Hi Jim et al,

The test unit here will use 2 CCDs, one to record fine motion,
say 10 or 12 bits of resolution. The second would be physically
offset to read the "strong" motion now that the "fine" CCD has
"saturated". Using a laser keeps the reflected beam from diverging
very much in the small CCD physical offset distance. Yes, the incoming
laser beam to the mirror needs to be very small, so some optics to
collimate and perhaps a rectangular spatial filter might help. The real trick
is glueing the two CCD signals together, but this could be accomplished in
software. Needless to say, the seismo frame and the CCD head must be
rigidly locked together and temp. stable. This is where my small stash of
Invar metal comes in. And yes, all this is a major effort-but as Carole King sang
"everything good in life you have to pay for". So with 2 10bit staggered CCD arrays,
we have 20bits resolution, which not too bad! Having cobbled lots of logic
circuits over the last 28 years, clocking and detecting which pixel is hot  is not
as bad as it sounds.

DL



Jim Hannon wrote:

> Having looked into building a spectrophotometer using a linerar CCD array,
> I know a bit about how they work. This web site
> http://www.kodak.com/cgi-bin/webCatalog.pl?product=KODAK+KLI+Series+Image+Sensors&cc=US&lc=en
> shows Kodaks offerings. There are up to 14000 pixels and the pixel spacing
> is around 5um. I would assume to uses this sort of sensor you would project
> the deflected light beam onto the sensor and use the pixel indicating the
> brightest light to indicated the position of the beam. It takes a lot of
> circuitry to clock the data out of the array and then you have to have some
> sort of circuit or computer software to sort out which pixel is the one the
> beam is on. since there are only a few thousand pixels you cannot get much
> resolution from your sensor. In the case of Kodaks biggest 14000 pixels
> less than 12 bits of resolution. This is truly a not so simple solution.
> Thanks, but I will stick with capacitive and inductive sensors. :)
>
> At 02:50 p.m. 30/05/01 -0400, you wrote:
> >In a message dated 30/05/01, blottobear@.......... writes:
> >
> >>LINEAR ARRAY CCD!
> >
> >Dear Dave Latsch,
> >
> >       You have our attention. Please tell us:-
> >
> >      What are the physical sizes of the CCD pixels and how many can you get
> >in a single strip?
> >
> >       Remembering that the movement of the arm on a Lehman may be less than
> >1/3,600 of a degree for the 6 sec background movements, how do you suggest we
> >use these wonderful strips please?
> >
> >       Regards,
> >
> >       Chris
>
> Jim Hannon
> http://soli.inav.net/~jmhannon/
> 42,11.90N,91,39.26W
> WB0TXL
>
> __________________________________________________________
>
> Public Seismic Network Mailing List (PSN-L)
>
> To leave this list email PSN-L-REQUEST@.............. with
> the body of the message (first line only): unsubscribe
> See http://www.seismicnet.com/maillist.html for more information.


 

Hi Jim et al,
The test unit here will use 2 CCDs, one to record fine motion,

say 10 or 12 bits of resolution. The second would be physically

offset to read the "strong" motion now that the "fine" CCD has

"saturated". Using a laser keeps the reflected beam from diverging

very much in the small CCD physical offset distance. Yes, the incoming

laser beam to the mirror needs to be very small, so some optics to

collimate and perhaps a rectangular spatial filter might help. The
real trick

is glueing the two CCD signals together, but this could be accomplished
in

software. Needless to say, the seismo frame and the CCD head must be

rigidly locked together and temp. stable. This is where my small stash
of

Invar metal comes in. And yes, all this is a major effort-but as Carole
King sang

"everything good in life you have to pay for". So with 2 10bit staggered
CCD arrays,

we have 20bits resolution, which not too bad! Having cobbled lots of
logic

circuits over the last 28 years, clocking and detecting which pixel
is hot  is not

as bad as it sounds.
DL

 

 
Jim Hannon wrote:
Having looked into building a spectrophotometer using
a linerar CCD array,

I know a bit about how they work. This web site

http://www.kodak.co
m/cgi-bin/webCatalog.pl?product=KODAK+KLI+Series+Image+Sensors&cc=US&lc=en

shows Kodaks offerings. There are up to 14000 pixels and the pixel
spacing

is around 5um. I would assume to uses this sort of sensor you would
project

the deflected light beam onto the sensor and use the pixel indicating
the

brightest light to indicated the position of the beam. It takes a lot
of

circuitry to clock the data out of the array and then you have to have
some

sort of circuit or computer software to sort out which pixel is the
one the

beam is on. since there are only a few thousand pixels you cannot get
much

resolution from your sensor. In the case of Kodaks biggest 14000 pixels

less than 12 bits of resolution. This is truly a not so simple solution.

Thanks, but I will stick with capacitive and inductive sensors. :)
At 02:50 p.m. 30/05/01 -0400, you wrote:

>In a message dated 30/05/01, blottobear@.......... writes:

>

>>LINEAR ARRAY CCD!

>

>Dear Dave Latsch,

>

>       You have our attention. Please
tell us:-

>

>      What are the physical sizes of the
CCD pixels and how many can you get

>in a single strip?

>

>       Remembering that the movement
of the arm on a Lehman may be less than

>1/3,600 of a degree for the 6 sec background movements, how do you
suggest we

>use these wonderful strips please?

>

>       Regards,

>

>       Chris
Jim Hannon

http://soli.inav.net/~jmhannon/

42,11.90N,91,39.26W

WB0TXL
__________________________________________________________
Public Seismic Network Mailing List (PSN-L)
To leave this list email PSN-L-REQUEST@.............. with

the body of the message (first line only): unsubscribe

See http://www.seismicnet.com/maillist.html
for more information.