EP0649512A1

EP0649512A1 - Device for measuring straightness

Info

Publication number: EP0649512A1
Application number: EP94915602A
Authority: EP
Inventors: Jean-Louis Delastre
Original assignee: SA Prodys Equipement; EXA Ingenierie
Current assignee: SA Prodys Equipement
Priority date: 1993-05-10
Filing date: 1994-05-09
Publication date: 1995-04-26
Also published as: HU9500025D0; US5519944A; FR2705145A1; WO1994027114A1; KR950702701A; JPH08503777A; FR2705145B1; HUT71145A

Abstract

Device for measuring the straightness of a conductive object (12), comprising a plurality of contactless distance sensors (22) aligned in the axis of a straight-edge (20) to be placed on the object. Means (24-1) are provided for transmitting the measurements supplied by each sensor.

Description

STRAIGHTNESS MEASURING DEVICE

The present invention relates to straightness measuring devices, in particular for measuring the straightness of a rail at its weld.

There are many ways in the laboratory to measure or control the straightness of an object. These methods use systems that are too complex or too bulky to be used outside the laboratory, for example to measure the straightness of a rail on site, or of any other fixed object or too bulky to transport it in a measurement laboratory.

It is important, after welding and grinding a rail, to check its straightness to ensure that excessive deformations do not risk causing vibrations when passing a train. FIG. 1 very schematically represents a conventional device for measuring rail straightness. This is a device known as the GEISMAR rule. This device includes a chassis, not shown, provided with two feet 10 which are placed on a rail 12. a toothed belt 14 is stretched between two pulleys 16 and 17 parallel to the rails 12. One of the pulleys, 16, is provided with a crank 16 -1 allowing to drive, parallel to the rails 12, a carriage 18 fixed to the belt 14. The carriage 18 is guided by slides, not shown, which must be manufactured with great care because their straightness must be particularly precise. The carriage 18 is provided with a probe 18-1 urged elastically towards the rail 12. To measure the straightness of the rail 12, in particular at a weld 12-1, an operator actuates the crank 16-1 to move the carriage 18 from one pulley to the other. During the movement, a stylus, not shown, actuated by the probe 18-1 via a link system, traces the profile of the rail on a strip of unrolled paper while the carriage 18 moves.

A first drawback of this device is that it is particularly heavy (it takes at least two people to transport it) because, to guarantee suitable precision, it is rigid in construction, which requires the use of massive steel parts.

Another drawback of such a device is that it is particularly fragile, that is to say that it can quickly lose its precision, for example as a result of shocks or the fact that it has been placed in a wobbly position during transport or storage. Once such a device has lost its precision, it is practically impossible to correct it. It is then necessary to establish a standard profile of the device which must be subtracted from each profile taken from a rail to obtain the real profile of the rail.

An object of the present invention is to provide a portable device for measuring the straightness of a conductive object, in particular of a rail, which is particularly light and easy to transport. Another object of the present invention is to provide such a device which is particularly easy to calibrate even after a significant deformation.

Another object of the present invention is to provide such a device which is particularly practical to use.

These objects are achieved thanks to a device for measuring the straightness of a conductive object, comprising a plurality of non-contact distance sensors aligned along an axis of a ruler to be placed on the object, and means of communication of the measurements provided by each sensor.

According to an embodiment of the present invention, the sensors are capacitive sensors. The rule comprises at each end a shim to be placed on the object, at least one of which is conductive and provides the object with an electrical signal necessary for the operation of the capacitive sensors.

According to an embodiment of the present invention, the capacitive sensors are produced on rectangular printed circuits comprising on a face opposite to that where the sensors are made of surface-mount electronic circuits intended to process the signals from the sensors.

According to an embodiment of the present invention, the device comprises means for displaying in a coordinate system the amplitudes of the measurements supplied by the sensors as a function of the respective positions of the sensors. According to an embodiment of the present invention, said object is a rail.

According to an embodiment of the present invention, each end of the rule is removably attached to a block which includes jaws intended to clamp the rail while aligning the axis of the sensors on the axis of the rail, means being further provided for, once the blocks are fixed on the rail, reposition the ruler laterally so as to be able to measure the straightness of a side of the rail at a predetermined position. According to an embodiment of the present invention, the capacitive sensors are formed by the central part of aligned rings engraved on a layer of copper on the printed circuits, the copper remaining outside the rings playing the role of guard rings. .

According to an embodiment of the present invention, the printed circuits are held in two internal internal grooves of a U-shaped portion of the rule, one of the sides of each groove being straight and serving as a support for the printed circuits. and the other of the sides of each groove being inclined so as to come into contact with the printed circuit without the latter reaching the bottom of the groove. Clamping means are provided to bring the two flanks closer to the "U" portion.

