Problem with 645D CCD ?

larkis · 01-18-2012, 11:13 PM

I have noticed this today on my image when I adjusted the contrast and you can see some of it even on the unadjusted image. Has anyone seen this before ?

Adam · 01-18-2012, 11:27 PM

I've seen it before, but not on a 645D. It generally indicated a flaw in the sensor, and it doesn't look too good- I'd send the camera in and get the sensor replaced if it's still covered by warranty.

larkis · 01-19-2012, 02:37 AM

I went out to do another test and upon boosting the exposure this is what I see:

Shuttershane · 01-19-2012, 11:05 AM

what you have there is sensor split. I actually have a thread here in the forums on the issue.

My sensor also had this issue. I sent it to pentax and they upgraded the firmware. For whatever reason that seemed to have taken care of the issue. I can no longer reproduce it....but if it comes back up its going back for a sensor replacement. Funny thing is, the shot i noticed my sensor split issue was also taken at Valley View in Yosemite. I noticed it while doing a black and white conversion of this picture: http://www.flickr.com/photos/14288454@N07/5498274600/

Also what ISO are those shots taken at?

larkis · 01-19-2012, 11:09 AM

At 100iso, I have another issue with a dead row of pixels in the blue channel above 800iso I need to investigate. It does not show up in most shots but once in a while it's visible.

Shuttershane · 01-19-2012, 11:12 AM

Call up Pentax. Ask to speak with Mark Davis. I literally got my camera back this week but haven't really had a chance to fully test it out. I may actually do a long exposure like you did to see if its an issue.

BTW my issue also seemed to happen in the blue channel. I really only noticed it when boosting the exposure on blue skies

larkis · 01-19-2012, 11:29 AM

Do you have a link to your split sensor thread ? I'm currently In Yosemite and won't be back home in Canada until the end of the month so I guess i will have to live with the problem somehow. I have done a few other long exposures before and can't see the issue on those images.

Yamanobori · 01-19-2012, 07:27 PM

It has happen with my 645D on long exposure at night from time to time. I think it might have something to do with the readouts are independent on each half of the sensor and when there is a color differential in the scene from one side to the other, you get a split.

This never happens with daylight exposures. And most night exposures are fine.

Shuttershane, did Pentax give a reason for the uneven readout?

dgpaterson2000 · 01-20-2012, 04:17 AM

This has also happened with my 645D - it's starting to seem like a pretty common problem. The camera has been in Japan for two months and I'm getting it back next week after "service, cleaning and re-calibration".

In my case, the problem also appeared with long night exposures (of the aurora, in Norway) and 2 shots out of about 60 were affected.

672 · 01-23-2012, 12:05 AM

I also had this problem and one or two threads here last year before getting the problem resolved. And by resolved i mean brand new camera, not "sensor replacement".

ziyah.gafic · 11-29-2012, 07:36 AM

i also have similar thing...towards the edge of the frame, here is detail...it appears from 200iso onwards, on 100iso it's gone, which is not really consolation...someone said that repeated mapping solved it but i tried it dozen times...CRISCAM said i can send it directly to Japan to save some time. did anyone sent to Japan directly, any contacts? i have US warranty but they said it should be fine...

DandA · 11-29-2012, 12:21 PM

This is a very uneducated guess....but since the readouts of each half of the sensor are on seperate channels, is it possible that it takes the combination of long exposures and cold climate, to effect this phenominon (issue). In other words, very cold temps cause each channel to be read with slightly different timings...therefore they are not in sync" with respect to the other and also the longer the exposure goes for, the more the descrepancy between the respective readout timings (or differences). Very short daylight exposures in cold weather are not long enough to to effectively see theses miniscle timing differences but longer exposures are. Likewise normal ambient temps do not effect these timing diffferences of the channels, hence, no line down the middle of the senor is seen.

This is not an absolute...as I'm sure there are many excepts and other reasons for this sensor issue.

Dave (D&A)

AstroMagus · 09-11-2019, 10:13 AM - **3 Likes**

Hello All,

I realize this is a new response to an old thread but I recently encountered this similar/same issue while traveling with my 645D and thought I would share my experience for those 645D users out there.

I didn't encounter a split sensor issue like the initial image query but the later EXIF attachment and other images on a separate related post demonstrated exactly what I encountered.

On daytime images, it was faint but magnification on post processing demonstrated a bright red dot with a vertical line extending down to the edge of the frame. Evening shots demonstrated it best for me. Sorry couldn't upload images to post here.

Anyway being technically versed, I searched online on numerous physics sites dealing with CCDs and found that it's due to hot pixel with associated "pixel bleeding" down the column. A lot of Astrophotography websites have this info with clear obvious examples. The photography sites I looked at didn't really offer any examples and basically rehashed each other.

