Dust.

TGphoto

TGphoto

Messages
3,398
Name
Tom
Edit My Images
Yes
One of my D750’s is displaying dust in images. Quite a few specs, all round and tiny. Mostly to the top and some to the right of the frame.

Only usually visible in long exposures and shots above f11 but very annoying and lots of them.

Have looked at the sensor and used a rocket blower but to me the sensor looks spotless. Any ideas as to anywhere else this could be coming from or should I get the sensor wet cleaned anyway?

Definitely the body, same results with several lenses.

Thanks
 
Last edited:
Get a kit and do it yourself, I was paying around £35 for a clean every few months as you get lots of dust near aircraft, I now buy the Dust patrol kits and do it myself for about £8 a clean
 
I agree, do t your self and save loadsamoney
 
As above - DIY it. I have just done my D810 with a kit off Amazon (swabs and fluid - £18). Took around 5 swabs to clean it, but it is spotless now. Thinking about it is more stressful than actually doing it ! Just take your time, ensure the battery is on full charge (so the mirror stays locked up!) and use a clean swab after every 'back and forth' sweep.
 
Buy yourself a sensor loupe so you can see the dust on the sensor more clearly. Don't forget that what is top right in your photos is actually bottom left on the sensor.

Otherwise as others have said take your time and don't overload the swab with fluid. It is a lot easier than you might think.
 
Canon Bob said:
Not quite......top right is bottom right.
Click to expand...

It is 55 years since I did physics at school so I am perfectly willing to accept my memory may be faulty. I seem to recall my physics teacher demonstrating that light entering a lens from the top is inverted when it reaches the focal point behind the lens and is at the bottom. Equally light entering the lens from the left is horizontally inverted to the right. if you watch the video below it shows that an image when seen through a lens is completely reversed not just flipped vertically. Almost the last thing the presenter says is that this applies to digital camera lenses as well.

No doubt we have some physics experts out there who can give a definitive answer. I for one would appreciate clarification.

Edit

It would help if I included the link

View: https://www.youtube.com/watch?v=7D2QYJBzrh4
 
Last edited:
Grumps1974 said:
......Equally light entering the lens from the left is horizontally inverted to the right.
Click to expand...

That would be true if you were looking through the sensor (as per the image) and not at it. By 'walking around' to the face of the sensor then you've cancelled out the left-right swap.

Bob
 
Thanks for all the advice guys. Thought that’s what you might say.

Sensor cleaning loupe may be a good idea - these particcles of dust must be minuscule but are obviously being amplified in images.

Have cleaned sensors before so shouldn’t be too much of an issue!
 
Canon Bob said:
That would be true if you were looking through the sensor (as per the image) and not at it. By 'walking around' to the face of the sensor then you've cancelled out the left-right swap.

Bob
Click to expand...

This is now making my brain hurt! The sensor is in a fixed position in relation to the lens. I cannot get my head around how "walking around" can negate the laws of physics. I am going to stop thinking about it now before I blow a gasket.:facepalm:

TGphoto said:
Thanks for all the advice guys. Thought that’s what you might say.

Sensor cleaning loupe may be a good idea - these particcles of dust must be minuscule but are obviously being amplified in images.

Have cleaned sensors before so shouldn’t be too much of an issue!
Click to expand...

Good luck I am sure all will be fine.
 
Grumps1974 said:
This is now making my brain hurt! The sensor is in a fixed position in relation to the lens. I cannot get my head around how "walking around" can negate the laws of physics. I am going to stop thinking about it now before I blow a gasket.:facepalm:

Good luck I am sure all will be fine.
Click to expand...

I think the confusion is perhaps that as photographers we are used to looking 'through' images from behind, on a ground glass focusing screen or with a film transparency for projection etc. That's not the case with the image on a sensor when viewed from the front through the lens mount.

Bob is correct. The image is simply upside-down, eg a dust mark on the top-right of the image is on the bottom-right of the sensor :)
 
Last edited:
HoppyUK said:
I think the confusion is perhaps that as photographers we are used to looking 'through' images from behind, on a ground glass focusing screen or with a film transparency for projection etc. That's not the case with the image on a sensor when viewed from the front through the lens mount.

Bob is correct. The image is simply upside-down, eg a dust mark on the top-right of the image is on the bottom-right of the sensor :)
Click to expand...

