D800 AF Dutch repair procedure

Started Jul 14, 2012 | Discussions
Marc Heijligers Contributing Member • Posts: 590
D800 AF Dutch repair procedure

For those who are interested, here is an impression on the steps and equipment used for the D800 AF repair as performed by Nikon Service Point in the Netherlands.

http://hifivoice.wordpress.com/2012/07/14/d800-autofocus-repair-overview/

 Marc Heijligers's gear list:Marc Heijligers's gear list
Sony RX100 III Sony a7R III Sony FE 35mm F2.8 Sony FE 55mm F1.8 Sony FE 28mm F2 +5 more
Nikon D800
If you believe there are incorrect tags, please send us this post using our feedback form.
Jolleyman Regular Member • Posts: 290
Re: D800 AF Dutch repair procedure

I wonder about the vertical lines being used to test the outer AF points, it is my understanding and experience testing that the outer AF points only detect horizontal lines.
Dave

 Jolleyman's gear list:Jolleyman's gear list
Olympus OM-D E-M10 II Nikon D850 Nikon AF-S Nikkor 16-35mm F4G ED VR Nikon AF-S Nikkor 24-70mm f/2.8G ED Nikon AF-S Micro-Nikkor 105mm F2.8G IF-ED VR +5 more
OP Marc Heijligers Contributing Member • Posts: 590
Re: D800 AF Dutch repair procedure

The camera is tilted 90 degrees in this case.

There is also a card with horizontal lines.

 Marc Heijligers's gear list:Marc Heijligers's gear list
Sony RX100 III Sony a7R III Sony FE 35mm F2.8 Sony FE 55mm F1.8 Sony FE 28mm F2 +5 more
michaeladawson Forum Pro • Posts: 13,326
Oh, oh!

Not a Leonard Shepherd approved AF target. It seems even Nikon does not have access to the super secret information that Leonard does. LOL!!

Marc Heijligers wrote:

For those who are interested, here is an impression on the steps and equipment used for the D800 AF repair as performed by Nikon Service Point in the Netherlands.

http://hifivoice.wordpress.com/2012/07/14/d800-autofocus-repair-overview/

-- hide signature --

Mike Dawson

 michaeladawson's gear list:michaeladawson's gear list
Nikon D810 Nikon D7200 Nikon D5 Fujifilm X-T2 Fujifilm X-H1 +30 more
Leonard Shepherd
Leonard Shepherd Forum Pro • Posts: 17,611
Re: Oh, oh!

michaeladawson wrote:

Not a Leonard Shepherd approved AF target. It seems even Nikon does not have access to the super secret information that Leonard does.

You imply you may not know much about AF

The "super secret information" you refer to is in every DSLR instruction book and on Nikon web site. Maybe you have yet to read or understand Nikon's guidance!.

Looking specifically at this target provided the pattern is an appropriate size relative to the AF detection area, which it will be in a Nikon lab, and provided the target is centered on the optical axis, which it will be in a Nikon lab, it will work good

Assuming this is genuine Nikon perhaps copyright target (it differs from the suggested 24-70 service link) it needs rotating 90 degrees to check outer AF points.

One challenge using a target like this is getting the focus distance right relative to the AF detection area. With many commercial targets and resolution charts the focus distance is in the region of 25 times focal length

Another challenge outside a laboratory is accurately aligning the camera on the optical axis. Although LensAlign uses a target which falls short on Nikon's guidance, lining up the central recessed circle ensures the camera is focused on the optical axis.

Get either of these set up details wrong when testing can lead to situations (Nikon's words) where the following applies Please note that the camera may believe focus has been acquired and still display the in-focus indicator and the camera may still make a beep indicating correct focus.

The point I make (which you seem to object to even though it is accurate) is all of the 51 D800 cameras with images I have looked at at dpreview are ones where (based on Nikon's examples or the vertical AF outer detail requirement) used targets where Nikon say the camera may believe focus has been achieved when the image is out of focus.

Having used Nikon AF for 13 years I do sometimes get wrong focus - but only with subjects where Nikon say AF may not be reliable. This implies the "super secret information" which is hard to miss if you look for it explains why some targets can result in wrong focus

-
Leonard Shepherd

Many problems turn out to be a lack of intimate knowledge of complex modern camera equipment.

