How to Disable Red LED on 5D Mark II - HELP!

[clarksbrother]

Not purplish, please see example above.

Leaves a purplish blotch - nothing to do with the LED.
KP
--



http://www.ahomls.com/photo.htm
http://www.phillipsphotographer.com
'The urge to save humanity is always a false front for the urge to
rule it.', H. L. Mencken

 

[clarksbrother]

I'm familiar with what sensor heating looks like and this is definitely not sensor heating. (Also, above examples are with dark frame subtraction)

Yes, it's local heating. This affects pretty much every DSLR known to
man, and as Ken posted, it is not due to the LED.
I do quite a bit of long exposure work with the original 5D, and it
is much better than the 10D in this regard, FWIW.
The problem will go away if you enable long exposure noise reduction
(in-camera dark frame subtraction). Note that this will double the
time required to capture an exposure. Manual dark frame subtraction
will also work.
--
'Life is pleasant, death is peaceful; it's the transition that's
troublesome' - Isaac Asimov

 

[Lemming51]

I cannot fathom a way the rear LED would leave that streak. A light leak could not have such well defined borders. Is it repeatable? Always the same size/position? Does it appear if you leave the lens cap on?

I'd guess a sensor flaw or even a memory card flaw.

 

[Cal Dawson]

can't find it there is a reference to "improved the camera's dust and water resistance" but nothing about weather proof just resistent same as the xxD series not the same as the 1 series... Too Bad....

I didn't know the 5Dmk2 was weatherproof, that was always my biggest complaint about the 5 series and the xxD's.

See also: http://www.dpreview.com/previews/canoneos5dmarkII/page5.asp

--
"Passion will make you crazy, but is there any other way to live?"
—Kara Saun

--
Cal

Put a Canon to your head, You deserve it....

http://funshots.smugmug.com/

 

[Scott Larson]

The example below is with dark frame subtraction.

That proves that the streak is somewhere beyond the shutter otherwise it would have been subtracted.

 

[clarksbrother]

I disagree. When it generates the dark frame all that happens is the mirror flips back down. That doesnt impact light coming from the viewfinder or other internal areas within the camera.

The example below is with dark frame subtraction.

That proves that the streak is somewhere beyond the shutter otherwise
it would have been subtracted.

 

[clarksbrother]

I cannot fathom a way the rear LED would leave that streak. A light
leak could not have such well defined borders. Is it repeatable?

Yes

Always the same size/position?

Yes

Does it appear if you leave the lens
cap on?

Yes

I'd guess a sensor flaw or even a memory card flaw.

I've repeated this over 2 5DM2 cameras and over different memory cards.

 

[Its RKM]

I disagree. When it generates the dark frame all that happens is the
mirror flips back down. That doesnt impact light coming from the
viewfinder or other internal areas within the camera.

No. On two counts.

Firstly, the shutter closes when you do dark frame subtraction (or "long exposure noise reduction" as Canon term it). Under no circumstances does the mirror flip down with the shutter open! The sensor is always completely sealed from light coming in through the viewfinder, whether the mirror is up or down, the shutter open or closed. Try it - take a long esposure with a small flashlight shining into the viewfinder.

Also, when you do dark frame subtraction the end result is the difference between the true exposure and the dark frame. If something is common on both then it doesn't get to the final result. Your red streak indicates something that is present in either the main image (as red) or the dark frame (as cyan), but not both.

I have to say that when I read your OP I also thought you were talking about the hot amplifier effect, which is on every Canon sensor. However, having seen your example image it is clear that this is NOT a hot amplifier, but indeed a red light leakage from some source. This could quite possibly be the red LED you initially suggested - but it would have to be getting into the camera in front of the shutter (which is not impossible) just as Scott suggested or else it would indeed be removed by the dark frame subtraction. Perhaps a light baffle hasn't been properly seated in the construction of your camera, permitting the red LED to get into the mirror box, reflecting off one or more edges in there.

Either way, it certainly should not be present. I would run a test exposure of the same time, but keeping the lens cap in place and stopping the lens down to the highest f/# just to be sure. If you get the same red streak then the camera is clearly faulty and should be returned to Canon with a copy of your sample images.

--
Its RKM

 

[clarksbrother]

I disagree. When it generates the dark frame all that happens is the
mirror flips back down. That doesnt impact light coming from the
viewfinder or other internal areas within the camera.

No. On two counts.

Firstly, the shutter closes when you do dark frame subtraction (or
"long exposure noise reduction" as Canon term it). Under no
circumstances does the mirror flip down with the shutter open! The
sensor is always completely sealed from light coming in through the
viewfinder, whether the mirror is up or down, the shutter open or
closed. Try it - take a long esposure with a small flashlight
shining into the viewfinder.

I took your advice and did a 15 second exposure. ISO 3200. Lens cap on, F22.
The results are below...thoughts?

Also, when you do dark frame subtraction the end result is the
difference between the true exposure and the dark frame. If
something is common on both then it doesn't get to the final result.
Your red streak indicates something that is present in either the
main image (as red) or the dark frame (as cyan), but not both.

The red streak has shown up in long exposures both with and without dark frame subtraction.

