Dirty Pictures

[bvf]

Does anyone here have a technique to overcome the tyranny of dust on the sensor? It drives me crazy when using focus stacking and/or macro bellows, when the effect of dust becomes very noticeable.

 

I'm following all the advice I've found:

• Point the camera downwards when changing lenses
  
• Use a filtered hand-held puffer bulb to clear dust from sensor and the rear of the lens.
  
• Use appropriate-size swabs to wipe across the sensor (that always makes me nervous), with the occasional single drop of sensor cleaning fluid.
  
• I've recently acquired one of those little motorised brushes which spin the brush head to give the fibres an electromagnetic charge so that when you drag them (not rotating now) across the sensor the charge attracts the dust. (The instructions don't tell you how long you need to spin the brush to charge the fibres or, afterwards, to centrifuge any dust off them.)
  

Any suggestions on technique or tools?

[colinbm]

"Use appropriate-size swabs to wipe across the sensor (that always makes me nervous), with the occasional single drop of sensor cleaning fluid."

 

Nothing to be nervous about Bernard, just like cleaning any other piece of glass that you have ever cleaned before.

Just follow your cameras instructions....

[bvf]

"Use appropriate-size swabs to wipe across the sensor (that always makes me nervous), with the occasional single drop of sensor cleaning fluid."

 

Nothing to be nervous about Bernard, just like cleaning any other piece of glass that you have ever cleaned before.

Just follow your cameras instructions....

 

My nervousness is because of various warnings I've seen about potential damage to the sensor, perhaps through too much pressure.

 

BTW - I'd be happy to pay for professional cleaning from time to time - did that once and it did the job. But the local photo shop that provided the service has closed down, and I don't want to send my cameras off for a few days to be cleaned.

[colinbm]

Bernard you are not digging a hold in the back yard....

Just like cleaning any other piece of glass that you have ever cleaned before.

The sensor glasses I have tried to remove were like using a jack hammer, they are tough on purpose.

You know the story of the 'Big' UV glass con, that you 'Must' put on your lens....well the 'Big' sensor glass cleaning con is another, just to get you in the shop spending....

[dabateman]

Bernard before you take dirty photos you should consult with your wife. Also don't know if you should post them here.

 

But to clean a sensor there are two good ways. I have only done number one so far.

 

1. Most common not hard is buy some 99.9% isopropyl alcohol. It used to be cheap and easily available, but now maybe harder to find. But it must be at least this pure. Add small amount to clean container, then with genuine Q-tip, wet it slightly, dab off excess IPA on the container side. Then move in consistent pattern to clean the sensor. The IPA will rapidly evaporate. Check in light to ensure clean streak free sensor. May need touch up if held too long at start or end allowing for pooling.

 

2. Use Scott tape 811 to clean the sensor following this protocol:

https://www.dpreview.com/forums/post/16542910

 

I haven't tried method 2 yet. But since I once worked for 3M, I know their products and would try as it does make sense.

 

Of course before you start cleaning use either a rocket blower or a household vacuum cleaner to pull off the easiest to remove dust/dirt. Never make contact with the sensor with either the vacuum or rocket blower. Never use an air can.

[bvf]

Thanks, Dave.

 

I've actually got a litre of isopropanol, so I'll give that method a try. I don't know what's special about genuine Q-tips as opposed to other brands of cotton buds, and it's not a brand that I recognise, but there are a few suppliers on Amazon UK so I'll get some of those.

 

I'd like to see some other testimonials about the Scotch Tape 810 & 811 before I give that a try.

 

BTW, talking about 3M - the first time I saw that as a brand name was about 50 years ago when I used to process Ferraniacolor CR50 colour slide film - couldn't afford Ektachrome or Kodachrome! The Ferrania packaging said it was a 3M company.

[otoien]

I had a bad experience way back cleaning the sensor of my D40x with those pre-moistened sensor swabs from Photographic Solutions inc. They left a glue like residue on the sensor, and in the effort to remove it the sensor became scratched, or perhaps the glue would not come off. They would not honor their guarantee against sensor damage, and it took a lot of quarrel plus intervention from Henry Posner at B&H before they even refunded the swabs I bought. I think what happened was that the generous amount of isopropanol on those swabs dissolved the rubber band that held the pec pads in place on the swabs, so it really was rubber glue that was deposited on the sensor. Luckily my D200 fared better as the other version of the swabs (with methanol) was so dried out that it really became dry cleaning. Never again with those pre-moistened swabs, and I have not ventured into wet cleaning since - probably should - my D7100 is in dire need, too much dust for the sensor pen.

