meta equipment choices in 2021

Just bought that first macro lens? Post here to get helpful feedback and answers to any questions you might have.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

tkphot
Posts: 8
Joined: Wed Mar 31, 2021 1:07 pm

meta equipment choices in 2021

Post by tkphot »

First of all, thanks to this community - I've learned a lot in reading forum posts over the past few days.

I'm looking to create a macro photography setup to acquire images of small fossilized insect inclusions in amber pieces, along with other small plant, mineral, and insect samples (budgeting around $500-$2k, depending on what it would take). From what I have learned here it seems the best technique is to set up a motorized rail for focus stacking, and use a microscope objective attached to a DSLR. I was hoping some folks could answer a few questions about equipment for such a setup:

1) Is there any major advantage/disadvantage for choosing the other approach - buying a used major name microscope and attaching a DSLR to the trinocular mount?

2) I currently have an old Canon t2i camera with 18 MP - is it worth investing in a better camera with higher resolution / better sensor or some feature that would be useful? I wouldn't mind an excuse to upgrade, just not sure what it would get me

3) In your opinion, what are the "meta" top 2-3 options for these categories, to get me pointed in the right direction on equipment purchases for a fresh build:
-DSLR camera
-Motorized rail for focus stacking
-2x, 5x, 10x+ microscope objectives (I am comfortable with image stitching and stacking, but not sure how high of magnification would be useful/feasible/affordable)

Thanks for your help and recommendations!

Lou Jost
Posts: 5401
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: meta equipment choices in 2021

Post by Lou Jost »

I spend a lot of time on photographing specimens submerged in 4-6mm of water, and this is optically almost the same problem as shooting inclusions. The answer to your questions depends very strongly on the magnifications you need. Low-magnification photography (say, 2x) is undemanding and requires no special equipment, but the higher the magnification, the shallower your inclusions will have to be. If you need to shoot inclusions deeper than 2-3mm at magnifications around 5x or higher, for best results you should use lenses or objectives that are designed to shoot through glass. There are a number of options for that. Mitutoyo (G-series) and Olympus and some others make objectives designed to shoot through LCD glass. These are very good though they are each designed to shoot through a specific thickness. There are also many 20x objectives with coverslip correction collars, including some that can be adjusted to any coverslip thickness between 0 and 2mm. These are very good.

Reversed Micro-Four-Thirds lenses are also designed to shoot through about 4mm of glass, so these are great choices for lower-magnification work.

I strongly prefer a microscope with a WeMacro Micro-Mate to turn the fine-focusing knob of the scope. This is very sturdy and stable.

However, if your desired m is not too high (say, 5x or less) and your subjects are not too thick, it is a zillion times faster and easier to use focus stacking rather than camera movement. You don't need a microscope or rail or MicroMate or anything else, and your stacks will be almost instantaneous. You would need to buy a camera and tube lens that does focus stacking, and an infinity-corrected microscope objective, or reversed MFT lens. You would also need a geared manual rail to position the camera precisely.

tkphot
Posts: 8
Joined: Wed Mar 31, 2021 1:07 pm

Re: meta equipment choices in 2021

Post by tkphot »

Thank you for the info and recommendations!

I can see how our setups would be similar - I believe the refractive index of Amber is around 1.54 so glycerol would be a decent immersion media if I went that route. Immersion lenses seem pretty expensive, but would probably give the best results. I also plan to shoot botanical samples, so I'll probably start with an air objective, 5 or 10x.

I really like the technique of focus stacking and hope to work towards 20x or higher at some point in the future, but will also probably buy a decent regular macro lens at some point for 1-5x magnification.

Any idea how I can determine what sort of camera resolution I should aim for to match the optical resolution of something like a 10x 0.28 NA objective?

