ScanMate drum scanner DIY maintenance, troubleshooting, mods

meloV8 said:
In my test no. In tiff 16bit mode i have worst shadow detail, blured image. This 16bit TIFF is simply destructive mode for overal scan quality. I scan 8bit Raw, then go to Photoshop and convert to 16 bit mode for editing and save file. Rest i do in Silverfast in HDR 48 bit mode.

I bet, that Color Quartet simply scan to 8 bit, add blur for "wow and flutter" and saved to 16 bit tiff without any 16 bit data from scanner.
Click to expand...

Its just pointless to save 8bit to 16bit file... you are not getting any new information. It does offer you more space to work with, but you are loosing a lot of color data compared to the 16bit tiff that includes 12bits or 14bits of actual data. Its a big difference.
 
Gilles78 said:
Thanks a lot meloV8
I also read the service manual ( 5.6 A - adjustement zero point vertical ) and your method, i will try to understand better 😉
I don't need to do this for the moment i think, but may be one day and i would like to be able to do that myself, if needed.
Thanks again 😉
Click to expand...

The zero point is adjusted by software in SM11000. Via the maintenance terminal program.
 
🙂 thanks ... Yes, for me it is the same.




I also have to thanks a lot Margus, who motivated me to get interested in ScanMate scanners. His work and his explanations are great and very interesting.
A big THANK to Fernando and Jack also !!! 😉
 
monkeyfist said:
Its just pointless to save 8bit to 16bit file... you are not getting any new information. It does offer you more space to work with, but you are loosing a lot of color data compared to the 16bit tiff that includes 12bits or 14bits of actual data. Its a big difference.
Click to expand...

Try use 8 bit file vs 8bit->16bit in Silverfast. U can't do anything in Negafix in with 8bit file, same as Color Perfect for negative inversion.
Make, a test. Scan slide in deep shadows in 8 bit raw, and 16 bit tiff. Then try overexpose shadows. U can see cleary that 16 bit Tiff has worst shadows, and bigger noise.

quick comparison :
raw 8bit vs tiff 16bit

25648726545_1bd8c254c4_o.jpg
 
monkeyfist said:
The zero point is adjusted by software in SM11000. Via the maintenance terminal program.
Click to expand...
Yes I know 😉
MeloV8 wrote : "i had problem with Zero Point Vertical.... I thing this is the most important, to minimalization "wow and flutter" effec"
I was just surprised about this point because Michael from ABC Scan told me that there is no link between "Zero Point Vertical adjustement" & wow and flutter.
Wow and flutter problem must be solved with Encoder cleaning, good motor condition ( ball bearing ) and lubrication axes.
 
Zero Point Vertical do not minimize "wow and flutter", but i think that unscrewing encoder and changing position make effect on better overall balance motor axle and encoder axle. I cleaned after encoder but this do nothing better to my scans. Encoder itself is very sensitive for any vibration.

I made a test. Left scan is ok, right scan had some big flutter, because when scanner making scan, i hit gently into encoder enclosure.

25027357623_7301551dec_o.jpg
 
Yes i understand perfectly 😉
On my SM3K, I also found the same problems with the encoder.
Very sensitive to vibrations. And vibrations which were also related to the coupling with the motor, a few 1/100 mm exentrements... It was difficult to solve. It is OK now.
 
Great! My English is very limited 😉
Now i must service my SM3000. Scanner is in very good condition, low working hours, but this is old machine. Time to clean optics, some adjustment and encoder "calibration".
 
meloV8 said:
Try use 8 bit file vs 8bit->16bit in Silverfast. U can't do anything in Negafix in with 8bit file, same as Color Perfect for negative inversion.
Make, a test. Scan slide in deep shadows in 8 bit raw, and 16 bit tiff. Then try overexpose shadows. U can see cleary that 16 bit Tiff has worst shadows, and bigger noise.

quick comparison :
raw 8bit vs tiff 16bit

25648726545_1bd8c254c4_o.jpg
Click to expand...

I dont think its smart to use these types of adjustments, like exposure or deep shadows of whatever. Just use levels & curves, this way you know what you are actually doing. There is basically no other ways to adjust an image, all of these weird slides adjustment usually just mess things up.

This could be also a colorprofile issue, as the 16bit tiff file is embedded with some color profile. You would need to make a custom profile for the scanner.

Or just check that you are using the same colorprofile for both files.

I had no problems with 16bit tiff on SM5000.
 
