D90 focus issues?

[gardy]

hi all, haven't really posted much on here, daily reader though. anyway, i believe my d90 is having some issues. I've done searching on the topic and have come up with nothing that relates to the specific problem im having. When using both auto and manual focus there is major miss focus when focusing on subjects past about 15' away. I am using single point AF with normal zone (not wide zone) and no focus point wrap around, mostly manual exposure mode. The thing is the results its giving are not predictable as far as a constant back focus or front focus, and i have tried all lenses i own on it, it will constantly focus wither several feet in front of the subject or several feet behind, as far back as 7-10 feed in either direction. i have shot some samples but wont be on my home computer for several days so will upload samples when i can. anyway does anyone see anything i have overlooked or an obvious problem i could be facing with this thing? I'm really starting to get frustrated to the point i almost skipped buying the lady a new computer to buy a d600 about 2 days ago, but don't want to spend the money if its something that can be resolved. ooh and everything has been cleaned including sensor lenses and mirror.

thanks for any help
-Tim

 

[Derrel]

Take off the lens. Set the shutter to B, in manual mode, and press the shutter release and HOLD it firmly so the mirror stays in the UP position, then blow out the AF sensor area, which is located at the bottom of the mirror box; you might very well have some fibers caught in there. Seriously.

Here is a link to an excellent article entitled Autofocus Troubleshooting. Autofocus troubleshooting by Thom Hogan

 

[jwbryson1]

And you are certain that the single point is on the spot you wish to focus?

 

[gardy]

Derrel,
thanks for the quick reply, and the link. i have not blown out the AF sensor area specifically but before i cleaned the sensor went through everything i could see without the lens on and blew it out with some rocket air. i will go back in and do so focusing on cleaning the AF sensor area.

 

[gardy]

And you are certain that the single point is on the spot you wish to focus?

100% sure

 

[KmH]

Focus point wrap around only applies when you are selecting a focus point. It does nothing when the AF module actually tries to achieve focus.

Are you using wide open or stopped down lens apertures?

You don't make paragraphs, and inconsistently use the shift key when you type, so I'm inclined to suspect user error rather than a camera error.

If the camera is indeed back focusing a repair shop can re-calibrate the AF module. Re-calibrating the AF module is best done with the lens you use most often.

You can check for back focusing by using a focus target - Nikon D70 Focus Chart

DSLR Camera Focus Test

If you weren't already aware of it, the View NX software that is included with a D90 can show where in a shot the focus was. ViewNX from Nikon

There are some scenes AF will always have difficulty with and they are noted on page 55 of the D90 User's Manual - User's Manual - D90 - Guide to Digital Photography

 

[gardy]

Focus point wrap around only applies when you are selecting a focus point. It does nothing when the AF module actually tries to achieve focus.

Are you using wide open or stopped down lens apertures?

You don't make paragraphs, and inconsistently use the shift key when you type, so I'm inclined to suspect user error rather than a camera error.

If the camera is indeed back focusing a repair shop can re-calibrate the AF module. Re-calibrating the AF module is best done with the lens you use most often.

You can check for back focusing by using a focus target - Nikon D70 Focus Chart

DSLR Camera Focus Test

If you weren't already aware of it, the View NX software that is included with a D90 can show where in a shot the focus was. ViewNX from Nikon

There are some scenes AF will always have difficulty with and they are noted on page 55 of the D90 User's Manual - User's Manual - D90 - Guide to Digital Photography

i have tried stopped down as well as wide open, and increments in between.
and just because my grammar and typing isn't that of a college professor or newspaper edited does not mean i don't know how to use my camera, and that I'm incapable of understanding something.
also, no i didnt know view NX would do that, i will check it out and see what i come up with.

 

[pixmedic]

Focus point wrap around only applies when you are selecting a focus point. It does nothing when the AF module actually tries to achieve focus.

Are you using wide open or stopped down lens apertures?

You don't make paragraphs, and inconsistently use the shift key when you type, so I'm inclined to suspect user error rather than a camera error.

If the camera is indeed back focusing a repair shop can re-calibrate the AF module. Re-calibrating the AF module is best done with the lens you use most often.

You can check for back focusing by using a focus target - Nikon D70 Focus Chart

DSLR Camera Focus Test

If you weren't already aware of it, the View NX software that is included with a D90 can show where in a shot the focus was. ViewNX from Nikon

There are some scenes AF will always have difficulty with and they are noted on page 55 of the D90 User's Manual - User's Manual - D90 - Guide to Digital Photography

wait...your extrapolating a theory on the cause of his camera issue based on his typing skills?? Classy.
By the way, hitting enter twice after every sentence doesn't constitute "paragraphs" either.

 

[gardy]

Update: blew out where the AF sensor lies, did not see much if anything fall free, so far i am seeing no change. i will check back later this evening after i am able to print a test chart and do some samples.

