WO1991010948A1

WO1991010948A1 - System for controlling the position of sensor elements

Info

Publication number: WO1991010948A1
Application number: PCT/GB1991/000005
Authority: WO
Inventors: David Bryn Edwards
Original assignee: Lk Limited
Priority date: 1990-01-06
Filing date: 1991-01-03
Publication date: 1991-07-25
Also published as: GB9000314D0; JPH05502725A; EP0509013A1; DE509013T1

Abstract

A ten axes control system controls the position of two respectively independent machines with associated sensors, whereby the sensors are arranged to move relative to and in opposite sides of a component in timed sequence. The control system defines a multi-dimensional major node matrix and generates a plurality of mini nodes equally time spaced between adjacent pairs of major nodes, and a plurality of micro nodes equally time spaced between adjacent pairs of mini nodes. The plurality of micro nodes define points upon the component in a mutual plane substantially orthogonal to respective opposite component sides.

Description

SYSTEM FOR CONTROLLING THE

POSITION OF SENSOR ELEMENTS

This invention relates to a control system and more particularly, but not exclusively, to a system for controlling the position of associate sensor elements respectively arranged to move relative to a component upon opposite sides of that component in timed order.

In order to inspect components it is common to employ non-destructive means, for example ultra-sonic inspection, this requires alignment of associated sensor elements about either side of the component. This is to ensure adequate coupling between the associated sensors and to prevent distortions in component signals that are not due to discontinuities in the component.

Previously, alignment of associated sensors has been achieved using accurate mechanical control elements or simple adjustment of the associated sensor elements to achieve the best signal response. This has severely limited the accuracy and speed with which component inspection can be made.

It is an objective of the present invention to provide a control system that ill allow adequate inspection of a component whilst obviating or mitigating the problems described above. According to the present invention there is provided a control system for controlling the position of associated sensor elements respectively arranged to move relative to and on opposite sides of, a component in timed sequence, the system comprising means to define a multi-dimensional major node matrix, means to generate a plurality of mini nodes equally time spaced between adjacent pairs of major nodes, and means to generate a plurality of micro nodes equally time spaced between adjacent pairs of mini nodes whereby the plurality of micro nodes define points upon a component in a mutual plane substantially orthogonal to respective opposite component sides.

Preferably, the associated sensors are mounted in separate independent housings and the multi-dimensional major node matrix defines five movement axes for each housing, three linear (X,Y and Z) and two rotational axes (A and B). Alternatively, the associated sensor elements may be mounted in the same sensor housing with each sensor having respective manipulation means to allow the five axes movement.

Preferably, movement of a first associated sensor element and movement of a second associated sensor element is synchronised in time such that each sensor element ,<es through their corresponding major, nodes at the same time.

Preferably, the multi-dimensional major node matrix is derived from either its computer aided design information or through machine teaching.

Preferably, the control system is arranged to allow either for the component to be submerged in a coupling medium or to incorporate in the associated sensor elements coupling means such as squirters with appropriate compensation factors accounted for.

An embodiment of the present invention will now be described by way of example only with reference to the accompanying drawings in which:-

Fig. 1 illustrates, in schematic form, a control system in accordance with the present invention,

Fig. 2 illustrates a major node matrix (in one dimension) and an associated plurality of mini nodes,

Fig. 3 illustrates for a curved surface mini nodes as illustrated in fig. 2 and a plurality of micro nodes equi-time spaced therebetween, Fig. li illustrates for a curved surface major nodes as illustrated in fig. 1 upon both sides of a subject component,

Fig. 5 illustrates a projected computer aided design "wire" frame with datum points,

Fig. 6 illustrates, in schematic form, the computer aided design wire frame world divided into respective WORLD A and WORLD B for each opposite side of the component ,

Fig. 7 illustrates, in part cross-section and part schematic, the computer aided design data converted to data sets for each side of the component, and

Fig. 8 illustrates, in part side elevation and part schematic elevation, compensation for squirter droop from associated sensor elements.

Referring to fig. 1, there is illustrated a ten axes control system consisting of two respectively independent machines each having five axes of mechanical independence with respect to each other. These five axes consist of three linear axes (X, Y and Z) and two rotary axes (A and B) which are mutually orthogonal to each other. It will be appreciated that the control system of the associated machine can be designed to run as two independent fi e axes machines or one ten axes machine. Fig. 1 illustrates the configuration that would be used for the ten axes synchronized mode whilst for two independent five axes machines the control system would be augmented with an additional host interface, data collection interface and data collection processor.

A component is mounted with respect to the machine and the control system and a major node matrix determined for that component. This major node matrix is most conveniently considered as a wire frame about the component with intersections considered major nodes. The major node matrix may be programmed from computer aided design (CAD) data or be taught in an interactive manner.

Considering firstly side A of the control system with five axes of motion determined by respective sensor elements, the position demands for each sensor to define each major node is communicated to the main central processor unit (CPU) 1. These demands (X 1 , Y1 , Z1 , A 1 , B1 ) are communicated along a local bus network 3 to a path generation central processor unit 5. The path generation CPU 5 is arranged to generate in accordance with a specified "real time" curve fitting procedure at the position of a plurality of mini nodes 7 (fig. 2). These mini nodes are equi-time spaced with a time interval of t1. Once calculated these mini nodes and their respective span therebetween is returned along the bus 3 to the processor unit 1 where, as illustrated in fig. 3, micro nodes 9 are determined again equi-time spaced with a set time interval t2 therebetween. These micro nodes are motion control points for relative manipulation between a component and associated sensor elements controlled by the control system. These micro nodes are determined by a "straight line" projection between adjacent mini nodes 7. The time interval t1 is sufficiently small such that the motion appears smooth and continuous rather than as previously a number of straight line vectors.

