GB2112944A

GB2112944A - Calibration of thickness gauges

Info

Publication number: GB2112944A
Application number: GB08200173A
Authority: GB
Inventors: James C Taylor
Original assignee: Individual
Current assignee: Individual
Priority date: 1982-01-05
Filing date: 1982-01-05
Publication date: 1983-07-27

Abstract

A probe contains a measurement coil system (8) using induction or eddy current effects to measure the thickness of a coating over a conductive or magnetic substrate. The probe is first used to form a calibration chart in a memory system (16) by being raised physically above a suitable substrate, simultaneous measurements being made of the motion of the probe e.g. by displacement transducer (6, 7) and of the measurement coil (8) output. The probe is then positioned on the coating to be measured and the output applied to the memory system where the calibration chart is used to recover the values of physical movement (and hence coating thickness) which give any particular value of measurement coil output. <IMAGE>

Description

SPECIFICATION A self-calibrating thickness gauge THE PROBLEM There is a widespread requirement in the painting and plating industries for the measurement of the thickness of a coating layer.

Inevitably the demands for accuracy are becoming more exigent, particularly where the coating material is expensive and so must be guaranteed to exceed a certain thickness, but where excess thickness represents an unjustifiable expense.

DEFINITIONS The following definitions will be used throughout this paper.

COATING -The topmost layer of which it is desired to know the thickness.

SUBSTRATE -The underlying base material to which the coating is applied.

PRESENT ART There are many instruments on the market; these fall into four categories which are: 1. Destructive 2. Eddy Current 3. Magnetic pull-off 4. Electromagnetic induction Destructive tests, involving either cutting through the coating, or total removal of an area of coating, while of reasonable accuracy, probably of the order of +3%, are of limited value for the very reason that they are destructive.

EDDY CURRENT INSTRUMENTS Eddy current tests are performed using high frequency currents; the effect of eddy currents, induced in a surface placed adjacent to the measuring coil, can be used as a measure of the separation of the coil from the surface, and thus of the thickness of any coating which intervenes.

Eddy current tests are usually used for the measurements of non-conductive coatings on conductive but non-ferromagnetic substrates, though they are capable of measuring certain other combinations as well. They have, in particular, only limited application to coatings on magnetic substrates.

Eddy current tests offer their highest accuracy when measuring non-conductive coatings on nonferromagnetic, conductive substrates, where their accuracy is limited by the variation in properties between the many and several substrated materials. Owing to the large range of suitable substrates, it is not possible to place a limit on the accuracy attainable, though commercial instruments claim about 3% on samples of aluminium. Owing to the degree of variability of the effect of different substrates on the output from a measuring head, it is customary to calibrate an instrument on one substrate only, often 'aluminium, and to use various expedients in the treatment of the signal to render the results from different substrates as like as possible.

PERMANENT MAGNET GAUGES Magnetic pull-off gauges are used for nonmagnetic coatings on ferromagnetic substrates.

The accuracy of measurement attainable is limited by variations in the permeability of a substrate and is generally accepted to be not better than 10%.

INDUCTION GAUGES Electromagnetic induction instruments work at a low frequency, typically'less than 500 c/s and use the effect of an adjacent magnetic surface on the inductance of a coil (usually a balance coil system), as a measure of the separation between the coil system and the substrate and thus the thickness of the intervening coating. Once again, accuracy is limited by the variation in properties, both permeability and conductivity, of the substrate. A figure of 3% is claimed for accuracy, but this is probably only achieved on mild steel substrates.

A method of intrinsically unlimited accuracy is known where two separate coil systems are used, the first is placed over the coating on a substrate and a second is separated by an adjustable amount from a separate, uncoated smaple of an indentical substrate. When the separation of the second is varied by an adjustment mechanism and the output from both coils is compared, a balance point will be observed when the separation of the second coil from its substrate is the same as the first.

The means of adjustment used for the second coil may be calibrated to show the separation from its substrate and thus to indicate directly the thickness of the coating which it is desired to know. The system is now limited to use on inductive systems only, but could also be used on eddy current systems.

