Can sensor depth can shift in a DSLR?

Raymond Lin

I am Groot
Messages
10,030
Name
Raymond
Edit My Images
No
So the little spring/lock that holds the SD card in place broke and I sent it in for repair, just got the invoice back. What is interesting isn't that they replace the mainboard...but the comment regarding the sensor.

Questions springs to mind.

1 - Sensor depth sifting? Isn't it fixed?
2 - I have 2 bodies, how does matching the camera to the lenses help with the other body?

u0udrxu.png
 
Probably questions you should be asking them? Thinking out loud, perhaps if you send both bodies in they'll calibrate the lenses to 'dead on' factory spec and then match the sensors on both camera bodies to them? Once again, a question for the person who wrote the comments on the invoice.
 
And sensor depth is what exactly? Are we talking here about actual physical movement of the sensor?
 
I remember reading a comment from Max Lyons author of PTAssembler and other programs. who stated that when he complained about the repeat accuracy of canon focusing and returned his lenses for adjusting, he was informed that they were within the specified tolerances, which are based on the depth of field at a given aperture.
so supposedly the smaller the aperture the wider the tolerance. which is hardly critical focusing.
 
My guess is the sensor has to be at a set position, within certain tolerances which is presumably specified as "depth" (from lens flange perhaps?) and yours has shifted from specified position, however because you have micro adjusted your lens(es)* and you are happy with the focusing if they set it back to in-spec position your lens(es) would no longer focus as it had and you'd complain, understandably, so they have given you some advice and a "warning".

* I assume lens micro adjustment is a s/w thing that the camera stores as opposed to some physical change to lens or sensor position, so it can store several different settings to cover different lenses.

I think when Canon match your body/lens combo they adjust the lens slightly, so it focuses correctly with a particular body, however they will use one of their stock lenses to ensure the camera focuses within spec if you dont send in your lens, which rather suggests there is "some" sensor position adjustment possible, of course if you use a lens calibrated to a particular body it may no longer be in-focus on the other body.
 
I think I read in one of the camera tear downs of the Nikon Z7 or Sony a7III that the sensors have either shims or adjustment screws to adjust the sensor to the exact focal plane. I would assume that all cameras have something similar. If it hasn't been done well in manufacture, or something has been knocked or moved through wear and tear, it may need adjustment for those that are aware of the inaccuracy because of the use of wide aperture lenses, which others using slower lenses may not see, but unknowingly may be in a similar situation. Just an thought. :)
 
To me I would assume that the sensor board is attached to the main pcb, which they have replaced, therefore the sensor will not be in the exact same position.
You will need to adjust your micro focus setting possibly.

Pete
 
To me I would assume that the sensor board is attached to the main pcb, which they have replaced, therefore the sensor will not be in the exact same position.
You will need to adjust your micro focus setting possibly.
That's what I thought when I read it, too.
 
To me I would assume that the sensor board is attached to the main pcb, which they have replaced, therefore the sensor will not be in the exact same position.
You will need to adjust your micro focus setting possibly.

Pete

Make sense....that's a job for the weekend then !
 
Make sense....that's a job for the weekend then !
Despite Canon saying better to send camera and lens for recalibrating, notwithstanding you have 2 bodies. Would it not be better to send off both bodies and the lenses you use and get the best settings for one body then micro adjust the other one. Tedious (and possibly expensive) I know but for optimum AF that's what Canon are recommending rather than a DIY approach?
 
Despite Canon saying better to send camera and lens for recalibrating, notwithstanding you have 2 bodies. Would it not be better to send off both bodies and the lenses you use and get the best settings for one body then micro adjust the other one. Tedious (and possibly expensive) I know but for optimum AF that's what Canon are recommending rather than a DIY approach?

But how would sending in the body and lenses to be calibrated actually work?

Think of it like this.

Every camera and lens as a tolerance, lets say like the settings, if 0 is absolute accuracy and -20 to + 20 is the range, the sensor can be anywhere in between, same as the lens.

So a -10 sensor and a + 10 lens will require ZERO calibration.

If the second body has a -5 sensor then you then micro adjust this camera to -5, taking it to -10 total.

Bearing in mind, there is an acceptable tolerance, they are not going to be able to make everything at zero, so even if they calibrate body 1 to lenses, body 2 will need to be micro adjust.
 