These objects, characteristics and advantages as well as others of the present invention will be explained in detail in the following description of particular embodiments made in relation to the attached figures, among which: FIG. 1, previously described, represents a positive device. conventional portable rail straightness measurement; FIG. 2 represents an embodiment of a portable device for measuring the straightness of a conductive object according to the present invention; FIG. 3 schematically represents an architecture of an information processing circuit supplied by distance sensors used in the device according to the invention; FIG. 4 represents capacitive distance sensors produced on printed circuit; 5 shows a rule for maintaining a set of printed circuits of the type of that of Figure 4; FIG. 6 represents an external perspective view of an embodiment of a measuring device according to the invention; and Figure 7 shows a sectional view of the device of Figure 6 placed on a rail, a second position of an element of the device being shown in dotted lines. FIG. 2 schematically represents a device for measuring straightness according to the invention placed above a weld 12-1 of a rail 12. This device comprises an elongated part 20, hereinafter rule, provided on its underside a plurality of non-contact distance sensors 22, for example inductive or capacitive, aligned along an axis corresponding to the axis along which the straightness is to be measured. Contactless sensors can be used thanks to the fact that the rail 12 is electrically conductive. The measuring device further comprises a housing

24 provided with a screen 24-1 making it possible to display, thanks to the electronic circuits described later, the measurements provided by each of the sensors 22. Preferably, the display takes place in a coordinate system where the the values of the measurements are represented as a function of the positions of the corresponding sensors. The difference between sensors 22 is constant according to one embodiment. It could also be variable in other embodiments; it is possible to have more sensors per unit of length in the center of the rule 20 in order to raise the profile of the rail more precisely at the level of the weld 12-1. The coordinate axis assigned to the positions of the sensors is then graduated accordingly on display 24-1.

The rule 20 is provided at each end with a wedge 26 to be placed on the rail 12. These wedges are made of a hard material, for example metal carbide, to minimize their wear during the many uses of the device.

Such a measuring device is particularly easy to produce because it is not necessary to position the sensors 22 with great precision, in particular perpendicular to the rail 12. In fact, as will be seen later, the sensors 22 are managed by a microprocessor (which is the easiest way to do this) which, on request, runs a calibration program to store values of correction for each sensor 22. Thus, if it is feared that the positions of the sensors 22 have been modified following a shock, for example, it suffices to rerun the calibration program. In addition, since it is not necessary to provide precision mechanical members which must withstand forces and wear over time (carriage slides in the device of FIG. 1), the measuring device according to the invention does not have heavy parts. Of course, the position of the sensors 22 must be stable over time, but these sensors are not subjected to any force and it is not necessary to make the rule 20 which keeps them particularly rigidly. Consequently, a measuring device according to the invention is particularly light and portable by a single person (for a measuring length of approximately 1 m, the device weighs approximately 8 kg).

It goes without saying that a device according to the invention is particularly practical to use since the micrαprOσesseur performs most of the measurement steps. The user only has to press a button to view the rail profile.

FIG. 3 schematically represents an embodiment of the architecture of the sensor management circuit 22 of the capacitive type. Preferably, capacitive sensors are chosen because they make it possible to provide an amplitude signal proportional to the distance measured. The same elements as in FIG. 2 are designated by the same references. Each sensor 22 consists of a metal plate surrounded by a guard ring 22-1 which is connected to a reference potential. The plates 22 are arranged close to and parallel to the rail 12 to be measured. Each plate 22 is connected to a terminal with an adjustable capacity C and to the input of a differential amplifier 28, another input of which is connected to the guard ring. Capacities C are adjusted once and for all during manufacture. The outputs of the amplifiers 28 are supplied to an analog multiplexer 30 which switches only one of the outputs of these amplifiers 28 to the input of an amplifier 32. The multiplexer 30 is controlled by a microprocessor 34. The output of the amplifier 32 is supplied to a demodulator 36, another input of which receives a sinusoidal signal supplied by a generator 38. The output of the demodulator 36 is supplied to a filter 40 which supplies a practically continuous voltage Vd corresponding to the distance measured by the sensor 22 selected by the multiplexer 30. The voltage Vd is applied to a modulator 42 whose other input receives the sinusoidal signal supplied by the generator 38. The generator 38 also supplies the adjustable capacitors C. The rail 12 receives the output of the demodulator 42 , which is connected to one of the shims 26 on which the measuring device is placed.

