Testing the limits

Starting out in microscopy? Post images and ask questions relating to the microscope and get answers from our more advanced users on the subject.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Pitufo
Posts: 259
Joined: Sun Jun 21, 2015 4:32 am
Location: United Kingdom

Testing the limits

Post by Pitufo »

My interests are microscopy and field mycology. I have a decent trinocular with compound with a 5MP cam - this works very well. I also have a reasonable 1:10 stereo which is good for visual use but I have found it to be disappointing for photography.

There are clearly some amazing rigs on here and I am considering building one for fungal microscopy.

My question is this - what are the practical resolving limits with an objective/lens/SLR system?

My goal is to be able to photograph some of the 3D structures found on the gills of fungi - basidia, pleurocystidia and spores. These are in the 1 - 100um range and are usually observed mounted in water, under a coverslip with a x100 oil immersion lens and x10 eyepieces (total magnification = x1000) with a microscope. I wish to photograph them in their natural state i.e. not mounted in water under a coverslip.

As far as I understand it, the usual top limit for microscopy without oil is a x60 lens (x10 eyepieces = x600). Higher-magnification lenses require oil but I think this is just to ensure enough light passes through the subject by reducing light loss due to refraction. I wonder if it the same for objective/lens/camera rigs or does the fact that it is reflected light (and possibly more of it) mean that the x600 limit is not the case?

I know that the absolute limit is determined by the Abbe diffraction limit is about 0.25um for conventional microscopy and I suppose the same for camera systems.

I already have a series (x4, 10, 40, 100) of infinity-corrected Nikon BE Plan objectives. From what I have seen on here already, I would need a x10 objective, an objective adapter, a MP-E 65mm macro lens, a Canon camera (the 6D?), a stackshot to mount it on and suitable lighting.

This is the kind of thing I am trying to photograph http://www.photomacrography.net/forum/v ... ht=basidia

Would the x10 objective/lens/camera kit as specified above be powerful enough and what sort of magnification would it produce (compared with a standard x1000 microscope as the benchmark)?

Does anyone fancy putting a mature gill from a shop-bought mushroom into their rig and seeing what comes out before I embark on a very expensive test?

Alternatively, I could stick with the compound microscope but try good quality 60x objectives and reflected light. The problem with this method is a lack of automated rail for focusing.

Apologies in advance if these questions have already been asked on here and I have failed to spot the answers.

Any help greatly appreciated.

Thanks.

John
Last edited by Pitufo on Tue Jun 23, 2015 6:28 am, edited 13 times in total.

Pitufo
Posts: 259
Joined: Sun Jun 21, 2015 4:32 am
Location: United Kingdom

Photos

Post by Pitufo »

Please see below for a few recent photos from my compound microscope. First two are helical threads/spores of a myxomycete. The spores are from a basidiomycete and show spore ornamentation (stained with Melzer's). Spores are approx 10um.
Metatrichia floriformis x 400
Image
Metatrichia floriformis x 1000
Image
Russula ochroleuca spores x 1000 (cropped)
Image

Pau
Site Admin
Posts: 6051
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

Hi Pitufo, welcome aboard!

You are asking some interesting questions on microscopy, all of them have been discussed over the time but I don't recall a comprehensive thread for all them.

First, some clarifications:

- About magnification: you're talking about visual magnification at the microscope with standard 10X eyepieces, while in macro and photomicrography is more usual -and accurate- to refer to the actual magnification on camera sensor, the final magnification will be dependent of your monitor or print size. A 100X objective produces 100X magnification at the eyepiece field stop (the round limit of the image) and 100X on sensor if you place it at that position (10mm lower that the end of the microscope tube in most cases), but 250X if you use the a 2.5X camera adapter pretty usual for FF film or digital SLRs.

- Resolution in microscopy refers to resolving power, the minimum distance that the system is able to differenciate (many people refers resolution to camera megapixels or to final image sharpness but despite related they are very different beasts).
Resolution is related to magnification, but is different: a 100X 1.4 oil plan apo objective has the same magnification but clearly higher resolution than a 100X 0.9 dry metallographic objective. The microscope resolution mostly depends of the objective NA, providing that the subject is adequately illuminated (condenser NA or even diffused light in reflected macro ...) and has enough contrast.

- In a dry objective the maximum NA is 0.95 (it's a physical limit due to the refractive index of air much lower than glass and oil). For higher resolution you need to use immersion, water immersion objectives usualy have up to 1.2 NA and oil up to 1.4 (well, there are some extremely expensive objectives up to 1.51 NA), so you need oil to have higher NA, not only to capture more light.

