Re: Digitizing glass plates.
You can try scanning as a positive, then invert the image later in PhotoShop. To do that, just set-up the scan based on the histogram, giving an even overall spread to the tonal range. The idea is to keep enough information to allow easier adjustments.
Here is one I scanned a while ago. This is an old family photo glass plate.
Ciao!
Gordon Moat Photography
Re: Digitizing glass plates.
The i900 wouldn't be my first choice but I have one question .... are you putting the glass plate on the glass tray in the film path or putting it on the reflective scanning bed on the top of the machine? If you are doing the latter that is your biggest problem.
Re: Digitizing glass plates.
on the glass tray, glass down emulsion side up. same with other negatives.
Re: Digitizing glass plates.
thanks, I'll give it a try.
Re: Digitizing glass plates.
I scan glass plates on my Epson 4990 regularly. Just place them on the glass emulsion down, adjust output levels etc, and scan. I'm not sure how the i900 works, but are you adjusting the output values in the scanning software? Also, could it be flare around the plate? Can you set your crop lines to the edges of the plate?
Re: Digitizing glass plates.
I'm scanning some 5x7 glass negs right now on a fairly ancient Microtek 8700. I lay them on the glass patten (for 10x8 film) rather than on the reflective scan surface and they work just fine. Wish the negs were better quality though! Good luck.
Re: Digitizing glass plates.
I forgot to mention that I'm scanning the plates as b/w negatives, using the film area guides and the transparency lamp cover.
Re: Digitizing glass plates.
A further thought. I use SilverFast software with the 8700. I make adjustments to white and black points/curves etc, if only to observe if the emulsion on the glass neg responds to SilverFast better that I think it will to Photoshop CS, post-scanning.
Also, ancient glass plates (in my experience) can have a limited tonal range, particularly in high contrast scenes although I suspect that anyone taking the trouble to use 10x8, no matter how long ago, would have a pretty good idea how to get the best out of the emulsion/exposure/development process. Perhaps you might have to do quite a bit of post-scanning work on various areas of the images.
Re: Digitizing glass plates.
Back to the original question and I apologize for not dealing with this right up front.
Are you manually setting your white and black points? If you aren;t this alone can cause the sort of problems you describe. Second, you are always better off scanning b&w as rgb and then converting in PS by dropping out the red and green channels. A small bit of blatant self promotion here .... we have a scanning workshop coming up in April at Midwest Photo, another at foto3 in June and are now working out details for a West Coast Workshop at either Samy's or Freestyle in LA.
You should have no more difficulty scanning glass plates than scanning b&w negatives (well maybe a bit more but not much).