 

As it comes this mishap contributed to bringing me into photography in the invisible spectrum, as I "saved" my D40x by having it converted to IR, and I am still using it as my only IR body.

[colinbm]

Peeps, you are never cleaning the sensor, it has a clear glass window over it.

Usually you are cleaning the UV / IR cut filter in front of that again, if you are using a un-modified camera.

If the camera has been modified, it is the clear fused silica replacement that you are cleaning.

You are just cleaning a piece of glass.

[dabateman]

Colin,

Most cases your cleaning a durable glass like dust shaker cover on modern camera.

Older cameras you are cleaning a slightly sensitive BG type glass or Dichroic UV/IR block glass on sensor.

But for a converted camera you maybe cleaning bare sensor. Depends whom did the conversion. Or very hard durable fused silica, or plastic IR filter, or ete.

 

Can't assume.

 

What I have found with fused silica is that its very scratch resistant. But not to chipping. The glas chips super easily. That a problem poping glass elements in and out of holders. But once in is really durable.

 

Bernard,

Only hear say, not supported. But why you want genuine Q-tip is the quality of the cotton is better. Softer and lest risk of bleaching or other chemicals present.

But I have no idea if actually true. I haven't analyzed various brands.

[Alaun]

The problem with Q-tips is, they loose fibres. You clean a surface and later it looks worse. Better is special paper for optical cleaning. This is often sold in small quantities at high prices (like on the swabs). If you buy a buch of 1000 or so it becomes a few cent per piece. A couple of years ago I was upset about the price of "cleaning sets", so I looked for that stuff. You can find it on eBay from japanese companies. I paid almost the same for shipping as for the paper, but still, on a long term, it is far far more cheap.

[dabateman]

My genuine Q-tips are very tightly wound.

I have never seen a fiber issue.

[OlDoinyo]

For prevention, I suggest

 

-Leaving the camera exposed (no lens or body cap) for as short a period as possible;

 

-Pointing the camera downwind when changing lenses;

 

-Stepping out of the wind if possible when doing the same.

 

If the dust gets on the sensor, use the built-in vibrator (if functional) to clear what you can; otherwise, open the shutter or guard curtain (depending on camera design,) put the camera on a box at appropriate height, get one of those cans of foof gas that is no more than 1/3 full (so it cannot spray liquid,) hold the can (without extension tube) so the nozzle sits at mount center height with the can resting on the counter, angle slightly from one side, and direct a few short, sharp blasts of gas at the sensor from either side with the nozzle just outside the mount. I find this will clear almost everything.

 

Caution: many pundits advise against doing the above and would be horrified at this suggestion. This is because if the can is too full, or sloshing inside, or if the trigger is depressed too long so internal contents boil vigorously, the can will emit a spray of liquid that contains dissolved impurities. If droplets hit the sensor and then evaporate, you will end up with a much worse mess than what you had previously! I have so far avoided this fate by observing the protective measures noted. A few cameras may have internal shutter mechanisms so delicate that they can be damaged with overzealous gas blasting. This is more of a problem with older SLR designs, but know your camera and don't blast more than needed.

 

Once I had a particle that stuck fast to the sensor cover glass so that gas would not budge it. I was able to poke it loose carefully with the end of a soft brush.

[bvf]

Thanks, OIDonyio.

 

I don't think I've got the courage to try the apreay can (although I've got a couple of them).

 

I'm just doing some tests of the methods I'm prepared to try. So far my attempts at cleaning (using filtered puffer, swabs, rotating anti-static brush, Q-tip with isoporpanol) have all made the sensor more dirty! (As someone said on another forum, all these methods have been used effectively, so the problem is with the technique.) Just waiting for some Scotch 811 tape to see if I can get that to work.

[ulf]

One very important thing when cleaning the sensor is to do it in a really clean environment, with as little dust present as possible.

Avoid dusty environments with fabrics.

[bostwickenator]

LensPen SensorKlear II. It is absolute magic. Will wipe anything off your sensor and leaves no residue.