Whilst you and CanonBob have this view I still cannot see how a camera does not comply with the laws of physics. With any lens light at the top of the "image" (e.g the sky) goes through the lens to the bottom of the sensor. Light rays at the bottom of the "image (e.g the ground) goes through the lens to the top of the sensor. To me it makes absolutely no sense that light from the sides do not also follow this law of physics. Walking round the lens or viewing from the front through the lens mount does not change this. The sensor at all time remains in a fixed plane to the lens. The image is corrected in camera for the eye to see either by the viewfinder prism or the camera's software (liveview and playback).

Now you both may be right but until I read a scientific explanation to support your views I am still firmly of the view that the entire image is inverted on the sensor.

I am not the only person who has this view - you might like to read the attached link which contains a reply to this question from a researcher with Intel.

https://www.quora.com/Are-photos-taken-by-a-camera-inverted-on-the-film-or-sensor-chip
 
As you shoot your subject, Martin, then it's correct that its left side is projected onto the right side of your sensor (and the top of the subject to the bottom). You then turn your camera around to look at the sensor and what was the right side when looking from behind is now the left side as you peer into the mirror box.
 
  • Like
Reactions: Nod
I understand what you are saying but the sensor is fixed in the camera. So what is recorded on the sensor stays static. You see the surface of the sensor as the lens “would see” the sensor. The fact that you move from behind the sensor to in front of it does change the sensor orientation with regards to the subject or the lens.

Tomorrow I will check my sensors with my loupe to see if I can see any dust. If there is I will then take some photos and see where this dust appears. This is the only way I will resolve this in my own mind.
 
Grumps1974 said:
Whilst you and CanonBob have this view I still cannot see how a camera does not comply with the laws of physics. With any lens light at the top of the "image" (e.g the sky) goes through the lens to the bottom of the sensor. Light rays at the bottom of the "image (e.g the ground) goes through the lens to the top of the sensor. To me it makes absolutely no sense that light from the sides do not also follow this law of physics. Walking round the lens or viewing from the front through the lens mount does not change this. The sensor at all time remains in a fixed plane to the lens. The image is corrected in camera for the eye to see either by the viewfinder prism or the camera's software (liveview and playback).

Now you both may be right but until I read a scientific explanation to support your views I am still firmly of the view that the entire image is inverted on the sensor.

I am not the only person who has this view - you might like to read the attached link which contains a reply to this question from a researcher with Intel.

https://www.quora.com/Are-photos-taken-by-a-camera-inverted-on-the-film-or-sensor-chip
Click to expand...
Canon Bob said:
As you shoot your subject, Martin, then it's correct that its left side is projected onto the right side of your sensor (and the top of the subject to the bottom). You then turn your camera around to look at the sensor and what was the right side when looking from behind is now the left side as you peer into the mirror box.
Click to expand...
Grumps1974 said:
I understand what you are saying but the sensor is fixed in the camera. So what is recorded on the sensor stays static. You see the surface of the sensor as the lens “would see” the sensor. The fact that you move from behind the sensor to in front of it does change the sensor orientation with regards to the subject or the lens.

Tomorrow I will check my sensors with my loupe to see if I can see any dust. If there is I will then take some photos and see where this dust appears. This is the only way I will resolve this in my own mind.
Click to expand...
You are not actually turning the sensor around, you are not looking at the sensor at all - you are looking at the image on a screen, corrected for both left-right and top-bottom anomalies. If you were looking at a film negative, Grumps would be right. If you were looking at a positive print made from that negative, Grumps would be wrong.
 
Grumps1974 said:
Whilst you and CanonBob have this view I still cannot see how a camera does not comply with the laws of physics. With any lens light at the top of the "image" (e.g the sky) goes through the lens to the bottom of the sensor. Light rays at the bottom of the "image (e.g the ground) goes through the lens to the top of the sensor. To me it makes absolutely no sense that light from the sides do not also follow this law of physics. Walking round the lens or viewing from the front through the lens mount does not change this. The sensor at all time remains in a fixed plane to the lens. The image is corrected in camera for the eye to see either by the viewfinder prism or the camera's software (liveview and playback).

Now you both may be right but until I read a scientific explanation to support your views I am still firmly of the view that the entire image is inverted on the sensor.

I am not the only person who has this view - you might like to read the attached link which contains a reply to this question from a researcher with Intel.

https://www.quora.com/Are-photos-taken-by-a-camera-inverted-on-the-film-or-sensor-chip
Click to expand...

