SUPPORT US. If mindat.org is important to you, click here to donate to our Fall 2019 fundraiser!
Log InRegister
Home PageAbout MindatThe Mindat ManualHistory of MindatCopyright StatusWho We AreContact UsAdvertise on Mindat
Donate to MindatCorporate SponsorshipSponsor a PageSponsored PagesMindat AdvertisersAdvertise on Mindat
Learning CenterWhat is a mineral?The most common minerals on earthInformation for EducatorsMindat ArticlesThe ElementsBooks & Magazines
Minerals by PropertiesMinerals by ChemistryAdvanced Locality SearchRandom MineralRandom LocalitySearch by minIDLocalities Near MeSearch ArticlesSearch GlossaryMore Search Options
Search For:
Mineral Name:
Locality Name:
Keyword(s):
 
The Mindat ManualAdd a New PhotoRate PhotosLocality Edit ReportCoordinate Completion ReportAdd Glossary Item
Mining CompaniesStatisticsUsersMineral MuseumsMineral Shows & EventsThe Mindat DirectoryDevice Settings
Photo SearchPhoto GalleriesNew Photos TodayNew Photos YesterdayMembers' Photo GalleriesPast Photo of the Day GalleryMineral Photography

Mineral PhotographyDust on camera components

27th Jul 2017 08:29 BSTJoel Dyer

03770530015651844927545.jpg
Hi,


I've been exasperated lately by dust inside my cameras used for macro and photomicrography. I know there is an old thread concerning dust on sensors, but I seem to have some other problem. I am sure than the sensor has been properly cleaned now.


I use Canon 650D and EOS 6D, with automatic sensor cleaning constantly on. At some point, I noticed that I was getting dust spots showing on my 1-3mm FOV pictures in particular. When stacking pictures, this is a real nightmare, as other Mindaters surely know.


The dust is not on any microscope components, because when rotating the camera the dust particles move with the camera. I have tried to clean the cameras now 10 times in the last month, without good enough results :-(


I can see dust on the mirror and the hinged components behind the mirror. When tilting the mirror with protective gloves, I can see dust also between the hinged elements behind the main mirror. The last mentioned fact seems a real disaster to me, because it is nearly impossible to properly clean these small inter-connected surfaces with existing tools. The whole mechanism is scarily fragile, too.


I've tried to clean the sensor and main mirror using:

- blower only

- pre-wetted swab only, single swipe and multiple swipe methods

- pre-wetted swab and then dry swap, single swipe and multiple swipe methods

- dry swab with cleaning solution

- dry swab with cleaning solution + dry swab, with / without blower




I have also changed the setup in my "lab" by removing almost all larger rocks into closed cabinets, re-arraging a lot of items and cleaning the floor and other office surfaces more frequently. I have also repeatedly cleaned the inside and outside of my microscope(s).


I wonder if there are any other methods of cleaning the fragile components in the cameras, or if I'm now doomed to cloning away 20-40 microscopic dust particles after taking every single photo or photo stack..?


Cheers,


Joel

27th Jul 2017 09:42 BSTJolyon Ralph Founder

Having cleaned the sensor directly many times over the years, I have found it is actually relatively easy to clean. I tend to use the following procedure.



a) Make sure there's as little dust in the air as possible. This means keeping windows closed, sitting down in a clean place and waiting for a while for dust to settle down.


b) Take lens off, hold camera with lens mount down, gently tap the camera a few times to release any loose dust and particles.


c) Keeping the camera in that orientation, use the blower to blow inside, but only sparingly. I'm not a great fan of using blowers as I think they just tend to push the dust further into the camera.


c) wait a minute for that dust to settle


d) using either a pre-wetted swab or a dry swab with solution (no big difference I've found) clean the mirror carefully.


e) set the mirror to 'cleaning' mode to lock up the mirror and reveal the sensor.


f) I've had great luck with the semi-sticky pads that you stamp onto the sensor rather than wiping, eg this https://photographylife.com/product/sensor-gel-stick, but I've also found that using the swabs is fine, and sometimes you do need to actually scrub a little bit with it to remove any stubborn greasy particles on the sensor!


g) Put on a good lens, put it to a small aperture eg f/11, defocus it in manual, put on live view and point at a clear white wall, move around and you should see any sensor marks quite easily.


h) repeat until these marks are gone.

27th Jul 2017 09:44 BSTJolyon Ralph Founder

I also use a loupe to help see the marks directly on the sensor.

27th Jul 2017 12:26 BSTPaul Brandes Manager

I've tried the pre-wetted swabs and the air method with mixed results.

For me, it's just easier to have a professional clean my equipment once a year.

27th Jul 2017 12:43 BSTJoel Dyer

Hi Joylon, Paul for the input so far.


