IMPORTANT MESSAGE. We need your support now to keep mindat.org running. Click here to find out why.
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 Articles
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 CompaniesStatisticsThe ElementsUsersBooks & MagazinesMineral MuseumsMineral Shows & EventsThe Mindat DirectoryDevice Settings
Photo SearchPhoto GalleriesNew Photos TodayNew Photos YesterdayMembers' Photo GalleriesPast Photo of the Day Gallery

My setup chronology

Last Updated: 8th May 2016

By Stephan Wolfsried

My first microscope I got in 1972 had'nt either a zoom nor a foto-tubus. I used it for about ten years. Taking photographs through the ocular was not really satisfying.

I bought my second microscope from the first bonus I earned. It was an Olympus SZ60 and had both, a zoom and a photo tubus. I used that for a long period of time. I built myself an adapter for a SLR-camera and started with taking photos using special tungsten film for diapositives. From 36 photos of one single film had I could usually make use of only one or two. The rest was scrap. Several days passed between taking pictures and getting the processed film from the lab. I didn’t really remember what was wrong with the parameter settings. The exposure time was normally between 30 and 60 seconds, and I could only have a rough idea, how the photo would probably look like while looking through the viewfinder. This efficiency was finally too poor and costly for me, and I gave up photographing for some years.

The breakthrough came with the invention of digital cameras. Immediately after taking a photo I could see if it was good or not. From then on the camera was not longer the limitation, but the microscope itself and at least as important - the illumination.

Another breakthrough came with depth of view enhancing software using the multi-layer technique. The challenge now is making use of this and although let a photo appear naturally and and give it 3-dimensional properties. End of 2004 I bought my third microscope, the Carl Zeiss Discovery.V12. I used a Nikon coolpix 8400 camera, attempts with a Canon D5 were not successful. SLR cameras on phototubus of a telescope type bino suffer from dust on the imager in combination with multi-layer technique and long exposure times.

From the very beginning I use halogen light sources (250W from Schott). Meanwhile I take two or three of those, one with Ringlight around the objective (not very often) and two furthermore with two gooseneck fibre optic illuminators each. Although I have a high consumption of light bulbs (5-10 a month) I don't use LED light sources because of the high light density which spoils almost every picture, at least without diffusing. In addition, the colour lightroom isn't comparable to bulb light, see also Tony Petersons remarks on the micromounting message board. http://www.mindat.org/forum.php?read,6,145171,145289#msg-145289. The colour temperature of my light sources is 3300 K, and white balance is not so difficult to set at the camera. Using different types of light sources I always had different colour temperature values by dimming one of them more than the other, and had to correct afterwards which is time consuming and not recommendable.

In May 2008 I got the upgrade of my microscope to the CZ Stereo Discovery V.20 (FOV 20...1 mm) and the adapter for a Canon G9. 12 Mipxels instead of previously 8 and a spread of the Microscope to a max magnification of 225x instead of 150x before brought some improvement.

In December 2008 I got a brand new Nikon Coolpix P6000. In comparison to the Canon G9 there is a minor improvement in resolution and noise behaviour. However, my main reason for taking the P6000 was the remote shutter release feature with an infrared remote control device. This gives the possibility to take lots of layer pictures without having a hand on the camera or the microscope either, with improves quality at high magnifications. Before, every touch on the camera in order to release the shutter (with 2 sec. delay of course) gave a micro impact to the whole system with vibrations and the risk of a movement of the object. While the Nikon Coolpix 8400 had some hot pixels and the Canon G9 lenses were not fully sealed to atmosphere and suffered from dust in the lens I had to work with dust maps then. With the P6000 there is actually no need for cleaning artifacts or anything else.

