GB2395030A

GB2395030A - Inspection results processing method

Info

Publication number: GB2395030A
Application number: GB0225865A
Authority: GB
Inventors: Thomas David Welsh; Philip Edward Galloway
Original assignee: Roke Manor Research Ltd
Current assignee: Roke Manor Research Ltd
Priority date: 2002-11-06
Filing date: 2002-11-06
Publication date: 2004-05-12
Also published as: GB0225865D0

Abstract

A method of testing a manufactured product comprises determining external properties of the manufactured product, inputting the determined external properties into an analysis tool, comparing the properties with properties predicted by a design simulation of the product; and outputting an indication of the result of the comparison. The comparison may include comparing the products' performance with the original design specification. The properties are preferably determined by means of laser scanning.

Description

( TESTING METHOD

This invention relates to a method of testing a manufactured article.

There are various known methods of testing manufactured articles to ensure that 5 what has been built satisfies a set of predetermined criteria. For example, on a production line for rivets, samples may be taken from each batch and destruction tested to ensure that they are strong enough and if the samples meet the criteria, then the batch is assumed to do so. Alternatively, for an electrical component, for example, then the sample may be inserted into a circuit to check that the output is the expected one.

10 Again, if the sample works as expected, the whole batch are assumed to be acceptable.

This arrangement is Me for mass produced items which are susceptible to direct testing and which should follow statistical norms. However, there can be situations where a bespoke product is manufactured for a client which may be so expensive to produce that it is not possible to manufacture a second example simply to destruction test it, or it may 15 be manufactured using a distributed manufacturing technique which does not bring all the components together in a form which can be tested until the manufacturing process has been substantially completed. Examples of these cases might be satellites or space equipment, such as the Hubble space telescope.

In accordance with the present invention a method of testing a manufactured 20 product comprises determining external properties of the manufactured product, inputting the determined external properties into an analysis tool, comparing the properties with properties predicted by a design simulation of the product; and outputting an indication of the result of the comparison.

The present invention determines the external shape of the manufactured product 25 and then compares this with the result expected if the product had been manufactured to its original design specification. In its simplest form, it is able to detect whether the

product is as designed or not, without interpreting that result.

However, preferably, the method further comprises analysing the determined properties and predicting performance for the manufactured article, comparing the 30 predicted performance with an expected performance derived from the original design specification; and determining whether to accept or reject the product according to the

nature of the deviation from its expected performance.

: l: 2 ''' ''

The present invention has a further advantage, that it is possible to determine, not only whether the product was manufactured correctly, but also, if it was not manufactured exactly as intended, whether or not this makes a material difference. For example, in a distributed manufacturing technique, there may be parts of the product 5 which need to interact with other parts manufactured elsewhere or which need to l produce a particular effect, e.g. to bend light through a certain no. of degrees so that it then enters the next component correctly. However, other parts have no particular effect, e.g. the corners of an outer casing, so if they are a little longer than intended, that may still be acceptable.

10 Preferably, the method further comprises allocating a grading to the result of the comparison of the predicted performance and expected performance and outputing an indication of acceptance and the grading under which the article can be accepted.

Having determined the degree to which the performance of the manufactured article deviates from its expected performance, the article is allocated a grading, such as 15 excellent, average or satisfies a reduced specification, so that subsequent decisions on

whether to use the article can be made.

The invention is applicable to testing of any manufactured product where design parameters are available, but it is particularly suited to products manufactured using distributed manufacturing techniques.

20 Where distributed manufacturing techniques are used, the ability to grade the article, so that all options can be weighed up to see which articles need modification before the complete product is assembled, is particularly useful. If all articles with which one article interacts have an excellent grading, then it may be possible to accept the one article which only meets a reduced spec. However, if all the other articles are 25 only graded average, the one article may need to be average or excellent to allow the product to operate correctly.

Preferably, the predicted design performance is one of mechanical, electromagnetic, optical or hydrodynamic performance.

Many techniques exist for scanning and digitising the 3-D shape of an object, 30 but preferably, the external properties are determined by laser scanning techniques.

The digital data is then input directly to the analysis tool for further processing.

.'.'..,., 1

3,,'.,'.

An example of a method of testing a manufactured article according to the present invention will now be described with reference to the accompanying drawings in which: Figure 1 is a block diagram illustrating one example of the method of the present 5 invention; and, Figure 2 is a block diagram illustrating a second example of the method of the present invention.

Fig. I is a block diagram of the method of the present invention. The first stage 10 of the design process for a complex or expensive article, such as a space telescope is to determine what components are required and what design these will have. The design process is commonly undertaken using a computer aided design (CAD) package and frequently the design will go through many iterations before it is fixed. There are a number of existing proprietary software simulations which allow a user to design a 15 product in a particular field of activity. Examples of these include, NASTRAN (MSC

Software Corporation) a computer aided engineering simulation tool for mechanical design; HESS (Ansoft Corporation) a 3-D electromagnetic simulation software; AMROC multi-level simulation code for optical design; and, Fluent (Fluent Inc) flow and heat transfer modelling software for hydrodynamic design, in particular for aircraft 20 design.