 Leonard Shepherd's gear list:Leonard Shepherd's gear list
Nikon D810 Nikon D7200 Nikon D500 Nikon D850 Nikon Z7 +24 more
michaeladawson Forum Pro • Posts: 13,326
Re: Oh, oh!

Leonard Shepherd wrote:

The point I make (which you seem to object to even though it is accurate) is all of the 51 D800 cameras with images I have looked at at dpreview are ones where (based on Nikon's examples or the vertical AF outer detail requirement) used targets where Nikon say the camera may believe focus has been achieved when the image is out of focus.

Having used Nikon AF for 13 years I do sometimes get wrong focus - but only with subjects where Nikon say AF may not be reliable. This implies the "super secret information" which is hard to miss if you look for it explains why some targets can result in wrong focus

And I and many others will continue to resond with "hogwash". These targets work. I won't attest to ALL of them, but the vast majority of targets are fine. The targets show left side problems reproducibly where no issues are seen on the left. Cameras come back from Nikon with the AF issue is resolved, testing against these same targets.

-
Leonard Shepherd

Many problems turn out to be a lack of intimate knowledge of complex modern camera equipment.

-- hide signature --

Mike Dawson

 michaeladawson's gear list:michaeladawson's gear list
Nikon D810 Nikon D7200 Nikon D5 Fujifilm X-T2 Fujifilm X-H1 +30 more
vincent__l Regular Member • Posts: 336
Re: Oh, oh!

1) D800 using target A does not focus correctly
2) Send D800 to Nikon service center
3) After repair, D800 using target A now works

Multiple choice:

a) D800 AF was faulty
b) target A was the problem

michaeladawson wrote:

Leonard Shepherd wrote:

The point I make (which you seem to object to even though it is accurate) is all of the 51 D800 cameras with images I have looked at at dpreview are ones where (based on Nikon's examples or the vertical AF outer detail requirement) used targets where Nikon say the camera may believe focus has been achieved when the image is out of focus.

Having used Nikon AF for 13 years I do sometimes get wrong focus - but only with subjects where Nikon say AF may not be reliable. This implies the "super secret information" which is hard to miss if you look for it explains why some targets can result in wrong focus

And I and many others will continue to resond with "hogwash". These targets work. I won't attest to ALL of them, but the vast majority of targets are fine. The targets show left side problems reproducibly where no issues are seen on the left. Cameras come back from Nikon with the AF issue is resolved, testing against these same targets.

-
Leonard Shepherd

Many problems turn out to be a lack of intimate knowledge of complex modern camera equipment.

alabaster Senior Member • Posts: 1,581
Re: Oh, oh!

Leonard Shepherd wrote:

michaeladawson wrote:

Not a Leonard Shepherd approved AF target. It seems even Nikon does not have access to the super secret information that Leonard does.

You imply you may not know much about AF

The "super secret information" you refer to is in every DSLR instruction book and on Nikon web site. Maybe you have yet to read or understand Nikon's guidance!.

Looking specifically at this target provided the pattern is an appropriate size relative to the AF detection area, which it will be in a Nikon lab, and provided the target is centered on the optical axis, which it will be in a Nikon lab, it will work good

Assuming this is genuine Nikon perhaps copyright target (it differs from the suggested 24-70 service link) it needs rotating 90 degrees to check outer AF points.

One challenge using a target like this is getting the focus distance right relative to the AF detection area. With many commercial targets and resolution charts the focus distance is in the region of 25 times focal length

Another challenge outside a laboratory is accurately aligning the camera on the optical axis. Although LensAlign uses a target which falls short on Nikon's guidance, lining up the central recessed circle ensures the camera is focused on the optical axis.

Get either of these set up details wrong when testing can lead to situations (Nikon's words) where the following applies Please note that the camera may believe focus has been acquired and still display the in-focus indicator and the camera may still make a beep indicating correct focus.

The point I make (which you seem to object to even though it is accurate) is all of the 51 D800 cameras with images I have looked at at dpreview are ones where (based on Nikon's examples or the vertical AF outer detail requirement) used targets where Nikon say the camera may believe focus has been achieved when the image is out of focus.