The data collection rate is established in relation to the linear surface distance between micro nodes. A system host computer 15 defines the required number of data collection nodes required between major nodes to ensure adequate inspection of the component by the sensor elements. The data collection requirement is communicated to the processor 1 which subsequently triggers the data collection interface 17 to stimulate collection of the data as the sensor elements are moved relative to the component. Side B is operated in a similar manner to side A with position demands (X², Y², Z², A², B²) defining necessary node positions.

In the ten axes machine mode, all ten position demands are communicated to a single side, for example side A through its main processor unit 1 (see fig. 4). These major nodes are then sorted into major node demands for respective side A and side B, the major node demands for side B being transmitted through a communications central processor unit in each respective side A and B. Synchronization between side A and side B of the machine is effected using the communication CPU 21. However, in all other respects operation of the ten axes mode is similar to operation in respective independent five axes machine sides.

Each servo is arranged to manipulate a respective component and sensor element's relative position in accordance with instructions from the control system. Relative movement between the component and the sensor elements is continuous and data collected at specified points. It is the determination of exactly the position of these data acquisition points that is of an importance in the present invention in order that sensor element orientation can be kept substantially orthogonal to respective component sides on either side of the component. By determining micro node positions as motion control target points for incremental movement of each sensor.

In order to programme the initial major node matrix with a defined geometry in three dimensional space either the computer aided design (CAD) data from component fabrication or an interactive teach technique may be used.

When using computer aided design data a "wire frame" of the part as defined in the CAD data is passed through the host computer 15 and datums identify upon the component and in the "wire frame" are used to accurately determine component position upon the machine from which the "wire frame" can be utilised to provide a major node matrix. The machine is thus able to define the major node matrix which will map the co-ordinates of the CAD "wire frame" from the CAD reference frame. This now common reference frame can be called world axes (see fig. 6). The position demand data can be shifted to compensate for standoff distance and material thickness to produce two sets of data points, namely world A and world B. The data in respective world A and world B can then be transferred through another transformation matrix, mapping it into respective major node sets for machine side A and machine side B (see fig. 7). These major nodes sets are utilised as previously described to provide mini nodes and micro nodes for the present control system.

To iteratively teach the component to the control system the component is first learned by collecting discrete location points on the component. A "wire frame" is provided by grouping sets of these discrete locations to form lines across the component and then fitting a two dimensional curve through the points. When all lines are defined, straight line ruling is used between the locations to finally define the "wire frame". The locations upon the component are normally learnt in world axes co-ordinates but if data is available to be learnt to act as fixed references upon the component to be learnt, the locations can be considered to be component co-ordinates relative to this data. It is beneficial to have component co-ordinates as this will allow the same component to be replaced in a different position on the machine and only the reference points found before rescanning the component.

It will be appreciated that the present control system may be applied to many types of associated sensor element machines, for example where one sensor element is a transmitter and the second sensor element is a receiver arranged on opposite sides of the component, whilst ensuring there is adequate coupling between the component and sensor elements. To ensure adequate coupling, for example with ultra-sonic non-destructing testing machine with an ultra-sonic probe it is usual to either use a water squirter system or submerge the component pay load in water. The present control system is capable of controlling the component and sensor element relative movement in these circumstances. With a water squirter system the control system is arranged to compensate for real time water droop (see fig. 8), thus ensuring that the squirter angle is determined to ensure that a projected column of water 31 strikes the component surface at an orthogonal angle.

Whilst endeavouring in the foregoing specification to draw attention to those features of the invention believed to be of particular importance it should be understood that the Applicant claims protection in respect of any patentable feature or combination of features hereinbefore referred to and/or shown in the drawings whether or not particular emphasis has been placed thereon.

Claims

Claims :

1. A control system for controlling the position of associated sensor elements respectively arranged to move relative to and on opposite sides of, a component in timed sequence, the system comprising means to define a multi-dimensional major node matrix, means to generate a plurality of mini nodes equally time spaced between adjacent pairs of major nodes, and means to generate a plurality of micro nodes equally time spaced between adjacent pairs of mini nodes whereby the plurality of micro nodes define points upon a component in a mutual plane substantially orthogonal to respective opposite component sides.

2. A system according to claim 1, wherein the associated sensors are mounted in separate independent housings and the multi-dimensional major node matrix defines five movement axes for each housing, three linear (X,Y and Z) and two rotational axes (A and B).

3. A system according to claim 1, wherein the associated sensor elements are mounted in the same sensor housing with each sensor having respective manipulation means to allow the five axes movement.

4. A system according to any of claims 1 to 3, wherein movement of a first associated sensor element and movement of a second associated sensor element is synchronised in time such that each sensor element moves through their corresponding major nodes at the same time.

5. A system according to any of the preceding claims, wherein the multi-dimensional major node matrix is derived from either its computer aided design information or through machine teaching.

6. A system according to any of the preceding claims, which is arranged to allow either for the component to be submerged in a coupling medium or to incorporate in the associated sensor elements coupling means such as squirters with appropriate compensation factors accounted for.