Image stabilization for the "Lap Telescope"

OK, this is going beyond what I originally intended but maybe it could
work. From a Wiki article, image stabilizing systems in lenses work
by a piezo detector and then electro-magnets that actually move the
lens that floats in a cell. I had thought it was some sort of gyro
but apparently not. Such a thing might be implemented for a small
"lap telescope". The mirror could rest on actuators that move it.
Small piezo actuators are available ( I have no idea of cost). The
actuators could also be used as the sensing element.
Based on what I see in stabilized binocs, you might expect to get
about double the mag you'd get without the stabilization.

On 2008-01-09, wrote:
OK, this is going beyond what I originally intended but maybe it could
work. From a Wiki article, image stabilizing systems in lenses work
by a piezo detector and then electro-magnets that actually move the
lens that floats in a cell. I had thought it was some sort of gyro
but apparently not. Such a thing might be implemented for a small
"lap telescope". The mirror could rest on actuators that move it.
Small piezo actuators are available ( I have no idea of cost). The
actuators could also be used as the sensing element.

My first thoughts were that this would be completely impractical
for a hobby project, but after some thought I've come to the
conclusion that it is possible, depending on your skill set. I'm
looking at a copy of Philip's Stargazing With Binoculars right now,
ISBN 978-0-540-09022-8, which briefly discusses the various
technologies used in binoculars, although it doesn't go into any
detail it outlines the general approaches taken and relative merits.
I'll send you scans of the relevant pages (there's only two of
them) if you want, although the book isn't too dear anyway - $13.57
at Amazon.

Apparently, there are three principal technologies. Canon's system
works similar to you describe, movement is detected electronically
and and the prism is then moved to compensate for it. The Fujinon
system uses gyros to stabilize things in the manner you speculated
about. There is also a compeltely passive system which uses gimbal
mounted prism used in some Russian binoculars. While this might
sound the easiest solution I would hazzard a guess that making it
work in practice would be fiendishly difficult. It isn't supposed
to be very good at any rate.

Personally, I see a Canon-style solution as the way to go, but then
again that may be simply because that approach best suits my
particular skill set. You would need some kind of mirror or prism
somewhere in the system to shift the image - I can't see how you
would do it simply by moving lenses. This would be attached to a
couple of servos to move it. I believe R/C servos available cheap
should be good enough for this. I initially had concerns that they
may not be fast enough but now I think this is probably unfounded.

To detect the movement the sensors you are after are either
accelerometers or inclinometers - which depends on your exact
circumstances (they are similar technologies and often lumped
together anyway). In the middle you would have a microcontroller
running the whole show. That's where your skill set comes in since
that implies reasonable electronics (building the controller) and
computer skills (writing the control program).

The nice thing about this approach, though, is that generally if
you use the right controller you can replace the software as often
as needed. This would allow a iterative approach to be taken,
modifying the program subtly each time and experimenting until you
have the best results. The cost of doing all this would largely
depend on your sensors - they can very in price from a few pounds
to several hundred pounds apiece here in the UK. I hazzard a guess
that you should be able to do it for under £100 though, maybe half
that if you find suitable sensors at the very low end of the price
range.

In short, yes it probably is possible given enough time and effort.
Whether it is possible for you with the skills you possess only you
can determine.

--
Andrew Smallshaw

On Wed, 9 Jan 2008 22:34:41 +0100 (CET), Andrew Smallshaw
wrote:

You would need some kind of mirror or prism
somewhere in the system to shift the image - I can't see how you
would do it simply by moving lenses.

When you move a lens laterally, so its optical center is off the optical
axis of your system, the image shifts as well. That's how SBIG's latest
tip/tilt corrector works. It's a relatively simple system to implement
mechanically.

Canon's original system (I don't know if they still use it) had variable
prisms. These are also pretty easy to make- an oil filled silicone tube
sealed with a glass window at each end, distorted by a motor or piezo
element. I made one for a tip/tilt corrector, and it was reasonably
easy.

This would be attached to a
couple of servos to move it. I believe R/C servos available cheap
should be good enough for this. I initially had concerns that they
may not be fast enough but now I think this is probably unfounded.

To get enough speed, you'd want to limit your rotation range. And the
problem with RC servos is that their position resolution is pretty poor.
I don't think you'd be able to move finely enough, or fast enough, to
make servos work.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

Chris L Peterson wrote in
:

On Wed, 9 Jan 2008 22:34:41 +0100 (CET), Andrew Smallshaw
wrote:

You would need some kind of mirror or prism
somewhere in the system to shift the image - I can't see how you
would do it simply by moving lenses.

When you move a lens laterally, so its optical center is off the optical
axis of your system, the image shifts as well. That's how SBIG's latest
tip/tilt corrector works. It's a relatively simple system to implement
mechanically.

Canon's original system (I don't know if they still use it) had variable
prisms. These are also pretty easy to make- an oil filled silicone tube
sealed with a glass window at each end, distorted by a motor or piezo
element. I made one for a tip/tilt corrector, and it was reasonably
easy.

This would be attached to a
couple of servos to move it. I believe R/C servos available cheap
should be good enough for this. I initially had concerns that they
may not be fast enough but now I think this is probably unfounded.

To get enough speed, you'd want to limit your rotation range. And the
problem with RC servos is that their position resolution is pretty poor.
I don't think you'd be able to move finely enough, or fast enough, to
make servos work.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com


If you put all that in a lap scope...you would get a free lap dance at the
same time...

/////
^ ^
* *
v
\_/

wrote:

OK, this is going beyond what I originally intended

Yepper! Giddyy Up! Might as well add anti lock brakes,
CG4 stereo DHL, and a higgs boson drive. Higgs boson has
yet to be found but SAA exports will give that to ya free!
Keep posting...
Enjoy ...



but maybe it could
work. From a Wiki article, image stabilizing systems in lenses work
by a piezo detector and then electro-magnets that actually move the
lens that floats in a cell. I had thought it was some sort of gyro
but apparently not. Such a thing might be implemented for a small
"lap telescope". The mirror could rest on actuators that move it.
Small piezo actuators are available ( I have no idea of cost). The
actuators could also be used as the sensing element.
Based on what I see in stabilized binocs, you might expect to get
about double the mag you'd get without the stabilization.