Having used Nikon AF for 13 years I do sometimes get wrong focus - but only with subjects where Nikon say AF may not be reliable. This implies the "super secret information" which is hard to miss if you look for it explains why some targets can result in wrong focus

-
Leonard Shepherd

Many problems turn out to be a lack of intimate knowledge of complex modern camera equipment.

It's impossible to trust any of this. You've been consistently caught out not knowing fundamental physics.

Some few examples of many:
http://forums.dpreview.com/forums/read.asp?forum=1021&message=41995324
http://forums.dpreview.com/forums/read.asp?forum=1021&message=41953715
http://forums.dpreview.com/forums/read.asp?forum=1021&message=41520031
http://forums.dpreview.com/forums/read.asp?forum=1021&message=41233226

-- hide signature --
 alabaster's gear list:alabaster's gear list
Nikon D810 +1 more
em_dee_aitch Veteran Member • Posts: 3,675
Re: D800 AF Dutch repair procedure

Marc Heijligers wrote:

For those who are interested, here is an impression on the steps and equipment used for the D800 AF repair as performed by Nikon Service Point in the Netherlands.

http://hifivoice.wordpress.com/2012/07/14/d800-autofocus-repair-overview/

Interestingly, the NikonUSA employees I have been corresponding with from the Melville office have disclaimed the reports about the European procedure being disseminated globally. I really don't know who to believe, and really it doesn't matter. But looking at this procedure, I feel like my camera got the hardware part but not the software part, because clearly the general range of my results improved, while the AF fine tune values across the bracket still vary too much to be considered fully fixed.

Great post, especially the photo showing that the laser rig exists.

-- hide signature --

David Hill
http://www.sanfranciscoweddingphotographer.com
San Francisco & San Jose, CA | Austin, TX
Wedding Photographer and Apparent Gearhead

OP Marc Heijligers Contributing Member • Posts: 590
Re: D800 AF Dutch repair procedure

Some more news on the repair procedure.

Yesterday I was told that a reference D800 camera (sent by Nikon) is used to analyse the software test setup that is used to re-program the AF field offsets. This reference camera is used to determine the measured values for all the AF points for the specific test setup of the service centre, and these values are stored in the software. So it essentially calibrates the test-setup. Subsequently, a victim camera goes through the test, the measured values are compared to the values measured by the reference camera, and adapted accordingly.

In conclusion, the measurement of a vicitim camera is a relative measurement to a reference camera.

I can imagine (this is not verified, so don't take this as given!!!!) that this may explain why the various service centres do report different results. Nikon has sent a new reference camera to NSP in the Netherlands, to further improve on the repair procedure. I can imagine this will also be done with other Service Centres world-wide. Let's hope this gives a better solution to the people that still report problems after their repair.

 Marc Heijligers's gear list:Marc Heijligers's gear list
Sony RX100 III Sony a7R III Sony FE 35mm F2.8 Sony FE 55mm F1.8 Sony FE 28mm F2 +5 more
Leonard Shepherd
Leonard Shepherd Forum Pro • Posts: 17,611
Re: Oh, oh!

michaeladawson wrote:

And I and many others will continue to resond with "hogwash". These targets work. I won't attest to ALL of them, but the vast majority of targets are fine.

Call it "hogwash" if you want - but Nikon say on page 100 of the D800 instructions AF does not perform well with most of the targets used in the images posted at dpreview.

The remainder contain no vertical detail - where outer AF points on 51 point AF systems are unlikely to detect the target.

If it was hogwash some-one would surely have posted an image at dpreview in the 4 months the D800 has been out using a target Nikon do not caution may cause poor AF in the D800 instructions.
--
Leonard Shepherd

Many problems turn out to be a lack of intimate knowledge of complex modern camera equipment.

 Leonard Shepherd's gear list:Leonard Shepherd's gear list
Nikon D810 Nikon D7200 Nikon D500 Nikon D850 Nikon Z7 +24 more
gl2k
gl2k Veteran Member • Posts: 3,232
Re: Oh, oh!

Leonard Shepherd wrote:

michaeladawson wrote:

And I and many others will continue to resond with "hogwash". These targets work. I won't attest to ALL of them, but the vast majority of targets are fine.

