Wing scales

[jin]

Hi, I tried shooting skipper butterfly's wing scale but not happy with the sharpness and I need help to tell what could be wrong or what need to be done in order to best I can with the setup.

EQ info:
Olympus BHM with ~170mm tube with Nikon 10x, 160/- WD5.6 objective and 450D wired remote, live view and 580EX flash, 110 stacks at 2um interval.

Below is complete pic with sharpening:


the attache image is a unprocessed, 100% resolution, small crop of upper corner area of the shot for your reference:

[rjlittlefield]

1. Is the center of the field sharp? (Show us a crop, please.)

2. Do you have any other optics in the system, or is it just direct projection from the objective onto the sensor?

--Rik

[Charles Krebs]

A couple of questions and thoughts come to mind...

Which Nikon 10X objective is this? There are dozens out there with various specifications. (A picture might be best). Your sensor format diagonal is about 27mm. Most finite Nikon (160mm tube length) objectives were intended to provide, at best, a quality image of about 20mm diagonal. Certain ones that were intended for use with super wide-field heads were designed to provide a quality image circle of about 26-27mm. That's not to say it is not possible to get a good image out to the corners with a non-super wide-field design, but it is not to be taken for granted, and must be tested to see how well it works.

When you say "~170 tube" are you referring to the distance from the objective threads to the camera sensor? The objective was designed for a 150mm distance in that regard, but using slightly more (like 170mm) might actually be better if the corner quality is not that great. Compare the center to a corner. The center almost always will look somewhat better, but the difference should not be extreme. If the center is good but all four corners are not, it is most likely a case where the objective is not providing an adequate size image.

... live view and 580EX flash...

Flash should eliminate most vibration problems (if the power level is turned down resulting in a very short flash duration). But even with flash it is still a very good idea to use technique that will minimize any potential sources of mechanical vibration. Are you shooting directly from from live view? If so, you should be aware that when doing this with a Canon flash attached, the shutter must first close and then re-open to start the exposure. From an "anti-vibration" standpoint is is better to not shoot directly from live-view when using flash. After getting everything set, turn off live view. Use the camera mirror lock-up. For each shot release the mirror up, pause, then take the picture. This way you only have one mechanical shutter motion before the flash fires.

If you are working in a room where you can turn down the ambient light levels it is also possible to set the flash to "second-curtain sync" and use a long exposure (such as 1/2 or 1 second) which then would allow any shutter vibration to dampen out before the actual flash exposes the picture. (Just be sure room light levels are low enough to avoid registering an "ambient light image" on the sensor. All registering exposure light should come from the flash).

110 stacks at 2um interval

This is "overkill" for a 10X objective. The steps can be larger and fewer images used. (This should not be the cause of any "unsharpness" you are seeing, but you are making more work for yourself). It can vary somewhat depending on the NA of the objective, but at 10X you are certainly safe at a 4 micron interval, and probably just fine at an 8 micron interval. Run your stack using every other image or every third image and compare results.

[jin]

1. Is the center of the field sharp? (Show us a crop, please.)

2. Do you have any other optics in the system, or is it just direct projection from the objective onto the sensor?

--Rik

Hi Rik,

Here is the crop of the center of the image, 100% resolution and unprocessed and there is also no other optics in the optical path.


Kindly advise

[jin]

A couple of questions and thoughts come to mind...

Which Nikon 10X objective is this? There are dozens out there with various specifications. (A picture might be best). Your sensor format diagonal is about 27mm. Most finite Nikon (160mm tube length) objectives were intended to provide, at best, a quality image of about 20mm diagonal. Certain ones that were intended for use with super wide-field heads were designed to provide a quality image circle of about 26-27mm. That's not to say it is not possible to get a good image out to the corners with a non-super wide-field design, but it is not to be taken for granted, and must be tested to see how well it works.
It is the one from Edmun Optics, item #59-935 at http://www.edmundoptics.com/microscopy/ ... ives/59935
When you say "~170 tube" are you referring to the distance from the objective threads to the camera sensor? The objective was designed for a 150mm distance in that regard, but using slightly more (like 170mm) might actually be better if the corner quality is not that great. Compare the center to a corner. The center almost always will look somewhat better, but the difference should not be extreme. If the center is good but all four corners are not, it is most likely a case where the objective is not providing an adequate size image.
See pic for the dimension and advice is this correct?