I think I'll try to order sensor gel stick in the near future. I agree that using a blower mostly causes more trouble than benefit.


I noticed there is a "reverse pressure can vacuum": I wonder if anyone has any experience of these type of tools?

https://www.bhphotovideo.com/c/product/818263-REG/Green_Clean_GCSV3000_V_3000_Mini_Vacuum_Attachment.html


With such a device perhaps I could get some of the dust removed from the mirror-element-group, so the dust doesn't fall back on the cleaned sensor... And one could suck up any dust around the lense attachment and other inside parts of the camera, if one was careful?


Paul, I think I might end up getting at least my 6D professionally cleaned, if it's not too expensive an operation. I'm interested in hearing if there are any more comments on this tricky issue.


Cheers,


Joel

27th Jul 2017 12:50 BSTOwen Lewis

There are so many differences in the hows and ways in the day-to-day living of all of us as to makes the perfect cleaning advice impossible, I think. all I'd add to the good advice already put here is as follows:


- Keep camera and supporting optics as free of dust as possible at all times. Investment in custom-made dust cover(s) for your setup(s) is best considered a must and not an option. Even in a small town, it is usually possible to find a seamstress who can make these for one at a reasonablle price out of fine weave nylon material. Even using large polythene bags is better than using nothing at all. Handwash, dry and then vacuum clean the covers) (I have five) every six months or so. Keep the equipment covered at all times when not in actual use.Most of us can't do much to control the dust in our domestic premises, the way it can be done in a dust-free lab, which has full environmental controls and even the complete separation of filthy humans from sensitive equipment :-) - but we can do a great deal to keep our equipment dust-free.


- My experience of cleaning camera internals and optic systems may very occasionally be necessary but that over-vigourous cleaning can do more harm than good. Its all too easy to scratch a mirror or damage a sensor plate. Wet cleaning may be OK for prisms and lenses by I use only a dry fine anti-static nylon brush (LensPen) on my sensor plate and mirror.


- Canon EOS (and some others?) offer a 'dust delete' get-over on software. Pages 185-185 of the EOS manual. This can be found on line at the Canon site by any curious and do not have this manual in hard copy. Reading between the lines of Joel's post, using this might save him some grief for almost no work and no expense at all.


But, yes, dust is a devil for all of us.

27th Jul 2017 12:57 BSTJolyon Ralph Founder

Dust delete is only going to work if you have a very small aperture otherwise you end up with large blurry marks on your image, not tiny dots.


Most modern cameras have an 'auto clean' feature. This usually involves the sensor head vibrating, Again, do this with the camera angled so the vibrating dust can fall out of the camera - absolutely no point doing this with a lens attached!


Jolyon

27th Jul 2017 16:00 BSTOwen Lewis

Jolyon & Katya Ralph Wrote:

-------------------------------------------------------

> Dust delete is only going to work if you have a

> very small aperture otherwise you end up with

> large blurry marks on your image, not tiny dots.


My understanding is that the aperture restriction applies only to the one-off collection of 'dust delete data' (DDD), imaging simply a plain white background. That single data set is stored in the camera and automatically added all the working image files collected thereafter and with the making of all those working images being free of any aperture restriction. It seems to me that this is a great way of cancelling out specks rather then physically removing them.


Re-making the DDD set can be done at intervals as may required if new specks occur that were not 'collected' in the initial data set - or have been physically removed since the last DDD was collected. So this should be considered a useful procedure to carry out occasionally, rather than a 'once only' procedure. he only tricky bit seems to me to be that the white background used for making the DDD set must be flawless and without marks or pattern. I don't think that small differences in shade of whiteness matter.

28th Jul 2017 05:40 BSTStephan Wolfsried Expert

Joel,


If You use the helicon retouching feature You don't get the blurry marks Jolyon is talking about but only dots. They are easily erasable in seconds. I didn't clean the sensor of my Sony Alpha 7R for more than two years now and 1.3 Mio shots and I live with the dust on its sensor. Working with dust maps Owen is reporting is also possible, but my way is quicker, because the dust is "living". Particles come and disappear. So You have always trouble to adapt the dust map.


But if You want to clean the sensor buy an aerosol can with pressurized air and an ejection pump. That is a kind of miniaturized vacuum cleaner, which You can use under the scope. After the can is empty do not buy a new one but connect the ejection pump to an (oil free) compressor, reduce the flow to an appropriate level and You have no more cost for that.


This method is working quite well. All other sensor cleaning equipment with fluids which I also tried is not satisfying from my perspective.


Last but not least I must say a mirror camera is not the best choice for stacking due to vibrations caused by the mirror.

I recommend a mirrorless camera with a so-called electronic first curtain.

