Lance Allen wrote:
> On 17 Oct 00 at 12:58, Robert Rolf wrote:
> > Yes, they use gyros. Ultrasonic gyros where the u/s mechanical
> > wave bounces around inside a piezo crystal. They sense the phase shift
> > resulting from motion, and then change the clock timing on the CCD chip
> > so that the image gets stabilized as it gets shifted out. This
> > technique can only compensate for horizontal/vertical motion, not
> > rotation, but that is sufficient for most users.
> >
> Yes they all use piezo gyros but at least one camera manufacturer
> uses a 2-axis-coil-lense-in-a-servo arrangement, like a cd lense to
> keep the image stable. That is... electro-mechanical stabilization.
>
> Lance Allen

That would be the old Sony TR101. Much better stabilization than
the new 'all digital' ones. Did have problems with bellows leakage
though. IMS JVC also had a EM stabilized version around the time (1995).

Now, if you're talking about the professional CAMERA stabilizers
(AeroCam
and the like), that unit does use a spinning gyro, and some clever
mechanics for vibration isolation (including an 'active' suspension).

> On 17 Oct 00 at 15:52, Stephen B Webb wrote:
>
> > > I thought they used some type of gyro, and I was
> >
> > Hmm.  I don't really know much about the cameras, but I didn't think that
> > they used gyros.

They use piezo gyros to sense the motion vectors.
Just look in the service manual for descriptions & part numbers.
The now discontinued 'Gyropoint mouse (about $50.US)' used piezo
gryos. A colleague of mine gutted about 50 of them for the gyros because
it was cheaper to do that the buy them from Murrata. There are many
sources now.

> > My understanding was that they had an "oversized" ccd, and only used a
> > (NTSC sized) subregion of the full ccd to form the image.

That is exactly how the newer Sony system works. With gyros to sense
the motion correction required. Lower overhead than DSP methods.

There is also an 'all digital' system that commercial broadcasters
use that basically does a 10% digital zoom on the incoming video, and
then uses DSP's (motion vector extraction) to correct the motion. $$$.
You can find out more about this by digging up back issues of the SMPTE
journal (circa 1995).

> Come to think of it (pontificate in haste, backpedal at leisure) all
> the steady cameras I have seen the workings of use piezo gyros BUT
> that doesnt mean there has since been far cheaper/better solutions
> developed.
> There is a world of variations out there.    Like the (at least) 2
> ways of steadying the image itself.

3 if you count the DSP method. The DSP method can also correct for
rotation, something the other two can't do.

> Lance Allen

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
use mailto:listserv@mitvma.mit.edu?body=SET%20PICList%20DIGEST