Well, this is interesting. I knew that most optical image stabilisation systems in lenses used electromagnetic fields in order to try and keep the elements steady. But what I hadn’t considered was how these fields might affect noise on the sensor and in the images. As it turns out, it affects them enough to warrant Canon developing a new system to help reduce it.
So, according to the patent, the way IS works is something like this…
Various imaging devices have been proposed each having an image shake correction mechanism for reducing an image blur on an image formation surface caused by a camera shake when capturing still images or moving images. In particular, the optical image shake correction mechanism can be considered to have a highly effective camera shake correction effect. A general optical image shake correction mechanism detects shaking of the optical axis caused by the camera shake or the like by means of a vibration gyroscope sensor, etc., and move an image shake correcting action unit such as a correction lens in the imaging lens so as to cancel the detected shaking of the optical axis. In order to move the image shake correcting action unit, electromagnetic force generated by supplying a current to a coil arranged opposite to a permanent magnet is used.
So, put simply, a vibration gyroscope sensor in the lens monitors its movements as you handhold it. It then sends signals to a correcting unit which uses electromagnetic fields to help keep things in place and counter your movement. This is generated by applying a current to a coil (and why having it turned on will drain your batteries more quickly). There are also permanent magnets inside the lens for the electromagnetic coils to repel or attract depending on the movement.
There is potential, enough potential to come up with a solution, for these magnetic fields to interfere with the electronic circuitry in the sensor, and generating noise. Think of those old CRT televisions, and what would happen to the picture when you held a magnet up to the screen.
Yeah, it’s a bit like that, although not quite as extreme.
So why haven’t lens manufacturers done something about this sooner? Well, it appears that shielding the sensor from these magnets isn’t so easy. But Canon have come up with a solution. Again, from the patent…
According to one aspect of the present invention, there is provided an imaging lens including: a lens; an image shake correcting action unit provided movably in a direction perpendicular to an optical axis of the lens; a stationary unit for supporting the image shake correcting action unit; a permanent magnet provided on one of the image shake correcting action unit and the stationary unit and a coil provided on an other; a drive circuit for moving the image shake correcting action unit relative to the stationary unit; a mount section for being connected to an imaging unit having an imaging element; and a conductive member which is nonmagnetically conductive and disposed between the coil and the mount section so as to include a facing surface facing a surface formed by a binding wire of the coil and having a larger area than a surface formed by an inner periphery of the coil.
So, essentially it seems that through a combination of non-magnetically conductive materials and clever shielding they’ve managed to solve, or at least reduce the issue.
Now, I don’t really shoot Canon. I do occasionally shoot it, but it’s not something I’ve noticed myself (with any brand), and don’t really have any kit here to experiment with so I don’t know how big of an issue it really is. But I’ve never seen anybody post “Why are my images noisier when I turn IS on?” to any Facebook groups. So, Canon shooters, you tell me, is this really an issue? Is noise due to IS noticeable as you creep higher up the ISO range?
Check out the full patent here.
[via Canon Watch]