More details see here:


https://www.mindat.org/article.php/2089/My+setup+chronology


Cheers Stephan

28th Jul 2017 11:55 BSTJoel Dyer

03895540015651845092779.jpg
Hi Owen, of course I could try your method but's I'd have to do a dust erase in Canon software for each photo in the stack. Then I'd have to import the photos into the Helicon Focus. I'd rather not have much dust on my sensor at all.

...


Stephan, I don't understand what you mean with this:
"If You use the helicon retouching feature You don't get the blurry marks Jolyon is talking about but only dots"


Before I build the stack with Helicon, I have a whole series of photos with various sized, horrible dust particles showing. I would have to clone away the dozens of dust spots in each stack photo (30-100 usually), and that would take an eternity...


When I build the stack, Helicon moves each dust particle slightly, leaving a messy trail of minute dots. These trails are a real pain to try and clone away, and often one is left with a useless or unsatisfactory picture.


I'll attach here a quick test picture below to demonstrate the matter, displaying the first photo in the stack and the stack created from 70 photos, using Method A. If I use Method C, the results are often so bad, that I want to give up photography work for a long time... Zoom into the stacked photo and you'll see what I mean :-(




As for myself giving up a perfectly working - though dusty - 1-year-old EOS 6d (that will take me another year to pay for), this option is completely out of the question. I always use mirror lock-up and Live View, use heavy old microscopes with vibration-surppessing platforms and have had no major problems with vibration. I don't take World Class stack photos with 300-600 pictures, so vibration really is not a big issue.

There are also countless people out there who use DLSR cameras to take superb stack photos, not that I do not highly admire your "mirrorless" pictures, Stephan, who wouldn't :-)


I do have a oil-free compressor in my garage, with the hose hanging outside of my IT Service company office. Compressed air is absolutely vital in computer servicing. But how one could manage to fit the specialised pump with its very small aerosol tube fitting onto a standard European large pressure tube, I have no idea... although it's a tempting idea....


PS: I repeat this is a quck test photo, and I'ved used jpg source pictures, not RAW, which take too long to stack on my current computer and haven't given that much better results.


Cheers,


Joel

28th Jul 2017 14:07 BSTOwen Lewis

Joel,


It does sound as though you are over-egging your pudding. And perhaps confusing dust problems with photographic artefacts that are prone to be created when using photo-stacking software?


Crudely, there are two kinds of dust speck, (1) permanent and (2) transient. (1) is a particle of foreign matter too firmly attached to a sensor plate to be removed by the auto-clean feature. (2) is stuff that is airborne and can float around in the camera from shot to shot, excited by shutter movement. The EOS DDD feature is an effective countermeasure to (1). There is absolutely no need ever to re-make the DDD correction more than once per session and most find they do not need to refresh the data set anywhere as near as frequently as that. Keep dust away from your equipment (and your fingers out of it) and a remake of DDD two or three times a year may well be sufficient. As for (2), if this is a serious problem to you, you absolutely require a professional clean, start using dust covers at all times when not actually using your eqipment - and to stop poking/looking/breathing into your camera, which IMHO causes more problems that it ever has fixed for anyone.


To show to yourself and others just how serious (or not) a dust problem you have, take a shot of a plain, clean, unpatterned and flawless white background and nothing else. That will identify the extent of any permanent problem for you. Making and comparing say three or four such shots will show the extent to which transient particles are causing you a problem.


There is no magic to this, only commonsense and the elimination of alternative causes of degraded imagery.

29th Jul 2017 05:37 BSTStephan Wolfsried Expert

Joel,


Your example You showed us is typical. I have similar results very often.

First: You get the more trouble the unstable Your specimen is mounted. Creeping of the object is an issue and try to avoid it.

Second: There are some typical insufficiencies of stacking algorithms. One You can see easily in Your example. The background in the upper left corner of Your picture shows more information than the stacked result. Helicon offers You to seamless overlay the stacked result with that part of a layer of Your Stack You want to have. With transparent objects there is no other way than to make use of that because otherwise You will get plenty of artifacts. Removing dust traces is a welcome side effect. Try it.

Third: Use a blowing gun as a compressor tool and reduce the diameter with appropriate hoses to the ejection pump. Not really rocket science.

And last but not least: The cheapest mirrorless camera I can really recommend is the Sony Nex 7. From time to time You get new ones in Ebay for 400-600€. If You want to stay with Your Mirror camera You can try to make the pause after the mirror prerelease as long as possible. My experience with the Canon 5D was that was even not enough. Typically You need five seconds, the prerelease time is less than one. Even a very stiff stand doesn't equalize that.


Hope that helps.


Cheers Stephan

29th Jul 2017 07:16 BSTJoel Dyer

Hi Stephan, you answers are excellent and informative. Yes, I must try to start cloning spots using specific layers, after getting rid of more dust particles in one way or other.