KmH, i checked out the feature in view NX2, this can be a handy tool in evaluating my own work, however on a particular series of photos taken of my boys on the front porch the focus point is right on them with none of it hanging over the edge of their face however the brick on the house about 5 feet away is tack sharp but nothing else.

 

[Derrel]

CHeck the screws on the lens mount on the body to make sure they are tight. It sounds like there might be a significant issue with the AF calibration on the camera...perhaps it has been knocked severely, and is wayyyy out of whack.

Have you tried this: place the camera in a fixed position. On a tripod is best. Take a long focal length lens, or a lens that is otherwise EASY to focus visually, using hand-and-eye dertermination based on viewfinder clarity...and take a few shots, using manual focus. Take three careful shots. Then, shoot a blank frame, by capping the lens and shooting on M mode at 1/125 second or whatever.

Then, take three shots from the same place,m allowing the Autofocus system to determine the right focus.

Compare the three MANUALLY focused shots against the three autofocused shots.

 

[gardy]

CHeck the screws on the lens mount on the body to make sure they are tight. It sounds like there might be a significant issue with the AF calibration on the camera...perhaps it has been knocked severely, and is wayyyy out of whack.

Have you tried this: place the camera in a fixed position. On a tripod is best. Take a long focal length lens, or a lens that is otherwise EASY to focus visually, using hand-and-eye dertermination based on viewfinder clarity...and take a few shots, using manual focus. Take three careful shots. Then, shoot a blank frame, by capping the lens and shooting on M mode at 1/125 second or whatever.

Then, take three shots from the same place,m allowing the Autofocus system to determine the right focus.

Compare the three MANUALLY focused shots against the three autofocused shots.

i have tried this to some extent, i guess when i did it i was using the indicator in the viewfinder with manual focus, Duh moment, if theres something wrong with the auto focus then that indicator will be off as well

 

[Derrel]

EXACTLY...your logic is impeccable. That is why I suggested comparing the VIEWFINDER screen's optical image using your own eyeballs as the determiner of focus/out of focus, and then allowing the AF system to have a go at it! Trying to determine where the issue is...is it the AF system??? Or what? It's sooo hard to diagnose many problems "over the wire"...

Good luck with it. Sounds like a regal PITA to have to deal with this.

 

[gardy]

Derrel, i used a tree about 30 feet away as a focus point, on a tripod, aperture priority, and 3 different lenses, every time the auto focus was used it was off, due to the snow i couldnt tell where the focus was though, and every shot except for one using manual focus and ignoring the indicator in the viewfinder yielded in focus results..
now i downloaded a test chart and took some samples of it as well

85mm 1.8g at 1.8

50mm 1.8d at 1.8

60mm micro 2.8 at 2.8

 

[gardy]

update, checked the 85mm and 60mm on the girls d3000, focus is spot on on the charts.. granted my setup wasn't lab quality

 

[Derrel]

Looking at the charts, I think this is within specification. Why do I say this??? Because the error at close range appears to be very,very slight back focus. At least at first glance. But...(there's always a but, isn't there?) I think we need to keep in mind that the AF brackets in the viewfinder are quite often NOT EXACTLY in accordance with the EXACT location where the AF system's sensor's actually take their information from...most ultra-critical testing will show that the ****exact*** area that an AF sensor's in-viewfinder scribed marks represents a very good approximation of the actual AF area--but not a totally accurate, ultra-precise indication of where the actual AF data is accepted from. In other words, in plain English, the AF marks in the viewfinder are not 100 percent accurate. In simple terms, the AF areas in the viewfinder are often not quite aligned with the bracket marks. And as a result, in really critical situations, there can easily be slight mis-focusing.

In these test...a 3/8 inch AF error toward back-focusing is pretty good...that could EASILY be attributable to a bracket/AF sensor mis-alignment, that results just from the sheer size of the AF bracket when it subtends a close-range target. At longer ranges, the AF brackets will, naturally, and inevitably, cover a wider and wider area of physical real estate. And as such, it makes it even more difficult to ***precisely*** lay the AF sensor's actual area right onto an ***exact*** target of smaller size. In other words, with say a moderate wide-angle, at 30 feet, the AF bracket might appear to be on a human's arm...buuuuuut...if the actual AF data is collected from say the LEFT EDGE of the outside AF bracket, what you think is the left arm might in actuality be...the house behind, some 10-12 feet back...

Anyway...I'm not saying this is what is going on in your situation, but the discrepancy in the AF bracket areas and where an AF system actually gets its data from is an issue that has been covered at length in many articles. Thom Hogan had some articles and posts years back on diagnosing AF problems, and the above points are ones he has made many times. I am only going in to this at this length because your test chart results make me think that the body is what Nikon would call "within specification" for function. And because you've presented some real,empirical evidence, and it deserves real commentary. From the above sample pics, I think with that lens,at close range, the AF system is doing relatively well.