December 2010 I finished assembling a second stand with a motor driven stative from Carl Zeiss like the Discovery V.20, but with a bellows and CZ Luminar lenses in front instead of a microscope. See thread at the Photography topic. I choose the Panasonic Lumix GH2 for that purpose: 16 MPixel sensor with 5,5 MPixel per square centimeter imager, no mirror, tilt and swivel display, radio controlled wireless shutter release. With no practice at all I could achieve pretty good pictures at the very first attempts. Their resolution is remarkably better than taken with the Discovery V.20, but only down to FOV of appr. 1,5 mm. The whole setup seems to be more robust against reflexions and bright light spots. Maybe the PlanApo 1,5 of the V.20 gets too much diffuse light. The 25 mm Luminar seems to become the most appropriate lens with a FOV from 3,5 to 2,5 mm with the bellows only and down to 1,5 mm with a bellows extension. 21 mm working distance or more allows enough external illumination with gooseneck fiber lighting. For FOVs smaller than 1,5 mm the Discovery V.20 for the moment remains benchmark. Pictures taken with the Luminar 16mm or the Nikon M Plan 40x have a good resolution but suffer from very small depth of view, which lets the transition from sharp to unsharp areas appear very prompt and artificial. The Luminar 25 and 40 mm lenses are my favorites now and I take a lot of photos from my personal mineral collection once again. Some of them I could improve over years step by step. I don't appreciate my very first attempts now any longer but don't erase them all to see and show the progress made. From now on I only take photographs with the bellows and take the microscope for handling and choosing motives.

January 2012 I changed my old bellows to a new Novoflex one and bought a Sony Nex 7.
The APS-C sensor of the MFT camera and the new bellows allowed a FOV down to appr. 2 mm with the Luminar 25. A 4000x3000 pixel crop of the picture delivers slightly better results than the related FOV taken by my Microscope. However, the number of layer pictures required increases to typically 100-250 depending on magnification. The minimal z increment of my motor driven stative is 0.01 mm was now no longer satisfying.

For that reason I implemented a Cognisys Stack Shot in May 2013 between stative and bellows. Increments of 0,002 mm are now feasable under some conditions. The hysteresis when reversing the stack shot has to be eliminated before beginning a new stack. However the overall stiffness of the assembly decreased significantly. The Sony NEX7 is a MFT camera without mirror, but even the shutter causes a vibration which requires a delay time of some seconds to calm down the whole assembly before releasing the camera once again. Stick slip effects in the bearing and/or the tread of the Stack Shot cause some unsharp pictures randomly scattered in the stack. Usually Helicon Focus struggles with that, so I have to erase the unsharp pictures in a stack manually. The memory size of unsharp elements is significantly less than the corresponding size of the valid ones, so it is possible to erase them by only reviewing the directory.

Since Luminar lenses now are about fifty years old I always wonder if there are modern alternatives. Inspired from a microfotografic Forum and some blogs of the icon of Microphotography Charles Krebs I tested the full set of Mitutoyo objectives, 2x, 5x, 10x, 20xSL, 50xSL. SL stands for more working distance. Even the 50x SL has comfortable 20 mm free working distance. As a difference to the Luminars all Mitutoyo lenses are infinite objectives and need an additional tubus lens. I use several Nikkor 105 mm manual focus lenses for that purpose. They work from 50x to 5x pretty well, with 2x I got vignetting. The Mitutoyos have a high light sensitiveness and require high f-stop values to be set with the tubus lens. Diffuse light sources and rather long exposure times avoid bright light reflexions which may spoil a picture otherwise. Using the Stack Shot I achieve pretty good results with the 20x SL, the 10x and the 5x and exposure times of 0,5...2 sec. FOV of min 0,9 mm are now feasible with reasonable resolution. The automization allows accepting even half an hour for taking a stack of 250 pictures because I can do something else during that time.

There are actually two different Luminars (40 and 63 mm) and three Mitutoyo objectives to work with in parallel. My solution for a reasonable quick change is using a bellows adapter Novoflex PROLEI and LEINIK to Nikon bajonet, Nikon to M42 adapter and M42 to RMS for the Luminars. Novoflex LEINIK fits with the Nikkors and the Mitutoyo objectives. The dust problem I experienced with my full frame Canon D5 DSLR already is the same or even worse with the Sony Nex 7 MFT camera now. Frequent change of objectives comes along with more and more dust over time on the sensor. Due to a very volatile particle status on the sensor I prefer cleaning the sensor by myself from time to time and correct remaining dust artifacts with a program like photoshop. Dust map correction routines like e.g. from Helicon take a lot of time for processing and are always "old" because of getting permanently new particles over time. Different to the earlier attempts with the foto tubus on the microscope the artifacts are now constantly sticking at one place on the picture, while the lateral movement of the microscope due to stacking let become the artifacts stripes, which are much more difficult if not impossible to erase in the final stacked picture.

