New Horizontal Setup

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

microcollector
Posts: 261
Joined: Wed Mar 14, 2007 7:19 pm
Location: Port Orchard, Washington

Post by microcollector »

Rik,

The carpet tape is from my local, non big box hardware store, and is double sided. It should hold everything in place when I transport the system. In fact, it may be a problem to remove the rail if needed.

Doug
micro minerals - the the unseen beauty of the mineral kingdom
Canon T5i with Canon 70 - 200 mm f4L zoom as tube lens set at 200mm, StacK Shot rail, and Mitutoyo 5X or 10X M plan apo objectives.

My Mindat Mineral Photos
http://www.mindat.org/user-362.html#2

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

DaveW wrote:I can only quote the equipment previously used for lens testing years ago by a firm in London (hardly a vibration and traffic free area) who claimed it was set up in their cellar on a very heavy cast vibration isolated concrete block.
And therein lies the key: isolation.

I received the following PM from Doug this morning.
microcollector wrote:Before changing out the objective I tried the bump test and even the 4x objective shows a jump when the beam is bumped, if the bench is bumped, or if I jump on the concrete floor. Normal walking does not affect it like it would when I was shooting through the microscope in the house.
Naively one might think that a concrete floor and a 1000-pound bench would be sufficient isolation. Obviously it helps, but it's not a cure-all. I think the sensitivities of these systems are not appreciated by most people.

As I have tried to explain before, not always successfully, it is very difficult to couple camera, lens, and subject so that the group moves as a unit. Success is easier to achieve by isolating sensitive stuff from the vibration sources so that they don't have to move in the first place.

In the case of a DSLR and microscope, a demonstrated effective solution is to eliminate tight coupling between the camera and the scope, as shown by Charles Krebs' setup (HERE) using a separate support for the camera and a non-contact bellows between the camera and scope.

I'm pretty sure that a similar approach would be equally effective in bellows-based systems like Doug's, mine, and so many others. The approach is conceptually straightforward: isolate the vibration causing camera from the rest of the rig so that the camera can shake itself while not shaking the lens and subject that are many times more sensitive. Further improvements could be made by coupling the camera to a large mass (also shock isolated from the lens and subject), and by using shock-absorbing supports such as Sorbothane so that the camera's own vibrations would die down quickly.

As usual, the question is whether that effort would be worth the trouble. "Conceptually straightforward" is not the same as "easy". Some time and expense would be needed to work out the mechanics, and there are other simpler solutions that make significant improvements also. I used to drape a few beanbags over the camera, lens, and/or subject stage when shutter shock was troublesome. I don't need those often now that I'm able to use electronic shutter, but people without that feature might test to see if something simple like beanbag shock absorbers is helpful in their setups. Sometimes it is, sometimes it isn't, depends on things like the exposure time.

--Rik

Mitch640
Posts: 2137
Joined: Sun Aug 15, 2010 1:43 pm

Post by Mitch640 »

The approach is conceptually straightforward: isolate the vibration causing camera from the rest of the rig so that the camera can shake itself while not shaking the lens and subject that are many times more sensitive.
But the sensor or film is still part of the camera body. Use an image stabilized camera?

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Mitch640 wrote:But the sensor or film is still part of the camera body. Use an image stabilized camera?
Now you're getting into two other cans of worms.

The first can of worms is the effect of different kinds of vibration on the image. When the camera body twitches, it can shift in any direction while also rotating on any axis (X,Y,Z; pitch,roll,yaw). At high magnification, if any of these motions couples to say the objective but not the subject, then the effect on the image will be much larger than if the motion affected the camera alone. Commonly the twitch is a rotation, in which case the effect is MUCH larger. The limiting case is when the center of rotation happens to coincide with the sensor, in which case there's essentially zero effect if only the camera rotates, versus a large effect if that rotation at the camera couples down to become a shift at the objective. Usually the difference is less extreme than that, but still you're better off to isolate the camera unless you can couple the twitch very very well to both lens and subject.