Call it "hogwash" if you want - but Nikon say on page 100 of the D800 instructions AF does not perform well with most of the targets used in the images posted at dpreview.

The remainder contain no vertical detail - where outer AF points on 51 point AF systems are unlikely to detect the target.

If it was hogwash some-one would surely have posted an image at dpreview in the 4 months the D800 has been out using a target Nikon do not caution may cause poor AF in the D800 instructions.
--
Leonard Shepherd

Many problems turn out to be a lack of intimate knowledge of complex modern camera equipment.

According to this b...s... a Nikon cam can hardly take in focus images at all with AF.

Except for extreme backlight my D300 easily handles all of the mentioned scenes. Although the lens makes a big difference.

OP Marc Heijligers Contributing Member • Posts: 590
Re: Oh, oh!

Leonard Shepherd wrote:

Nikon say on page 100 of the D800 instructions AF does not perform well with most of the targets used in the images posted at dpreview.

It is not Nikon, but you who says this. You structurally deny the logic explaining why page 100 is NOT applicable to the most of the tests shown in this forum.

If it was hogwash some-one would surely have posted an image at dpreview in the 4 months the D800 has been out using a target Nikon do not caution may cause poor AF in the D800 instructions.

Leonard uses a writing style where he uses the word "Nikon", whereas his observations are done on personal grounds. This is a trick used by many politicians, to widen the context of their assertions and make them "more true" than they are. The best thing to do is to read the word "Leonard" or "I/me" where he writes Nikon:

If it was hogwash some-one would surely have posted an image at dpreview in the 4 months the D800 has been out using a target I do not caution may cause poor AF as I did understand from the D800 instructions.

You now see how his statement weakens, because it is clear that we're looking at Leonard's observations, and NOT Nikon's observations.

 Marc Heijligers's gear list:Marc Heijligers's gear list
Sony RX100 III Sony a7R III Sony FE 35mm F2.8 Sony FE 55mm F1.8 Sony FE 28mm F2 +5 more
Alejandro Daz del Ro Fery
Alejandro Daz del Ro Fery Veteran Member • Posts: 3,735
What about lenses bayonet align ? ... Re: D800 AF Dutch repair procedure

The only sin committed by the autofocus system of the D800 is its best IQ ever high resolution sensor.

Marc Heijligers wrote:

For those who are interested, here is an impression on the steps and equipment used for the D800 AF repair as performed by Nikon Service Point in the Netherlands.
http://hifivoice.wordpress.com/2012/07/14/d800-autofocus-repair-overview/

-- hide signature --

Un saludo.

GMack Senior Member • Posts: 2,928
Re: D800 AF Dutch repair procedure

I'm somewhat confused about the "reference camera" part. Is it used just to demonstrate "Here is what a good one is like. Check it if you want." or are they dumping that camera's data to the customer's body which could account for the deviations in L-R focus errors?

Somehow I get the feeling Nikon is programming these things quickly with a general program from one that passed QC and not individually. Seems they try and align mechanically close to the reference camera, and then dump the resultant software from the QC camera into them.

Sounds like this is not a standard either if each country's lead tech comes up with their own means to do this (Germany makes the mirror-alignment jig?) and not from the factory either. No wonder there is a Hodge-podge of fixes from various facilities around the world, along with varying designs of focus charts too (e.g. lines, Siemens star, lettering, boxes, etc.).

Mack

Shotcents
Shotcents Veteran Member • Posts: 4,472
Leonard...so wrong

It's pretty amazing how wrong you've been regarding this issue.

I have two D800 bodies and the 1st one focuses like a champ. The second has the dreaded left AF issue. Using ANY target the second body will miss the focus 100% of the time while the first body NEVER misses.

Arguing over what's in the manual is rather silly at this point. Just like my 2nd D800 you've missed 100% of the time. Give it a rest.

Robert

 Shotcents's gear list:Shotcents's gear list
Nikon Coolpix P7700 Nikon D800 Nikon D5200 Nikon D5300 Nikon Df +11 more
OP Marc Heijligers Contributing Member • Posts: 590
Re: D800 AF Dutch repair procedure

GMack wrote:

I'm somewhat confused about the "reference camera" part. Is it used just to demonstrate "Here is what a good one is like. Check it if you want." or are they dumping that camera's data to the customer's body which could account for the deviations in L-R focus errors?