Since the artifact wasn't black (which would suggest a DEAD PIXEL which requires a definite expert technical repair), I realized it was a hot pixel (these are always colored, Red in this case). This is attributed to a "full well" of that pixel with subsequent "overflow" and an associated SAME COLOR "column" extending below it.

Unfortunately Ricoh just referred me to a repair site in U.S., but their response was fast so they get points for that. I also spoke to someone at B&H in NYC (a great store). Even though he was a Pentax MF technical expert, he wasn't able to offer an initial owner fix. Lesson here is importance of reading manual AND understanding CCD (or CMOS for that matter) technical aspects, including artifacts.

Fortunately I found a simple fix. Reading the manual, I got to "pixel mapping function" in Tools 4 menu after checking for "dust alert" in same menu (there was NONE) and before attempting to clean the sensor (which was unnecessary in my case- it was spotless). I ran the mapping function and after awhile it showed the last photo with the defect. I ran it again just to be sure and there was no subsequent photo show on LCD display. Then I took several photos with the camera, and it was gone on magnified download images.

No one has offered this initial fix attempt beforehand anywhere on this site, on my search, and the 645D manuals (main and menu) don't really discuss the function in depth but it seems Ricoh/Pentax has an internal fix for this issue in the tools menu. Thankfully I only had one artifact. It may not be so simple with multiple artifacts

Hope this helps others.

johnha · 09-12-2019, 05:49 AM

Thanks for the info. I've seen a hollow green square artefact (looks like the outline of a pixel) appear on an in-camera JPEG conversion of a night shot, it doesn't look like a sensor issue as it's not in (or at least not visible in) the DNG file. I re-ran the JPEG conversion with the same result, but a different DNG converter did not reproduce it.

01-18-2012, 11:13 PM	#1
larkis Senior Member Join Date: Dec 2011 Posts: 130	Problem with 645D CCD ? I have noticed this today on my image when I adjusted the contrast and you can see some of it even on the unadjusted image. Has anyone seen this before ?

01-18-2012, 11:27 PM	#2
Adam Administrator Site Webmaster Join Date: Sep 2006 Location: Arizona Photos: Gallery \| Albums Posts: 51,595	I've seen it before, but not on a 645D. It generally indicated a flaw in the sensor, and it doesn't look too good- I'd send the camera in and get the sensor replaced if it's still covered by warranty. Adam PentaxForums.com Webmaster (Site Usage Guide \| Site Help \| My Photography) PentaxForums.com server and development costs are user-supported. You can help cover these costs by donating or purchasing one of our Pentax eBooks. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, KEH, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
645D color problem at 1600	672	Pentax Medium Format	18	07-01-2011 01:52 AM
smudge on my ccd mirror	fevbusch	Pentax DSLR Discussion	3	12-25-2010 12:32 PM
k200d infrared ir ccd	jon pafford	Pentax DSLR Discussion	3	02-18-2010 11:58 AM
K-7 Viewfinder and Mirror or CCD problem	Wildspirit	Pentax DSLR Discussion	2	08-18-2009 06:41 AM
Dust or CCD Problem?	Twinky	Pentax DSLR Discussion	9	07-09-2007 11:27 AM

01-19-2012, 02:37 AM	#3
larkis Senior Member Join Date: Dec 2011 Posts: 130 Original Poster	I went out to do another test and upon boosting the exposure this is what I see:

01-19-2012, 11:05 AM	#4
Shuttershane Senior Member Join Date: Dec 2010 Photos: Gallery Posts: 164	what you have there is sensor split. I actually have a thread here in the forums on the issue. My sensor also had this issue. I sent it to pentax and they upgraded the firmware. For whatever reason that seemed to have taken care of the issue. I can no longer reproduce it....but if it comes back up its going back for a sensor replacement. Funny thing is, the shot i noticed my sensor split issue was also taken at Valley View in Yosemite. I noticed it while doing a black and white conversion of this picture: http://www.flickr.com/photos/14288454@N07/5498274600/ Also what ISO are those shots taken at?

01-19-2012, 11:09 AM	#5
larkis Senior Member Join Date: Dec 2011 Posts: 130 Original Poster	At 100iso, I have another issue with a dead row of pixels in the blue channel above 800iso I need to investigate. It does not show up in most shots but once in a while it's visible.

01-19-2012, 11:12 AM	#6
Shuttershane Senior Member Join Date: Dec 2010 Photos: Gallery Posts: 164	Call up Pentax. Ask to speak with Mark Davis. I literally got my camera back this week but haven't really had a chance to fully test it out. I may actually do a long exposure like you did to see if its an issue. BTW my issue also seemed to happen in the blue channel. I really only noticed it when boosting the exposure on blue skies

01-19-2012, 11:29 AM	#7
larkis Senior Member Join Date: Dec 2011 Posts: 130 Original Poster	Do you have a link to your split sensor thread ? I'm currently In Yosemite and won't be back home in Canada until the end of the month so I guess i will have to live with the problem somehow. I have done a few other long exposures before and can't see the issue on those images.