... live view and 580EX flash...

Flash should eliminate most vibration problems (if the power level is turned down resulting in a very short flash duration). But even with flash it is still a very good idea to use technique that will minimize any potential sources of mechanical vibration. Are you shooting directly from from live view? If so, you should be aware that when doing this with a Canon flash attached, the shutter must first close and then re-open to start the exposure. From an "anti-vibration" standpoint is is better to not shoot directly from live-view when using flash. After getting everything set, turn off live view. Use the camera mirror lock-up. For each shot release the mirror up, pause, then take the picture. This way you only have one mechanical shutter motion before the flash fires.
Yes, I shot with Live View on (but mirror-up disabled)as I thought this gives the least vibration after reading your article here at http://krebsmicro.com/Canon_EFSC/index.html , but it seems like I misunderstood it . I’ll try to re shoot without Live view and mirror-up enabled.
If you are working in a room where you can turn down the ambient light levels it is also possible to set the flash to "second-curtain sync" and use a long exposure (such as 1/2 or 1 second) which then would allow any shutter vibration to dampen out before the actual flash exposes the picture. (Just be sure room light levels are low enough to avoid registering an "ambient light image" on the sensor. All registering exposure light should come from the flash).

110 stacks at 2um interval

This is "overkill" for a 10X objective. The steps can be larger and fewer images used. (This should not be the cause of any "unsharpness" you are seeing, but you are making more work for yourself). It can vary somewhat depending on the NA of the objective, but at 10X you are certainly safe at a 4 micron interval, and probably just fine at an 8 micron interval. Run your stack using every other image or every third image and compare results.
I’m aware that there is a DOF calculation but I have yet to figure it out yet at this beginning state. Since my wing is not flat so I just turn the knob from the higher point to lower point with 2um interval and that ended up ~110 steps.
Still a lot to learn understand and apply so kindly bear with me.

[rjlittlefield]

1. Your understanding of Live View was correct in the first place. When using Live View, mirror lockup selection does not matter because the mirror is always locked up during Live View and it just stays there during the actual exposure.

2. See http://www.photomacrography.net/forum/v ... php?t=9393 for some indication of the center-vs-corner difference that I've seen with that same type of objective.

3. Comparing your center and corner, I think the objective is working OK and you are being misled by choice of subject. Fine striations on the scales are visible but low contrast at image center. The same fine striations are also visible, though less sharp, at the image corner. So the difference from center to edge is not huge, about the same as I saw. True, comparing your image with other people's images of different butterflies, this one looks pretty "blah". But I think most of that is due to the subject. It's common for skipper scales to not have strong detail. See for example my Glassy spot on the wing of a skipper.

I suggest looking around for another subject with stronger detail. Many moths are good.

--Rik

[Charles Krebs]

Sorry for any confusion about mirror lock-up...

If you are using continuous light to take the photofgraph (not using flash) then you should always shoot directly from from live-view. (And whatever the mirror lockup is set to do does not matter at all. It has no effect from live-view). This provides the least vibration with this Canon camera. (The camera starts the exposure totally electronically with no mechanical shutter movement).

If you are shooting with an electronic flash from live-view, then the camera will do a mechanical double shutter action (close and re-open) before the flash exposure is made. This is what I was referring too. It happens very fast and many people are not even aware of it. While I doubt very much that it would have much of an effect at 10X with electronic flash, it you are trying to "troubleshoot" a resolution question it is a good idea to eliminate as many sources of vibration as possible. When using a Canon flash, if you turn off live-view after setting up the shot, and then use the mirror lock-up with pause for each shot, you will only get a single mechanical shutter operation (shutter opens) before the exposure begins. (And using the technique of a long exposure with second curtain sync is perhaps the "ultimate" in avoiding equipment induced vibration. Again... I seriously doubt going to that extreme is necessary in your case, but when checking all the potential problem areas in a new set-up, trying this can effectively eliminate camera induced vibration as a possible problem. You then have a "no-vibration" baseline for comparison with other methods).

