I do realize that elastomers like Sorbothane are designed to be “lossy” and absorb vibration energy by dissipating it as heat. But my point is that they are not perfect at isolating all vibration, especially at low frequencies.
Depending on the durometer of the Sorbothane, the supported mass and other factors, I contend that the system most definitely will exhibit resonance at some particular (low) frequency. And when exposed to vibration at this frequency the system will oscillate. Also, Sorbothane will indeed act like a highly dampened spring when subjected to a mechanical impulse---so I believe the “spring constant” (or perhaps elastic constant) reference is correct. These concepts are shown in the following plots from the Sorbothane web site (upper “ringing” spring effect, lower resonance):
http://www.sorbothane.com/material-properties.php
ChrisR wrote:If you've got the losses you don't get resonances.
See above.
ChrisR wrote:The "C"s have to be sized appropriately, they don't have to be huge. Complex vibration absorbing structures have multi-frequency components, but Sorbothane is pretty wide spectrum.
Sorbothane is very good at absorbing high-frequency energy, but becomes much less efficient at lower frequencies. So if the excitation frequency in your diagram contains components near the resonance point of the two systems (subject and camera), they will begin to wobble, and likely at different rates.
ChrisR wrote:Bob^3 wrote:There is another configuration option which can also reduce the key issue here, relative motion between the subject and camera (or more precisely the image of the subject on the sensor). That is to remove the lossy small "c" supports in your diagram and tightly lock both camera and subject to a massive, very rigid support base.
That works at low frequencies - like the slide copier. But it's hopeless at high frequencies. Think about the executive toy with the bouncing balls. The row of balls is nice rigid steel, but though it doesn't appear to move it transmits (impulse = fast) energy very effectively to the other end. Bigger balls would do it well too. You'd have to have lots of massive fixings to do as good a job of stopping it as a piece of blotting paper.
If you to connect your large mass to the generator via a rigid but springy cantilever (like a vertical rail, to a table) you're asking for trouble.
You ever so definitely don't want shutter energy transmitted to your subject. It has low mass, it'll move easily, and a darned site faster than the sensor, and you're magnifying the image of that movement.
Well, let’s see if I can answer those issues. I would agree that this
could be an issue with some (most) DSLRs (e.g. Nikon) that still employ the original film SLR mirror/shutter sequence. But in light of the current discussion and specifically Richard’s use of the Canon DSLR in Silent Mode, where is your high-frequency “impulse” going to come from? Yes, there is an impulse when the mirror is flipped up and when the shutter opens and closes. But as has been stated, as long as the delay between those actions and the actual electronically initiated exposure period (which is all that counts) is long enough to allow any vibration in the system to die-out to a level that it doesn’t effect the image capture period, there shouldn’t be any motion blur (except as Rik and Charles stated, the slight possible effect induced when the shutter closes, at least on the Canon T1i).
If we agree that there should not be much high-frequency energy coupled through the base and Sorbothane supports, we are left with two key parameters to consider: the mirror and shutter delay periods (2 and 4 in my sequence) and the damping period of the system. Assuming the camera is in LiveView mode, the mirror should already be locked up. So only the shutter delay would matter. I don’t know what this period is or if it can be adjusted. But since the shutter action usually imparts much less shock than the mirror, even a short delay may be adequate? If not, I would think that a little more mass added to the base of the camera and if necessary dampening material added to the column would eliminate any significant vibration.
Another potential source of vibration in Richard’s rig to consider is from movement of the StackShot stage. But again, as long as the delay between the stage movement and image capture is long enough, there should be no issue.
ChrisR wrote:The schema I sketched with the masses and C's is a standard lab set-up. …
Books go on ( or they used to) about controlling transmissions from generators, and susceptibilities of sensors. Keep em separate!
Don’t know what “standard lab set-up” you refer to, but all the ones I’ve seen for high magnification imaging, precision laser benches for optical disc recording and holography all use some variation of the concept where all optical components are rigidly mounted to a low-vibration transmitting base (granite plate, honeycomb optical table, etc.), which is in turn isolated from earth by a sophisticated active or passive base isolation system.
One book referenced many times on this forum that I know and trust is “The Manual of Close-Up Photography” by Lester Lefkowitz (BTW Richard, it you don’t have a copy, I would highly recommend you get one---inexpensive used on Amazon). Lefkowitz’ opinion is very clear on this issue. His homemade “ultimate solution” for magnifications greater than 5x (shown on page 158) consists of a 100 lb patio stone supported on the floor by four inner tubes partially filled with air. The camera is rigidly mounted to a 75 lb stand by heavy-gauge pipe.
At the risk of drifting too far of topic and getting overly wordy, I’ll offer one more “ultimate” example from my own professional experience. I’ve done extensive work with an instrument called an “Atomic Force Microscope” (AFM).
http://en.wikipedia.org/wiki/Atomic_force_microscopy
It uses a silicon probe etched to a nanometer-sharp tip to scan a surface in the x-y plane, while following the vertical features and recording contours. The resulting image can resolve feature below 1nm (nanometer)---three orders of magnitude higher magnification than we are discussing here. Since the scanning tip travels (or oscillates in “tapping mode”) only a few nanometers above the surface, near perfect vibration isolation of the subject and piezoelectric-driven scanning head is mandatory. In these systems, the subject and heavy support column to which the head is attached are mounted on a massive granite base block (don’t know the mass). The base is suspended on a 3-axis vibration isolation mechanism using passive air or fluidic dampening. The whole platform is mounted inside a thick acoustic enclosure to attenuate sound and airflow during the scanning process.
The AFM scanning head is analogous to the camera in a macro photography setup, but with much higher sensitivity to vibration. The slightest relative movement between the subject stage and the head causes noise in the servo loop to the scanning tip and results in artifacts in the final image. Of necessity, the subject, support column and head must be rigidly coupled together with vibration dampening materials. Just having the door to the enclosure open during scanning while talking near the unit, can cause artifacts in the image.
