Scanning artefacts when including negative borders on Epson V850 Pro

[MikeOates]

I have had this problem before when scanning Fuji Instax prints where I want to include the white border in the scanned image. I got/get what looks like flare from the white border spilling onto the image area and looked like shadows cast along the full length of the edge. I never sorted that one out, so had to leave borders out.

Now I am scanning 4x5 negs and I am getting the same effect. I show 2 images below, one shows the banding at the left and right edges, more obvious in the highlight areas, so you can see it better on the right hand border.

If I mask with black card as the other image shows along one edge, the problem goes away.

But is there a way round this when you want to show the full negative scan?

I am scanning with an Epson V850 Pro, using an inverted dry/fluid mounted system very similar to "The BetterScanning.com Variable Height Mounting Station"

So do others have this issue? Is there a workaround?



Note the top and bottom of the image does not suffer from this at all, so it has something to do with the way the scanning head moves.

Thanks,

Mike

[Corran]

Looks to me like a simple global change in the black point, induced by "seeing" the opaque black card in the scanning frame.

I see a change throughout the whole image, not just the middle. Try your test again but lock the settings for the scan and see if there's any difference. Start with the black card in place, or just make sure your black point isn't clipping your inverted highlights on the scan.

[MikeOates]

Hi Corran,

No it's not that, but I did just do another scan to confirm it, even just scanning the image with a mask revealing the border still gives the same result. (I am using SilverFast AI Studio, and I selected a small area just from the image to scan)

Looking at the size of the images I posted, there is no way you will be able to see the artefact clearly, so I have done a higher res version showing the bottom right corner.



You will see it consists of two bands of different grey levels. I think it has something to do with refraction / reflection of the light through the glass.

In this case I put a mask above and below the negative, which made no difference.

So to be clear, the layers being scanned are: (in order from the scanning head)

Scanner glass
mask
mylar
negative
glass (2mm anti reflection glass)
mask (this was just added to for this example)
light source in lid

In practice of course the scanner glass is the lowest component.

Mike

PS: I should add that this issue was the same with my previous Canon scanner with Instax prints, just as this scanner does. And it's only on the longer edges, i.e. along the length of the scanner. So it's not related to transmissive scans, it is any scan that has a bright edge.

[paulbarden]

Mike,
This is a problem that Epson has acknowledged that their scanners have. Any time the first frame the scanner encounters shows unexposed film border next to a bright (dense) area on the image, it will create a band of density that travels the length of the side of the film. With 120 negative strips, this can affect the whole strip. Its unclear from my correspondence with Epson whether this is entirely a software problem, or a combination of hardware/software. They have a solution to offer their users, though: they will ship you new film holders that have a smaller opening, so you can't scan the clear margins!

Here is a very obvious example of the problem I encountered when I bought a V750 a few years ago:



What is the fix for this? Don't scan the clear film margins if there is a dense image area adjacent to it. Sometimes you can rotate the negative so that the first part of the negative the scanner encounters doesn't have this strong margin-to-image contrast. Otherwise, you have to omit the clear margins, or be prepared to photoshop the snot out of the resulting image. I would have hoped that Epson would fix this problem with the newer models, but it seems not.

[Corran]

Ah, I see now. Interesting, good info above. Back when I used a V700 I never scanned the frame edges so never encountered this.

[MikeOates]

Thank you Paul,

Looks like I will have to look at scanning twice and combining for those images, a bit of a pain though.

Where you say "they will ship you new film holders that have a smaller opening" did you mean a 'larger' opening?

The frames that are supplied with the scanner, actually crop off quite a bit of the image, at least it does on the 4x5 frame.

Regards,

Mike

(Paul, That's a rather nice image)

[jonby]

This is quite a well-known problem with all of the V-series Epson scanners (and probably lots of others too). I believe the light bands are essentially reflections of the bright film base at the edge of the frame. They are caused by bright light passing through the frame edge, bouncing off the scanner glass back up to the film (or in this case mylar), and then towards the lens. In most cases there are two or three reflections due to the multiple surfaces involved (eg top and bottom of the scanner glass).

I think you'll find that it is worse towards the edge (right or left) of the scanner bed than in the center, in that the reflections move further into the image area. In fact, if you line one edge of the negative up with the centre of the bed, there should be almost no banding on that edge. Looking at Paul's image above, notice the banding is only on one side. I suspect that was the side on the outer edge of the bed.

You have already found that the problem doesn't occur strongly along the horizontal edges. This will be because at the time of capturing the image just inside these edges, there is not much light passing through the adjacent clear base, because the light source is a moving thin strip. Therefore, if you place your negative so that the edge/s which have the most density lie either horizontally or down the central axis of the bed, you should avoid the worst effects of this.

In theory also, this effect should be much reduced if you placed the negative directly on the scanner glass (or at least closer to it). Of course, you won't capture as much detail doing this, but in some cases this may be the lesser of two evils.

Finally, are you scanning with the emulsion face down or face up? In theory, reflections could be reduced slightly with the less reflective emulsion side down, though since you have a layer of mylar there also, I doubt this would make a lot of difference.

[MikeOates]

Thank you very much jonby for that extra information, most helpful.

The negative is about 2mm away from the glass, the best distance I found to get maximum sharpness and yes the negative has the emulsion side facing down, but like you say, the mylar means it remains reflective.

Looks like I have some experimental scans ahead, it's a good job that I have more time on my hands at the moment as many of us do across the world!

Regards,

Mike

[jkober@mac.com]

Hey guys, thanks for this discussion. Mike, can you tell me where you found "an inverted dry/fluid mounted system very similar to "The BetterScanning.com Variable Height Mounting Station"? Betterscanning.com no longer seems to be a live site. at least not that I can get any kind of response from them.
thanks.
Jeff

[MikeOates]

Jeff,

I could not find any so I made my own. I got a picture framing shop to supply me with 2mm anti reflection glass and I made my own holder using plastic sheet. I had to experiment to get the right distance from the scanner glass for best focus.

I got my plastic sheets from ebay, just search for "plastic modelling card". I used superglue to join the pieces after first using tape until I got the focus right.

Mike