I have to say that when I read your OP I also thought you were
talking about the hot amplifier effect, which is on every Canon
sensor. However, having seen your example image it is clear that
this is NOT a hot amplifier, but indeed a red light leakage from some
source. This could quite possibly be the red LED you initially
suggested - but it would have to be getting into the camera in front
of the shutter (which is not impossible) just as Scott suggested or
else it would indeed be removed by the dark frame subtraction.
Perhaps a light baffle hasn't been properly seated in the
construction of your camera, permitting the red LED to get into the
mirror box, reflecting off one or more edges in there.

Either way, it certainly should not be present. I would run a test
exposure of the same time, but keeping the lens cap in place and
stopping the lens down to the highest f/# just to be sure. If you
get the same red streak then the camera is clearly faulty and should
be returned to Canon with a copy of your sample images.

--
Its RKM

I hate to do it because it means I'll be without a camera for 2 weeks but it looks like I have no other choice. Much thanks for the good info and input.

 

[Its RKM]

Try it - take a long esposure with a small flashlight
shining into the viewfinder.

I took your advice and did a 15 second exposure. ISO 3200. Lens cap
on, F22.
The results are below...thoughts?

Wow! You certainly shouldn't be getting this sort of result, which indicates serious light leak internally in the camera. I am not surprised you are getting some LED leakage onto the image now. This isn't right.

I would run a test
exposure of the same time, but keeping the lens cap in place and
stopping the lens down to the highest f/# just to be sure. If you
get the same red streak then the camera is clearly faulty and should
be returned to Canon with a copy of your sample images.

I hate to do it because it means I'll be without a camera for 2 weeks
but it looks like I have no other choice. Much thanks for the good
info and input.

I understand how you feel, but its better to be a couple of weeks waiting to get the camera fixed rather than sticking with an insurmountable problem and never being able to get the shots you really want.

An intermediate path may be to send your example images of the red streak, including shots of the star field and with the lens cap on, together with the viewfinder light leakage test you just did to Canon and get them to confirm that they can fix it before sending it in.
--
Its RKM

 

[Petter Stahre]

I took your advice and did a 15 second exposure. ISO 3200. Lens cap
on, F22.
The results are below...thoughts?

I did the same thing with my 5DII and ended up with a black frame (histogram with a thin peak glued to the left).

Seems there is some internal light leakage in your 5DII, or something other strange. Send it in. Sorry.

Petter

--
| http://www.studiostahre.se |

 

[Hojong Lin]

when you do long exposure. Black duct tape would do.

I've been doing astrophotography from time to time. Covering
the rear view finder window is critical in preventing any external
light leak. Hope this tip helps.

Hojong Lin

 

[clarksbrother]

I used the included eyepiece cover that is on the strap and still got the red streak. Unfortunately it seems as if the problem lies deeper in the camera frame which allows light to leak in from elsewhere.

when you do long exposure. Black duct tape would do.

I've been doing astrophotography from time to time. Covering
the rear view finder window is critical in preventing any external
light leak. Hope this tip helps.

Hojong Lin

 

[Scott Larson]

Please, for the love of all that is sane, let this not be a light leak in a $2,000 camera. That's the thing you get when the seals on the door of your thirty year old AE-1 start to fail, not something you get with a brand new digital SLR.

 

[agavephoto]

the red streak on the photos posted is the exactly what i get when i use my 24-105L at 24mm and shoot long exposures . . . with the 5D2 AND the 5D. no other lenses i use show this. canon tech support couldn't help me over the phone, and suggested i send the lens in, not the body. i needed the lens for other purposes, and additionally it is a very good copy that i don't want replaced or messed up.

an example with the 5D (original) with the 24-105 @ 24mm, 400 second exposure at ISO 800. the red streak on the left side of the is image, and is always in the same relative place with the 5D and 5D2 but ONLY when i use the 24-105L at 24mm.



this red streak is NOT localized sensor heating . . . as already mentioned, that will appear as a more or less purple blob. this is light incident upon the sensor.

 

[RichW]

http://www.usa.canon.com/consumer/controller?act=PgComSmModDisplayAct&keycode=2112&fcategoryid=216&modelid=11924

Rich

 

[Its RKM]

an example with the 5D (original) with the 24-105 @ 24mm, 400 second
exposure at ISO 800. the red streak on the left side of the is
image, and is always in the same relative place with the 5D and 5D2
but ONLY when i use the 24-105L at 24mm.

Interesting. I notice that the EXIF data on the OP's example image was shot under similar conditions:
ISO800
24mm
f/4
902seconds.

Was the OP image shot with a 24-105L4 lens at 24mm?

Strange that this ghost red image should only appear at 24mm though - does it appear constant at all apertures?
--
Its RKM

 

[agavephoto]

no, my lens was NOT recalled: it has a date code WELL after those with that problem, nor does it show the flare problem for that recall on light hitting the FRONT of the lens. the issues here are NOT from light coming in through the front of the lens, but something else internal to the camera that happens to be with this lens. i would guess that somehow light is being reflected off the rear element of the lens, as the rear element is closest to the sensor when set to 24mm.

i get this flare with any ISO, and any longer exposures (5 min at ISO 800, or 1 hour at ISO 100, or in between and beyond . . . just two specific examples i happen to recall.

i do NOT see any flare with the 17-40L at any focal length, including 24mm.

 

[Rubio]

Do you remember if IS was turned on or off during your exposure?

 

[agavephoto]

IS was definitely NOT on during my exposures, as they were on an equatorial mount to track the stars. I do not use IS when using any type of tripod . . . ever.