This method, while of unlimited accuracy.

requires the use of two separate coil systems which must be manipulated manually, and of a second sample of uncoated substrate. The accuracy of the method is limited in practice by variations between the two sets of coils caused by extraneous factors such as environment, since the two coils could be at quite different temperatures, and by variations between nominally, but not actually, identical samples of substrate.

THE NEW INVENTION This invention reveals a method which obviates these difficulties and offers a convenient instrument of inherently unlimited accuracy.

The invention consits of a self-calibrating system wherein the relationship between the output of a measuring probe and its separation from a substrate is determined automatically by moving the probe physically away from an uncoated sample of the substrate by means of a driving system, typically pneumatic, and measuring the motion of the probe, typically by means of a Linear Variable-Displacement Transofmer (L.V.D.T.) whilst at the same time measuring the output from a measuring probe, typically of a three coil balanced construction. The two sets of measurements if not already in digital form are digitised by means of an analogue to digital converter. The digital measurements are then used to form a look-up table in memory in the computer such that the measured signal may be related to a known separation throughout the full range of the instrument.This couid be done for instance by using the motion signal to form an address in memory and storing output from the measuring head at that address. This calibration chart is independently determined for each substrate and may be constructed with any required degree of accuracy.

It is necessary to set the zero point of the L.V.D.T. for each calibration and this together with the control of all signals is performed by a computer system, typically a microprocessor.

One form of the invention is described by reference to the figure.

Referring to Fig. 1: An outer housing 1 is placed in contact with an uncoated substrate.

Inside the housing, a rod 5, attached at its upper end to a flexible diaphragm 4, carries the core 7 of a linear-variable-displacementtransducer 6. (L.V.D.T.) and a coil system 8, which is wound round a core, whose tip 11, is held lightly in contact with the substrate by means of the resilience of diaphragm 4. A cavity 3 is formed in the housing with one wall bounded by the diaphragm 4. The cavity communicates by means of a flexible tube 2 with a means of communication.

An excitation signal from a generator 9 is connected to the input coil of the L.V.D.T.

A second excitation signal from a generator 10 is connected to a measurement coil system 8.

The output signal from the L.V.D.T. is passed through signal conditioning circuits 12 to render it into a direct voltage signal which is then converted into a digital form. This signal will be referred to as the distance signal.

The output signal from the coil system 8 is passed through signal conditioning circuits 14 which render it into a direct voltage and permit the removal of initial offsets and temperature effects.

This signal is then converted into a digital form.

This signal will be referred to as the probe output signal.

The distance signal is used to form an address for a Random Access Memory 16 in such a manner that at convenient discrete increments of the spacing of tip 11 from the substrate, the momentary value of the probe output signal is entered into an address defined by the control computer as corresponding to a known spacing.

In use, an initiation signal is given by the operator when the outer housing is placed on an uncoated substrate; this causes the initial value of the distance signal to be used as a zero distance measurement, and then causes the means of evacuation to start to reduce the pressure in cavity 3, thus raising rod 5 and increasing the distance between tip 11 and the substrate. As each increment of distance selected by the computer is reached, the momentary value of the probe output signal is entered into an appropriate memory location. When tip 11 has been moved to its maximum range the calibration phase is complete, the evacuation of cavity 3 is ceased and air is admitted.

When subsequently it is desired to make a thickness measurement, the outer housing is placed on a coated substrate, tip 11 now resting on the top of the coating, urged by the pressure of diaphragm 4 acting simply as a spring. The probe signal obtained in this condition now corresponds to that of a separation equal to the coating thickness and this is found by scanning the table held in memory for the value most nearly corresponding: the position of this value in memory now gives the thickness of the coating and is presented to the operator, suitably scaled on a display.

In practice, to avoid the approximation inherent in the above procedure, an interpolation, linear or otherwise, would usually be performed between the two nearest values found during the comparison of probe output signal with values obtained on calibration.

A first development of this system is the incorporation of-a means of measurement of the probe temperature in order that the temperature coefficient of the probe may be determined automatically by the storage of sets of graphs made at different temperatures.

This measurement may typically be made by passing a direct current through the coils of the probe and using this as a resistance thermometer.

A second development of this system is to use the computer to calculate an equation for each graph as required, which is then used to calculate the separation thus permitting interpolation between measured points and, with care, extrapolation also.