The second can of worms is how and how well a camera's image stabilizer actually works. What those devices do is to measure physical movement, predict what effect that will have on the image, and use that prediction to drive some other component such as shifting the sensor by a compensating amount. Notice first that this is a prediction, and the accuracy of that prediction depends on the camera having a correct model of the stuff attached to it. When you remove the ordinary lens and stick the camera in a bellows rig, the camera's model becomes wrong and it makes lousy predictions. In addition, stabilizer systems are tuned to compensate for slow vibrations typical of hand-holding tremor. They are not fast enough to compensate for a shock followed by fast vibrations like you typically get in these focus stacking rigs.

Bottom line, it's an attractive idea but it doesn't work in practice.

--Rik

microcollector
Posts: 261
Joined: Wed Mar 14, 2007 7:19 pm
Location: Port Orchard, Washington

Post by microcollector »

For the Nikon D300, the image stabilization is done by VR lenses, not in the camera.

Doug
micro minerals - the the unseen beauty of the mineral kingdom
Canon T5i with Canon 70 - 200 mm f4L zoom as tube lens set at 200mm, StacK Shot rail, and Mitutoyo 5X or 10X M plan apo objectives.

My Mindat Mineral Photos
http://www.mindat.org/user-362.html#2

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

microcollector wrote:For the Nikon D300, the image stabilization is done by VR lenses, not in the camera.
Good point. That's the case for Canon's IS lenses as well. I was thinking of the sensor-shift models, which were the only ones that made sense in the context of Mitch's question.

--Rik

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

A suggestion has been made to split this thread, separating general issues of vibration reduction from the specific details of Doug's system.

I'm going to defer to Doug on whether to do that or not, since he owns the thread. From a standpoint of forum organization, there's no clear preference. Many of the later postings are quite general, so from that standpoint it would make great sense to split. But they derive from earlier postings that were directly related to Doug's system, so from that standpoint, some continuity could be lost.

Doug, based on what you see so far, what say ye -- split or not?

--Rik

microcollector
Posts: 261
Joined: Wed Mar 14, 2007 7:19 pm
Location: Port Orchard, Washington

Post by microcollector »

Rik,

I do not have problem keeping it as one thread. I did spend some time yesterday trying to get an image with a good, sharp corner for you. Once I looked at them under actual pixels, none were all that great. It appears to me that the mirror is the problem. I have a few other things to try to see if I can improve the sharpness.

Doug
micro minerals - the the unseen beauty of the mineral kingdom
Canon T5i with Canon 70 - 200 mm f4L zoom as tube lens set at 200mm, StacK Shot rail, and Mitutoyo 5X or 10X M plan apo objectives.

My Mindat Mineral Photos
http://www.mindat.org/user-362.html#2

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

microcollector wrote:It appears to me that the mirror is the problem.
Could be the mirror, could be the shutter itself. If I recall correctly, Charles posted some tests showing that (for one particular camera & scope) they were about equal contributors.

Since you're using continuous illumination, one attack is to turn the brightness way down and use a long exposure -- several seconds -- so that most of the exposure time occurs after vibrations have had a chance to die down. You'll have to do a custom WB, of course. Beanbags can be used to make the vibrations die down faster, as mentioned earlier.

I used to use this trick on all my high mag stacks,with typical exposure times in the 5 second range.

--Rik

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

Doug,

I think the comparison Rik is referring to is this one:
http://krebsmicro.com/Canon_EFSC/index.html

It's probably not directly applicable to your set-up, but it does point to a few things that universally need be considered (and really only confirm what was generally already known or suspected).

When using a continuous light source...

"Normal" mirror action together with the shutter action is best avoided in a high magnification set-up. It has the real potential to seriously degrade image quality. ("Live view" methods where, when an exposure is initiated the sequence is: [ -the mirror drops down, -shutter closes, -mirror goes up, -shutter opens ] are even worse since you have multiple mirror and shutter movements before the exposure even starts)

Mirror lock-up is a great help, but there can still be a considerable vibration "contribution" from the mechanical first shutter curtain. If you use mirror lock-up but determine that the shutter is still causing vibration problems, using a long shutter speed as Rik suggested is about the best way of dealing with it.

Post Reply Previous topicNext topic