Epson V700 vertical banding / streaks issue

Discussion in 'Digital Darkroom' started by artflei, May 18, 2008.

  1. Hello,
    I recently got an Epson V700 scanner to process medium format negatives. So far I'm very satisfied with it but one problem I have is I get subtle vertical banding near the edge of the frame. These streaks visible in the scan are not on the negative (when I turn the negative around and re-scan they appear on the other side of the picture) and are most apparent in pictures with large, monotonous areas - see attached sample picture, on the left. I really hope someone can help me with this - Thanks in advance!
    [​IMG]
     
  2. You write, "when I turn the negative around and re-scan they appear on the other side of the picture".

    That means the streaks are associated with one particular side of the negative; they follow as you turn the negative. That would imply that the streaks are, indeed, on the negative, but not apparent in your visual inspection.

    Try inspecting the negative with a brighter back light and under higher magnification.
     
  3. No, the streaks cannot be on the negative. Maybe I was unclear in my explanation, here are two scans of the same negative (darkened deliberately to make banding more visible), turned 180 degrees. The banding is on the left or right in the picture, the same scanning area is affected, but not the same area of the negative.
    [​IMG]
     
  4. What software are you using? Is it up to date? If it's Vuescan, can you manually recalibrate at the start of every scan session?
     
  5. Roger, I'm using Vuescan indeed (8.4.62) but I tried the Epson Scan software as well and the effect is the same.
     
  6. This is a classic symptom of dust on the "calibration area" of the scanner glass (usually near the end where the lid attaches to the scanner body). The calibration area has to be absolutely spotless. Clean it off and see if your problem is solved.
     
  7. Are you using a frame or are you on the glass directly?
     
  8. Gary, I'm using a negative holder. I bought one from betterscanning.com with Anti-Newton glass (works very well, btw). The banding occurs regardless of whether I use this holder with or without glass or the original Epson holder. I think dust in the calibration area is not the cause of the problem. Right now I'm suspecting light reflecting from the holder might be... I'll have to do some further testing.
     
  9. Arthur, I've recently found the same problem on my Epson V750 scanner, which is very similar to your V700, but supposedly has an extra anti-reflection coating on the flatbed glass. Unfortunately, even with this coating, I'm having the same problem as you. I agree that it's not dust, nor the software, and I'm also not sure it's a reflection from the negative holder. (I have the problem with color and BW negs, 35mm and medium format, and with Epson Scan as well as SilverFast, the pro software with the V750.) I also find the same streaking using the film holder, and when scanning without it by placing my negatives directly on the glassラand rotating the film 180 degrees simply produces streaks on opposite sides of the imageラsame location in relation to the scanner. It seems to be something about the way a very dark (clear film) area is read by the scanner, and it becomes more pronounced the further the film is placed from the center of the flatbed/lightsource. In my tests, I've been using a medium format image that is almost all sky and I find that when the negative is placed in the center of the glass, there are narrow streaks along both sides of the image. (The light sky makes it easier to see the streaks, as with the image you've shown, and darkening the image in Levels makes it even easier to see.) But when the negative is moved toward one side of the glass or the other, the streaks become wider and seem to only be on the outer side/edge, away from the center. Thinking that it might be some sort of "light leak" from scanning the very dark edge of the film (just outside of the image area), I have tried cropping my images more tightly into the image area (in preview) and find that streaks are still there. Iメm attaching a sample test image, where you can even see the double notches in the streak (from a Hasselblad film back) suggesting the streak is a ghost of the image edge and not a reflection of the film holder. And to further confirm my fears, in another test I've found the streaks even in the middle of the image area, when the there is a very dark object in the image, such as a tree. So there is definitely a problem with a ghost-like streaking from any very dark (under- or unexposed film) area being scanned. I wonder if having a film holder that actually masks off everything outside of the image area (thus blocking out all unexposed film along the edge) might be a solution, as the scanner would not モseeヤ and react to that area and thus not ghost it? I wrote Epson Customer Service yesterday, but have yet to hear back from them. I'm hoping they (or someone) will have a solution to this problem. Hopefully itメs a malfunction of an occasional individual scannerナor else it would seem to be a serious design flawナ.
    00PjgZ-47383584.jpg
     