Gilles78 said:
Hi Monkeyfist, thanks ! I understand... but ...
You wrote : "I also switched the internal color gel filter to a proper multicoated glass filter."
when I dismount my P2, I do not remember seeing a filter or gelatin ...
Where is it situated ?
Is it a circular screwing filter ?
Regarding SM11K I tried to scan with the mode "TIFF RAW 16B" but the result was very strange, very dark image, and heavy USM in the picture, strong contrast, all this was not neutral at all.
Is this the same on your side?
Click to expand...
@ Monkeyfist, I'm not sure you read me 😉
 
Gilles78 said:
Hi Monkeyfist, thanks ! I understand... but ...
You wrote : "I also switched the internal color gel filter to a proper multicoated glass filter."
when I dismount my P2, I do not remember seeing a filter or gelatin ...
Where is it situated ?
Is it a circular screwing filter ?
Regarding SM11K I tried to scan with the mode "TIFF RAW 16B" but the result was very strange, very dark image, and heavy USM in the picture, strong contrast, all this was not neutral at all.
Is this the same on your side?
Click to expand...

There is a filter at the top of the lens, blue piece of color gel. It's under the glass UV/IR filter.

On my SM11000 16bit raw works fine, sounds weird if you have USM in a raw. Though i always switch sharpening off even when scanning in raw, so i do not know will it affect it. Best to turn all sharpening off. Very dark image is probably normal, as its 14bit in 16bit file.. so all the data is in one end of the file. This is corrected via levels, by just moving the middle grey point. White calibration also might affects this, its sometimes good to do the white calibration on the base sheet of the film. If you wish to get all the data possible out of the film.
 
Thanks, monkeyfist 🙂
About P2, next overhal time ( soon ), i will check if there is a blue piece of color gel 😉

For SM11K and Tiff 16bit Raw, yes, switcched off USM ... but looks like USM inside ... I don't know why, i need to pratice more the CQ raw mode to understand better. For the moment, i use to scan with Tiff 16 bit, without any adjustement and USM, no embed file. And assign after with PS a correction profil ( generate with X-Rite i1 Profiler ).
About the dark image via Tiff 16bit raw, i think you are right with the white calibration on the base sheet of the film. It should be better
 
Hey guys,
I have a question regarding the output in the 11K. A friend sent me a RAW file with a scan of an IT8 target. I noticed in PS, the RGB values for the fields L17, L18 and L19 do not match the info in the reference txt file for the IT8 target. It seems like the app (I think he's using CQ) is "touching" the data.

Any idea why the data is being processed and how to disable it?
 
There is no ideal scanners, so you need to calibrate the scanner. Scanmate have typical color casts and you need to calibrate it.
 
A quick update regarding the red cast in the shadows for these scanner. I connected with a fellow drum scanner enthusiast, Armando Vergara, who is an expert with this technology. After reviewing a few of my scans, he suggested that I should add a new UV-IR filter to cut down extra red light. I received the filter he suggested earlier this week and installed it it today and it has completely resolved the red cast issue for me in my Scanmate 11000.

There is a square glass filter already in place right in front of light receiving fiber optic cable where the bulb projects the light onto the fiber. There are three screws holding the fiber cable to the bulb frame that you can easily open. Once opened, the end of the fiber cable is exposed and you can see the square filter that was right infront of the cable. I replaced this square filter with the one show in the link below. I had to make a quick tape DIY mount to make sure the circular filter sat in center of the fiber cable.

So far, I haven't seen any other impact on the scans from doing this but the red cast issue is completely gone. I had to re-calibrate (IT8) my scanner after installing the filter.

eBay UV-IR filter link

Hope this helps someone. I'll be doing this with my Scanmate 5000 next.

Regards, Pali
 
Netsoft2k said:
After reviewing a few of my scans, he suggested that I should add a new UV-IR filter to cut down extra red light. I received the filter he suggested earlier this week and installed it it today and it has completely resolved the red cast issue for me in my Scanmate 11000.
Click to expand...

Pali,

Sound very interesting, especially considering how easy the mod is.

Did you compare - does it maybe kill deeper reds too much (i.e. Velvia reds) or did it affect any of the shorter wavelenght performance, i.e. how are the blue hues, did you notice any difference?

Also if there's more "halo" effect (glow on the corners of very contrasty details - scanning sprocket holes are good test for this)? Usually more (uncoated-) glass on the light path increases the halo which can be annoying.

The Hamamatsu 1617 PMTs used in ScanMate 11000 go from near UV of 350 nm to quite far infrared 850 nm, hence reds and blues are rendered beyond to most film emulsion capabilities. For UV most organic glass is enough to block it and I'm sure the fiber optic path already serves this function, but infrared is more "penetrating" type of light, hence it goes pass it I reckon.

Let me know how this mod works in practice, I might do this on mine as well.

Margus
 
You must log in or register to reply here.
Back
Top Bottom