- A very high NA objective has very short working distance (less than 1mm, often much less) to be able to provide a wide light entrance cone, so it's extremely difficult to use with a not full flat subject. Long working distance objectives (often named LWD, SLWD, ELWD.. and so) of high magnification have pretty modest NA for that reason, for example, the Mitutoyo M plan Apo 100 0.70, and can't deliver a very sharp image when compared with a typical oil immersion objective even a modest one.
Would the x10 objective/lens/camera kit as specified above be powerful enough and what sort of magnification would it produce (compared with a standard x1000 microscope as the benchmark)?
No, you can't have the same resolution with a 10X dry objective than with a 100X oil one, not even close.
Alternatively, I could stick with the compound microscope but try good quality 60x objectives and reflected light. The problem with this method is a lack of automated rail for focusing.

You can automatize the microscope fine focus movement, in fact it will be more precise than a screw driven rail like the Stackshot, there are many examples at the forum, the main issues for you application are working distance and illumination
Does anyone fancy putting a mature gill from a shop-bought mushroom into their rig and seeing what comes out before I embark on a very expensive test?
Take a look:
http://www.photomacrography.net/forum/v ... 4299#94299
http://www.photomacrography.net/forum/v ... =9061#9061
Pau

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Pitufo, welcome aboard! :D

Adding to what Pau has written...

First, let me explain a little more about "numerical aperture" (NA). This single number actually represents the combined effect that angle and wavelength have on achievable resolution. The reason that maximum NA gets so much larger with oil immersion is that the high refractive index (RI) of oil causes the wavelength of light to get proportionally shorter within the medium. For example green light with wavelength 550 nm in air becomes green light with wavelength 363 nm in oil with RI 1.515 . That's because 550/1.515=363. As a result, imaging a spore with green light in oil gives the same maximum resolution as UV light in air, but at much lower cost and risk. When the spore is in water instead of oil, the effect is less pronounced but still significant: 550/1.333 = 416.

Now, for your main question "what are the practical resolving limits with an objective/lens/SLR system?", the answer is that they are identical.

Lots of us like to use "open" systems rather than ordinary microscopes because 1) they offer more options to manipulate and illuminate the subject, and 2) they scale gracefully from very high magnification such as 100X on sensor down to 0.1X with a macro lens.

None of this bears on maximum resolution except that with a typical open rig, it's more difficult to get the full benefit of oil or water immersion because those require a specialized condenser which is also coupled to the slide with oil or water.

You've mentioned that http://www.photomacrography.net/forum/v ... hp?t=20556 shows exactly what you are trying to photograph. That particular image is described as having been shot with an Olympus LWD CDPlan 20 objective. I see from other sources that that objective has NA 0.40, which is typical for low end and long working distance 20X objectives. Your CFI BE 20X would have the same resolution but substantially shorter working distance, only 3.7 mm according to spec.
Alternatively, I could stick with the compound microscope but try good quality 60x objectives and reflected light. The problem with this method is a lack of automated rail for focusing.
This problem is easily fixed by coupling a stepper motor to the fine focus knob of your microscope. See http://www.photomacrography.net/forum/v ... hp?t=11008 for a particularly simple and risk-free approach.

However, to run the experiment it's not necessary to automate the rail. While it's a bit tedious, it's also very practical to shoot focus stacks by manually tweaking the fine focus knob between shots.

I have no idea whether it's valuable for your work, but one nice feature of focus stacking at high magnification is that it allows to make stereo pairs or rocking animations by post-processing a single stack. For a recent example, see my images of fruit fly maggots at http://www.photomacrography.net/forum/v ... 017#171017. Those were shot using a 20X objective, the Mitutoyo M Plan Apo 20X NA 0.42, using the open setup shown at http://www.photomacrography.net/forum/v ... hp?t=27714. But you could get exactly the same results with your microscope. More can be read about this "synthetic stereo" approach at http://zerenesystems.com/cms/stacker/do ... eticstereo.

One last point, the topics by Charles Krebs, referenced at the end of Pau's post, were shot at quite low magnification and small NA, using 2X and 4X objectives. They show much less resolution than what would be produced by using a higher magnification objective like you're thinking. The reference that you give to Ikudzma's Coprinopsis gill is a closer match for you.

--Rik

Pitufo
Posts: 259
Joined: Sun Jun 21, 2015 4:32 am
Location: United Kingdom

Lots to learn!

Post by Pitufo »

Thank you very much for your welcomes and comprehensive replies :D

As you can tell, I am coming to this from the biology side rather than the photography side and therefore some blundering through the new terminology is inenvitable.

There is quite a lot of information to absorb for both your answers and I'm sure I will come back to read them many times..

So, I think my initial approach will be to stick with the microscope for the timebeing and manually adjust focus.

My first little test using a scope with Euromex Plan infinity objectives fitted immediately hit the problems outlined by Pau.