Using the Stack Shot with high magnification creates another problem: Creeping of the whole specimen while stacking. Now it is essential to have no movement of the specimen at all while a stacking routine is ongoing. Even some tenth of a millimeter creeping over half an hour may spoil the result. Having the specimen totally balanced is now crucial.

I use Helicon Focus for years but up to April 2014 never took notice of the very useful retouching function. I got aware of it after testing the fusion of two different algorithms with the Zerene Stacker (without satisfying results). Once getting aquainted it is easy to work with and I believe it is crucial for good results of transparent probes. Especially such ones which have inclusions or through which You can see to the bottom. I also noticed that areas with little contrast and/or large topologigal differences show not really good results having had the default stacking algorithm running only. In this case that feature is also very helpful.

Another success factor for me is the hardware to display photos. For stacking, retouching and postprocessing photos I use the 27" cinema screen from apple which allows a sufficient resolution displaying a whole 24Mp picture. Adjusting brightness, reducing noise and erasing artefacts works fine with that. Artefacts due to dust on the sensor are somehow unavoidable and erasing them is very time consuming. Therefore such a big display gives a chance to cope with that problem with a reasonable time consumption because navigation in a photo from the one edge to the other is not necessary.

End of 2013 I bought an IPad air and migrated my gallery there. The so called retina display with its high resolution brings a picture quality which I didn't see before anywhere. Its really amazing and makes me even more satisfied with my pictures. The only downside of using apple hardware is the limited database organization functionality offered with IPhoto. Using cyber duck as FTP server and Foto Manager Pro as App on the IPad solved that and I can have my folder structure on the IPad as I am used to it under Windows.

As I mentioned previously the focus lenses used so far in combination with the Mitutoyo M Plan objectives were the Nikkor 105 mm manual objectives. In May 2014 I tried a Nikkor 180 mm f/2.8 ED instead. The FOV is now significant smaller, with the 20x I achieve 0.55...0.9 mm depending on bellows extension. With the 10x 1.05...1.8 mm and with the 5x 2.1...3.6 mm. The resolution is now significant higher and in that combi the result is better than with the Luminar lenses.

In June 2014 I replaced the original stack shot rail with a linear actuator from THK which is much stiffer. Those linear actuators are normally used in CNC machines. I added a second sledge to have a base of 100 mm for the bellows rig. This is quite tricky, because the bearing balls fall apart when disassembling that second sledge from a different actuator and I had to reassemble the slegde in its new bed by refilling the balls. At the end I had some balls more than I could reassemble, but anyway the whole assembly works fine and is now as stiff as possible. Pressing some grease into the sledges nipple with a self adapted grease gun (everything is a little bit smaller than elsewhere) overcomes stick slip effects due to the few lacking balls. One can see the improvement in regarding the stack while processing in helicon focus. No lateral movement from one picture to the next is visible (except if the probe itself is creeping). The results are now convincing. Using Objectives with a FOV of 3 mm and more weren't even problematic before. But the 10x, 20x and 50x objectives require very small increments down to 1/1000 mm which are now feasable. The good thing is that the control unit can be maintained without change. Only the ratio of 1.59 due to the different pitch of the ballscrew which is now 1 mm per rotation vs 1.59 mm before has to be taken into account. A dedicated article of that setup is published, have a look on my article list.