Unfortunately, the "small" sum of 400-600€ is completely out of my range in any near future (I can email you to explain better, if you're really interested in my long-term difficulties...). Otherwise, I'd spend immediately a few hundred on new parts to enhance my MicroRaman system, with which I'm desperately trying to earn some pocket money.


Yes, I use a pause after mirror prerelease and between stacks, naturally this is necessary. I'll see if I can adapt a special vacuum to my pressured air system, once I'm able to obtain such a vacuum system this year if lucky. First, I'd like to try the Gel Stick method, though.


Again, thanks to everyone for all the various advise and help: it is much appreciated & I'll tuck away all options in my mind's cupboard.


Cheers,


Joel

29th Jul 2017 22:58 BSTRonald J. Pellar Expert

A couple of thoughts on the DDD process mentioned by Owen. It was developed by the camera manufacturers to suppress "dead" or black pixels. But dust on the sensor blocks light and appears as a "dead" pixel(s). To optimize the DDD, use a glossy white photo paper for inkjet printers and evenly illuminate it. A small aperture, e.g., f/32, reduces the cone of light from the lens and sharpens the shadow of the dust particle on the sensor array, otherwise you get a "penumbra" that can affect adjacent pixels. Defocusing the lens suppresses any remaining structure in the white surface. The DDD allows the software internal to the camera to fill in the "dead" pixels with some kind of appropriate algorithm that can differ between manufacturers. Because of limited processing power within the camera it will not be as sophisticated as the "spot healing" tool in Photoshop particularly for larger dust particles.


Dust on the mirror is irrelevant since the mirror is not part of the image making process. In my photography which is 98% macro, I keep the macro lens on the camera at all times. For the 2% of my photos which are photomicrographs, I remove the macro lens from the camera with the camera facing down and keep it that way until it is mounted on the microscope adapter. When the camera is off the microscope, I keep a rear lens cap on the microscope adapter to prevent any dust from entering. In five years of using my camera I have had to issues with dust on the sensor. The worst case for dust entry is interchangeable lenses. This makes it hard for photographers that must use their cameras for multiple purposes.


Cleaning dust of the sensor is a very delicate process due to the various coatings that are applied to the sensors, especially anti-aliasing and color filters that may not be very hard and/or secure. Professional cleaning is preferred when possible.

5th Nov 2017 15:12 GMTStephan Wolfsried Expert

05072910015098940737821.jpg



This is my sensor cleaning set. Very effective, much better than any method with fluids.

It is a ejection pump driven by a compressor. You see the full frame sensor under the scope and You can eliminate each dust particle without generating new ones.


Stephan

5th Nov 2017 20:43 GMTRonald J. Pellar Expert

I have looked at your example photos again. The so called dust spot is round and white and I think that it may be a spurious reflection of your light source within you microscope. Dust specks on the sensor block light and are generally black and are in the focal plane and would be sharp silhouettes. White round spots are generally not indicative of dust on the sensor. In my experience with photomicrography, white spots are extraneous reflections of light within the setup. Sometimes an appropriate baffle, or movement of the light source(s), will solve the problem.



Sensor dust particles can be compensated for with the DDD as Owen points out. However, the f/ is only important in taking the DDD to eliminate penumbra around the dust particle by using the highest f/ for the lens. The DDD is then used by software (in the camera?) to create a mask that software then fills in appropriately.

6th Nov 2017 05:03 GMTStephan Wolfsried Expert

Ronald,

the dust spots on Joels example photo are on the upper left side. In the stacked version they are not dots but traces. This results from lateral movement of the specimen while stacking.

So you can be sure Joel talks not about light reflections. Pros and Cons of dust maps are discussed above. Just read it.


Cheers Stephan

6th Nov 2017 08:41 GMTJoel Dyer

Hi Ronald and Stephan,


Thanks for your further efforts to help, very kind of you.

Yes, naturally some dust spots can be caused from sample creeping, but not nearly all.

I've tried using 2 different type microscope, and laying down the sample firmly on a hard surface. My Epityp2 inverted metallograph is solid and weighs about 30 kg, and the sample cannot move anywhere. When photographing, I micro-step the platform up or down 1 or 2um or more / step, depending on the objective used.


In the end I bought the Sensor Gel-stick that was recommended - thanks Jolyon! Jolyon is absolutely correct that one has to first scrub the sensor wtihs some other (damp+dry) method. Then the gel-stick has to be used repeatedly to get the sensor more or less clean.


My problem is the space I have to work in. It is a very small company office, where people bring in their computers to be serviced. We have a gravel-covered yard outside. And I have rock samples in the same room, not all behind glass doors. So, my sensor is dirty again...