Also looking for info on camera features/brands that might make working on focus stacking macro photography easier

Lou Jost
Posts: 5401
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: meta equipment choices in 2021

Post by Lou Jost »

"Immersion lenses seem pretty expensive, but would probably give the best results."
I think you might have misunderstood what I was saying or I may have misunderstood what you said. I was not suggesting immersion lenses, which usually have short working distances that would not be able to focus on an inclusion. I was suggesting working in air with objectives that are designed to shoot through glass, which is optically similar enough to amber.
"I really like the technique of focus stacking and hope to work towards 20x or higher at some point in the future"
Stacking is essential no matter what you get. I was speaking specifically of stacking by changing the focus of the tube lens, as opposed to stacking by using a rail to move the whole camera. This requires an informed choice of camera and tube lens. This technique does not work well for three-dimensional objects at magnifications higher than about 10x because there is not enough focus throw in most tube lenses to cover the whole depth of the subject.

tkphot
Posts: 8
Joined: Wed Mar 31, 2021 1:07 pm

Re: meta equipment choices in 2021

Post by tkphot »

I've read that stacking using a rail (moving the camera) is better for above 5x mag, and was planning to get this raynox 150 tube lens setup with mitutoyo objective 5x or 10x to start (https://www.wemacro.com/?product=raynox ... e-lens-pro)

Later on I will probably get a good normal macro lens for under 1-5x mag / shooting in the field.

There are a lot of cameras on the market and they all seem great. I'm having a hard time figuring out if there are any features/brands that stand out for macro shooting, and what resolution I should be aiming for ( I suspect even though the microscope objectives have high optical resolution, at some point I would just be gaining empty magnification by going with higher MP, maybe I'm wrong)

Thanks again

Lou Jost
Posts: 5401
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: meta equipment choices in 2021

Post by Lou Jost »

I've read that stacking using a rail (moving the camera) is better for above 5x mag
Yes, that's right.
There are a lot of cameras on the market and they all seem great.
You should get one that can do focus-stacking.
what resolution I should be aiming for
Get the highest resolution you can afford, so you can have it when you need it (a good macro lens can outresolve most sensors).

rjlittlefield
Site Admin
Posts: 22519
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: meta equipment choices in 2021

Post by rjlittlefield »

I see that Lou has also answered, while I ate dinner with this posting in draft. His bottom-line guidance will guarantee that you have plenty of resolution for microscope objectives, which put less information on sensor than a good macro lens does.

But since I have this written already, I'll hit Submit so it goes in the record.
tkphot wrote:
Fri Apr 02, 2021 1:28 pm
Any idea how I can determine what sort of camera resolution I should aim for to match the optical resolution of something like a 10x 0.28 NA objective?
The standard rule for "matching" is that the sensor's pixel pitch should be just enough to reach the Nyquist sample limit of 2 pixels per cycle, at the same spatial frequency where diffraction cutoff gives MTF=0 no matter what else you do, for green light.

The algebra goes like this:
wavelength of green light, lambda = 0.00055 [in units of mm/cycle]
effective f-number Feff = magnification/(2*NA) [dimensionless]
cutoff frequency nu_0 = 1/(lambda*Feff) [in units of cycles/mm]
pixel_pitch = 1/(2*nu_0) [in units of mm]

For 10X NA 0.28, the numbers work out as
Feff = 10/(2*0.28) = 17.86 [that is, f/17.86]
nu_0 = 101.8 cycles/mm
pixel_pitch = 0.00491 mm = 4.91 microns

To get megapixels, the formula would be
megapixels = (sensor_width*sensor_height) / (pixel_pitch^2) / 1000000

Assuming a nominal APS-C sensor size, the resulting number would be
megapixels = (25.1*16.7) / (0.00491^2) / 1000000 = 17.4 megapixels

Note that this requirement is not very challenging for modern cameras. The reason is because microscope objectives give small effective apertures (large effective F-numbers), which means they get hit hard by diffraction.

You'll have to re-work the numbers if you change the sensor size or run the objective at a different magnification, say by changing the focal length of the tube lens with an infinity objective. For example it turns out that a Mitutoyo 10X NA 0.28 can fill a full-frame sensor, 36x24mm, while still running at 10X. In that case the megapixel requirement computes as
megapixels = (36*24) / (0.00491^2) / 1000000 = 35.8 megapixels

With a smaller sensor, say Micro Four-Thirds, you'll probably also want to push down the magnification. In the end it will work out the same, that any modern sensor will have enough pixels to capture whatever image the objective can form.