A third development of this system is to use the facility to calculate separation using an equation to permit the user to enter a previously known equation, determined previously using this instrument, or otherwise.

A fourth development of this system is to permit the presentation of discrete coated samples, of known coating thickness, to the measurement probe, with manual insertion to the computer system of this known thickness, thus permitting a graph to be stored of measurements of combinations of coating and substrate where the coating does have an effect on the output from .the probe. This graph will have a limited number of known points, so the facility of the instrument to calculate the equation to a curve fitting these points, is essential in order to allow interpolation to a sufficient resolution and accuracy, for the instrument to be of utility.

A fifth development is to use the computing ability of the system to extrapolate 'backwards,' i.e. towards the zero thickness measurement, in order that measurements may be made on coated samples for which an uncoated substrate is not available.

Claims

1. I claim a method of measuring the thickness of a coating applied to a substrate wherein calibration of a measuring probe is made on an uncoated substrate by physical movement of the probe relative to the substrate and simultaneous measurement of the position of the probe and of the electrical output from a coil system forming a part of the probe which is affected by the proximity of the substrate. The two measurements are stored in a digital memory in such a manner that in subsequent measurements where the probe is spaced away from the substrate by the presence of a coating, the stored values of electrical output can be used to recover the physical separation of the probe which gave that value of electrical output when the probe/ substrate system was calibrated.

2. I claim a system for measuring coating thickness as in claim 1 where the measurement of motion is made by means of a Linear Variable Displacement Transducer.

3. A claim a system for measuring coating thickness as in claim 1 where the measurement of motion is made by means of a grating.

4. I claim a system for measuring coating thickness as in Claim 1 where the measurement of motion is made by means of wave interference.

5. I claim a system for measuring coating thickness as in Claim 1 where the means of causing movement is pnuematic.

6. 1 claim a system for measuring coating thickness as in Claim 1 where the means of causing movement is an electric motor.

7. I claim a system for measuring coating thickness as in Claim 1 where the means of causing movement is a solenoid.

8. 1 claim a system for measuring coating thickness as in Claim 1 where the coil system consists of one coil only.

9. 1 claim a system for measuring coating thickness as in Claim 1 where the coil system consists of two coils, an excitation coil and a search coil.

10. I claim a system for measuring coating thickness as in Claim 1 where the coil system consists of a three coil balanced system, where two coils are connected in opposition so as to reduce the standing output to a lower level.

11. I claim a system for measuring coating thickness as in Claims No's 1-10 where correction for varying temperature of the probe is made by the measurement of the temperature of the probe and the storage of compensating values in the digital memory system.

12. I claim a system for measuring coating thickness as in Claim No. 11 where the measurement of temperature is made by the passage of current through the coils and the use of those coils as a resistance thermometer.

13. I claim a system for measuring coating thickness as in Claims No'd 1 to 1 2 above where the simultaneous measurements of separation of the probe from the substrate and the electrical output form the probe coil system as stored in the digital memory are used to calculate an equation relating to three or more of these measurements and the use of this equation to interpolate between these points in order to provide an improved accuracy in measurement of coating thickness.

14. I claim a system for measuring coating thickness as in Claim No. 13 above, where the calculated equation is used to extrapolate outside the values obtained on calibration.

1 5. I claim a system for measuring coating thickness as in Claims Numbered 1 to 14 above, where the form and constants of an equation relating separation and electrical signal from a coil system may be entered by an operator by a keyboard or switches.

1 6. 1 claim a system for measuring coating thickness as in Claim No. 1 5 above where the form and constants may be entered from a storage medium.

17. 1 claim a system for measuring coating thickness as in Claim No. 13 or 14 above where the measurements of separation from the substrate are made other than by physical movement of the probe and where these measurements of separation are entered on a suitable keyboard or switches.

1 8. 1 claim a system for measuring coating thickness as in Claim No. 17 above where the measurements of separation are entered from a storage medium.

19. 1 claim a system for measuring coating thickness as in Claims numbered 1 to 1 8 above where the initial calibration is made on already coated substrate and the thickness of this coating is found by extrapolation towards the zero coating thickness by means of an equation or otherwise.