The 10x Achro plan objective (NA 0.25) has a reasonable working distance and I was able to light it from above with an IKEA LED light. The image is OK but the magnification/resolution is not high enough to see the structures I am looking for.

Switching to the next objective (x40 Achro N.A. 0.65) means that the lens is so close to the flat sample of gill that it cannot be lit.

Sooo.. as I understand it.. a decent long working-distance objective would be a good start.. They don't look cheap!

And then some pretty potent illumination would be the next step. Ikea lamps for starters (but then fibre-optic illumination or similar)?

Next on the list would be the stepper-motor, kindly outlined by Rik.

Am I heading in the right direction?

Thank you again - help very much appreciated.

Cheers,

John

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Am I heading in the right direction?
Yes, definitely. To use front illumination at high magnification will require a long working distance objective, and those are not cheap at 20X and above.

One objective that might work nicely with your current setup is the Nikon CF Plan 50X NA 0.55 ELWD inf/0, shown at http://www.photomacrography.net/forum/v ... hp?t=15581. I'm pretty sure that is optically and mechanically compatible with your CFI BE's, so you could just swap it in place of your current 40X and get 8.7 mm working distance. That objective is often available on eBay for a few hundred dollars. I see several there now.

I also see that I glitched in reading your list of objectives, since I wrote about "Your CFI BE 20X" but you don't actually have one of those.

About illumination, another option to consider is flash. A couple of inexpensive Yongnuo flashes positioned close to your subject will give plenty of light with very short effective exposure times (1/5000 second typical). This will freeze out whatever vibration there may be in your setup.

--Rik

Pitufo
Posts: 259
Joined: Sun Jun 21, 2015 4:32 am
Location: United Kingdom

Post by Pitufo »

Thank you Rik.

The Nikon CF Plan 50X NA 0.55 ELWD looks like a good option. I will keep an eye on UK sites to see if one turns up.

It would probably be mounted on my Euromex iScope http://www.euromex.com/gb/catalog/eurom ... ield/8393/ rather than my Nikon E100 (as the latter is binocular).

The Euromex objectives are DIN, so should match the RMS threads on the Nikon CF Plan 50X NA 0.55 ELWD if I understand correctly.

The trinocular head has currently got a c-mount 0.5x tube and 5MP USB cam on it. Would I get better images replacing this with an SLR adapter (x2.5?) and Canon SLR (Model 6D?)?

Apologies for all the very basic questions but I don't want to make any expensive mistakes!

Thank you again - your help is very much appreciated.

John

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Is that made for "infinite" objectives??

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

"DIN" can mean a lot of different things. (Basically it just means "Deutsches Institut für Normung e.V.", the name of a German standardization institute. There are DIN standards for everything from concrete on up.)

In microscopes, "DIN" used to most commonly mean a finite objective, designed for use with a 160 mm tube length, having RMS threads and a 45mm parfocal length.

Hopefully with your scope, it means an infinite objective, also with RMS threads and a 45mm parfocal length, since those are the specifications of the Nikon CF Plan 50X NA 0.55 ELWD. In addition there may possibly be an issue with aberration correction, if the Euromex iScope is designed to do some of that correction in the tube lens.

Before purchasing the Nikon CF Plan 50X for use in the iScope, I suggest to move one of your Nikon CFI BE objectives to the iScope and confirm that it works properly there.

You can also check the markings on the iScope objectives. If they say something like "160/0.17", then they're finite and that's bad for your plan.

--Rik

Edit: insert missing word
Last edited by rjlittlefield on Tue Jun 23, 2015 3:19 pm, edited 1 time in total.

Pitufo
Posts: 259
Joined: Sun Jun 21, 2015 4:32 am
Location: United Kingdom

Post by Pitufo »

Excellent idea - I will switch the objectives tomorrow and see what happens. The Euromex objectives are infinity-corrected, so in theory.... :)

Do you think switching the adapter/camera would make a big difference?

Pau
Site Admin
Posts: 6051
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

The trinocular head has currently got a c-mount 0.5x tube and 5MP USB cam on it. Would I get better images replacing this with an SLR adapter (x2.5?) and Canon SLR (Model 6D?)?
Yes and no: The DSLR sensor is likely much better (color, noise, dinamic range, ability to shot RAW, more megapixels...) and for low magnification it can capture more detail but for high magnification you will be more limited by the objective NA and rarely you can resolve more details than with the 5mpx camera. At high magnification microscopy you easily reach the resolution limit and you enter in the empty magnification territory, when more magnification doesn't allow for more actual resolution but for a larger and less sharp image.

That said, I personally much prefer the DSLR. I've tested the 6D at my microscope and it's excellent.
Pau

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

The Euromex objectives are infinity-corrected, so in theory...
This is good... probably RMS threaded and 45mm parfocal. (It would be nice to know the tube lens focal length... probably 200mm or 180mm).