August 2014 I got a Nikon D810 with full frame sensor. This one should replace the Sony Nex 7, which was a very good companion for more than two years. As the Nikon DSLR has a mirror in difference to the mirrorless Sony Nex7 I had to learn that the stiffness of the whole assembly is even more important than before not to influence picture sharpness with vibrations from the camera. A twin bellows to have a seamless FOV variety lived therefore only two days, because the whole assembly became too weak. Increased amounts of data with that 36 MPixel camera result immideately in an exchange of my computer hardware. Now I use an IMac with 27" screen and a 3,5 GHz i7 processor with 32 Gb RAM. SSD memory cares for quick data transfer. Now a 200 picture stack typically lasts about 15 Minutes for taking. Processing with Helicon Focus needs less than 10 Minutes with Rel. 6.2.0 for rendering. The majority of the postprocessing time is now to be invested in retouching. The higher the magnification, the more time is needed for eliminating artifacts and correcting insufficiencies of the stacking algorithm.

Helicon Focus all the time is one of the key success factors of my workflow. However, there is no Helicon Filter under OS-X available, and so I switched back with both programs on my brand new IMac under Parallels to Windows. Rel. 6.2.2 is much more performant and accurate than 5.3 was and the results are really satisfying.
Retouching of probes which are not almost flat is mandatory. It becomes the more important, the less the free working distance between objective and probe remains. This comes more and more with higher magnifications. E.g. the Mitutoyo 20x has a FWD=free working distance of 15 mm and a FOV of around 1 mm. If a probe has also 1 mm range in vertical to be shown, the angle resulting from the vertical movement is tan 1/15 or appr 3 dregrees. This shadowing effect is visible in rendering results of stacks. The area behind looks artificial. Retouching helps here a lot to let pictures appear naturally.

After a couple of weeks dealing with mirror prerelease time and all kinds of different objectives I must admit that the Sony Nex 7 delivers by far better results than the Nikon D810. In September 2014 I disassembled the Nikon D810 and reinstalled my Sony Nex7 setup. What I am now focusing on are searching for high resolution objectives, more DOF, more working distance and alternative focus lenses. Actually I check Optem and Motic Plan Apo 10x objectives als alternative to Mitutoyo. The Motic MPlan Apo 10x with NA 0,28 and FWD 33,5 mm has identical spec data than the Mitutoyo 10x and appears to be equivalent to Mitutoyo by comparing the results. The spec data of Optem differ slightly with NA 0,30, a FWD of 34 mm and a appr. twice as high DOF with 6.1 1/1000mm. Compared with the other two objectives I cannot find the higher DOF with own measures with the Nikkor objectives as tubes lenses. It is more or less comparable. The resolution of the Optem 10x is visibly worse than with the other two objectives. The same result occured with the Optem 5x NA 0,225 high Resolution objective in comparison to the Mitutoyo 5x NA 0,14. I could not recognize any improvement of resolution in comparison to the Mitutoyo 5x NA 0,14 objective.
Obviously the tube lens for Optem infinite objectives has completely different properties to what I use in my setup or the spec is too optimistic.

October 2014 I tried Raynox macro lenses as tube lens and had good results with both, the DCR-150 and the DCR-250.
In combination with an adjustable aperture the results are really convincing. Stacking is easier and not so time consuming in postprocessing. The total length of the setup is significant shorter than with the Nikkor 180 and 200 mm manual focus lenses as tube lenses. The resolution ist higher than with any setup I ever used before. With the Mitutoyo M Plan 20x NA 0,42 I achieve 1100 LP/mm with a target of Carl Zeiss. This means a resolution of 0,0009 mm. The used lenses are now Mitutoyo 20x, 20xSL, 10x, 5x (each with a Raynox 250 in reverse Position as tube lens and a aperture between objective and Raynox) and a naked Luminar 63mm. An almost seamless FOV variety from 1 mm to 22 mm can so be achieved.

Setup from 11/2014


The Raynox DCR 150 and DCR 250 have the same threads, the difference in FOV is marginal. I choosed the DCR 250 in retro position. The difference in normal and retro position is also marginal. Adapters for the retro position are easier to obtain, and the whole setup is very compact, see below.

For this lens the specified optimum working distance from Sensor to lens is appr. 220 mm.
With the Novoflex Balpro 1 bellows as shown in the above picture the working distance varies from 120 mm to 270 mm. I cannot see any degradation in that range.

