Nikon CoolScan V ED Ghosting Problem

I scanned some slides I developed last weekend and ran into a problem. It looks as if I'm getting some internal reflections in the scanner that cause a bit of a ghost, or double, image for some of the brighter areas of the film. This is very pronounced where there is a sharp transition from a very bright area (clear acetate) to a very dark area (opaque acetate) of the slide. This is definitely not on the film. The film looks fantastic, with no ghosting and much higher local contrast at these areas. I really haven’t seen this problem before or if I have it may not have been as pronounced. It looks as if it may be caused by internal reflections in the scanner. The film is Kodak E100G I developed myself, cut in strips of 4 frames, and feed through the strip feeder like negative film. The film is not flat, having a bit of a curve laterally across the strip but not extreme and no more than any other type of film I develop and scan. I have scanned quite a bit of E100G before as well but probably not with transitions as sharp dark to light as these. The film feeds fine. Has anyone else seen this issue? If so are there any suggestions as to how to reduce its affect. Here are the shots. http://www.lamarlamb.com/On-Film/Technical-Stuff/35mm-Film-vs-Digital/9301207_vi24p#1151511811_gCGLo The first several pictures in the gallery (of the cameras) show the problem. There is a digital image for reference. The effect is most pronounced in the pictures of the F100 and the F4.

Thanks for any suggestions or help,
Lamar

My understanding is that this is not uncommon among Coolscan V and LS-5000 users, since I think they have similar (if not identical) optical paths.

I don't have links, but there are at least 2 fairly involved discussions on photo.net that describe this phenomenon (Searching LS-5000 and ghosting should get you to them). First thing, make sure the 45-degree mirror below the slide stage is absolutely clean. There are instructions online for how to do this. Cleaning it could make the flare less bad, but probably not make it go away. Make sure you're using the absolute minimum scanner exposure level needed for that image, and try re-orienting the slide so that the ghosts go a different direction and are less noticeable. Short of custom modifications to the optical path, like those discussed in one of the photo.net threads, I'm not sure that there's much else to be done about it.

Here's a good example I personally witnessed with a Stouffer step wedge. (Note: Step 20 is ~3.0 logD, and the clear text is ~0.06 logD, a pretty extreme example, but probably not too far off what Lamar is seeing with the specular highlights in his images.) The upper half is the straight scan, and the lower is after an aggressive, but not unreasonable, IMO, levels adjustment in PS.

Sorry I can't be more helpful.

--Greg

Thanks for the info, it really helped. After reading several of the threads it appears the most likley issue is dirty optics. I now have to decide how I want to handle cleaning, I either try it myself or just eat the cost and send it to Nikon, probably around $275 based on other posts. I usually do these things myself but considering the V-ED is out of production and can't be replaced and that the task is delicate I think I may just go ahead and send it off.