--Rik

tkphot
Posts: 8
Joined: Wed Mar 31, 2021 1:07 pm

Re: meta equipment choices in 2021

Post by tkphot »

Thank you Lou and Rik, finally starting to narrow down my parts list

The equations help a lot in understanding this - I was reading about nyquist sampling last night but didn't know where to go from there

rjlittlefield
Site Admin
Posts: 22519
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: meta equipment choices in 2021

Post by rjlittlefield »

tkphot wrote:
Fri Apr 02, 2021 9:51 pm
The equations help a lot in understanding this - I was reading about nyquist sampling last night but didn't know where to go from there
I should have linked to my source: https://www.photomacrography.net/forum/ ... 831#124831 .

But of course there is more background. The formulas that I've given correspond to Nikon's recommendations at https://www.microscopyu.com/tutorials/m ... resolution . It took me a long time to figure out what their recommendation really means. The answer is that it provides just barely enough pixels to meet the Nyquist limit (2 pixels per cycle) at a level of detail where MTF=0 anyway. So it's like reaching two limits at the same time. Having only two pixels per cycle does not provide good sampling, for the reasons explained at https://www.photomacrography.net/forum/ ... php?t=2439 . But on the other hand, having only two pixels per cycle at diffraction cutoff, where MTF=0, also means having four pixels per cycle at half that frequency, where the MTF will be around 40%. So another way of looking at Nikon's recommendation is that it guarantees good sampling for detail that is coarse enough to be well transmitted by the optics, while also guaranteeing that no detail is totally lost because of inadequate sampling.

You can always improve the quality of capture by having more pixels. The rule that I outlined is the minimum that you need to have, to not short-change the optics.

--Rik

Lou Jost
Posts: 5401
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: meta equipment choices in 2021

Post by Lou Jost »

You can always improve the quality of capture by having more pixels. The rule that I outlined is the minimum that you need to have, to not short-change the optics.
So there really is no need to worry about Nyquist etc, just get the most pixels you can afford. You plan on getting a dedicated macro lens, and good ones will outresolve standard sensors. You can alsways downsample your images to save disk space, but you can't upsample them.

ChrisR
Site Admin
Posts: 8637
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Re: meta equipment choices in 2021

Post by ChrisR »

Start with the light
Continuous or flash?
Continuous is better for control but invokes serious vibration issues. Those get complicated by what sort of shutter you have. You need investment in a stable platform, unless you live in a mountain.
Flash may kill vibrations but you can't always use it with tethering or auto focus-stepping modes in the camera. Such as, stacking by altering a tube lens focus. You may need hundreds of images in a stack, and more than one flash unit. Doable but not trivial.

Do look hard at the shutter on the camera. You probably don't want a mirror. You probably DO want an electronic-only shutter option but again, will that trigger flash? Are you constrained by the manufacturer's software, to drive it? Are you going to wear anything mechanical out before you want to upgrade anyway?

Lenses - At 5x say, there's nothing much which I can think of that's readily available at any price, and certainly not up to medium price (hundreds of dollars/pounds/euros) which is ideal. Macro/bellows lenses are at their limit. Microscope objectives don't cover very well. Some are happy with say a Mitutoyo 5x on APS-C, others aren't. You have to be realistic about your requirements here - inclusions probably don't often stretch to the corners of a frame. You could OTOH use a Nikon BE 4x NA 0.1, which for $50 is pretty good, and be happy. (There are similar at 10x).

Resolution isn't everything. You might get a 4x NA 0.2 objective which is sharp as heck in the middle, but introduces a lot of longitudinal CA (tends to come out in stacking but leaves purples fringes) or lateral CA (which doesn't, but is perhaps more addressable in post processing).


