The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

help.problem whit sun in the frame mamiya DM22

stpf8

New member
I don't have an answer, but I've seen very, very few good digital photos with the full sun in the frame. Digital just can't handle the tonal transitions in the vicinity of the sun like film is able to do.
 

kipling

New member
Had the same problem with my Aptus 22 and my Aptus 7 II.
I had to send mine in and they replaced something. I had other problems after that, but the blooming was fixed.

Good luck.
 

John Black

Active member
Had this same issue with the P25, Leica M8 and M9 (all Kodak CCD's). It wasn't something I set out to deliberately test for, so some of my other digital backs may have done it as well. For whatever reason, my Canon CMOS cameras have no issue with the sun. I routinely shoot into the sun with the Canon 1Ds3 and never have that issue.

As noted earlier in the thread, the sensor gets totally saturated with light and the photocells seem as if they completely err out - resulting in sensor blooming. In my experience, this was guaranteed to happen when shooting into the sun with the lens wide open. If the lens was stopped down to F8 or F11, that solved it. Not the same as your shot, but the Mamiya 200mm F2.8 APO @ F8 on a Leica S2 (Kodak CCD) -



I wasn't happy about having to stop down because that causes all sorts of issues at sunset if the wind is blowing (moving foliage) or by water and trying to capture waves, etc. But, stopping and using a tripod was the only work around I could figure out.
 

ondebanks

Member
Hello Paolo,

That's classic sensor blooming. Severe over-exposure -> too many photons are allowed to impinge on the pixel -> too many photo-electrons are released in the pixel -> the pixel electron well fills to maximum [aka sensor saturation] and then the excess electrons overflow. They would end up in adjacent pixels [aka blooming] above and below the saturated one, as if extra light had struck those pixels, but to prevent this, most CCDs (and certainly all CCDs designed for photographic cameras) devote some of the pixel area to an overflow "anti-blooming" channel.

However, in extreme cases, there are just so many excess electrons, released at such a rapid rate, that the anti-blooming drain can't conduct them away fast enough, and you get some blooming into adjacent pixels anyway. In really EXTREME cases, such as yours above, the bloomed pixels themselves saturate and bloom into still more nearby pixels, in a cascade effect, until all the electrons have found a non-saturated pixel they can call home.

Blooming can only occur vertically on the CCD: there are channel stops to prevent electrons moving laterally between columns, but electrons have to be able to move vertically along columns - as that is how they are read out at the end of an exposure.

You can find the degree of anti-blooming performance on the datasheet for each type of CCD. It is normally specified as a multiple of the full-well capacity.
Kodak will say:
Blooming Protection ...1600 x Esat
Dalsa will say:
Overexposure handling ...200 x Qmax level
Different terminology, same meaning: the overflow drain can cope with no more than 1600 (or 200) times the exposure which saturates the pixel.
Kodak, by the way, tend to have better anti-blooming than Dalsa, at least in the case of their medium format CCDs.

But it depends on time as well. If the exposure to hit a given amount of excess electron flux is longer, then the drain has more time to clear the overflow, and so it can achieve a much larger factor of anti-blooming. Kodak illustrate this clearly in their datasheets (see screen-grab below):



This plot explains why John found that stopping down helped: smaller aperture -> longer exposure -> more effective anti-blooming.

Hope this solves the mystery for you...

Ray

PS. Having to give over some of their "silicon real estate" to the overflow drain hurts the quantum efficiency of the pixel. So astronomical CCDs traditionally didn't have anti-blooming, in order to maintain maximum "fill-factor" and sensitivity to faint light. CCDs with microlenses solve this problem because they redirect light headed for the overflow drain into the sensitive area.
 

John Black

Active member
Thank you Ray - that was very informative. Why does this not happen with my Canon 1Ds3? Just curious - thinking of a CMOS M10 :)
 

torger

Active member
CMOSes are much less sensitive to blooming, that's why the 1Ds3 don't bloom.

As seen in the diagram above a workaround on CCDs is to lower the fill rate, then they get much better at avoiding blooming. In practice this means to use a very small aperture and/or strong ND filters when shooting into the sun so you can have longer shutter speeds.
 

ondebanks

Member
Thank you Ray - that was very informative.
My pleasure, John.

Why does this not happen with my Canon 1Ds3? Just curious - thinking of a CMOS M10 :)
As Torger says, CMOS is much better at anti-blooming. I suspect that it's connected to the same basic reason why CMOS can readout faster and with lower noise than CCDs - each pixel, or at least each line of pixels, has its own dedicated circuitry, so there could be multiple, parallel overflow drains. Whereas, on a CCD like the Kodak above, there will only be 2 or 4 pins dedicated to a common Lateral Overflow Drain for the whole sensor.

Ray
 

John Black

Active member
Thank you gentlemen. It is much more assuring to understand the technical reasons behind it rather than just accepting it as a fact of life.
 
Top