In this example of the present invention, data is input to the package 1 and a simulation 2 is run to determine what shape a mirror for a space telescope having the input parameters would take. When the simulation produces the desired outcome, then the design is fixed and stored as the final design 3. A suitable production line 4 is set up 25 and the product is produced. The manufactured product is laser scanned 5 to determine its external characteristics 6 and the external characteristics that would be expected from the final design are also determined 7. These characteristics are compared 8 and a threshold is set for the comparison process. Below the threshold, any differences in the external properties are ignored and an output 9 indicating that the product can be 30 accepted is made. Above the threshold, the difference in the external properties may have some impact on the performance of the product and so the output indicates that the product cannot be accepted without further review.

l: ( This review process may he undertaken manually by an operative checking the product and determining using his professional judgement whether the differences are significant. However, in Fig. 2 is illustrated an example of how this review step can be undertaken automatically.

5 The initial stages of design l, 2, 3 and production 4 are carried out in the same way. External properties are determined by laser scanning 5, 6 and directly from the design data 7, then compared 8 as before. An output 9 indicates the result of the comparison.. If that result is above the threshold, further steps are taken. A simulation 10 is run using the properties determined from the manufactured product to determine to the expected behaviour of the product in operation. Similarly, a simulation l l is run from the design properties to determine their expected behaviour. The two sets of expected performance are compared 12 with know performance requirements and the deviation from those is analysed to see whether it will have a significant impact on the resulting system which is created when all the products under construction are brought 15 together. This can give rise to an output 13 accepting or rejecting the article, or to an instruction to revise the system 14 which is communicated to other construction areas, so that errors in one (particularly complex or expensive part) can be corrected by modifying manufacture of a co-operating, but less complex other component.

The Hubble space telescope is an example of space technology which might have 20 benefited from the present invention. When it was launched it was found that the primary mirror was incorrectly shaped, leading to an inability to focus all the light from an object to a single point. Astronauts were able to correct this on a maintenance visit to the telescope, but it would have been cheaper to have discovered the potential problem before launch and corrected it then. The next generation space telescope will be built 25 with an even larger primary mirror which is constructed in segments because of the space constraints for launching. Predicting the behaviour of the parts which are actually manufactured when joined together to form the mirror and correcting any errors before launch could be seen as a desirable state of affairs.

Claims

CLAIMS 1. A method of testing a manufactured product, the method comprising determining external properties of the manufactured product, inputting the determined 5 external properties into an analysis tool, comparing the properties with properties predicted by a design simulation of the product; and outputting an indication of the result of the comparison. 2. A method according to claim 1, further comprising analysing the determined 10 properties and predicting performance for the manufactured article, comparing the predicted performance with an expected performance derived from the original design specification; and determining whether to accept or reject the product according to the nature of the deviation from its expected performance. 15 3. A method according to claim 2, further comprising allocating a grading to the result of the comparison of the predicted performance and expected performance and outputing an indication of acceptance and the grading under which the article can be accepted. 20 4 A method according to any preceding claim, for testing products manufactured using distributed manufacturing techniques 5. A method according to any preceding claim, wherein the predicted design performance is one of mechanical, electromagnetic, optical or hydrodynamic 25 performance. 6. A method according to any preceding claim, wherein the external properties are determined by laser scanning techniques. Amendments to the claims have been filed as follows CT ADI.S

1. A method of testing a m...:ufacturcd product the talc.,.] conprisii!= dot:rmining external properties of the rranufactured product inputting the 'etcrmined 5 external properties into an analysis tool; analysin, the dctcrnincd properties it, d predicting perfonnancc for the manufactured article; co;parin the predict d performance with an expected performance derived front the vi; in:1 design specification; and dctcrminhg whether to accept or rcicc t tile product a.co;-din to he

nature of the deviation from ILS C?CPeCLed PerfOrD.1]nCe.

2. A method according to claim 1 ftitl- r comprising allocating a gra ink to the À result of the comparison of the predicted performance and expected perfomla;,c-_. rid outputing an indication of accept nce and the. -. Olin. under \W ich the article can - _ acc ep! ed.

3. A method according to claim 1 or claim for testing procluc s; nanuf:turf d lisin; distributcdmamfacturingtcchniques.

4. Amcthodaccordingt.oanyprccedingclaim \.ilcrein he:predicteddesi?n 20 performance is onc of mechanical clectroanctic optic! orliydrody: chic performance.

5. A method according to any preceding claim wh.cin tl;c cstcrnai prrperi.cs re detcrn1incd by laser scanning techniques.