The parts list for the objective adaption is as follows:

M26x36tpi to SM1 Male thread Thorlabs
Aperture SM1 with SM1 female thread twice Thorlabs
Adapter SM1 to M34x0,5 female thread Thorlabs
Step up Ring M34x0,5 male- to M43x0,75 female thread Ebay
Raynox DCR 250 macro lens Amazon
M49 to Nikon F Bajonet Ebay

Raynox tube lens with aperture, 11/2014


November 2014 I bought a Nikon PB6 bellows which is maybe 30 years or more old but in mint condition with a bracket which is normally used to assemble a bellows extension. I took the bracket to mount it on the THK KR20 linear actor. The advantage of the Nikon PB6 is a range of 180 mm from zero to full extension, exact the 30 mm more I searched for and which I couldn't get from Novoflex. The mechanical properties are at least comparable to the Novoflex Balpro 1, I think it is even stiffer. The only disadvantage is that the camera is now 180 degrees rotated. A Kipon shift adapter manages the 180 degrees rotation. The big advantage is that now there is an really seamless FOV range between my 20x, 10x and 5x objectives. E.g. the 10x has now a FOV range from 2.1 to 4.2 mm (so far 2.1 to 3.8 mm), the 5x from 4.2 to 8.4 mm (so far 4.3 to 7.85 mm).

Working for a while with the new bellows I recognized more and more dust on the sensor, significant more than I am familiar with the Novoflex bellows. Obviously this comes from the bellows material when unfolding it. Therefore an action with a vacuum cleaner is necessary. When the PB6 was produced digital sensors have not yet been invented and the normal film was more or less robust against this problem by transporting dust outside the camera.

Stack-Master on the motor driven CZ stative


The THK linear actors I used for about one and a half years are much stiffer than the original Stack Shot rail, however increments of less than 0.0002 mm were even not feasible with this device. Rainer Ernst convinced me to use his stack-master unit and I got my own one in the week before X-Mas 2015 with the according adapters. The assembly was easy, it is electrically compatible to the stack shot control. Next Year a stand alone controller will be available from Rainer Ernst.

Bellows PB6 on the Stack-Master and the motor driven CZ stative


First results are really convincing, increments of 0.001 mm are feasible without any problem. The only condition is to let the assembly calm down appr. 3 sec. after each shot.

January 2016 I got a brand new Sony Alpha 7R Mk2. The Sony Alpha 7R Mk1 had no electronic first curtain which was prohibitive for stacking.
The shutter vibrations of the first curtain spoiled every high resolution picture. This was solved with the A7R Mk2. First of all I made some resolution measurements. The 42 Mpixel sensor of the Alpha 7R Mk2 has a slightly larger pixel pitch, 204 pixel per mm compared with the Nex 7 APS-C sensor with 250 Pixel/mm. This expectation could be confirmed with target measurements where the Nex 7 in combination with a Mitutoyo M Plan 10x has a resolution of 800 LP/mm, the Sony Alpha 7R Mk2 only 710 LP/mm.
The 11% lower resolution is overcompensated by the 1.5 times larger FOV.

Sony Alpha 7R ii with Mitutoyo M Plan 10x
Sony Nex7 with Mitutoyo M Plan 10x




















With the Nex 7 the FOV Range of the Mitutoyo M Plan 10x in the described assembly with the Nikon PB6 bellows is from 2 to 4 mm.
With the Sony Alpha 7R Mk2 and the M Plan 10x the FOV range is 3 to 6 mm. That means a FOV of 1.5 to 3 mm is covered by the Mitutoyo M Plan 20x. As a result the resolution per FOV is significantly higher with the Sony Alpha 7R Mk2.





Article has been viewed at least 12971 times.

Comments

In order to leave comments to this article, you must be registered
Mineral and/or Locality  
Mindat.org is an outreach project of the Hudson Institute of Mineralogy, a 501(c)(3) not-for-profit organization.
Copyright © mindat.org and the Hudson Institute of Mineralogy 1993-2018, except where stated. Mindat.org relies on the contributions of thousands of members and supporters.
Privacy Policy - Terms & Conditions - Contact Us Current server date and time: November 18, 2018 05:47:09
View slideshow - Go to top of page