I might just keep using it that way. Much better than what happened to my neighbor's D80: her mirror was completely locked up and the shutter stuck. It cost her something like $300.

Problem ( 2-3 fstop over-exposure ) repeats after keeping camera off for a few hours . No problem after the first shot in high speed. Must be a sticky shutter mechanism .

Good to hear. I'd still send it in for a look, though, unless I really needed to use it.

That also seemed like the most likely reason to me . After advising the guy to do some 40-50 high speed shots in continous mode, the problem seems to have mostly diminished . There is now barely 1/3-1/2 stop difference - a lot less than the 2-3 stops we started with . I say another 100 shots fired and should be back to normal :-)) .

The fact that the problem seems to get less severe at longer shutter speeds seems to suggest a shutter problem - I'd wager that the shutter was sticking in the initial exposures.

Was hast du gesagt?

lol .. but they may talk a different language...

Wait so you're giving advice in another forum based on what we tell you in this forum? Let me go and ask in another forum and see what those forum members suggest I tell you so you can tell your forum member...

I agree with those above that say to check a different lens and stop down some. When I first got my 35 1.8, The aperture blades were sticking, so by the time you would stop down to f11, the image would be completely white, even though the EXIF data looked fine. They could also manually move the aperture lever on the lens to see if the aperture blades are opening or closing as someone above mentioned.

"Right?"

Maybe - but not enough information yet. Your reasoning seems sound but your assumptions may not be.

You need to test with another lens to if the problem moves with the lens or stays with the body.

You are right clillja - I was thinking the lens off the body - don't know why ( 35 C - boiling here ) . When you put it on the body it opens up to the maximum aperture to allow metering . My mistake ...

But still ,it does not change what I'm trying to say .The lens is set at f3.5 at APERTURE PRIORTY by the user which is maximum aperture . So it is not supposed to stop down . f3.5 is "the choice" .

If it does by mistake , then we have UNDEREXPOSURE . I'm not quite sure how you relate it to OVEREXPOSURE.

" If you can see your lens is not at MAXIMUM aperture when metering/viewing ....."
If that happens , again the only way it can end is underexposure if the aperture becomes anything other than f3.5 . There is no wider aperture that can end in overexposure .

Right?

"Lens as you know are at minimum aperture when standing . When you take the shot , the body pushes the pin mechanism to open to the correct aperture."

Not true - lens is at MAXIMUM aperture when mounted on camera/viewing/metering. Lens stops down to the correct aperture as the mirror flips up. It does not "open up" to the correct aperture. (And if you remove the lens from the camera it should close down to MINIMUM aperture.)

If you can see your lens is not at MAXIMUM aperture when metering/viewing/no DOF preview - you have your answer right there. Again, switch lenses - if the problem follows the lens - the fault is in the lens. Otherwise, yes, of course it's the body/shutter/iso control.

One of the first things that come to mind is lens communication .However ,

- Shots taken at Aperture priorty , max aperture f3.5 selected .

- Lens as you know are at minimum aperture when standing . When you take the shot , the body pushes the pin mechanism to open to the correct aperture .

So considering the lens is set at f3.5 - if the mechanism does not work , the only thing that can happen is UNDEREXPOSURE ( diaphgram not opening up enough ) as seen many times in this forum .

We have 2-3 stop OVEREXPOSURE in the shots .

Your argument would have been OK if the lens was - let's say set to f 11 or so and the aperture opened more than enough ( again opposite of what you say ) all the way to f3.5 .

It may be still be originating from a lens communication problem that we are unable to relate to , though ...

I'm more inclined to thinking it is a sluggish/faulty/sticky shutter which can't reach the very high speeds at initial shots . Waiting for the guy to find a fan for test shots .

I believe the aperture mechanism is not stopping down. Could be the lens or the body. Switch lenses to confirm lens problem. It's not an electronic communications problem. The aperture linkage on Nikons is still mechanical. In the old days, this was usually caused by lubricant on the aperture blades.

I would test #2 anyway just to rule it out for certain. I know you cant open it any further but could it have closed it for the "proper" shot? At the moment we know nothing for certain as logically it should not be doing it in the first place. We need confirmation of the accuracy of the aperture to point the finger at ISO. If shutter turns out good.

Heartyfisher ,

I'll go with number one as I also was suggesting.

