Lines in Scans from the Minolta 5400: Noise from power line

[svein-frode]

PS! I am posting this as a follow up to this thread:

http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=006I3Q

 

In case anyone has had troubles read this:

 

One of the problems with my Minolta 5400 scanner has been solved. The

lines are gone, and the solution was tragically simple. The scanner

was situated near my HP 990CXi printer (after I had moved it as far

away from my monitor as possible). By unplugging the power chord on

the printer the lines were immediatly gone from the scans. Obviously

the transformator/power unit on the HP 990CXi interfeered with the

scanner.

 

I still can't use it with Firewire, even though the Nikon 8000ED and

Epson 2100 works like a charm - go figure. I downloaded the new

software 1.1.2 - By running it the scanner won't even boot.I'm on USB

at the moment and have ordered a Firewire/USB2 PCI card - I'll keep

you updated.

[geoff_shapiro]

Since my scanner suffers from these lines, I hope that it is really that simple. I'll test this out later today after I get home from work.

 

Something I wonder about - if the line artifacts are caused by powerline interference, would these 1 pixel wide linear artifacts really be occurring in the same places in the scanned images from scan to scan?

[tom_harvey1]

Here's a bit of a tip regarding signal interference: If you have a jumble of wires and cables behind your computer (as most of us do), try to segregate the AC line current-carrying cables from the signal-carrying cables. Don't bundle or tie-wrap these two groups together, because that alone will allow inductive coupling between the 120 volt lines and the low-level signal cables.

 

If the two groups of cables MUST be in close proximity to one another, then try to rearrange them so that the power cables group and the signal cables group cross each other at a 90 degree angle.

 

Also try checking the integrity (good solid connection) of the ground on your computer.

 

TH

[suchismit1]

<i>Don't bundle or tie-wrap these two groups together, because that alone will allow inductive coupling between the 120 volt lines and the low-level signal cables. </i> <p>

 

Spoken like a true EE major :-)

[tom_harvey2]

Aiyeeeee!

 

U-betcha! Went through h-e-double-toothpicks (that would be hell for the layman) with one of my systems a few years back trying to isolate 60 Hz interference....

 

TH

[michael_melford]

Thank you for the answer to this problem. I am new to photonet, and ended up here

as a result of trying to find the answer to this scan line problem. After upgrading to

1.1.2 (through Europe, thank you very much) and isolating the firewire cable, I still got

some lines with my G4 powerbook. However when I made a scan with the G5 and

isolating the cable some more, the problem is solved. Has everyone else found this

problem solved by islolating the cable?

[michael_melford]

Let me make a correction here: Both Svein-Frode and I were mistaken when we said

that isolating the cable solved the problem. I have the scan lines on all my scans from

the very first one, and have since returned my 5400.

Michael Melford

[ian_mclaughlin]

I too noticed scan lines using my 5400 in dark areas of 35mm transparencies but the problem has been resolved. Hopefully the following details will help others with this problem.

 

Initial problem: While, scanning at high resolution above 2700 dpi on a Minolta DimageScan Elite 5400, 1-pixel wide scan lines either red, green, or blue in color appear in dark areas typically but can extend into mid-tone areas and travel the length of the slide in the same orientation as the slide is scanned. These lines become very apparent when the image is even slightly sharpened. The lines may appear more readily in dark areas just because of their bright nature and probably all extend the length of the slide into the highlights. Noise removal software like Noise Ninja can partially cure the problem but the lines still remain to some degree.

 

When I first started using this scanner, I was on a Dell 2.4Ghz laptop with 1GB RAM and connected through the USB1.1 port (I haven't been able to try the Firewire400 port on this laptop). The lines always appeared whether using the latest DimageScan software 1.1.5 or the latest version of Silverfast 6.2.1r6.

 

Apparent Solution: I just recently switched platforms to a Mac G5 Dual 2GHz with 2.5GB RAM and the scan lines are gone for good. I've tried the same versions of scanner driver software and tried the USB2.0, USB1.1, and Firewire400 ports and the lines haven't appeared. As a test I tried the PC laptop again, and the lines came back.

 

While first troubleshooting the problem on the PC months ago I tried isolating cables, switching USB cables, isolating USB cables, and separating devices in case as someone suggested electrical interference was to blame. None of this worked.

 

From my experience, either the USB1.1 driver (software or hardware) on the PC or the scanner driver for the PC caused the line problems.

So, if you have this problem still, at least try your scanner on a different computer system with different USB connectivities.

 

I've only recently reported this situation to Minolta.

 

So, I hope this helps. By the way, there is a considerable speed increase in scans going to USB2.0 or Firewire400 as one would suspect for all who are still trying to scan using USB1.1.