The reason for the reference camera is to calibrate the measurement setup.

If you put a camera in place in a measurement setup, it is not exactly at 1m and not 100% aligned to the card holding the vertical stripes. Furthermore, the test is carried out with a reference 50mm/1.4D lens (at f2.2 if I understood correctly), that may also have variations, so those also need to be included and recorded.

So essentially a camera that is adjusted is made equal to the reference camera as much as possible. This means that the repair is at most as good as the reference camera will be.

 Marc Heijligers's gear list:Marc Heijligers's gear list
Sony RX100 III Sony a7R III Sony FE 35mm F2.8 Sony FE 55mm F1.8 Sony FE 28mm F2 +5 more
elliotn Senior Member • Posts: 2,009
Re: D800 AF Dutch repair procedure

Nikon seem to be learning how to repair their own camera in slow motion. The issue has been common knowledge for several months now. It seems like it will be another few months before a convincing repair procedure is devised. I only let Nikon have one go at repairing my D800 and then I got a refund. I think they should be exchanging defective D800s for good ones as a matter of course.

Gabriele Sartori Veteran Member • Posts: 4,422
Are you surprised?

It's called NIH (Not Invented Here) a very popular sport here in the US of boycotting anything smart that comes from other places

em_dee_aitch wrote:

Melville office have disclaimed the reports about the European procedure being disseminated globally. I really don't know who to believe, and really it doesn't matter. But looking at this procedure, I feel like my camera got the hardware part but not the software part, because clearly the general range of my results improved, while the AF fine tune values across the bracket still vary too much to be considered fully fixed.

-- hide signature --

Regards
Gabriele
California, CA

 Gabriele Sartori's gear list:Gabriele Sartori's gear list
Leica D-Lux (Typ 109) Nikon 1 J1 Nikon D800E Panasonic Lumix DMC-GH3 Leica M Typ 240 +3 more
GMack Senior Member • Posts: 2,928
Re: D800 AF Dutch repair procedure

Marc Heijligers wrote:

GMack wrote:

I'm somewhat confused about the "reference camera" part. Is it used just to demonstrate "Here is what a good one is like. Check it if you want." or are they dumping that camera's data to the customer's body which could account for the deviations in L-R focus errors?

The reason for the reference camera is to calibrate the measurement setup.

If you put a camera in place in a measurement setup, it is not exactly at 1m and not 100% aligned to the card holding the vertical stripes. Furthermore, the test is carried out with a reference 50mm/1.4D lens (at f2.2 if I understood correctly), that may also have variations, so those also need to be included and recorded.

So essentially a camera that is adjusted is made equal to the reference camera as much as possible. This means that the repair is at most as good as the reference camera will be.

I get the "Reference Lens" part since they use it on your body as it generally comes into service without a lens attached.

However, I suspect what the "Reference Body" is for is to use your lens on their reference body allowing them to calibrate your lens, or at least check it. Sort of the opposite to the "Reference Lens" scenario above.

A test setup should not require using a reference body to set it up, and then switch it out for a customer's body. Distances are all preset on a test bench if they use a reference lens.

From the write up, it seems this is still going through the service mill as to how to fix this L-R issue. I doubt if it will be solved 100% though as they cannot get a zoom lens to work perfectly over the zoom range as it is. They won't spend the time to do it all according to "The Book" anymore than your mechanic does everything 100% in their manufacturer's service procedures. Maybe the tech will do it to their own body if they own one, but that's about it.

I seriously doubt if they will test ALL 51 sensor points and adjust them individually either, unless they were to set up similar to FoCal's Multi-Point Sensor Test and let it run through each sensor one-by-one and then adjust the phasing (+ or-) for each which would be a major undertaking, unless they could automate it somehow. However, their post-repair QC check seems to deviate all over the board too which is suspect.

Might still be some time before all service centers and their techs get up to speed on this stuff.

Mack

Keyboard shortcuts:
FForum MMy threads