The only way is to always have all mineral samples in closed cabinets and closed sample boxes. And no paperwork, tools or other stuff should be laying around on desktops, atrtacting dust and making surface cleaning difficult.

Even if I clean my office every day more dust is carried from outside. And I have to work on both computers and mineral samples all tethe h time, so keeping everything in order is only a theoretical possibility ;-)


I will try to get some good covers for all the sensitive equipment some time next year. Hopefully this help a little bit.


Cheers,


Joel

6th Nov 2017 18:10 GMTRonald J. Pellar Expert

My apologies for not looking closer at the photos.


I would like to point out that the dust particle trails created in stacking are not due to specimen shifts but are due to the change in magnification that is required in processing the stack to correct for parallax changes in the stack progression.

6th Nov 2017 19:23 GMTJeff Weissman Expert

Ron is correct - the dust is usually on the filter that protects the camera sensor, sometimes dust is also on either the front or rear lens element. The best course of action, once everything is clean, is to never remove the lens from the camera body, but that is of course not possible, as I use 3-4 different lens/microscope objective combinations, as I suspect many of us use more than one lens. Usually, these dust streaks are minor or readily removed using Zerene Systems stacking software, or Photoshop. I'd suggest visiting http://www.photomacrography.net/forum/ for additional information, they are very friendly.


Another solution is to run a humidifier for 10-15 minutes prior to removing the lens, the humidity will help settle any dust in the air and prevent static attraction.


If the problem was dust on the mineral, it would move with the mineral as focal length or magnification changes during stacking, and appear just as sharp and/or out-of-focus as the surrounding areas of the specimen.


For cleaning the sensor filter, I use either a gentle blower, such as Giottos Rocket Blaster Dust-Removal Tool, or swabs or other devices made for the purpose - and like Stephen, I use my microscope to examine the element while cleaning. Be careful, as too aggressive cleaning can leave scratches on the filter element.

7th Nov 2017 05:13 GMTStephan Wolfsried Expert

Ron and Jeff, sorry but You are wrong. Creeping is a big issue I try to avoid as much as possible. Therefore I let the mounted specimen in the final position calm down for at least a day. Then You can avoid such traces more or less completely. I do not stack with focus adjusting of the camera but with incrementally increasing the distance of the whole setup. Yes, lens change is always an issue. A way to avoid that is described here:


https://www.mindat.org/article.php/2676/A+turret+for+my+setup


Cheers Stephan

7th Nov 2017 06:17 GMTJoel Dyer

04592700015651869488237.jpg
Hi Stefan, nice article there you have, thanks for it. I have a turret on my Epityp2, this does help to keep dust away from inside of objectives, but not from the outside. I don't use stereoscopes anymore for photography, except maybe at some (rare) fairs or meetings. The Lomo Min-8 petrological microscope uses detachable single objectives & indeed this is a major problem.


Stefan, You are a master photographer, so I know you must be right in many cases, especially if using blue-tack or other soft support material. But we alll have different setups. How does the creeping occur, if a sample is placed flat on an inverted microscope platform, and the movement mechanism is tight and working properly, can you explain to me?


Also, I don't understand why it would benefit to wait for a sample to "settle down" on a platform, if oneself and customers are walking around in the room during the day when one is not doing stack runs? You also have to have a lot of spare time to wait most of the day each time before you photograph - out of the quEstion for myself...maybe you're now retired? :-)


I, too do not stack using camera movement: my cameras are firmly attached to both microscopes. With the Lomo, the normal stage is moved up & down. This stage might creep a bit, so yes I'll have to keep a sharper eye on the tightness of the system, thanks.


With the Jena Epityp2 (30kg "tank"), the camera is completely unmoved in the right-hand camera port & only the very heavy stage with the sample on the top of it (inverted microscope) moves up & down. I have left samples on the stage for hours and no creeping has been noticed, because the focus is still very sharp - I've taken tests photos. You must remember that this device is an old - but excellent - DDR workhorse that was meant for heavy, repeated industrial work - and of course it's been serviced often by myself.


A picture of Epityp2 opened for servicing with turret showing in the background & platform system removed:



Surely different stacking processes move - have to move - the pixels in order to carry out the stacking properly? Naturally, the results vary depending on what method (A-C) is used in Helicon Focus?


Ron, nice system you have there & very good advise and information you have. I suspect that many master photographers on Mindat have low-dust areas specially setup for microscopy and photography, I have no such option at the moment & explained this above, so dust will always be a problem, until I retire(?). Owen rightly said that equipment should be properly covered when not is use, this must be addressed better at this end, to include all microscopes and other equipment.


I hope this additional information on my setups might give further light into the issue. I've now condensed all this advise from people together in a "reminder list " to stare at my from my laptop screen before I start stacking photos again :-) .