[bvf]

LensPen SensorKlear II. It's absolute magic. Will wipe anything off your sensor and leaves no residue.

 

Thanks - I've ordered one of those pens now.

[bostwickenator]

Nice I've had mine for about 12 years now and it's still going well. Probably should buy a second one as a backup one of these days.

[otoien]

LensPen SensorKlear II i the one I am using too. Works well if the sensor is not too dirty in combination with a filtered blower.

[bvf]

Right, I've tried out all the techniques I could find (apart from the LensPen device - as I understand it, this is to pick out individual bits of dust rather than clean the whole of the sensor, and you need a loupe to use it which I didn't have - but got one on order).

 

My conclusion is that all the techniques work to some extent. Except the 811 tape method - this left a huge amount of crud on the sensor, which fortunately came off after using Q-tip + Isopropanol and wet swab. My advice is do not use Scotch 810/811 tape to remove dust from your sensor.

 

However, none of the techniques got anywhere near to achieving what I needed. But what I've realised is that I was probably trying to achieve the impossible - to get the sensor clean enough for "extreme" macro work (say 4x magnification) when using focus stacking. I can get the sensor clean enough for general photography and for magnifications up to about 2:1, but at greater magnifications the problems start appearing.

 

The images below illustrate this. All were taken during the same session, so with the same level of dirt on the sensor. All at an indicated aperture of f/8. All focus stacked.

 

The first image is using an El Nik 50mm at max. bellows extension + a 4cm extension tube. So a magnification of around 4x. The wiggly lines are dust spots which have been stretched out as a result of focus stacking.

 

The second image is using an El Nik 105mm at max. bellows extension + a 4cm extension tube. This must be around 2x magnification. No sign of dust! The absolute size of the aperture would have been larger than for the 50mm lens because of the longer focal length, and this would have helped reduce the effect of dust. (Incidentally, I also tried an El Nik 80mm lens, and this gave results in between - there were dust marks but less proniunced than with the 50mm lens.)

 

The third image is using the El Nik 50mm lens again, but at minimum bellows extension and with no extension tube, to give a magnification similar to that in the 105mm lens image. Again, no dust.

 

 

 

[Andy Perrin]

Bernard, I think you need a pressurized clean room for the level of cleanliness you are after. Second best is probably what you have plus one of those plugins to fix the dirt on each image before stacking.

[bvf]

Second best is probably what you have plus one of those plugins to fix the dirt on each image before stacking.

 

Do you have a plugin in mind? It would have to be able to work on a batch of images rather than just one image at a time.

 

I've tried Canon's Dust Delete Data option, but the software that processes this (Canon Digital Photo Professional) always reports it can't process my macro images. Guess I'll just have to build that clean room .... Actually, I was wondering about trying to build a clean box (you know, like they use in chemistry labs when handling dangerous materials).

[nfoto]

Retouching stacked photomacrographs is an inevitable part of the process. Sorry but that's a fact of life. I've seen much worse than the examples you posted though.

 

If you use, for example, Zerene Stacker, you can do a lot by retouching directly there using for example an individual frame or DMap composite to touch up spots on a Pmax image.

[Andy Perrin]

Bernard, I would just take a photo of a blank surface using the same settings as the rest of the photos, use it to make a mask by thresholding, and use content aware fill on the masked regions of each photo. In Photoshop this could even be automated.

 

I haven’t googled but i bet someone has something like that commercially available somewhere.

[nfoto]

Following this approach, *every* image for stacking needs to be processed with the mask before the stacking commence. Still, there will be imperfections shown by "moving dots". So a lot of work beforehand and still more to do afterwards.

 

I don't know rhe stack "depth" which Bernard applies, but for my own purposes number of images in the stack can often exceed 100.

[bvf]

At the extreme macro end I am limited to a focus stack of about 25 images by the minimum amount I can manually adjust focus or move the whole bellows unit. But I want to create stereo pairs, so that is 50 images. So I need a tool that can remove the dust streaks in the final image or a batch process that removes the dust spots in all the component images.

 

I was intrigued by Birna's comment that "[in] Zerene Stacker, you can do a lot by retouching directly there using for example an individual frame or DMap composite to touch up spots on a Pmax image. " I wasn't aware of that, so I'll have a look at whether that can help me.