How big an image do you need? If there was ever an 80/20 rule it applies here. I don't think I've ever needed a single image which was limited by an outresolved sensor of a mere 12MP. I mean, that's quite a big print. If you're going for awards for comparisons with others' 100% enlargements on 40MP camera, then it's a different game, but remember they're less than twice as demanding, in a linear scale.

How clear are your subjects? Are any views of things buried in amber, all that clear?

Base - microscopes are best over say 5x, but many folk here construct a hybrid. Some lenses e.g. Mitutoyos don't go on any ordinary microscope. Below about 5x, microscopes aren't very good. Working distances and NA's are small.

So, it's easy really. Let me know when you have the perfect solution!! It's even worse when you have legacy hardware to think of, and photos of other than your current passion, to dream of perfecting.
Chris R

tkphot
Posts: 8
Joined: Wed Mar 31, 2021 1:07 pm

Re: meta equipment choices in 2021

Post by tkphot »

Thank you all again, I have learned a lot from your thoughts. Ive attached a picture of an example insect inclusion in amber (taken by the seller with cheap cell phone macro clip on), most of the insects I have are 1-4mm in length and are fairly close to the surface. The amber seems fairly clear, and will become more so when I remove some of the lacquer that came on the surface. The beetle in the picture below is from the cretaceous period ~95 MYA !

Aside from amber inclusions, I plan to use this new camera for more general macro work and nature/travel shooting. I've increased my budget for this since it's been about 10 years since I've had a camera upgrade =) I am almost set on getting the Sony a7r III or IV, but not sure if it makes sense to consider Canon over Sony (for tethering / remote PC control purposes, or other features I haven't considered).

You all said to go for more pixels - is 61 MP (higher with pixel shift) going to be super overkill if I eventually pair it with something like a Mitutoyo 5x or 10x? For regular non-macro shooting I would be fine with the a7r III's 42 MP, but for extreme macro work my goal is to tease out as much detail as I can. If the extra 20 MP isn't going to get me much as I think, I might be better off spending the extra money on something like an mjkzz stack and stitch (otherwise I plan to get a more basic wemacro motorized rail)

Thanks again
coleop.jpg
ChrisR wrote:
Mon Apr 05, 2021 7:58 am
Lou Jost wrote:
Sat Apr 03, 2021 6:20 am
rjlittlefield wrote:
Fri Apr 02, 2021 10:39 pm

CharlesT
Posts: 18
Joined: Sat Feb 01, 2020 9:02 am

Re: meta equipment choices in 2021

Post by CharlesT »

Speaking as someone who operates at the lower-quality end of the macro world (it is still lots of fun!) I would suggest that your camera is pretty usable, though I have found that a twist and flip rear screen is very useful and just about essential if you are thinking that you might want to pop the camera onto/into the top of a microscope (without having to tether it to a phone or PC). And if you do think of getting a new camera ponder on whether you would like to be able to use a plug-in shutter release cable.
If its any help I use a short focal length macro lens... most people advise a long reach lens but the light weight of a short one [and its near-normal focal length] means that it is on the camera body most of the time.
[The 35mm efs macro, if you stick with canon APSC ]
The one advantage you have got is that your subjects are not going to move about ;-)
Canon 600d
Watson Service 1
Beck Epimax

Lou Jost
Posts: 5401
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: meta equipment choices in 2021

Post by Lou Jost »

I'll just repeat myself, GET MORE PIXELS, you will never regret it and you will still outresolve the sensor with good lenses. See Beatsy's recent tests. If those don't make you want to get an A7R4, I will be surprised.

viewtopic.php?p=272808#p272808

ChrisR
Site Admin
Posts: 8637
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Re: meta equipment choices in 2021

Post by ChrisR »

tkphot wrote:
Mon Apr 05, 2021 8:26 pm
You all said to go for more pixels -
With respect, you have misread.

If want to multiply your budget by 3 or many more, and spend a while getting everything else right, then perhaps spend more time peeping at pixels rather than producing useful images - the answer is yes. Go for the highest number of pixels you can find.
In the quest for great image it's the weakest link which shows, not the strongest.
Chris R

Post Reply Previous topicNext topic