The measuring principle of a capacitive sensor is classic and will not be described. On the other hand, one aspect of the invention is to group the amplifier 32, the demodulator 36, the generator 38, the filter 40 and the modulator 42 for all of the sensors 22 which may be in high number. These elements being grouped together being the most costly in a capacitive measurement system, significant savings are made.

The output Vd of the filter 40 is supplied to an analog / digital converter 44 connected to the central unit 34. The microprocessor 34 is further associated with a memory 46 (ROM ROM and RAM), with a keyboard 48 and in the above-mentioned screen 24-1, which is for example a liquid crystal matrix display.

To use the measuring device according to the invention, a user presses a specific key on the keyboard 48. A program, stored in ROM memory, is then executed by the microprocessor 34. This program successively selects the sensors 22, reads the corresponding Vd values supplied by the converter 44, and stores these values in RAM memory. Then, or simultaneously, the values read are corrected by values stored in a non-volatile memory, for example during a calibration step, and displayed adequately on the screen 24-1. As previously indicated, the measured values can be displayed in a coordinate system as a function of the relative positions of the respective sensors, which directly gives the profile of the measured rail. The microprocessor 34 can also perform numerous operations on the stored values, for example smoothing or any other operation deemed useful.

As previously indicated, calibration values are stored for each sensor. Indeed, the distance d to be measured by each sensor is expressed by d≈ A Vd + B, where A and B are constants. A calibration therefore consists in memorizing the values A and B for each sensor. To do this, we proceed as follows.

A first calibration phase is carried out with the rule placed directly on a conductive reference plane. This plane can be a steel plate of precise flatness. Preferably, the reference plane is a body of water whose flatness has the advantage of being perfect. It is possible, according to the invention, to use a water body as a reference plane because the sensors are non-contact, that is to say that they do not disturb the surface of the water which stays perfectly flat. The wedges 26 are then placed on metal pillars which bathe in the water. The user then selects, by pressing a specific key, this first calibration phase. The microprocessor performs a first series of measurements by measuring the voltages Vd for each sensor 22, these voltages Vd corresponding to distances d assumed to be zero.

A second calibration phase is then carried out which consists in placing the rule on the reference plane by placing reference blocks of known height between the reference plane and the blocks 26. The user selects the second calibration phase which consists in carrying out a second series of measurements by memorizing the voltages Vd, which then correspond to distances d equal to the height of the reference blocks. There is thus, for each sensor, a system of equations with two unknowns whose unknowns are the coefficients A and B. At the end of the second calibration phase, the microprocessor calculates and stores the coefficients A and B for each sensor. The realization of the programs allowing to carry out the various operations described above is within the reach of any programmer.

FIG. 4 represents an embodiment of a set of sensors 22. These sensors are produced on one face of a rectangular printed circuit 50. The sensors 22 are formed by etching in the copper layer of one of the faces of the printed circuit of the rings whose internal zones constitute the sensors 22. The remaining copper surface constitutes the set of guard rings 22-1. FIG. 5 represents an embodiment of a measurement rule according to the invention, that is to say the part serving as support for the sensor 22. This rule is formed from a profile comprising a part 20-0 to inverted "U" section. One or more printed circuits 50 of the type of FIG. 4 are slid into internal grooves 20-1 of the sides of the U-shaped part, near the lower part. As is shown, one of the sides of each groove 20-1 is horizontal and serves as a reference surface and as a support for the printed circuit 50. The other side of the grooves 20-1 is inclined so that the printed circuit 50 is supported on the two sides of the grooves without however reaching the bottoms of the grooves. Screws 52, distributed over the length of the rule, pass through one of the sides of the U-shaped part and are screwed into the other side. By tightening these screws 52, the two sides of the U-shaped part approach and jam the printed circuit 50 in the grooves 20-1, the inclined sides of these grooves press the printed circuit 50 on the reference plane formed by the right sides of the grooves. As shown, to further increase the flexural rigidity of the rule 20, it includes an ascending part 20-2 so that the rule has a section in "h". This ascending part may have a folded end, as shown, which is used to fix elements such as printed circuits.

