Sensor noise in film scanner

[pentaxpete]

I have TWO Epson flatbed scanners -- a 1650 which has a broken wire somewhere and scans with loads of coloured lines only and a 2480 with a smaller light in it and will scan only three 35mm frames against the 6 for the 1650 . i have some 'home-made masks' for Medium Format films,Now I seem to have strange 'scan lines' using the 2480 on medium format and 35mm . i HAVE cleaned the glass top and underneath but have not found the sensor to check if it is 'dirty' -- can anyone help me what is happening? I use 'Vuescan Pro' which a won FREE some years ago. I set 150 for 'pre-view' and 1600 for 'scan' . I attach some samples of scan lines ---

The Medium Format is Fuji 160 NPC in Minolta Autocord, the Rainbow one is outdated Fuji 'Surveicolor 400 rated 200 ASA ' in Olympus XA

 

[Saganich]

Can anyone comment on sensor noise in a film scanner? I have a Canon FS4000us scanner, which is considered to be an excellent scanner. I have done some rough testing of the scanner, and it looks like the noise level is several bits, even when scanning in 8 bit mode.

I still haven't yet done a rigorous test, but I thought I would see if anyone here has done a rigorous test of their scanner, or even a non-rigorous test, or even has an informed opinion, or even a non-informed opinion.

I think the overlooked question is if vuescan is functioning correctly. I've had some odd issues with updates in 2018 needing to revert to previous versions with older scanners. Might be good to try different software or older version of vuescan just to eliminate it as a variable.

 

[pentaxpete]

Oh Dear ! I've just 'upgraded' to the latest version 9.6.20 !

 

[shutterfinger]

Those vintage scanners gas out and the residue that settle on the mirrors and lens cause the colored lines.
Start a scan then pull the power cord from the rear of the scanner once it has moved a few inches from the home position. The scan block will now be positionable anywhere on the track until pushed back to the home position where it will lock. The scan block will return to the home position on next power up.
Remove the bed, the cover over the lamp, the mirrors if possible. Clean the lamp, the mirrors, and the lens. They are on the front of the scan block.
The sensor is on the rear of the scan block. DO NOT remove the sensor unless it is absolutely necessary to clean it.
You will need a very good linearity target to scan to align the sensor on reassembly. The procedure is secure the sensor, install the bed, scan the linearity target, remove the bed, reposition the sensor, install the bed, scan the target until the scan is the same as the target.

 

[pentaxpete]

thanks -- i took off the cover again and cleaned the underneath of glass BUT I could not see any mirrors and scan lens -- it all 'seemed' dust free but I will try again .

 

[Ted Baker]

thanks -- i took off the cover again and cleaned the underneath of glass BUT I could not see any mirrors and scan lens -- it all 'seemed' dust free but I will try again .

I would not touch anything to do with the mirror, sensor or lens on these units. Unless of course you have decided its already dead. BTW is all in the little box that moves along, is just a little digital camera really

The fact that you are seeing colour banding suggests to me at least this is hardware fault, or even software fault, not dust. But this is just a guess!

 

[shutterfinger]

Once the cover on the carriage unit is removed ( 2 screws) the lamp and mirrors will be visible. All Epson scanners use the same basic build and disassemble similarly.
The mirrors are front surface.
I use microfiber lens cleaning cloth to clean the mirrors.

 

[nmp]

Then I did a scan and looked ad the region in the frame around the image. That region is imaging thick black plastic, so no light should be getting through. That too had some width in the distribution of values. It is also not quite definitive because the sensor could be picking up some stray light from somewhere.

There is indeed light reflected from the neighboring areas which introduces error in the measured value for a given spot. The amount of error is dependent on the luminosity of the neighboring areas. This is studied extensively by Doug Gray as presented at the Luminous Landscape forum:

see for example: https://forum.luminous-landscape.com/index.php?topic=126590.msg1065018#msg1065018

Just another complicating factor when any quantitative analysis is done on a scanner.

:Niranjan.