#2 - not correct aperture - not possible ...The setting is already on f3.5 , the max aperture for the lens . But it still manages to overexpose another 3 stops in the first shot. ( In the following shot , it exposes correctly at f3.5 ) ...Can't be opening the diaphgram more than f3.5 .

#3 - I don't see any noise that could relate to ISO 1600 ( to give the 3 stop overexposure ).

If I were having the problem or if I had the camera ,I would be running hundreds of tests myself but the D90 seems to be in the hands of someone in our forum who has to be just told what to do . Does not have the knowledge to run tests shots and make own analysis .

The fan test is a good idea - if he can find one to test . Will check with him...

is this problem repeatable.

if he can cause it to occure then you can test for the 3 posible causes

1) Shutter speed not as teh setting says- photograph a rotating fan and you will see if the shutter speed is different between the correct and incorectly exposed picture by the motion blur.

2) Aperture is not as requred. do a close up shot (with close up filter?) and see if the DOF is different between the shots.

3) ISO, not sure how to test this . maybe shoot at high ISO and see if noise level is different. I guess if its not the first 2 then it must be this.

http://img138.imageshack.us/i/dsc1399j.jpg/
http://img412.imageshack.us/i/dsc1400d.jpg/

Slower speed this time - 1/160 in both ...Everything the same .
I'd say no more than 1/2 stop difference visually .

The camera kept on doing the same problematic shots after these again at high shutter speeds .

Definitely something wrong ...I asked the guy to refit the lens and try - in case it is a lens communication problem ...

Anyone heard of such problem ?? Camera is new - 1900 shots or so ...

http://img189.imageshack.us/i/dsc1346r.jpg/
http://img713.imageshack.us/i/dsc1347.jpg/

You can check EXIFS yourself but exposure at 1/2000 f3.5 @ ISO 200 ( yeah,weird choice :-) ) in both shots . Aperture priorty , matrix metering, no bracket, no bias -everything identical . I say they look 2 fstops apart .

------------------

http://img837.imageshack.us/i/dsc1353o.jpg/
http://img696.imageshack.us/i/dsc1354p.jpg/

Everything is again the same except the correct shot is not at 1/2000 but at 1/2500 this time . N ot enought to count for the 3 stop difference .

Not a one time thing - happens every time he tries . He just uploaded some shots at slower speeds - the problem is a lot less ...I'd say 1/2 to 2/3 stops but still there ...

Sorry for not being able to load the photos direct - as I said I really don't know much about using this forum. We usually have icons etc doing that elsewhere .

He is using a Nikon 18-105mm so that can't be it ...

Some more pictures came this morning and it definitely looks like a camera problem. I'd say shutter staying too slow in initial shots . Or maybe firmware ...He's got same exposure values/EXIF in shots I would say that are easily 3 fstops apart .

The slower the speed , less the exposure difference - interesting . Or maybe just coincidence . I will not upload ( explained why above ) but will give links to photos in a while . If you are really interested you can have a look .

I have had this happen a few times with both my D80 and D300 - the only thing that I found in common was a Sigma lens. I have never had it happen with Nikon lenses but have had it happen with 3rd party "G" style lenses. It is odd for sure but it doesn't happen often to me so I have just learned to live with it.

I certainly don't have the knowledge to make a very good guess, but the shutter speed issue sounds right.
Especially in terms of light frequencies, I was thinking in terms of indoor even if shot at the same shutterspeed, the exposures could be quite different. But if its only happening with the first shot it sounds like what you are on to is more likely, unless its just a firmware/hardware issue. who knows, sounds bizarre.

Does the user live in a cold climate?

Jonnyapple , it is apparently only in the first shot when the camera is turned on .

Jorpet , can't be because of spot metering as the exposure in EXIF is same for both shots . So the camera is actually reading the correct exposure for both shots ( however ends up somehow managing to overexpose the first shot ) Checked anyways and saw it is matrix metering anyway .

Just one clue though ...I see both problematic sets of shots were on very high speeds - 1/2000 and 1/800 . Can it be the camera shutter is a bit sluggish in the first shot after a period of no use ?? The camera is used little - still at 1900 shutters or so . Just a wild guess - don't know if it makes amy sense .

I will ask the guy to make the same tests on slower speeds . Will also ask permission if I can upload his photos here ...

Do you mean spot metering, JorPet?

It sounds like center weighted metering to me. One shot is on something darker than the overall picture second is on something that is about the mid-point for the picture. If hand held, it might take just a fraction of a degree to change the exposure that amount.