  10. Hi,

    I'm having the exact same issues with my Epson V700. More and more people are becoming aware of this problem.
    Has anyone found a solution yet? Epson support is all, "Have you tried turning your computer on and off, sir?" This is
    driving me mad...
    <br>
    <img src="http://www.marius-rustad.com/3linjer.jpg">
     
  11. i am having the exact same issues with my V700, except mine are yellow blurred streaks.

    http://www.photo.net/digital-darkroom-forum/00PwRj
     
  12. Forgive me for the bumping of such an old photo, I came across this while trying to find a way to improve the luminance in the V700 scans.
    Anyway, the last scan posted by Laura Ryan shows something interesting, it looks like it is a result of a scanner error, as the 'ghost/vertical banding' appears to be an exact mirror image of the film frame (it has the indents in exactly the same places, exact same width etc).
    I'm not sure if newer versions of the Epson software have fixed this, but I'd suggest trying new Epson software/drivers, or lodging a service request with them.
     
  13. I'm now experiencing this same vertical banding problem on my Epson V700. Wondering if any of the above posters ever found a solution to this?
     
  14. After a lot of discussion with Epson Customer Service over many months (where I found the techs were very sympathetic and nice to deal with), and a lot of trial and error testing with subsequent discussions with Espon Support, it seems that this banding issue is inherent to the way the scanner operates. It's not a software issue, and no hardware fix is on the way soon. It may just be a problem inherent in flatbed scanners, or at least the ones that use the high-quality dual scan optics as on the Epson V700/750. The better news though, is that it's an otherwise decent scanner, and some things can be done to minimize the banding/streaking problem.
    Basically, the ghost banding occurs when the clear edge of the film (color and B&W) is visible--and scanned--and shows up most prominently when the clear film edge directly adjacent to an area of the image area which is dense, such as a bright sky. The banding also is more visible on an even image area/subject like the sky much more so than a busy part of an image, where there is a lot of value change and the subject isn't as bright (such as grass, trees, etc.). In short, it's mostly bright and even tone areas that show the streaks. The streaks also tend to be most prominent the further away from the center of the flatbed/glass one scans. That means that scanning with the 35mm film holders, the streaks are widest in slots 1 or 4, and less wide in slots 2 or 3, closer to the middle of the glass. Additionally, if one rotates the film 180 or 90 degrees depending on where on the image the sky or dense area is, then the streaks can be further minimized and sometime eliminated. When doing this, one needs to rotate the film so that the sky is toward the center of the scanning bed, as opposed to the outer edge.
    In sum, putting the film toward the center slots on the film holder and then rotating film so that the sky is toward the center seems to give the best results. Not always perfect, but better.
    (And with 120 film may need to be rotated 90 degrees if it's shot with tops of images toward one end of the film or the other, as opposed to 35mm where the images usually run "left to right" more horizontally in relation to the length of the film strip. Thus, I've adapted my film holder to sit perpendicular to the length of the flatbed glass, by cutting it in half. This allows me to have the film in the holder, rotate it so that the sky in a landscape is facing in toward the center of the flatbed, and still close the scanner lid completely.)
    The only other fix that Epson now has is to use a modified film holder, which has narrower openings for the film. This simply crops the edge of the film so the clear edges are not visible to the scanner. This stops the problem of banding pretty effectively, but I found it unsuitable because I want to print the full image without any cropping at all. (In theory the smaller film holder opening could correspond perfectly with the edge of the image area on a strip of film, but I've found that there is always a little bit cropped from my image.) And sometimes I do want to be able to print the "black border" around the image, which was what I liked about the flatbed scanner in the first place.
    Overall, I find the quality of the scans on my V750 to be quite good, especially for the cost of this scanner compared to dedicated film scanners. The banding problem persists, and unfortunately this means that sometimes one just has to hand correct (remove) the bands in PS or other software. On some images the banding isn't as apparent at first, but temporarily adjusting levels (going darker) or darkening highlights helps make the streaks more easily visible if they are there. Then one can go figure out the worst areas and go to work....
    Hope this is helpful!
     