01-19-2012, 07:27 PM	#8
Yamanobori Veteran Member Join Date: Feb 2011 Photos: Gallery Posts: 573	It has happen with my 645D on long exposure at night from time to time. I think it might have something to do with the readouts are independent on each half of the sensor and when there is a color differential in the scene from one side to the other, you get a split. This never happens with daylight exposures. And most night exposures are fine. Shuttershane, did Pentax give a reason for the uneven readout?

01-20-2012, 04:17 AM	#9
dgpaterson2000 Junior Member Join Date: Jan 2011 Location: Perthshire Posts: 29	This has also happened with my 645D - it's starting to seem like a pretty common problem. The camera has been in Japan for two months and I'm getting it back next week after "service, cleaning and re-calibration". In my case, the problem also appeared with long night exposures (of the aurora, in Norway) and 2 shots out of about 60 were affected.

01-23-2012, 12:05 AM	#10
672 Veteran Member Join Date: Dec 2010 Location: santa monica Photos: Gallery Posts: 486	I also had this problem and one or two threads here last year before getting the problem resolved. And by resolved i mean brand new camera, not "sensor replacement".

11-29-2012, 12:21 PM	#12
DandA Senior Member Join Date: Feb 2011 Posts: 129	This is a very uneducated guess....but since the readouts of each half of the sensor are on seperate channels, is it possible that it takes the combination of long exposures and cold climate, to effect this phenominon (issue). In other words, very cold temps cause each channel to be read with slightly different timings...therefore they are not in sync" with respect to the other and also the longer the exposure goes for, the more the descrepancy between the respective readout timings (or differences). Very short daylight exposures in cold weather are not long enough to to effectively see theses miniscle timing differences but longer exposures are. Likewise normal ambient temps do not effect these timing diffferences of the channels, hence, no line down the middle of the senor is seen. This is not an absolute...as I'm sure there are many excepts and other reasons for this sensor issue. Dave (D&A)

09-11-2019, 10:13 AM - 3 Likes	#13
AstroMagus New Member Join Date: Sep 2019 Posts: 1	Hello All, I realize this is a new response to an old thread but I recently encountered this similar/same issue while traveling with my 645D and thought I would share my experience for those 645D users out there. I didn't encounter a split sensor issue like the initial image query but the later EXIF attachment and other images on a separate related post demonstrated exactly what I encountered. On daytime images, it was faint but magnification on post processing demonstrated a bright red dot with a vertical line extending down to the edge of the frame. Evening shots demonstrated it best for me. Sorry couldn't upload images to post here. Anyway being technically versed, I searched online on numerous physics sites dealing with CCDs and found that it's due to hot pixel with associated "pixel bleeding" down the column. A lot of Astrophotography websites have this info with clear obvious examples. The photography sites I looked at didn't really offer any examples and basically rehashed each other. Since the artifact wasn't black (which would suggest a DEAD PIXEL which requires a definite expert technical repair), I realized it was a hot pixel (these are always colored, Red in this case). This is attributed to a "full well" of that pixel with subsequent "overflow" and an associated SAME COLOR "column" extending below it. Unfortunately Ricoh just referred me to a repair site in U.S., but their response was fast so they get points for that. I also spoke to someone at B&H in NYC (a great store). Even though he was a Pentax MF technical expert, he wasn't able to offer an initial owner fix. Lesson here is importance of reading manual AND understanding CCD (or CMOS for that matter) technical aspects, including artifacts. Fortunately I found a simple fix. Reading the manual, I got to "pixel mapping function" in Tools 4 menu after checking for "dust alert" in same menu (there was NONE) and before attempting to clean the sensor (which was unnecessary in my case- it was spotless). I ran the mapping function and after awhile it showed the last photo with the defect. I ran it again just to be sure and there was no subsequent photo show on LCD display. Then I took several photos with the camera, and it was gone on magnified download images. No one has offered this initial fix attempt beforehand anywhere on this site, on my search, and the 645D manuals (main and menu) don't really discuss the function in depth but it seems Ricoh/Pentax has an internal fix for this issue in the tools menu. Thankfully I only had one artifact. It may not be so simple with multiple artifacts Hope this helps others.

09-12-2019, 05:49 AM	#14
johnha Veteran Member Join Date: Apr 2012 Location: Lancashire, UK Photos: Albums Posts: 1,155	Thanks for the info. I've seen a hollow green square artefact (looks like the outline of a pixel) appear on an in-camera JPEG conversion of a night shot, it doesn't look like a sensor issue as it's not in (or at least not visible in) the DNG file. I re-ran the JPEG conversion with the same result, but a different DNG converter did not reproduce it.