Man as the dominant factor in the measurement process

The specialist conference on production metrology at the NTB in Buchs (Canton St. Gallen) has become a fixed date for metrology specialists. It is now being held for the eighth time and, in addition to a highly technical programme of presentations, also offers a trade exhibition to enable direct contact with manufacturers of measuring equipment and service providers. We spoke to Prof. Dr.-Ing. Michael Marxer, head of the event, about the most important trends in production measurement technology.

Production metrology and Industry 4.0: How are these buzzwords related?
Michael MarxerAt the heart of Industry 4.0 is the Internet of Things and the interacting "cyber physical systems". Manufacturing systems, manufacturing processes, administrative processes and engineering are virtually mapped, networked and fully integrated. Production metrology is required to obtain information about manufacturing systems and processes. Production metrology is a decisive enabler of Industry 4.0 in that, on the one hand, information about individual processes is generated and, on the other hand, dependencies in functional chains can be examined with the help of production metrology.

In the presentation topics, one finds terms such as "fast and robust", "complex", "ultra-precision": To what extent is the impression correct that measurement technology has to deal with ever smaller dimensions at ever higher speeds?
Two trends can be observed in manufacturing processes. One trend is towards ever-larger workpieces, which are caused by the mega-trends of energy and mobility. Workpieces such as those required for wind turbines or aircraft parts present measurement technology with challenges that are being addressed in current research projects. Another trend is towards workpieces with ever smaller dimensions and often at the same time greater complexity. Particularly in the case of small workpieces, metrology must break new ground in order to be able to determine the required characteristics with sufficiently low measurement uncertainty.

"Production metrology must provide information to make decisions. »

What "new ways" are we talking about here?
The main issue here is the dimensions of the measuring heads. In tactile measurement processes, for example, sensing elements must be able to detect dimensions in the nanometer range. Or in the case of optical measuring devices, we work with light points of ever smaller diameters. The interactions between the measuring system and the workpiece must not be ignored with these small dimensions: A thin wall thickness on a workpiece must not be deformed to an inadmissible degree by contact.

What role does automation play?
In measurement technology, the reduction of measurement time plays an increasingly decisive role in many areas. In order to achieve this, automation of measurements is helpful in many cases. The automated assembly of measuring devices with robots and/or the use of pallet systems can support this approach.

Keyword "batch size 1": What does that mean for production metrology? Back again from the production plant to the laboratory?
Production metrology must provide information in order to make decisions, e.g. to control manufacturing processes. For time or logistical reasons, it is not always possible or practical to remove a workpiece from the machine tool and bring it to a measuring laboratory for measurements. In order to be able to manufacture economically with batch size 1, in-situ measurement technology or in-process measurement technology is necessary. Here, the workpiece is measured close to or in the manufacturing process.

Your institute is also a service provider for industry. Which inquiries are you most frequently confronted with at the moment?
We are accredited as a testing laboratory for dimensional measurements and as an accredited calibration laboratory for roughness, gauges and reference bodies. We carry out measurements for our customers with our coordinate measuring machines and increasingly advise them on the definition of measurement strategies. A particularly exciting aspect of this is the interpretation of design drawings, in which we advise and support our customers. This subject area, Geometric Product Specification and Verification, has become very important for us and our customers against the background of the rapid development of standards in this field. We also offer training courses that help to deal with enquiries in this direction.

We have now talked about the technology and about devices and their dimensioning. What role does the human factor play?
The human being is the dominant factor in the entire measurement process. This is less in connection with influencing a measurement, but rather in relation to the definition of an overall measurement strategy and the interpretation of results. E.g. it is a crucial task of the measurement technician to determine the kind of assignment of geometry elements into measurement points, which can have a decisive influence on measurement results. If one wants to express the weighting of the factors man - environment - measuring system in the measuring process in figures, the ratio can be 100 : 10 : 1.

For the September 5, 2019 symposium, what can attendees expect?
At the symposium, we will discuss the latest developments in measurement technology and the requirements placed on measurement technology. The symposium is a mixture of top-class lectures and a trade exhibition with about 20 exhibitors. In addition, we offer the opportunity to visit our measurement laboratories at NTB. The four sessions with two lectures each cover

"Software represents a crucial part of modern measurement systems."

The following sections deal with very important questions for practice and provide an outlook on future developments. Considerations about measurement uncertainty and the assessment of test processes have already occupied all measurement engineers. In the first session, we will present new, practical methods for performance verification and determination of measuring instruments and processes. In the second session we will deal with a new generation of measurement sensors for in-situ measurement and give an insight into the possibilities of multi-sensor technology in practice. In the third session we will have two presentations from the field of ultra-precision metrology. On the one hand we will take up the aspect of in-situ metrology and on the other hand we will have an insight into the latest developments in the field of nano-measuring machines. Software represents a decisive part of modern measuring systems. In the final fourth session we will hear a comparison of different software concepts and in a concluding lecture we will gain an insight into measurement technology in a Swiss high-speed train.

Your view of the future: Is there a limit to what can be measured?
In the lectures, we will hear about the accuracies to which it is already possible to carry out measurements today. We will also learn about new, promising approaches to further improve the performance of measurement systems and measurement processes. In the discussion about the limits of what can be measured, the comparability and traceability of measurement results play a decisive role. The challenge of having normal measurements available that can be economically produced and calibrated is crucial in this discussion.

Or would it be better to ask:
How much measurability is useful for practical production?
In order to be able to make rational and sufficiently reliable decisions, it is beneficial to have as good a database as possible. This means that the information used for the decision must be available in sufficient quantity and should be sufficiently reliable. Using the example of production control, this means that the smaller the uncertainty with which this process was considered, the more accurate the process can be evaluated. If the measurement technique is used to check specifications, the measurement uncertainty must be taken into account in each case against the person providing the evidence. Therefore, the correlation can be derived here as well: The smaller the measurement uncertainty, the more workpieces can be classified as good parts.