections, those sections will carry less of the normal reaction tensile loading on the exit bridle roll 8 than tight or under-rolled sections.

Thus, the loose and over-rolled wavy and buckled edges of a strip shown in FIG. 5A resuit in the uneven internal stress distribution and the uneven tensile distribution on the exit bridle shown in FIG. 58, while the tight and under-rolled edges of the strip shown in FIG. 6A result in the stress distribution of the exit bridle roll shown in FIG. 6B.

As the strip passes over pins 16 the differences in tensile loadings across the face of roll 8 produce different signals which are recorded on display unit 44. FIG. 4 shows such signals for three revolutions of roll 8 where the differences between the signals indicate that the strip has very low tensile loadings in the center as compared to the edges and the edges have slightly different tensile loadings.

Thus the differences between signals from load cells 14 indicate the relative degree of flatness of the strip S as it passes over roll 8. Whether or not the strip will in fact have buckles will depend on the thickness of the strip and the amount of nonuniform deformation.

While the differences between signals from load cells 14 indicate that there may be a departure from flatness, this difference by itself does not indicate the extent of the departure and therefore does not indicate whether or not the departure from flatness is within tolerance limits. This is determined by first finding the total tensile loading from load cells 52 and then determining the average load cell tensile load from the contact areas of pins 16 as compared to the total tensile load area, the roll face area from point A to point B of FIG. 3. The average could also be determined by averaging all the load cell signals sent to display unit 44. The average tension is shown for each of the five recordings in FIG. 4. This could show, for example, that the edges and center are out of tolerance while the quarter points are in tolerance.

FIG. 7 shows the flatness readout where display unit 44 is a multi-point chart recorder which records the peaks or rms values of the signals from load cells 14. FIG. 8 shows the flatness readout where display unit 44 is an oscilloscope in which the sweep and afterglow provide a continuous flatness pattern. Obviously the flatness deviation signals may also be used as control signals for the strip production equipment by connections 56 from display unit 44 to adjust rolls 6. These readouts require additional conventional circuitry in display unit 44 to determine and display averages and tolerance limits so that comparisons can be made.

Timer cam 30 has been indicated as closing points 28 from point A to point B, thereby defining the degree of wrap of strip S around the roll 8. Contacts 28 control the length of the signals in display unit 44 which may be wide as shown in FIG. 4 or the signals may be narrow (with appropriate adjustments to average determination) by a different cam surface to provide the signal shown in FlGS. 7 or 8.

While five load cells have been shown and described, ideally a continuous tensile determination would be made across the width of the strip instead of at discrete locations. More than one line of load cells may be placed on a bridle roll.

The tension required in the strip should be of suffcient magnitude to provide some tensile loading throughout the width of the strip so that there will be a signal from the load cell at the least strained section of strip. This method is particularly useful when the high exit tension in a cold rolled strip mill masks any visual observation of flatness deviation. The relative flatness determination and the magnitude of the deviation from flatness are therefore independent of the strip tension.

This method of flatness determination is obviously useful for any type of metal or plastic material, either in the form of sheet or strip, that may develop internal stresses in forming and to which tension may be applied and the normal reaction force caused by the tension measured at a location where the material changes longitudinal direction.

While the apparatus shown and described uses an electrical signal system, fluid systems using for example oil or water, could also be used.

We claim:

1. Apparatus for determining the flatness characteristics of a longitudinally moving strip under sufiicient tension to maintain said strip substantially flat which comprises a roll with its axis perpendicular to the direction of travel of said strip over which said strip passes with an angle of wrap sufficient to provide a measurable reaction force perpendicular to said strip against said roll at the least strained portion of said strip, a plurality of load cells mounted inside said roll along a line parallel to the roll axis, a plurality of transfer pins each connected to one of the load cells at one end and terminating slightly beyond the surface of the roll at the other end, said pins being adapted to transfer the reaction force of the strip on the end area of each pin to the associated load cell when the rotation of the roll causes a row of pin ends to contact the strip, said load cells providing negligible displacement under load, and means connected to the load cells for displaying the output of the load cells.

2. Apparatus according to claim 1 in which said roll is a bridle roll in a cold roll steel strip mill.

3. Apparatus according to claim 1 which includes a relatively thin resilient covering over said roll between the strip and the roll face and the roll face ends of said transfer pins.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US2674127 *May 17, 1952Apr 6, 1954Olin Ind IncFlatness indicator for cellophane film
US2809519 *Sep 22, 1954Oct 15, 1957Olin MathiesonWeb flatness indicator
US3324695 *Feb 10, 1965Jun 13, 1967Asea AbDevice at strip mill for deciding strip tension distribution across the strip
US3334508 *Nov 9, 1964Aug 8, 1967American Metal Climax IncMethod and apparatus for controlling flatness in sheet metal
US3413846 *Jul 12, 1966Dec 3, 1968Asea AbDevice in strip mills for determining the distribution of strip tension over the width of the strip
US3442104 *Jan 25, 1966May 6, 1969Sumito Metal Ind LtdControlling method and measuring instrument for the flatness of strips
US3481194 *Oct 5, 1966Dec 2, 1969Asea AbStrip flatness sensor
US3499306 *Jul 6, 1966Mar 10, 1970British Aluminium Co LtdMeasurement of the shape and flatness of sheet or strip material
US3526114 *Jul 7, 1969Sep 1, 1970British Iron Steel ResearchRolling of strip
US3554025 *Jan 31, 1968Jan 12, 1971Bofors AbForce measuring device
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US3738151 *May 2, 1972Jun 12, 1973United States Steel CorpStrip load simulator for shape-measuring roll
US3890834 *Apr 17, 1974Jun 24, 1975Asea AbForce or tension measuring means
US4016756 *Jun 8, 1972Apr 12, 1977Beloit CorporationNip load sensing device
US4062235 *Jun 9, 1976Dec 13, 1977Hazelett Strip-Casting CorporationTwin-belt continuous casting wherein the belts are sensed by mechanical probes
US4188809 *Mar 29, 1978Feb 19, 1980Hitachi, Ltd.Apparatus for detecting the shape of a strip
US4747911 *Oct 23, 1986May 31, 1988Boise Cascade CorporationLaser generator, detectors for comparison of shadow images
US5678447 *Apr 17, 1996Oct 21, 1997Eastman Kodak CompanyOn-line web planarity measurement apparatus and method
US5901591 *Apr 29, 1996May 11, 1999Tippins IncorporatedHot strip reversing mill
EP0262002A1 *Jul 29, 1987Mar 30, 1988Clecim SaFabrication method of a roller for checking a strip profile
EP0270442A1 *Nov 18, 1987Jun 8, 1988ClecimRoller for checking the profile of a moving strip
WO2007016773A2 *Aug 1, 2006Feb 15, 2007Dofasco IncMethod and system for detecting faults in sheet material
U.S. Classification73/862.46, 73/159, 72/11.4, 72/11.7, 72/31.7
International ClassificationG01B7/34, B21B38/02, B21B38/00
Cooperative ClassificationG01B7/345, B21B38/02
European ClassificationB21B38/02, G01B7/34B
Legal Events
Mar 31, 1989ASAssignment
Effective date: 19880112