Cheers,


Joel

7th Nov 2017 09:37 GMTStephan Wolfsried Expert

Joel,


Ron referred on Your example photo some posts above, and to this I also referred. I am talking about usual thumbnails mounted with tack. Dust on the objective is out of focus and would not cause artifacts You and I deal with.

As long as the artifact traces are parallel as in Your example photo they come from creeping of the specimen and nothing else.


If You look at a thin section this is a complete different story. I have no experience with that, but I would assume that there is no issue with dust traces at all. Stationary dust dots will be the normal issue than in other cases as well.


My setup weight is some 150 kg, despite of this the major source for vibrations is the shutter or the mirror if present.

If I walk around I will see that also when I use the 20x objective. Therefore You need a time for calming down of in my case typically 5 seconds between each layer photo release. And leaving the room for the processing time is helpful. If You have vibrations from people walking around this is really contraproductive. Unsharp layers spoil the whole stack.


For keeping my family life unaffected I use a high degree of automization. Only 5 to 10 % of the whole process time I am present. When a stack runs I am typically only present at the beginning and at the end, feeding Helicon and change specimens. I use several object holders which I can assemble hours earlier with regard to the creeping issue.

This is something which has nothing to do with my retirement. I produce so many photos which would be impossible without automization. A perfect stack runs like a slomo video at helicon. Unsharp single layers are easily recognizable in watching this slomo film.


What remains is postprocessing of the rendered stack in erasing dust, adjusting brightness and so on.


The original question raised from You what is the most appropriate way of sensor cleaning: As I described above, I trust on a miniaturized vacuum cleaner. Much better than anything else I tried before, and believe me I tried a lot.


And last but not least my setup I showed is completely sealed against room atmosphere.


The only way of avoiding influence of vibrations is a complete different lighting setup like insect microphotographers use.

Charles Krebs is one of the icons of insect microphotography. He uses flashlight. Sometimes in the past I considered following this path. I refrained because our objects are different, and a good photo requires appropriate lighting.

With a flash there is no topology, what is in front and what behind and so on.


Maybe the easiest solution would be in finding a more quiet place to take photos...


I wish You a good success


Stephan

7th Nov 2017 16:39 GMTJeff Weissman Expert

Please google "streaks site:photomacrography.net"


For example, this page explains sensor dust streaks - http://www.photomacrography.net/forum/viewtopic.php?t=35487&sid=5c7916b57c7cb1a6b1c38739f198c9ca


This is Joel's' problem -he has some dust on his sensor. If dealt with this on/off through the years of doing stacked macro, and is readily resolved by a good sensor (actually the filter in front of the sensor) cleaning.


The area in the upper left corner of Joel's stacked image shows that the background is out of focus, but yet the streaks are in focus. This is not due to specimen movement or dust on the specimen (as this would also be out of focus), but due to minute dust particles on the sensor filter, where they appear as being in focus; this is most noticeable at higher magnifications due to optics. For the top single image, the entire specimen is out of focus, as it should be being the first image in the stack, yet the dust particles appear as being in focus, as they are sitting on the sensor.


Joel - you may also have some issue with lighting and vibration, but I don't see any evidence of specimen movement - although I realize that this was just a quick stack for illustration purposes only. Most stacking software automatically corrects for specimen movement or rotation anyway, usually up to +/- 20% through the entire stack, so shouldn't be any issue. Basically - if the specimen hasn't shifted in a few minutes, which is the time needed to get a good view, get lights aligned, etc., then you should be good to go, no need to wait 24 hours for a single specimen.


I invite anyone who wants further information to visit photomacrography.net


By the way, you can make an excellent, sharp macro zoom lens covering from 0.7x to 1.4x by modifying a Canon EOS 35-80 zoom lens, by removing the front element - you lose the autofocus capability, but retain automatic aperture. These lens are usually cheap in the used market, $30 or so. See http://www.steamengine.com.au/index.php/photo/52-techniques/337-modification-of-the-canon-35-80-lens-for-macro for example. I did this and it works great, and delivers nice bokeh to boot!

7th Nov 2017 16:46 GMTRonald J. Pellar Expert

Stephen:


There are several reasons for "creeping spots". Besides the movement of the specimen itself, which can be eliminated by many means, there are two that are important when dust particles reside on the camera sensor.


First, and foremost is apparent magnification change the occurs when the focus is incremented, either by camer focus or camera movement, i.e., "focus" or "rail" stacking. To understand this better, the dust particle on the sensor is fixed in its pixel coordinates regardless of any camera settings, but as focus in incremented objects within the image will shift with focus adjustment, Thus the dust particle which appears static relative to a particular sensor pixel, will appear to move relative to an object in the image. This will cause a traveling, or creeping, dust particle in the stack of images. The stacking software corrects the image for this problem, but unfortunately it does not correct for the "creeping" dust particle, since the dust particle is fixed to a particular pixel(s) and is in relatively sharp focus. This subject of magation changes was discussed in this forum in past concerning Focus stacking vs Rail stacking and their possible differences in perspective. The change in magnification is due to the areas that are going in or out of focus are changing their distance from the camera lens and thus their size changes. Some would call this a parallax effect. The dust on the sensor is fixed behind the lens and stays fixed relative to a pixel of the sensor and is not affected by any focus/distance changes.