No disagreement with that link. Yes, the image is inverted - but in the dust-on-sensor scenario it is not flipped left/right (as it often is in other situations). A couple of simple tests:

Take a camera lens and hold it close to a wall opposite a window with a pot plant on the windowsill bottom right, move the lens back and forth so the image of the window is clearly focused on the wall. The plant will appear in the top right, just as a speck of dust on the sensor would.

Then take a sheet of plain copy paper and make a mark in the bottom right-hand corner. Now turn the paper around and look through the back - the mark is now on the bottom-left. And this is how we view film images - through the back - but it is not how we view dust on the sensor through the lens mount from the front ;)
 
john.margetts said:
You are not actually turning the sensor around, you are not looking at the sensor at all - you are looking at the image on a screen, corrected for both left-right and top-bottom anomalies. If you were looking at a film negative, Grumps would be right. If you were looking at a positive print made from that negative, Grumps would be wrong.
Click to expand...

I am not sure I understand this post as if you are cleaning the sensor then of course you are looking at the sensor! This post is all about cleaning sensors. As I said in a previous post the camera corrects the inverted image so anything shown on the back of the camera in the viewfinder or LCD screen is shown correctly.

HoppyUK said:
No disagreement with that link. Yes, the image is inverted - but in the dust-on-sensor scenario it is not flipped left/right (as it often is in other situations). A couple of simple tests:

Take a camera lens and hold it close to a wall opposite a window with a pot plant on the windowsill bottom right, move the lens back and forth so the image of the window is clearly focused on the wall. The plant will appear in the top right, just as a speck of dust on the sensor would.

Then take a sheet of plain copy paper and make a mark in the bottom right-hand corner. Now turn the paper around and look through the back - the mark is now on the bottom-left. And this is how we view film images - through the back - but it is not how we view dust on the sensor through the lens mount from the front ;)
Click to expand...

It must be my age but I cannot see how this gives any sort of scientific explanation of what is said by others to be occurring. I have seen nothing in these posts that in anyway explains to me why when you look at a sensor the image is inverted only vertically and not horizontally.

As demonstrated by the quote in the Quora link above the image is captured on the sensor both vertically and horizontally inverted so top right is captured bottom left of sensor. In my view this means that If you have dust on the bottom left of the sensor it would show top right in the image when viewed on a PC. Ergo if you see dust top right on the screened image you look bottom left on the sensor for this dust.

I think I will leave it there. I am still none the wiser and at my age I have better things to do with my time than worry about dust on sensors.:confused:

Thanks for all the contributions. I now need to go and lay down in a darkened room. My brain hurts!:)
 
Last edited:
Final post from me. I have just had a look at both my camera sensors through a loupe but no dust (thankfully)! So unfortunately I am still none the wiser.
 
Grumps1974 said:
Final post from me. I have just had a look at both my camera sensors through a loupe but no dust (thankfully)! So unfortunately I am still none the wiser.
Click to expand...

You won't see anything with a regular loupe - it has to be a high power magnifier with LED illumination specially designed for the purpose.

I think the confusion is because of our experience with film where viewing is always through the back, but that's not how we look at a sensor for cleaning from the front. And when you turn things around like that, right becomes left and left becomes right. The image is still inverted though.

Do those simple checks outlined in post #17. Never mind the science or old habits, just see what happens. There is no mirror or pentaprism or electronic jiggery-pokery to confuse things, just lens, image, sensor - what you see is what you get :)
 
Last attempt Martin. You're admiring the view through your lounge window when, splat.....a bird craps all down the lefthand side. You go out to clean it off and, bugger me, it's on the righthand side now.
 
Canon Bob said:
Last attempt Martin. You're admiring the view through your lounge window when, splat.....a bird craps all down the lefthand side. You go out to clean it off and, bugger me, it's on the righthand side now.
Click to expand...

I said I would not post anymore but once more I still consider this argument to be irrelevant. I am not talking about what you see from behind the camera (or window in your analogy) or the fact it is reversed vertically when you go to the other side. The sensor does not move - it stays looking forward through the lens and therefore has a fixed perspective. What appears to the sensor is a horizontally and vertically inverted image. When you take the lens off the camera you see the sensor in exactly the same way as if you were looking through the lens. If you could see the image on the lens you would see it is inverted across both axis. So dust on the bottom left of the sensor would show up on the top right of the on screen image.

This is what my physics master taught me many many years ago. I cannot see any reason why this should not still apply to camera lenses and until I see physical evidence for myself that this is wrong it is what I choose to continue to believe.

Thanks for trying to explain your point of view to me. I guess we will have to agree to differ.:)
 
You must log in or register to reply here.
Back
Top