Wanna see some cool copies?

[blackcat4866]

KM-3051ci, 1200 total prints.

Copies off the glass and prints are normal. Scans and copies through the DP-770 are focused, clear and totally off color. In the pix the lower half is the copy off the glass or print, upper half copied through the DP. I know it's hard to believe but it is the same original.

KM3051ci DP Color1.JPGKM3051ci DP Color2.JPG

As you can see I used scans and my Colorpic tool to break down the RGB/CMYK color numbers of each area. I expected to find some interesting pattern, for example: 22% too much cyan across the board or 12% less yellow across the board. Wrong, it's far more interesting. Each color defined in the pixs are uniquely off color. Most have 40% or more too much yellow, but some have 24% less yellow. It's as though the color catalog was randomly re-organized.

What have I done?

First I thought that it was possible that when light source #1 reached the slit glass, it was tipping a mirror, or perhaps a loose mirror ... something like that. I checked the wire ropes, tension, mirror mounts, spring clips, focus adjustments and screws, cleaned the ribbon cable ends and re-seated the connectors.

Second I thought that perhaps the RADF was somehow out of adjustment: height, rotation, etc. So I adjusted the rotation, checked the height, and moved back the black strip ~1.3mm in case the scanner was catching sight of the edge of the strip.

Third I printed off the Maintenance Report and picked out any setting or adjustment that might have anything to do with DP image quality, and read about each one.
U063 only made a difference at it's highest setting (18), then it started to shade worse.
U068 didn't really have any quality affect between -15 and +38. Below -15 light source #1 is starting to scan the underside of the white plate, rather than the slit glass. Surprisingly, this is the best way to align the RADF with the slit glass. Set to -14 you get shading that can be made even front to back by rotating the DP. It's a ton easier than using a small ruler and flashlight, and trying to read the small marks on the ruler from 15" away, in the dark.
U074 is designed to compensate for background from light blue originals. At it's highest setting +3 it removes most of the yellow/green background, but does nothing for the totally off color representation.
U425 stores some sort of color settings. I made some pretty radical changes that had no effect on the problem.
U429 didn't seem to change anything, off the glass or through the DP. The manual is a little confusing here. In one place you think the setting ranges from -5 to +5. In another place it seems to read 1 to 10. It's the latter.

Fourth I went through the user original settings to see if I could change the issue at all.
Original Type-defaulted to Text/Photo, Book/Magazine. Both the Text settings have less yellow/green background. The Graphics/Printer setting and the Custom1 setting made it much worse, more like a heavy lime green background.
Background Density Adjustment-defaulted to Off. Set to Auto entirely removes the yellow/green background but does nothing for the off colors.
Prevent Bleed Through-default to Off. Set to On removes some of the yellow/green background.

Fifth I reloaded firmware.

Conclusions?
I've decided that the mechanical scanner and CCD are OK.
I believe that the color problem is electronic. Image processing, at the ISC (scanner PWB) or Main PWB.
I've ordered the ISC PWB, and the two Charts to calibrate the CCD.

So maybe you read to the end of all my thoughts here. I feel like I'm missing something. What did I miss? =^..^=

[Hansoon]

Seeing that thread title I was hoping for some decent porn, but no.........

Sorry cant't help here but would like to know what model of machine this is.

Hans

[NeoMatrix]

Could it be something like an alpha channel grayscale adjustment?
User tools |calibration |grayscale ?

[wragsdale]

I'm not sure if you noticed this, but your problem copy looks hue-shifted somehow.

I overlayed an adjusted and cropped image of your original area above the problem copy (original on the top, modified image on bottom). The hue was shifted down 120 degrees and the saturation adjusted slightly to make up for some small lighting differences.

KM3051ci DP Color Shift.JPG

[JVergin]

U411 Needed.jpgU411 Needed!.jpg
We experienced that same exact thing when we replaced an ISC board in a 2550ci. There was a yellowish background that looked identical to yours and the colors were completely out of wack. Other than that, it looked great! After replacing the board, the scanner had no idea what colors were what.

In our case though, it was the same result from both the glass and the DP. Prints and internal prints were fine.

What we had to do was use that 90 dollar test chart # 7505000005. In Maint mode U411 (Adjusting the scanner automatically) you can set the target value by either manually entering the numbers on the bottom of the chart, or letting it scan it and set it up automatically.

I do notice in U411 that there is a separate adjustment for the DP, which says Do not use under the description for chart 1. Maybe if you have chart 2 302AC68243, you could try to run DP FaceUp (Chart 2) to set the target properly.

[blackcat4866]

@wrangsdale: That's really interesting! What sort of software did you use to alter the the image color?

@LVergin: The Charts didn't help with the original ISC PWB. With the board I also ordered a new set of charts, in case they were too dirty/worn. The new ISC improved the image immediately, but the directional lines in the gray were still invisible until I ran the U411 calibrations.
=^..^=

[Iowatech]

While it has been quite a while since I've ever worked on any Konica based equipment, could it be that the document feeder is now set up to default to fax or some other process for no apparent reason? I suppose scans from the document feeder could be taking a different path through image processing than scans off the platen glass if that's the case. While that's probably a pretty stupid idea, it could be a path you may wish to explore if nothing else works, I guess.
Sorry if that was a waste of time.

[blackcat4866]

While it has been quite a while since I've ever worked on any Konica based equipment, could it be that the document feeder is now set up to default to fax or some other process for no apparent reason? I suppose scans from the document feeder could be taking a different path through image processing than scans off the platen glass if that's the case. While that's probably a pretty stupid idea, it could be a path you may wish to explore if nothing else works, I guess.
Sorry if that was a waste of time.

It's fixed. It was the ISC PWB. My guess is that when the CCD registers the RGB data, it references some sort of color table. One table for the contact glass, and another table for the DP. Whatever form that data is stored as must somehow be corrupt, at least on the document processor side.

The new ISC improved the image immediately, but the directional lines in the gray were still invisible until I ran the U411 calibrations.

And it's a Kyocera, not Konica.

And yes, it's definitely an image processing issue. =^..^=

[Hansoon]

** Off Topic **

90 dollar test chart # 7505000005

Never heard of, would love to see a sample of that. What's so special about it?

Hans

[JVergin]

** Off Topic **



Never heard of, would love to see a sample of that. What's so special about it?

Hans

It's just a chart with color bars. The only special thing about it is on one side of the test chart, it has the exact values of each color in case you needed to manually enter each one. It even comes in a shiny cover sandwiched in between two cardboard pieces.