A second cause of creep, especially in some less expensive microscopes, is the mast that the microscope, or camera, is attached to, is not exactly perpendicular to the stage. Changes in focus by means of elevation of the rig, cause the image to creep across the sensor. The stacking software does a realignment of the images but the dust position is fixed relative to the sensor. Dust in or on elements of the microscope will behave similar to the dust on the sensor, i.e., stay fixed with respect to the imaging sensor. This causes an incremental shift in the image as the stack is accumulated. Better equipment and can solve this problem.

8th Nov 2017 04:11 GMTJoel Dyer

Well, it seems I may have to add some more information here about my setup etc, so there are as little possibilites for misunderstanding as possible, but the thread is getting a bit long :-) .


Stephan, you have an admirable setup, but then you are a perfectionist & I guess have been able to also invest a considerable sum & reserve a particular area more suitable for the photography. Also, you have the benefit of a lot of automation, which I don't have access to. Your one comment tells a lot: "And last but not least my setup I showed is completely sealed against room atmosphere."

Stefan, naturally I photograph only when there is no movement & no customers. I also cannot go away anywhere from my little room in the garage-office building, because I have to wait (and pray) for customers, run Raman analysis, and do other work on the side. Also, there is the complete lack of automation devices. I don't have any money to spend on any rail systems, motorised controls etc, none at all, and likely never will have, end of that discussion... ;-)


Ron, your last addition here is very illuminative, and this indeed is what I suspect is happening in my case.Only way out it to minimise dust occurence (nearly impossible currently) and protect equipment better (possible).


One solution woiuld be removing all my equipment upstairs into an unheated, incompletely set up area. It was used previously, but was eventually far too expensive to heat with available methods. Donations for a small heat exchange pump unit ("ilmalämpöpumppu" in Finnish) are welcomed :-). PS: We live close to horse stables & flies are also a problem....particularly upstairs it is a constant battle. With the upstairs unheated, the problem is much less severe.


I now regret having posted such a horrible quick-stack for all to see. But is a good demonstration of extreme cases.

Naturally, especially after having cleaned the cameras I get much better results, also using longer pauses. Yes, I always use pauses between photos. And yes, I have read a lot of articles on various macrophoto /photomicrography sites, but opened this message since I had trouble with removing dust.


I must clarify, that I don't use any lenses anymore when taking photomicrographs, except in rare exeptions. I haver already been been through this stage, and vibration was the problem there as you start to go over maybe 3x or 5x microscope objective class. I studied the matter & experimented, and draw the same conclusions as other people.

Now I have far, far less vibration issues, mostly dust issues. I actually take photos for documentaion, not to enter into competitions. But dust trails are horrible in any photos, and can obscure for instance some inclusion details etc..

I use camera bodies only, with special photo tube / photo port oculars/objectives bought separately for each microscope type.

But when for normal photography you dismount/mount the camera body from/onto the microscope, and attach/remove normal camera objectives to/from the camera body, dust can easily enter inside. This has been discussed repeatedly here.


This message thread sure has brought out a lot of interesting ideas and recommendations, much more than expected. A lot of dedication and helpfulness, typical of Mindat :-)


Cheerio,


joel

8th Nov 2017 18:27 GMTRonald J. Pellar Expert

Joel,


I don't have any problem with dust trails in macrophotos, since I don't change lenses, except to remove the lens to attach the camera body to the microscope. It is when using the microscope that I have most of my problems with dust trails. However, the dust is in the microscope and not on the camera sensor and is not in sharp focus. They appear as a nice round black circle and since they are fixed with respect to the camera, the show up as nice black trails of circles after the stacking software is finished. Spurious light reflections show up as nice round white circles. I remove these in photoshop with the "spot healing" tool. It replaces the spots with content that matches the area surrounding the spot. This tool has saved many of my photomicrographs. :-) There are times I have to move my specimen slightly so the spots are not in a critical area of the specimen and retake the stack.

9th Nov 2017 07:22 GMTJoel Dyer

09251090015651869862971.jpg
Hi Ron, thanks again for the information. I agree dust inside the microscope is an issue, definitely, also on the oculars, filters etc. when doing research. I just cleaned part of my Epityp2 again.

