Lensrentals’ flange-to-sensor measurement project finds fractured sensor mounts in Sony cameras
posted Friday, June 12, 2020 at 12:24 PM EDT
Last week we told you about a fairly daunting, time-consuming investigation from our friends over at Lensrentals, who were looking into the inconsistencies and variations in flange-to-sensor distances between different cinema cameras. For a quick recap, Roger & Co. discovered that flange distance variation could make cine lenses unable to achieve sharp focus at infinity, particularly with wide-angle lenses since most cine lenses have hard stops at infinity focus. High-end cine cameras with adjustable flange distance get around this issue by allowing a particular lens and camera to be precisely calibrated to one another. However, there are many non-adjustable cinema cameras nowadays, hence the issue.
But what about photo cameras, you ask? Unlike cine lenses, most still photography lenses do not have hard stops at either end of their focusing scales and can often focus past infinity. Roger and his team therefore initially assumed that flange-to-sensor variations aren't an issue, and that photo cameras would have more flange-distance variations than cinema cameras. However, as you can guess, considering Roger just posted their follow-up Part II article over on the Lensrentals Blog, their assumptions were incorrect.
Turning to photo cameras, Roger and his team fired up their Denz Flange Depth Controller machine and went through their collection of Canon DSLRs, Sony full-frame and crop-sensor mirrorless cameras, as well as some Micro Four Thirds cameras. Unfortunately, the Lensrentals testing machine doesn't support mounts for every manufacturer at this time, so data for Nikon, Fujifilm, Leica and Pentax cameras are not available.
The results discovered were quite interesting, though not in the same way as with cine cameras. The Lensrentals team did indeed find variations in flange distances with photo cameras, but they didn't vary more than cine cameras, as the team initially suspected. However, as expected, having flange distance variations with photo cameras doesn't negatively affect functionality with focusing, since photo lenses focus past infinity. Yet, by doing these measurements, the team was able to find fixable issues in a small percentage of their fleet that they were previously unaware of.
The most notable discovery came when measuring their fleet of Sony A7-series mirrorless cameras. When investigating the large outliers of Sony full-frame mirrorless cameras, those with higher flange distance variations than the average, they discovered that a number of these cameras had cracked mounts holding the sensor to the IBIS assembly. In Roger's teardown of the A7R III, he noted that the sensor is mounted to a plastic component that is then screwed into the image stabilization system. In some of these outlier cameras, one of the three mounting points on this plastic component had broken. On other Sony outliers, one had a loose screw holding the sensor in place and another showed a metal fracture on the sensor mount. What was most shocking is that these fractures apparently didn't affect the operation of the cameras. No error messages, no wildly soft or otherwise poor image quality, and all cameras had previously passed Lensrentals' rigorous inspection process. (Lensrentals have notified Sony of this fractured sensor mount issue.)
At the end of the day, flange-to-sensor distances aren't a critical issue for photo cameras it seems. However, by doing these measurements, the Lensrentals team were able to discover minor issues with a small percentage of cameras in their fleet, and likely discovered these issues early before they became major problems.
"We learned that while slightly over 2% of photo cameras had issues, they still worked well. So that first assumption, that flange-to-sensor distance of photo cameras doesn't matter, is largely true," Roger says. "Even broken sensor mounts don't seem to matter a whole lot. However, there may be some subtle problems. AF microadjustment on an SLR may be increased; there may be some subtle image softness or mild IBIS dysfunction. But you'd have to look hard and even then finding it is iffy."
It's important to keep in mind that these test results are all based on heavily used rental cameras, so your mileage may vary with your own personal cameras. But, if you do manage to drop your camera and at first it seems OK, there might be some subtle damage deep inside.
For the full rundown of their testing results and analysis, head over to the Lensrentals Blog for all the details.
(Images courtesy of Lensrentals.)