Lighting for macro photography of fern gametophytes
Moderators: rjlittlefield, ChrisR, Chris S., Pau
Olympus LMPlanFl objectives are infinity-corrected and need a tube lens for macro.Lou Jost wrote:Do those Olympus lenses need an eyepiece or projection lens?
I think Charlie Krebs uses Olympus SPlanApo objectives on his microscope, and these need an NFK photo eyepiece for photomicrography.
Alan Wood
At the BH microscope you're right, but he also uses infinite corrected UIS2 plan fluorite objectives at his hybrid Nikon MNI think Charlie Krebs uses Olympus SPlanApo objectives on his microscope, and these need an NFK photo eyepiece for photomicrography.
http://www.photomacrography.net/forum/v ... 389#144389
Pau
If I used those kind of objectives with a tube lens, would it be okay that I have this horizontal focus block arrangement? All the good images I've seen were taken with a proper microscope, but I really would prefer to use something that can be taken places to show to people.
I think they funding body would be much more willing to fund the objectives that Das Callaghan and Charles Krebs have been using as they have got good results that I can actually show. Thanks!
I'm going to move my setup to a permanent location and mess around a bit with my stack sizes and the movement of the block to see if I can get smaller cleaner stacks. I'm also going to try some very different ideas about lighting.
I think they funding body would be much more willing to fund the objectives that Das Callaghan and Charles Krebs have been using as they have got good results that I can actually show. Thanks!
I'm going to move my setup to a permanent location and mess around a bit with my stack sizes and the movement of the block to see if I can get smaller cleaner stacks. I'm also going to try some very different ideas about lighting.
I have set up my equipment in a permanent location now so I don't waste time setting getting it out of the cupboard and setting it up for each shot.
My previous images were fairly thick specimens with a 1000 slices each on f/2.8 with a 5um movement between slices.
The image below was 15 slices with a 10 um movement of the camera between slices. I used MP-E only with no teleconverter on x5, f/4 and 1/100 second exposure. There are two flashes, with no diffusion, sitting on either side of the camera body, pointing diagonally at the subject (6 inches away). The subject is also lit from the side by a window.
This is just a test to see if the image is sharper if I use thicker slices and a much much smaller number of slices. I'm finding that it's very hard to get really thin specimens, so I must work on that.
slice
stack:
50%
100%
My previous images were fairly thick specimens with a 1000 slices each on f/2.8 with a 5um movement between slices.
The image below was 15 slices with a 10 um movement of the camera between slices. I used MP-E only with no teleconverter on x5, f/4 and 1/100 second exposure. There are two flashes, with no diffusion, sitting on either side of the camera body, pointing diagonally at the subject (6 inches away). The subject is also lit from the side by a window.
This is just a test to see if the image is sharper if I use thicker slices and a much much smaller number of slices. I'm finding that it's very hard to get really thin specimens, so I must work on that.
slice
stack:
50%
100%
AbsolutelyIf I used those kind of objectives with a tube lens, would it be okay that I have this horizontal focus block arrangement?
I doubt it - you're excluding a lot. For a kickoff, all of Rik's I think.All the good images I've seen were taken with a proper microscope,
The image you just cited by Charles Krebs proves that it isn't the lens sharpness which makes the difference. It looks great but it's only 1000 pixels wide.
If you run the numbers, because of the 2.5x eyepiece it's 25x, and the EA is f/50 (working in a way like your converter). F/50 is about OK for 1000 pixels wide with a circle of confusion of a couple of pixels.
(Part of) what Charles has done is control the light very well, it would have been easy to let the highlights burn out using top light. He's let the light diffuse through the leaf, and the edges which would have been highlights with top light, are dark instead.
In your recent pictures, the detail you've zoomed in on shows the ridges as strong highlights (check the histogram). That comes from the small light source you used.
The risk you run with a small light source is that the light from it bounces particularly into a small part of the lens. That gives you a smaller effective aperture. We've come to use the term utilised aperture, for it when it occurs that way. The result is the same. A large bright blob, in focus in many frames in a stack, but without much resolution.
It does look sharp at small picture sizes.
If that's what you want to show up some structure, that's fine, but pictorially it may not be. If sharpened they can look horrible!
Chris R
This seems better. It's a raspberry, and I set the focus block to move 100um instead of 10um. I realised that the tiny movement was hardly changing the pane of focus relative to the fairly deep plane of focus.
