PaulBa Loc: Cardiff, Wales

I find this quite interesting I hope some of you will although I realise many may not, apologies to them, maybe I'™m just an odd Welshman but these things just keep bouncing around in my head :-)
A few months ago I asked for help with the design of version two of my stacking rig.
UHH member oldtigger, quite rightly, suggested that my stability problems could be caused by the number balls, slides, fulcrums and joints between lens and subject. To remedy this I've reduced their number and strengthened those that are left.
Uhh member naturepics43, also quite rightly, pointed out that the sensor dirt trails indicated that I had a problem with stability. So taking that on board I've compared the trails from version 1 to a test stack I took yesterday.
Warning, this is not a scientific comparison just a invitation for friendly discussion. While, to me, the Vr2 dust trail looks straighter it appears to have a evenly pitched rhythm to it.
My queries are:
Can I assume that the straighter dust trail indicate a more stable rig but maybe still not stable enough and, more interestingly, what do you make of the rhythm in the latest trail. Is it the result of some local rhythmic vibration or Zerene Stacker realigning itself (I estimate average shots between peaks too be 15 with 5 second interval between shots).
I will post pictures of my new rig and stacked examples as soon as it's completed and I've some examples I'm happy to share" Canon EOS 6D, Canon MP-E65, StackShot Rail, Canon 430 ex II speedlight, Zerene Stacker.

From Rig Version 1

(Download)

From Rig Version 2

(Download)

oldtigger Loc: Roanoke Virginia-USA

Are you sure your direction of movement is at right angles to your sensor plane?

PaulBa Loc: Cardiff, Wales

Just lined up by eye, I’ll have to think of a way to check optically. Thanks oldtigger.

oldtigger Loc: Roanoke Virginia-USA

Point your camera at a dot and move camera back and forth. If the dot shifts position you are not aligned.

pfrancke Loc: cold Maine

my thinking on these dust trails...
Assuming your camera on rail is moving closer to the subject for each frame:
1) the subject would become larger
2) but the dust on the frame remains at the exact same spot on the image
3) the stacker software is smart enough to recognized the subject as the subject and does a realignment and tightening process of the subject
4) and on the resultant image merge, the dust appears to move in an INWARD or OUTWARD direction (I keep confusing myself thinking about it).

So if the above is true.. then if we had dust on all sides of the subject and your rail is aimed perpendicular to the center:
Then... the lines would radiate from a center point
If on the other hand multiple line are in various places and they are roughly lined up with each other (all going in the same direction):
Then... likely the rail is sliding to the left or to the right instead of dead forward. But the stacking software still does a nice job aligning the images.

Anyway, that is my sad logic.. Also, as noted in another thread (the name escapes me right now), putting the rail under a load might help reduce backlash - I am eager to attempt this on my rail using rubber bands to see if I can achieve better stability.

Edit - so anyway, regarding stability, every trick in the book! Wait after the rail moves before allowing the shot so the camera settles - Use Live-View so that their is no mirror movement - use shortest reasonable flash time to freeze the subject, Use cement floor - perhaps foam under the subject and platform to soak up vibration.

PaulBa Loc: Cardiff, Wales

Thanks for taking the time to comment Piet. I think your logic is correct, what appears to be spots traveling across the image is in fact the result of the subject moving behind the spots, Zerene aligned the subject not the spots. However, your suggestion of extending the trails to see if they merge to the centre point is an excellent idea which I shall follow up.

mawyatt Loc: Clearwater, Florida

I think your logic is correct. Zerene Stacking software does a pretty good job of finding the right things to lock onto, Rik (author) has done a masterful job with the algorithms & software IMO.
BTW the other thread is mine, if regarding the rubber bands. Made a remarkable difference in my stacking uniformity with my particular Stackshot rail. Be sure to use the "off-axis" technique rather than the usual backlash removal on-axis loading. The "off-axis" helps reduce any rail wobble as well as backlash.

pfrancke Loc: cold Maine

LOL -- I don't know if there is great value in having a perfect lineup (so long as distance traveled is short). And there are (at least) two different line up issues. One is pointing the rail to the dead center and the other is having the camera mounted perfectly square on the rail - (and that's got to be tough if you take the camera off and return it - in other words it isn't dedicated on the rail).
But the more I think about it, the longer the lines are, either the greater distance traveled on the rail or perhaps... the greater the misalignment to center!

pfrancke Loc: cold Maine

I just want you to know that those results were inspiring. And thank you for the off-axis advise! Who would have thought - about rubber bands, but the additional clarity you achieved was fantastic.

mawyatt Loc: Clearwater, Florida

I routinely do stacks below 10 micron steps and sometimes down to 1~2 microns. I don't do anything other than visually align the optical camera axis to the rail axis, but these are shallow stacks maybe only a few mm in length.

oldtigger Loc: Roanoke Virginia-USA

if the dust spots remain in a tight little group then you were aligned pretty well. The radial issue will factor out when zerene scales.

mawyatt Loc: Clearwater, Florida

Thanks. Quite frankly I was surprised how well it worked on my Stackshot rail. Would be interesting to see how well it works for others.
I've send the images and details to Paul at Cognisys (Stackshot source) so he could share with others that use his rails.

pfrancke Loc: cold Maine

This makes sense to me. If the distance travelled doesn't increase the image size that much, then a tight group would mean good alignment. And perhaps the length of the line itself is the best indicator of not being aligned.

oldtigger Loc: Roanoke Virginia-USA

I hope you included shots of the little hooks fabricated from old coathangers. I'm going to send him my setup with the camera mounted to the stack shot and a roller bearing mounted to the lens foot and bearing on the guide rail to support the lens and prevent droop, etc.
If they would just sell us an additional carriage with an adjustable clamp for the lens.

pfrancke Loc: cold Maine

He absolutely should be told to make rubber bands standard issue with that wonderful rail he's got. LOL
Note to PaulBa - your next step in my mind, after stability, is a combination of a light box and possibly concave diffusion. You will find that the type of lighting you have will have a dramatic impact. I've just gotten a stronger flash that I wish to use - Oldtigger can attest that I have a problem with too little light and underexpose.