Emerald Group Publishing Limited
Copyright © 2007, Emerald Group Publishing Limited
Our theme for this issue is position measurement and our research has highlighted the arrival of a number of new magnetic rotary encoders, some offering 10 or even 12b accuracy (0.1°).
Measuring the angle of rotation of a bar magnet might not sound too difficult but it is actually rather tricky. Optical encoders are relatively straightforward by comparison – you just need to apply marks to a glass disk and use optical sensors to count the marks as the disk rotates – everything is effectively digital and based around on/off sensing. Some magnetic encoders work this way as well – many linear encoders used in machine tools also employ magnetic encoded strips, but the fundamental switching is digital or binary in nature. Of course there is nothing to stop you using a hard disk drive as a rotary encoder – simply write the coded and read it back on subsequent rotations.
The main difference with this new breed of magnetic encoders is that everything is analogue and the real advantage is that the separation between the bar magnet and the sensing head does not need to be the hair splitting distance required by hard disk drives or linear magnetic encoders. Also, the axial alignment between the magnet and the sensing head is not super critical. You can expect to loose accuracy if you are a bit out, but the system should still basically work.
The nature of the system eliminates the need for encoder bearings and because both halves of the system can be separate they can be encapsulated and made relatively immune to aggressive environments and will shrug off water, oil and most other substances you might throw at them.
The research emphasis at the moment seems to be directed towards achieving high resolutions and accuracies, however, I would expect that they will become most popular in applications that require a bit of rotary feedback but where absolute accuracy is not too important – such as music volume and other controls that are adjusted until you get the result you want rather than to some precise digital value.
I do, however, have one nagging doubt regarding their use which further pushes me towards their use in non-critical applications. Magnetic fields are everywhere and they are constantly changing and are all pervasive. Every wire or coil that has a current flowing through it generates a magnetic field that distorts the existing field created by the Earth (which itself is subject to local anomalies), and also by the magnet in our vulnerable rotary encoder. Screening a circuit from electrical interference is hard enough, but screening it from magnetic interference is an order of magnitude worse. In fact, it is so difficult that it usually becomes impractical and so magnetic interference is something you have to learn to live with rather than overcome.
The new generations of magnetoresistive materials are now opening new doors into a field of sensing that has previously been closed to us. We need imagination to exploit their properties and accommodate their deficiencies, but I can see great potential for this new family of devices and intend to cover these in more detail in a future issue.