The printed circuit 50 is provided, on its internal face, with components mounted on the surface 54, such as, for example, the amplifiers 28 and the capacitors C of FIG. 3. To perform rail straightness measurements, it is advisable to choose a rule about a meter long. Since it is diffi¬ cult to produce printed circuits of such a dimension, the printed circuit 50 is subdivided into, for example, five printed circuits 20 cm long. Each printed circuit 50 is provided with a surface-mounted connector making it possible to connect the outputs of the amplifiers 28 and the common connection of the capacitors C to a printed circuit, not shown, comprising the other elements of FIG. 3. The connectors of the printed circuits 50 are accessible through openings 20-3 made in the U-shaped part 20-0. In a practical embodiment, a gap of about 1 cm is provided between the sensors, in particular in the central zone intended to be located above a weld. In extreme areas, a gap of 2 cm between sensors is sufficient. However, to simplify manufacturing, the printed circuits 50 will be identical but only one sensor out of two will be equipped with its components mounted on the surface at the extreme zones of the rule. We can of course choose a difference of less than 1 cm. FIG. 6 represents a perspective view of an external embodiment of the straightness measuring device according to the invention. The housing 24 of this device is generally elongated parailélépipédique. Handles 62 are provided in three lateral recesses of the housing 24, one recess being located at the central part and the other two at the end parts. The screen 24-1 mentioned above is arranged between an extreme recess and the central recess, and the keyboard 48 is arranged between the central recess and the other extreme recess.

Each end of the device is provided with a removable fixing block 64 including a quick mounting system on a rail or other profile.

7 shows a sectional view of the device at one of its handles 62. The same references as in the previous figures denote the same elements. The housing 24 has an inverted "U" section closed at its lower part by a plate 65. The rule 20 is fixed by its longer part on a side of the housing 60, on the side opposite to the handle 62. The shims d 'support 26, only one of which is visible in Figure 7, are fixed at the respective ends of the rule 20. The wedges 26 are supported on a rail 12 and are electrically isolated from the rest of the device because one of they or both, as mentioned above, transmit an electrical signal to the rail 12 used to perform the distance measurements.

As shown, the block 64 comprises, below the rule 20, a housing for the rail 12 provided with jaws 66. These jaws 66 allow, using a mechanical or hydraulic control, not shown, to tighten the rail 12 for fixing the device on the rail while centering the rule 20 on the axis of the rail in order to carry out a measurement at the appropriate place. As previously indicated, the blocks 64 are removable. By actuating a mechanical control, the housing 24 of the measuring device can be detached from the blocks 64 which remain fixed on the rail. This possibility is provided so that the housing 24 can be repositioned according to the dotted representation to perform a straightness measurement of the side of the rail 12. For this purpose, the end of the housing 24 includes a groove 24-2 provided for sliding on a lug 64-1 of the block 64 and maintain the housing 24 in an adequate position. The straightness measuring device according to the present invention has been described in the context of the straightness measurement of rails. Of course, the invention applies to the measurement of straightness of any conductive object, even in a bad conductive material, such as water, graphite ...

Claims

1. Device for measuring the straightness of a conductive object (12), comprising a plurality of capacitive distance sensors (22) aligned along an axis of a rule (20) to be placed on the object, and means of communication (24-1) of the measurements supplied by each sensor, characterized in that the capacitive sensors (22) are formed on rectangular printed circuits (50) by the central part of aligned rings engraved on a copper layer of the printed circuits ( 50), the copper remaining outside the rings playing the role of guard rings (22-1).

2. Straightness measuring device according to revendi¬ cation 1, characterized in that the rule (20) comprises at each end a shim (26) to be placed on the object (12), at least one of these shims being conductive and providing the object with an electrical signal necessary for the operation of the capacitive sensors.

3. Straightness measuring device according to Reven¬ dication 1, characterized in that the printed circuits (50) have on a face opposite to that where the sensors of the electronic circuits are mounted on the surface (54) intended to exploit the sensor signals.

4. Straightness measuring device according to Reven¬ dication 1, characterized in that it comprises means (24-1) for displaying in a coordinate system the amplitudes of the measurements supplied by the sensors (22) as a function of the positions respective sensors.

5. Device for measuring straightness according to Reven¬ dication 1, characterized in that said object is a rail (12).

6. Straightness measuring device according to claim 5, characterized in that each end of the rule

(20) is removably attached to a block (64) which includes jaws (66) intended to clamp the rail (12) while aligning the axis of the sensors (22) on the axis of the rail, means (24-2, 64-1) being furthermore provided for, once the blocks (64) are fixed on the rail, reposition the ruler laterally so as to be able to measure the straightness of a side of the rail at a predetermined position.

7. Straightness measuring device according to revendi¬ cation 1, characterized in that the printed circuits (50) are held in two internal internal grooves (20-1) of a "U" part of the rule (20) , one of the sides of each groove being straight and serving as a support for the printed circuits and the other of the sides of each groove being inclined so as to come into contact with the printed circuit without the latter reaching the bottom of the groove, clamping means (52) being provided to bring the two flanks closer to the "U" portion.