Electronic shutters

[Mark Donnell]

Not sure where to put this thread, but I have recently become more educated on the details of electronic shutters and thought others might be interested. Occasionally some of our members with advanced technical knowledge have referred to some of these issues, but I never fully understood the importance of the sensor read-out time, rolling shutter artifacts, and for photography, flash sync time as it relates to electronic shutters. The Wikipedia articles on camera shutters and rolling shutters are good basic introductions. For video purposes, almost all shutters are electronic, although I believe that some high-end cameras shooting at 24p still employ mechanical shutters. I will post a few links here for those who are interested.

I can't get the link to work, but a good article is a PDF entitled "Shutter Operations for CCD and CMOS Image Sensors" at TrueSenseImaging.com (best way to get there is to Google the title and TrueSenseImaging). Another good basic reference is [PDF] "How an electronic shutter works in a CMOS camera" (again Google the title). The "rolling shutter" is explained well at www.ptgrey.com/support/kb/index.asp?a=4&q=115 For those interested in the Panasonic DMC-GH3 or the rolling shutter effect, this article is excellent - http://m43photo.blogspot.be/2012/12/...c-shutter.html. Rolling and global shutter operations, plus the advantages and disadvantages, are explained at : andor.com/learning-academy/faqs-on-rolling-and-global-exposure-what-you-need-to-know (sorry - it is www but the link will not work - try pasting it in).

[Jack Zhang]

The F65 I believe uses a mechanical shutter. The F55 is a global shutter.

My preference is the Global shutter, but we need an advancement like organic CMOS sensors (which improves sensitivity and dynamic range) before we can reduce noise to acceptable levels for proper acquisition. (the smaller the sensor, the more important this would be.)

[Lawrence Bansbach]

Originally Posted by Jack Zhang

The F65 I believe uses a mechanical shutter. The F55 is a global shutter.

My preference is the Global shutter, but we need an advancement like organic CMOS sensors (which improves sensitivity and dynamic range) before we can reduce noise to acceptable levels for proper acquisition. (the smaller the sensor, the more important this would be.)

Well, separating the global-shutter circuitry from the photodiodes and moving it to a different layer would accomplish much of that. See the comments by David Gilblom, president of Alternative Vision, on "Red Announces 6K 120fps Sensor with 15+ Stops DR" (Image Sensors World, April 16, 2012). He describes a hypothetical 5-μm-pixel global-shutter sensor that his company could make. With a full-well capacity of around 200,000e- and a read noise of 2-3e-, the theoretical dynamic range could be calculated as follows:

20 x log(200,000e- ÷ 3e-) = 96.5 dB (~16 stops), or 100 dB at 2e-.

Even with smaller sensors, fairly high dynamic ranges could be achieved with this design, especially if complemented by simultaneous dual readouts (à la the Arri Alexa sensor and the BAE/Fairchild CIS2051 in BMD's BMCC).

[Jack Zhang]

Problem is that the R&D of the big companies might be doing this research on their own, and it still may be a couple of years down the line before it's cost-effective to manufacture. This has been the biggest stumbling block for smaller global shutter cameras. Currently, the smallest sensor size you see global shutter cameras in practice is S35. Almost no one is using 2/3'' or smaller.