Last edited:
It sounds to me like they did not calibrate the sensor (update sensor depth) in an attempt to retain the accuracy of the AFMA settings... which seems dumb/lazy.
The other comment about calibration/matching being better than AFMA due to the fact that AFMA is only optimal for one subject distance doesn't really make sense to me... I don't see how factory calibration is/can be any different in that aspect.
 
Last edited:
It sounds to me like they did not calibrate the new sensor (update sensor depth) in an attempt to retain the accuracy of the AFMA settings... which seems dumb/lazy.
The other comment about calibration/matching being better than AFMA due to the fact that AFMA is only optimal for one subject distance doesn't really make sense to me... I don't see how factory calibration is/can be any different in that aspect.

If it requires MA again they might as well wiped the settings clean, but I suspect that also includes all my buttons assignments so its best to leave that alone.
 
If it requires MA again they might as well wiped the settings clean, but I suspect that also includes all my buttons assignments so its best to leave that alone.
It shouldn't. The sensor is not mounted to the main board on the 5D nor most (any?) DSLR's. Although the sensor does have it's own sub circuit/board... that should be noted as sensor replacement.

Sounds to me like that second sentence should read "calibration checked and sensor cleaned." It also sounds like they probably would have adjusted the calibration if there hadn't been AFMA settings... I don't think it's uncommon to update/refine the calibration beyond manufacture tolerance upon individual inspection/service.
 
Last edited:
The other comment about calibration/matching being better than AFMA due to the fact that AFMA is only optimal for one subject distance doesn't really make sense to me... I don't see how factory calibration is/can be any different in that aspect.
It makes sense to me, I think. I'll explain.

When you're using Canon lenses on a Canon body, or Nikon lenses on a Nikon body, the camera can recognise the lens. So when you want to do AFMA, the settings get stored in the camera. That way, if you have more than one camera, each one can maintain a full set of AFMA settings for all of your lenses. Happy days.

If you're using a modern Sigma lens, there is still the capability to fine-tune the autofocus, but this time the camera can't or won't recognise the lens, so the settings can't be stored in the camera. Therefore the settings have to be stored in the lens, and you use the Sigma software and Sigma USB dock to do it. (The situation is the same for a few of the newest Tamron lenses too, I believe.) But here's the interesting bit. The Sigma software allows you to store multiple different AF fine tune settings, which apply at different focus distances. For example here's a screen shot with a 135mm Art lens connected, and you can see that it allows 4 different adjustments to be stored: at distances of, roughly, 0.9m, 1.3m, 2.2m, and infinity. Obviously the settings can't be stored in the camera, so they are stored in the lens.

upload_2018-11-9_17-28-41.png

Now, the comment from the Canon Service Centre suggests very strongly to me that it's also possible to calibrate Canon lenses at multiple different distances, and presumably those settings would be stored in the camera.

Do you think I'm right?

If so, why don't Canon allow users to do this via the camera and/or lens firmware, as Sigma do?
 
Raymundo stop resisting you KNOW your secretly hankering to go mirrorless , don’t keep bottling up those deep inner feelings come out of the closet and lose them shifting mirrors :banana::banana::banana::banana::banana::beer:
 
If you're using a modern Sigma lens, there is still the capability to fine-tune the autofocus, but this time the camera can't or won't recognise the lens, so the settings can't be stored in the camera.
On Nikon bodies Sigma/Tamron/etc are recognized and AFMA adjustments stored in the camera (as well as recording lens info and AFMA setting in the exif)... I thought other cameras do as well?

It is quite possible that factory matching/calibration is able to store changes/settings in the lens like the Sigmas... and if it is for multiple distances/etc I can see that being a significant benefit over AFMA for use with the body it is matched to. But I can also see that being a potentially significant detriment for use on bodies it's not matched to.
Either way, that would be separate from calibration of the sensor position w/in the DOFocus, which is what was being referred to (I think).
 
Last edited:
I meant a canon one :clap:
 
But how would sending in the body and lenses to be calibrated actually work?

Think of it like this.

Every camera and lens as a tolerance, lets say like the settings, if 0 is absolute accuracy and -20 to + 20 is the range, the sensor can be anywhere in between, same as the lens.

So a -10 sensor and a + 10 lens will require ZERO calibration.

If the second body has a -5 sensor then you then micro adjust this camera to -5, taking it to -10 total.