The corner/center resolution comparison here looks good, so it seems the objective "covers" the APS format well. The ~170mm distance you show in your diagram is what I was referring to, and that is fine. As far as optical aberrations are concerned, an objective with a NA of 0.25 is quite "forgiving" of variations from the designed 150mm distance. (If you make that extension distance smaller you will get a lower magnification and you need to double check the corners/center resolution again since the "quality" image will be a little smaller. If you continue to reduce that extension distance you will eventually reach a point where the corners clearly deteriorate to an unacceptable degree, or the image is cut off entirely by vignetting. As you increase the extension over 150mm you get a little higher magnification. The "quality" image circle can get a little bigger... BUT... you also get a smaller effective aperture which will cause diffraction to begin to reduce the resolution. Keep extending it and you will eventually reach a point where the entire image gets obviously "softer" due due diffraction.

[jin]

I suggest looking around for another subject with stronger detail. Many moths are good.

--Rik

thanks and will certainly look for a dead moth to try it out.

[jin]

Hi Charles,

As advised I tried today to shoot the same subject with a continue LED light source from CCS, model PFB20SW see spec at: http://www.ccs-grp.com/s2_ps/src/pro_sr ... /e.html#sr with slight different wing area as shot before. However I noticed with a cont. light with similar light exposure on thh 450D with mirro-up the pic taken with this method seems soft and lacking detail as compared to image taken earlier with flash in LIVE VIEW or with just Mirro lock is up.

Can you share your thought and advice pls.

[rjlittlefield]

This looks motion blurred. Was this shot directly from Live View?

If it was taken direct from Live View, then probably you are seeing environmental vibration.

If it was just taken with mirror lockup but not direct from Live View, then probably you're seeing vibration from shutter movement.

For lens testing, it is usually better to use flash, placed close to the subject so that it can run at low power and thus give short flashes. With flash it is better to use mirror lockup and not shoot from Live View because of the double action that Charles mentions when using flash from Live View.

--Rik

[Charles Krebs]

Remember what I said earlier:

If you are using continuous light to take the photograph (not using flash) then you should always shoot directly from from live-view. (And whatever the mirror lockup is set to do does not matter at all. It has no effect from live-view). This provides the least vibration with this Canon camera. (The camera starts the exposure totally electronically with no mechanical shutter movement).

If you did shoot directly from live-view, then as Rik said there may be some external "environmental" vibration affecting the image. (Your image has no EXIF data so I don't know what shutter speed was used. Does that LED light source have a fan inside? Sometimes cooling fans create a slight vibration that can be transmitted to the set-up.)

If you did not shoot the continuous light image directly from live-view, but instead used the mirror lock-up and a pause... then you will almost certainly have some negative effect from vibration caused by the mechanical first shutter curtain.

When I compare shots taken using electronic flash, with shots taken using continuous light (the continuous light shots taken directly from live-view) I can see no difference in "sharpness".

[rjlittlefield]

For lens testing, it is usually better to use flash, placed close to the subject so that it can run at low power and thus give short flashes.

When I compare shots taken using electronic flash, with shots taken using continuous light (the continuous light shots taken directly from live-view) I can see no difference in "sharpness".

Despite how it might appear, there is no conflict between these two statements. I get the same results as Charles (no difference between flash and continuous), when I am careful to work in a place with no environmental vibration.

The advantage of flash for lens testing is that it means the results are less likely to be messed up by whatever environmental vibration might have slipped past my guard.

--Rik

[jin]

Remember what I said earlier:

If you are using continuous light to take the photograph (not using flash) then you should always shoot directly from from live-view. (And whatever the mirror lockup is set to do does not matter at all. It has no effect from live-view). This provides the least vibration with this Canon camera. (The camera starts the exposure totally electronically with no mechanical shutter movement).

Sorry Charles, I suppose to shoot continuous light with Live View as advised to compare the shot with flash with mirror-up, but somehow I had forgotten to switch back after I done with shooting with flash. I redo it last night with cont. light and i unplugged the fan connector to turn it off to avoid an vibration induces at higher SS of 1/125sec it seems better but still can't match the sharpness and light quality with flash in my setup seems better ie, more defused unlike this one taken using LED light with lot of shadows and with lot of reflection on the surface of the scales. othewise sharpness seem close?