I used two flashes with plastic diffusers and double layers of kitchen towel to make a big white flash.
Flashes are almost level with the camera body
MP-E x5
f/4
1 second exposure on camera and flash.
This was only 10 slices but because the camera was moving so far between each slice I got a much larger area of focus in the stack.
Setup:
This was still a bit dark but had levels adjusted in photoshop in the stack:
I used two flashes with plastic diffusers and double layers of kitchen towel to make a big white flash.
Flashes are almost level with the camera body
MP-E x5
f/4
1 second exposure on camera and flash.
This was only 10 slices but because the camera was moving so far between each slice I got a much larger area of focus in the stack.
Setup:
This was still a bit dark but had levels adjusted in photoshop in the stack:
Last edited by jsp on Sun Aug 21, 2016 10:19 am, edited 1 time in total.
-
- Posts: 187
- Joined: Tue Jan 20, 2009 2:59 pm
- Location: Berwickshire, Scotland
This is starting to look much better.
The exposure is still a little low. Another half stop or so on top of what you have already added in post would do no harm. Best to get that right by playing with the flash settings.
The light is still a bit hot for my taste. Can you add a styrene cup as well as the kitchen towel?
I am still not understanding why you do not use the maximum flash sync speed on the camera? You are not gaining anything with that 1 second exposure and may be losing some sharpness.
The exposure is still a little low. Another half stop or so on top of what you have already added in post would do no harm. Best to get that right by playing with the flash settings.
The light is still a bit hot for my taste. Can you add a styrene cup as well as the kitchen towel?
I am still not understanding why you do not use the maximum flash sync speed on the camera? You are not gaining anything with that 1 second exposure and may be losing some sharpness.
Thanks, I'm glad it's heading in the right direction. :-) I'll try the things you suggest on my next go.
I did try to use 1/250 second exposure as you suggested, but the bottom part of the frame came out black. The only way I can find to increase the power of the flash is to turn it up from 1/64 to 1/32 and then on to 1/1 which is the maximum. I assumed that that meant that the flash was then firing for one second and that to make the most of available light I would need to have the camera shutter open for 1 second. Is that right? I had a good rummage through the flash manual but I wasn't really any the wiser for it.
Thanks!
I did try to use 1/250 second exposure as you suggested, but the bottom part of the frame came out black. The only way I can find to increase the power of the flash is to turn it up from 1/64 to 1/32 and then on to 1/1 which is the maximum. I assumed that that meant that the flash was then firing for one second and that to make the most of available light I would need to have the camera shutter open for 1 second. Is that right? I had a good rummage through the flash manual but I wasn't really any the wiser for it.
Thanks!
Jennifer,
I suggest you try the approach with paper towels again, but after moving the paper towels much closer to the raspberry, farther from the flashes. Try flashing your lights a few times, observing by eye how much of the paper toweling is evenly lit. Then take pictures (don't bother stacking at first), and learn to correlate what portions of the paper towel are lit, vs. the lighting appearance of the raspberry. If you want evenly diffuse light on the berry, you need to apply even light to its entire "sky"--imagining that you are the raspberry and looking out at the world in the direction of the camera.
Also, I don't think your try with the polystyrene cup was so very far off. I agree with Pau that you want to make some adjustments, but rather small adjustments could reap big gains. For those images (of the fern), I'd add that a splash of backlighting would help a lot. Thin botanical subjects are often imaged best with a combination of back and front lighting. The backlighting makes them glow a bit, and provides visual information about their internal structure; the frontlighting acts as fill light, and provides visual information about the plant's surface.
This said, a one-second camera shutter speed won't hurt your images if your studio is very dark. The way to determine if you have a problem is to take a picture at your selected shutter speed, without any flash. If the resulting image is completely black, room light is not contributing to--and thus potentially blurring--your image. If you see any image formation at all, room light is polluting--and potentially blurring--your image.
--Chris S.
I suggest you try the approach with paper towels again, but after moving the paper towels much closer to the raspberry, farther from the flashes. Try flashing your lights a few times, observing by eye how much of the paper toweling is evenly lit. Then take pictures (don't bother stacking at first), and learn to correlate what portions of the paper towel are lit, vs. the lighting appearance of the raspberry. If you want evenly diffuse light on the berry, you need to apply even light to its entire "sky"--imagining that you are the raspberry and looking out at the world in the direction of the camera.
