Automated stack on new rig.. (blue crystal)

[phreakocious]

Unbranded 4x microscope objective
52mm->RMS step-down ring
Nikon F->52mm adapter
PB-6 bellows
Nikon D7000

Yesterdays ruby pictures were pretty uninspiring, but I have some new insights into possible causes. This set was done with a bellows and finite objective instead of the medium format camera (tube) lens and infinite objective. The results are markedly better.

The dark diagonal line at the bottom is a solid spot in the individual pictures, so perhaps there is some off-axis movement happening?

160 images, stacked in the demo version of Zerene.

Cropped and scaled:


100% crops:

[ChrisLilley]

Unbranded 4x microscope objective

Does it say what the NA is?

Yesterdays ruby pictures were pretty uninspiring, but I have some new insights into possible causes. This set was done with a bellows and finite objective instead of the medium format camera (tube) lens and infinite objective. The results are markedly better.

An interesting finding. Are your 'possible causes' related to the finite vs. infinite objectives, or are they unrelated (lighting, flare avoidance, choice of subject)?

And I would agree that these look better than the ruby, which seemed sort of greyish and smeary somehow.



There seems to be some 'batwing' aberrations (points turn into streaks or such like shapes, oriented at 90 degrees to a radial line running to the image centre, and worse the further from centre you go) which I assume is due to the unbranded objective reaching the limits of its price point or alternatively, being used with a wider image circle than designed to be seen.

But overall this looks pretty nice to me.

[phreakocious]

Does it say what the NA is?

I believe the NA was 0.10.. Will double check.

An interesting finding. Are your 'possible causes' related to the finite vs. infinite objectives, or are they unrelated (lighting, flare avoidance, choice of subject)?

And I would agree that these look better than the ruby, which seemed sort of greyish and smeary somehow.

Right now, I'm leaning towards a combination of lighting angles, and a bug in my Arduino code that may have caused the stepper motor to move while the shutter was open.

There seems to be some 'batwing' aberrations (points turn into streaks or such like shapes, oriented at 90 degrees to a radial line running to the image centre, and worse the further from centre you go) which I assume is due to the unbranded objective reaching the limits of its price point or alternatively, being used with a wider image circle than designed to be seen.

But overall this looks pretty nice to me.

Thanks, I'm feeling much better about this setup after seeing these! I suspect the choice of subjects are not the easiest to start with. Still looking for moths, as Rik suggested. =)

As for the bug, how are people maintaining synchronization between the controller and camera to prevent them stepping on each other without bidirectional communication? As far as I know, there are not separate IR codes for mirror up and shutter, so if the timing is off or a signal is missed, the timing of the whole thing can be wrecked. I'm thinking of reverse engineering the USB commands from Camera Control Pro to fix this...

[rjlittlefield]

I agree with Chris -- this looks quite good except for those "batwing" artifacts in the upper left corner. (Chris, is that a standard term?)

The dark diagonal line at the bottom is a solid spot in the individual pictures, so perhaps there is some off-axis movement happening?

Probably. But once in a while the alignment procedure will get misled by subject appearance into thinking that there's movement when there really isn't. That could have happened here because of the asymmetry of having one corner especially bright. A good test would be to shoot a subject with uniform brightness across the whole frame and see if you get the same amount and direction of dust trail.

See the Zerene Stacker FAQ for some other hints about how to evaluate movement.

If the alignment process is getting misled by subject appearance, but your setup really is stable and well centered, then you can just turn off alignment (uncheck all boxes at Options > Preferences > Alignment) to get the best result.

how are people maintaining synchronization between the controller and camera to prevent them stepping on each other without bidirectional communication

The StackShot uses fixed time delays, which are set by the user based on exposure time, flash recycle time, and so on.

--Rik

[ChrisLilley]

I agree with Chris -- this looks quite good except for those "batwing" artifacts in the upper left corner. (Chris, is that a standard term?)

Its a common non-technical term to describe the effect. I used it because I am not sure of using a more precise term correctly. It gets described as coma, or as sagittal comatic flare, but I have also read that calling it this is incorrect. Sagittal oblique spherical aberration may be the correct term.

The famous quote about "I can't define it, but I know it when I see it" applies here.

[phreakocious]

Objective markings are 4/0.1 160/0.17 .. I don't know if the coverslip correction has any real bearing on image quality when it's being used in this fashion.

[ChrisR]

Objective markings are 4/0.1 160/0.17 .. I don't know if the coverslip correction has any real bearing on image quality when it's being used in this fashion.

No worries on that one.
AS the text and links in
this thread
reveal, coverslip incorrectness doesn't matter below NA 0.3, or a little more.

NA 0.1 for a 4x objective isn't high though. 0.13-0.15 is common and there are APO versions with NA 0.2. They are a bit sharper, but may cost ten times as much.
Even at the same NA, some manufacturers' objectives are better than others. And some are designed for corrections in the eyepiece, which you aren't using.
At about 4x, there are many options, using reversed enlarger lenses or specialist macro lenses, which cover other magnifications as well.
It all depends on the tightness of your specification, and the same property, of your wallet!