The Nikon CF Plan 50X NA 0.55 ELWD Rik mentioned would be good choice with its 8.7mm working distance.

For the smallest subjects in your stated desired range the 0.55 NA might not get you the detail you want. The problem is that going to a significantly high NA to gain resolution almost always means a working distance that makes surface illumination very difficult or impossible. (Except for perhaps the expensive 100/0.70 Mitutoyo with a WD of 6mm. But that has different thread mount and parfocal distance so it would require a different microscope or equipment set-up. And there is still more resolution possible with a 0.80 to 0.95 "M" type objectivers, but the working distances are ridiculously small).

One thing to consider (perhaps in the future if you find you need more resolution) is a a vertical or "epi" illuminator used with an M-type objective in the 0.80-0.95 NA range. You might want to check with your equipment dealer to see if they have a vertical illuminator that can be used with your microscope. (Perhaps one from their metallurgical series might fit). The light from such an illuminator can be aesthetically "lacking" since it is quite frontally direct, but it is sometimes the only practical way to get high NA shots of opaque subjects. (I suggest you seriously consider this only if you eventually determine that a 50.0.55 can't handle your smallest subject features).

Frankly, with the 50/0.80 or 100/0.95 "M" type objectives I have it is imperative to have a pretty "flat" subject. If your subject is too "3-dimensional" you may actually hit the higher portions of the subject with the objective as you focus down to obtain an image stack. And it can be extremely difficult to get a clean stacked result with many 3D subjects. But since you are at a stage right now where you seem to be assessing your options, it is worth keeping this in mind.

Here are a few examples using an epi-illuminator, some with included scale bars which might be helpful:

http://www.photomacrography.net/forum/v ... php?t=6561
http://www.photomacrography.net/forum/v ... php?t=8711
http://www.photomacrography.net/forum/v ... hp?t=10493
http://www.photomacrography.net/forum/v ... php?t=6566

Pitufo
Posts: 259
Joined: Sun Jun 21, 2015 4:32 am
Location: United Kingdom

Post by Pitufo »

Excellent - thanks to all for your generous help.

I have a good idea where to start and the path to follow - and a much better understanding of the fundamentals behind it all. :D

Good news too regarding compatibility - the Nikon BE Plan objective fits my iScope (which has the advantage of LED, Kohler and trinoc head over the E100) and gives a sharp image.

Pitufo
Posts: 259
Joined: Sun Jun 21, 2015 4:32 am
Location: United Kingdom

Post by Pitufo »

Your advice has already paid off as I managed to take a couple of photos using my Nikon BE Plan x 10 mounted on my iScope today (with IKEA lighting from above and a bit of back lighting.

Much improved over my previous attempts at these Comatrichia sporangia using the stereomicroscope

Image
Image

I think I have located a potential CF plan x 50 ELWD, so will probably invest in one.

One thing I thought might be quite useful is some kind of LED ring light to go around the objective (a bit like a small version of a stereomicrocope ring LED). Does such a thing exist commercially or has anyone successfully bodged one? Apologies if I have missed a post which contains this info.

I have found reference to a few larger ones (60mm) such as Angel Eye Headlights http://www.photomacrography.net/forum/v ... t=led+ring- but I think I need about 25mm diameter to fit an objective.

This sort of thing may be OK (without the disco colours) https://www.adafruit.com/products/1463

It appears there are some commercial ones, but crazy prices http://warneronline.com/img_lg/720246,m ... glight.jpg

I may have answered my own question here http://www.aliexpress.com/item/42-28-mm ... 62585.html and here http://www.microscopesmall.com/microsco ... LED28.html I'm not sure if they are powerful enough though..
Last edited by Pitufo on Thu Jun 25, 2015 1:08 pm, edited 3 times in total.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

I'm happy to see such rapid progress. This is great!

The usual trick for illuminating around an objective is to use a wrap-around diffuser. See http://www.photomacrography1.net/forum/ ... 4478#24478 for illustration of the approach.

Probably the most frequently used diffuser is half of a ping-pong ball, with a hole cut it in for the objective to fit through.

Although that link shows the diffuser being lighted with a fiber optic illuminator, it works almost as well to use either flash units or LED desk lamps. (Ikea Jansjö are popular.) The advantage of a fiber optic illuminator is mostly that it allows finer control over the distribution of light, which is important for artistic purposes, not so much for scientific. Again, flash has the advantage of freezing out vibration, which may become important at higher magnifications with some cameras.

By the way, when you switch to diffused illumination, you may notice a striking difference in the appearance of "sparkly" artifacts around shiny parts of the subject. See HERE for some discussion of why that is. There's a more vivid demonstration HERE.

--Rik

Post Reply Previous topicNext topic