Also, I don't think your try with the polystyrene cup was so very far off. I agree with Pau that you want to make some adjustments, but rather small adjustments could reap big gains. For those images (of the fern), I'd add that a splash of backlighting would help a lot. Thin botanical subjects are often imaged best with a combination of back and front lighting. The backlighting makes them glow a bit, and provides visual information about their internal structure; the frontlighting acts as fill light, and provides visual information about the plant's surface.
This is likely due to slightly mismatched timing between your flash and camera. Lengthening your camera's shutter speed to 1/200 second, 1/180 second, or 1/150 second will likely get rid of the problem.jsp wrote:I did try to use 1/250 second exposure as you suggested, but the bottom part of the frame came out black.
No, not right at all! I don't know your particular flash, but typically, its output duration might be around 1/1000 second at full power, and 1/30,000 second at 1/128 power. Regardless, just set your camera's shutter speed as high as it can go without causing the bottoms of your frames to go black. As said above, this is likely somewhere between 1/150 second and 1/200 second.jsp wrote:The only way I can find to increase the power of the flash is to turn it up from 1/64 to 1/32 and then on to 1/1 which is the maximum. I assumed that that meant that the flash was then firing for one second and that to make the most of available light I would need to have the camera shutter open for 1 second. Is that right?
This said, a one-second camera shutter speed won't hurt your images if your studio is very dark. The way to determine if you have a problem is to take a picture at your selected shutter speed, without any flash. If the resulting image is completely black, room light is not contributing to--and thus potentially blurring--your image. If you see any image formation at all, room light is polluting--and potentially blurring--your image.
--Chris S.
I find it much easier to go too flat with the lighting rather than starting too hard.
Then you can move a flash closer to the bright-side diffuser to harden it up a bit,
If you take pictures with just one flash you can see what each one is contributing.
Try with the shadow a couple of stops darker than the brightest.
Flash sync speed on the 5DMk2 is 1/200th.
Then you can move a flash closer to the bright-side diffuser to harden it up a bit,
If you take pictures with just one flash you can see what each one is contributing.
Try with the shadow a couple of stops darker than the brightest.
Flash sync speed on the 5DMk2 is 1/200th.
Chris R
I have had a go with the room blacked out to minimal lighting.
The settings are:
f/4
1/200 exposure in camera
two flashes set to 1/1
MP-E only, 5x magnification.
The focus block moves 200um each slice, which I thought while doing it worked well for producing slightly overlapping areas of focus with the angle that the sample was sitting at. Looking at the stack, I'm not quite so sure.
The sample is a piece of cabbage with the cuticle peeled off.
I forgot to take the UV filter off (oops).
This is a shot of the ambient light on the sample with the flashes turned off. It seems like no light is leaking in:
This is the setup. It still needs a lot of work, but it seemed to be a bit of an improvement going by the histogram. The flat sheet on top is a silver reflector, and the flash on the right is turned off.
This is the histogram of a slice:
This is a slice. I resized the image so I'm not uploading huge files.
This is the stack:
This is the 50% version:
This is it at 100%
The specimen isn't nearly as interesting looking as I was hoping, but it does show up the fuzziness quite ruthlessly, which must be a good thing.
It's not really as sharp as I hoped when I zoom in, though the whole stack on screen without zooming in does look okay-ish.
The settings are:
f/4
1/200 exposure in camera
two flashes set to 1/1
MP-E only, 5x magnification.
The focus block moves 200um each slice, which I thought while doing it worked well for producing slightly overlapping areas of focus with the angle that the sample was sitting at. Looking at the stack, I'm not quite so sure.
The sample is a piece of cabbage with the cuticle peeled off.
I forgot to take the UV filter off (oops).
This is a shot of the ambient light on the sample with the flashes turned off. It seems like no light is leaking in:
This is the setup. It still needs a lot of work, but it seemed to be a bit of an improvement going by the histogram. The flat sheet on top is a silver reflector, and the flash on the right is turned off.
This is the histogram of a slice:
This is a slice. I resized the image so I'm not uploading huge files.
This is the stack:
This is the 50% version:
This is it at 100%
The specimen isn't nearly as interesting looking as I was hoping, but it does show up the fuzziness quite ruthlessly, which must be a good thing.
It's not really as sharp as I hoped when I zoom in, though the whole stack on screen without zooming in does look okay-ish.