[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Green Light



I would like to support a comment made by both Dave Corbitt and 
Christopher Bacon (both on 3/16/98).

Dave indicated that the shape of the CRT spectral response to be...

>very smooth and continuous, albeit somewhat slim in Blue and far Red.

We have recently measured several CRTs from two different manufactures 
and found them to be strikingly similar, and both consistent with Dave's
observations.  I will try to get a copy of the response to Rob for 
posting.

On the same topic of green light, Christopher wrote

>Of course there is a lot less light on the other side from a
>CRT than a xenon bulb, but that doesn't prove anything since some types of
>sensors need very little light to operate, while other types need more.

Mr. Bacon brings up a very critical point.  The photomultiplier tubes 
used to 
receive the light generated by the CRT have an extremely wide gain range, 
in
excess of four usable decades (10^4).  This gain is optimized for each 
transfer
by adjusting the voltage supplied to each PMT.  Adjusting the gain of the
sensor to match the film and CRT output is roughly equivalent to 
adjusting the
light output to match the film and the sensors.  In both systems it 
optimizes
the operation of the system.

As has previously been discussed, the advantage of adjusting the light 
output
is that the SNR of the sensors remains constant.  This is indeed true, 
but it
should be noted that over a significant portion of the usable gain of a 
PMT,
the noise does not increase linearly with gain.  There is a substantial 
portion
of the PMT operating region where the noise is effectively fixed, and is
determined more by the design and construction of the PMT than by the 
gain the
PMT is operated at.  

Our measurements of PMT noise using a low noise preamps and a modified 
bias
network indicate that the SNR from the PMTs typically used today can be 
better
than 80dB (computed for a 7.5MHz bandwidth). Even with very high gain 
(-1000V PMT
voltage) the PMT noise increased by just 10 to 12, still providing 
approximately 
70dB of SNR. Our evaluations of several high gain/low noise PMTs 
indicates that 
these devices may have 10 to 15 dB better performance than the PMT 
currently used
in most Cintel telecines.  Unfortunately, these high gain PMTs will not 
operate 
properly with the PMT bias networks which are currently available today.

I hope to be able to perform some specific performance tests based on the 
film 
types and conditions mentioned by previous TIG contributors. As I 
complete these
tests I will attempt to keep the TIG informed.

I provide this information in an attempt to augment  the discussion which 
has gone 
before.  I think it is important to recognize that there is a lot of 
signal
processing performed on the output of the sensors, and that this 
processing may
contribute as much, or more noise, to the final image than the sensor 
itself.

Jeff DesCombes

President, Director of Engineering
Sprocket Digital
Burbank, CA  USA

Sprocket Digital manufactures professional video equipment which can be 
used with
any type of telecine as well as system upgrades for Cintel Telecines.  
Sprocket 
Digital has no marketing or manufacturing arrangement with either Cintel 
or
Philips.  Anyone interested in the details of the above mentioned results 
should
contact me and I will provide the specific test conditions and results. 

---
Thanks to Complete Post L.A. for supporting the TIG in 1998..
No product marketing allowed on the main TIG.  Contact rob at alegria.com
957 subscribers in 36 countries on Mon Mar 16 18:47:32 PST 1998 
complete information on the TIG website http://www.alegria.com/tig3/