Interview with Chris Jones

Chris Jones, managing director of high precision sensors and measurement equipment supplier Micro Epsilon UK, talks to Justin Cunningham

JC: What are the current innovations within Micro Epsilon, and what benefits do these bring to your customers? CJ: We are looking at ways to make our sensors smarter and more intuitive. For example, having a laser displacement sensor that will automatically adapt its settings to the surface being measured ensures the output signal is the most stable and accurate it can be. Similarly, an eddy current sensor that has its calibration characteristics inside the sensor head, allows you to plug it into the control electronics and start to measure. This removes the need for linearisation and calibration against the target material. This can often be a lengthy process and is not always performed by the end user as it results in reduced sensor accuracy and performance. Additionally, we are moving into the realms of 3D measurement of large objects or surface areas. Imagine, for example, a large part moving down a production line. It would be a huge advantage to the manufacturer to be able to measure the whole component completely noncontact, to a high degree of accuracy and perhaps even drill into key areas for more detailed measurements and quality control. JC: What are the common pitfalls you see in industry when it comes to selecting the right sensor for the right application? CJ: With so many different sensor technologies available, it is not the fault of the end user, if he/she selects the wrong sensor for a task. Sensor selection is a detailed process and a wealth of knowledge is required of many different sensor technologies in order to be able to assess which technology is correct for each particular task. Additionally, many technical sales people do not have the correct level of knowledge to advise design engineers in this area, so quite often the user is recommended a product that may not be right for the task. I also have to mentioned price. I fully understand that issue of searching for a low price for sensors, especially with today's market conditions. However, I feel there has to be a price versus performance ratio that enables design engineers to consider innovative sensor features that may be at a higher price, but bring greater functionality to the application. Having said that, Micro Epsilon is fully aware of a 'price ceiling' in certain markets and applications, so we strive to be below these benchmarks whenever possible. JC: What advice would you give to design engineers when it comes to sensor selection? CJ: As before, review the different sensor technologies that are available, look for a proven sensor technology rather than a novel sensor design that has not yet been put into production; look for references into how this sensor has been applied before, if possible with reference sites. Finally - and perhaps most importantly - try to partner with a company that understands your requirements and can advise at a technical level throughout the whole project cycle. JC: What do you offer to your customers in terms of bespoke sensor design and can you give an example of an application? CJ: Initially, we have an introductory meeting with an open mind. Micro Epsilon has many proven sensor technologies so before we recommend a solution, we must fully understand the user's requirements. Then we will review this information and propose either an existing sensor design that can be adapted to suit the customer's needs or a bespoke solution. Most importantly, if a solution already exists for the requirement, it makes no sense to make modifications to a sensor for the sake of it. This costs time and money for both companies. If a sensor does not already exist, then we can work through the complete development cycle from functional prototypes to pre-production samples and through to final sensor production - in high quantities. Quite often, the solution lies between those two options. A recent example is the development of an OEM turbocharger speed sensor for the automotive industry. Micro Epsilon has produced turbo speed measurement systems for many years for R&D environments, but the brief here was to develop that system into a package suitable for high under bonnet environments that could be fitted to every turbocharger produced. This posed several problems, including sensor and electronics miniaturisation, developing a solution that would meet the strict environmental specifications for under bonnet use, developing a sensor that could be assembled automatically and of course reducing the system price to an acceptable level for the automotive industry. The complete development here was 18-24 months, and in that time we moved from a prototype solution with discrete electronic circuits, to an ASIC electronic circuit approx 20mm2 and a completely moulded, engineered plastic sensor which can be produced in volumes of up to 3million per annum. JC: What are the pros and cons of using contact versus noncontact sensors – and where should each be used? CJ: In simple terms, when looking at a sensor selection, contact sensors are generally simpler devices and will usually have a lower unit cost. They can be very accurate with resolutions to sub micron levels, but by the nature of their design will be a slower measurement than noncontact sensors. Of course they have to be in contact, ie: touching, with the target to be measured and so this can sometimes give problems and increase the cost of sensor integration. But a target may be damaged by contact or it may cause contamination, particularly in pharmaceutical or food production. Hot or continuously moving targets will also be difficult to measure with a contact sensor. Finally, contact sensors will tend to wear with time, resulting in reduced performance and repair or replacement costs. Noncontact sensor technologies, on the other hand, measure very well against virtually any target materials including glass and liquids. They have the same - or often much higher - resolutions as contact sensors and because they are usually solid state sensors they will measure much more quickly and will not wear. Therefore the cost of ownership of the product may be lower than a contact solution. Noncontact sensors usually have a higher unit cost and so for large volume OEM applications, such as automotive applications, this may restrict the use of these technologies. However, when the cost of ownership is considered in OEM applications, one can often find the noncontact sensor to be more favourable.