Bearing in mind, there is an acceptable tolerance, they are not going to be able to make everything at zero, so even if they calibrate body 1 to lenses, body 2 will need to be micro adjust.
As you say there are tolerances in both body and lens, but afaik the point at which the lens focuses can be adjusted slightly and presumably that is constant across all apertures. So when your sensor was new and you adjusted focus of your lens using ma you got it spot on, probably because the sensor was within the adjustment you could perform on that lens. Now assuming the sensor has somehow shifted, as stated by Canon, even if by a tiny amount it might mean that you can no longer get acceptable focus using ma, hence them suggesting if you have a problem then sending back lens and body would be a better option. Of course as you recognise camera to body adjustment for "perfect" focus might only work for one body, the other body might be at the opposite end of tolerance so you would have to live with that and ma might not be able to compensate enough and maybe you would have to send both bodies and lenses back so they could get best settings for both that would require ma on both, one going positive and one negative to get acceptable focus.
 
I would get some clarification to what they meant by what they wrote on the invoice. :thinking: I have read their note to be saying the sensor may have shifted, but you have lenses micro adjusted, so they didn't align it because of that. I would have wanted them to get the sensor set right, or at least to have the option to do it or not before it came back, in the knowledge that I may have to re-micro adjust my lenses.

You can see here (near the end) in a tear down of a 5DIV that the sensor has three screws "to level the sensor so that it’s in exactly the same plane as the lens mount." If the sensor were to be adjusted to be perfect then any misalignment would be with the lens(es) rather than the possibly the sensor and the lenses. In theory the lenses may need no micro adjustment, or less anyway, because they would be on a camera with the sensor in the position the lenses were designed for.
 
That teardown also shows that the sensor is not part of the main circuit board; and replacing the mainboard should not change/require calibration...
Mmm:thinking: The invoice says "Sensor depth has not been updated despite it shifting" which to me sounds like they knew it was not in the correct position, and like I said previously, rather than get in touch with the customer and give them the option to correct it or not, sent it back because there were micro adjustments set for lenses. If the customer (in this case Raymond) did now want to have the sensor shift corrected it, would be another trip back to the service department and a lot more time without a camera.

Imho, and feel free to disagree, :) but with such vital aspect of the camera, the precision placement of the sensor, if they found it was not right while doing another repair, they should have got in touch and asked whether the customer wants it corrected while they already have it with them, and you would presume, the camera already disassembled, even if there would be an extra cost involved. It may have turned out to be cheaper than now sending a camera back (if that is what the customer wanted, not just speaking about Raymond now) because now it would have to be taken apart again potentially adding to the labour charges for the repair, but definitely leading to more time without the camera.
 
On Nikon cameras on the top of the body is a mark of a circle with a line through it showing where the sensor is in the body. Now when they mentioned sensor depth I was wondering it that referred to the distance from the sensor to the back of lens which might have shifted (an angle) a bit? I don't know, just a thought
 
Last edited:
On Nikon cameras on the top of the body is a mark of a circle with a line through it showing where the sensor is in the body. Now when they mentioned sensor depth I was wondering it that referred to the distance from the sensor to the back of lens which might have shifted (an angle) a bit? I don't know, just a thought
I think most cameras, at least above compact cameras, have the same focal length mark, not just Nikon. It allows people to precisely measure to that point for certain types of photography.
 
The mark is a carry over from film days, it showed where the pressure plate pressing against the film was but I doubt if it's super accurate, but probably ok for rough alignment these days.
 
Imho, and feel free to disagree, :) but with such vital aspect of the camera, the precision placement of the sensor, if they found it was not right while doing another repair, they should have got in touch and asked whether the customer wants it corrected
If they found it to be out of tolerance, then they should have just fixed it... anything else is just lazy/ignorant IMO.
 
Well, it seems every settings has been wiped from the camera, including buttons, custom menus etc...so need to start from scratch.
 
LOL! So much for preserving your AFMA adjustments...

I would interpret it as meaning that the user has performed AFMA and has functional values (therefore the sensor wasn't moved) rather than the values being retained in camera.
 
Probably questions you should be asking them? Thinking out loud, perhaps if you send both bodies in they'll calibrate the lenses to 'dead on' factory spec and then match the sensors on both camera bodies to them? Once again, a question for the person who wrote the comments on the invoice.

First reply is still the best! Not sure why the OP isn't doing this?
 
Back
Top