I can assure eveyone that most of the irregular trail-causing elememnst in my stacks have been dust on the sensor. This is easy to test when changing the angle of the camera & re-shooting stacks. I got rid of most of the dust using the gel-stick plus damp/dry wiping first.

The mini-vacuum that Stephan mentioned is definitely on my wish list: maybe I will ask Santa Claus to get me an Amazon coupon this year for this :-) .


Ron, about "extranenous" or internal reflections of light , I guess it depends on what kind of light you are using? I normally use today fiber-optic lighting for oblique/transmitted light photography & clean the fiber-heads regularly.


When using my Lomo petrological microscope, I have to use the internal light for some transmitted light photos / stack. Here there maybe some dust or grease - or scratches in the coatings! - that refract the light. You have the objective elements, the possible wave plates, polariser and analyser surfaces, photo tube elements, photo tube/ocular view switch prism - swung out of the way when photographing with the Lomo - and other components. If there are constant reflections despite cleaning from inside the microscope, then the centering and angles need to be checked and corrected.

Yet when on other than perhaps the microsope objectives, shouldn't most of these white spots turn up as more or less un-focused defects in the photos or sometimes as a mild bluriness?? I have indeed seen this happen when the 'scopes need internal cleaning. I'm sorry, but I use mostly free software for editing photos, and don't have Photoshop. Helicon is the only paid software used currently. All costs must be constantly prioritised.


Stephan and Owen, your words got me re-thinking, thank you. I have some matte-surfaced Thorlabs Blackout Boards left from re-constructing my Raman Microscope / Spectrometer unit encasing. This is excellent, light-weight material that is easy to cut.




If you use Laser Blackout Tape from Thorlabs or other sources to connect or hinge the boards, then BOTH extraneous light and dust can be kept outside of the setup by "boxing in" the setups, with hinged doors etc. I'll let you know how I'm progessing with this project later on :-)


Cheers,


Joel

9th Nov 2017 18:30 GMTRonald J. Pellar Expert

Joel,


It can be very difficult to pin down the source and path of extraneous light spots. All bright or dark spots when out of focus take the shape of the Lens(es) involved, i.e., out of focus spots will be circular for circular shaped lenses and square for square shaped lenses. In addition the internal reflections off of components in the imaging path may cause reimaging by lenses after the reflection changing their size larger or smaller. In my case, my microscope is a Russian copy of a well known high end brand of microscope with a large entrance lens. So light reflections are not as controllable as I would like and may even be caused by bubbles in one of the lenses.


Of all the photo-editing software, I feel that Adobe Photoshop is well worth the investment. There is an old saying "you get what you pay for". :-) Not strictly true, there is some excellent freeware out there, the functionality and quality of PS undeniable. You do have a heavy investment in your equipment and one piece of software to match will pay dividends in the long run.

10th Nov 2017 06:43 GMTJoel Dyer

Hi Ron, I understand and appreciate what you say. Yes, there can be so many components in microscopes & all can cause troubles. This is why I didn't manage to connect my Raman Spectrometer to an existing microscope, even after spending a few hundred € (of donated money) and a couple of hundred hours experimenting. Almost wanted to throw myself in the nearby sea here.... I had to build my own microscope unit. I've heard similar difficulties from others. Raman has extremely small - or nil - tolerance for any extraneous disturbances at all, believe me.


As for investing in software, that's not possible, no funds for even most normal purchases & activities, sorry. The dividends side is much worse than a couple of persons predicted it would be, very small, occasional pocket money. It is interesting how many peope want to know what there samples are, but so very few people who are willing to spend even small sums like 20€ for finding this out with expensive gear, including specialised microscope as part of the whole. You can also ask John Attard what he thinks about this matter ;-) . A vey nice guy John is.

If' I'm lucky, the Raman costs will in best case be paid back in perhaps 11-12 years time, and lab lasers are only expected to last around 10.000 hours. By then or earlier a new laser unit (several €€€€) will have to be bought again. C'est la vie...


Cheers,


Joel

10th Nov 2017 18:14 GMTRonald J. Pellar Expert

Joel,


I wish you the best of luck!! You seem to have accomplished a lot on very little resources. Hang in there, things might get better in the future.


Best Regards,


Ron


PS: Say Hi to John for me next you see him and I will say Hi next February at the Tucson show for you.
 
Mineral and/or Locality  
Mindat.org is an outreach project of the Hudson Institute of Mineralogy, a 501(c)(3) not-for-profit organization. Public Relations by Blytheweigh.
Copyright © mindat.org and the Hudson Institute of Mineralogy 1993-2019, except where stated. Most political location boundaries are © OpenStreetMap contributors. Mindat.org relies on the contributions of thousands of members and supporters.
Privacy Policy - Terms & Conditions - Contact Us Current server date and time: August 25, 2019 05:57:12
Go to top of page