  15. Thanks Mori for a complete and accurate analysis of a tiresome and annoying problem. Before finding this thread, I had found that the only way to eliminate the banding on sky-to-clear-film edges of 8x10 negs is to mask the edge with black artist's tape.
    In the period the thread has been dormant I wonder if there have been any developments?
     
  16. Im just trying to find a way to mask one 35mm negative in the holder.. But thats really hard.. anyone have a tip with a step to step guide?
     
  17. I occasionally have linear vertical banding that is perfectly straight when scanning 6 x6 negatives on my Epson V500; the streaks continue throughout the scanned negatives. During "Preview" scan with Vuescan, it could be seen beginning before the sensor got to the negative area. It occurred with either Epson's scanning program or Vuescan, and cleaning the glass surfaces made no difference. Reinstalling driver made no difference. Negatives were without visual evidence of scratches or other defects. I scan strips of 4 negatives, so that means 2 are not in the scanning area of the negative holder; the unscanned negatives should be towards the front of the scanner, as there is a small cutout at the back of the negative holder that I guess is used for sensor calibration. Turns out when I inadverdently left that area covered with negative--when I had the strip sticking out of the holder the wrong way-- the vertical bands would appear. Reorienting the film fixed the problem for me. Such a simple cause, but so much frustration.
     
  18. I have an Epson V700 and did some research on this:
    1. It is not a software issue, it happens with Epson scan, Vue scan and SilverFast.
    2. It happens with both V700 and V750 scanners (there are samples on the net of both models), even though the V750 is supposed to have better anti-reflective coating.
    3. It happens with over-exposed, dense negatives, never happens with slides.
    4. The cause of the dark streaks is light reflections (extra light shows up as dark streaks on negative scans). These light reflections occur whenever the clear film base is adjacent to a very dark image area (for example overexposed sky.... the overexposed sky of course looks dark on the negative). The solution is to mask the negative and prevent light leaking between the edges of the scan holder and the film area. The mask can simply be placed on top of the Anti-Newton glass. Since it is difficult to position this mask with great precision, a little bit image loss may occur at the edge. I will try to post sample photos.
     
  19. 5. I have never seen the streaks occurring at the inner edge of the holder, only at the outer edge. So depending on your negative, you may get lucky simply rotating the film strip 180 degrees (upside down) in the film holder. If not, here is what I do: I move the film strip as close to the inner edge of the film holder as possible, to minimize light leaking from that side (this is just a precaution, again, I never saw reflections happening on that side) and then I place a mask on top of the AN glass to completely mask the clear base along the outer edge of the film.
    00buwM-541947784.JPG
     
  20. Here is a sample image. Notice the triple reflections along the right edge of the image. Film is Delta 100 (i.e. very clear film base, allowing for lots of light leaking around the image). Negative was very dark (i.e. over exposed).
    00buwU-541947984.jpg
     
  21. Here is same image with a mask placed on top of the AN glass of the film holder. It is necessary to completely block of light leaking from the side, so you see a little bit of image loss along the right side if the mask is placed to far into the image area. This image loss shows as white edge.
    00buwY-541948084.jpg
     
  22. Note to previous poster Kenneth Dodds: You have identified another cause of black bands, which happens when the calibration sensor on top of the film is covered or dirty. These bands show as perfectly even vertical bands and can occur at any place of the film strip.
    The shadows, i.e. reflections, shown in original poster's and my sample are different, they follow the outlines of the film edge, you can even see the triangular cut-outs of the Hasselblad film backs in the reflections.
     

Share This Page