scanning from ADF on Epson WF-7710 has page offset

Bug #2048433 reported by Bernard Moreton
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
simple-scan (Ubuntu)
New
Undecided
Unassigned

Bug Description

Scanning from the ADF on my Epson WF-7710 produces images with a left margin about 4 times as large as when scanning from the flatbed, so that the right-hand side of the page is lost.
Top and bottom margins are handled correctly.
The scanner is connected via the local ethernet.

Memory tells me that it used to scan ok from the ADF, and that the problem may have started sometime in 2023 - but memory is terribly fallible ...

$ lsb_release -rd
Description: Ubuntu 22.04.3 LTS
Release: 22.04

Synaptic gives the package version as 42.0-1, the latest available.

Tags: libsane
Revision history for this message
Bernard Moreton (bernard-moreton-1) wrote :

simple-scan -d gives
[+10.76s] DEBUG: scanner.vala:348: sane_get_devices () -> SANE_STATUS_GOOD
[+10.76s] DEBUG: scanner.vala:374: Device: name="escl:https://192.168.1.73:443" vendor="Epson" model="WF-7710 Series" type="platen,adf scanner"
[+10.76s] DEBUG: scanner.vala:374: Device: name="airscan:e0:EPSON WF-7710 Series" vendor="eSCL" model="EPSON WF-7710 Series" type="ip=192.168.1.73"
[+61.34s] DEBUG: simple-scan.vala:1820: Requesting scan at 300 dpi from device 'airscan:e0:EPSON WF-7710 Series'
[+61.34s] DEBUG: scanner.vala:1778: Scanner.scan ("airscan:e0:EPSON WF-7710 Series", dpi=300, scan_mode=ScanMode.GRAY, depth=2, type=adf, side=front, paper_width=2100, paper_height=2970, brightness=-11, contrast=10, delay=0ms)
[+61.34s] DEBUG: scanner.vala:863: Processing request
[+61.35s] DEBUG: scanner.vala:924: sane_open ("airscan:e0:EPSON WF-7710 Series") -> SANE_STATUS_GOOD

The first entry was requested (escl:https://192.168.1.73:443), but the second entry was obviously taken instead; and "type" for that device was discovered as an IP address, though in the request "type" is shown correctly as "adf".
It may be that "adf" is being ignored, and that the scan is actually assuming a use of 'platen'?

At least twice later in the output there are entries of the form
[+87.21s] DEBUG: simple-scan.vala:1669: Unable to find colord device airscan:e0:EPSON WF-7710 Series: property match 'Serial'='sane:airscan:e0:EPSON WF-7710 Series' does not exist

Revision history for this message
Bernard Moreton (bernard-moreton-1) wrote :

I commented out "airscan" in /etc/sane.d/dll.d/airscan, and now simple-scan sees just the one device, presumably discovered now simply from libsane and no longer from sane-airscan? -
but the result is still the same - an over-large left margin, and loss of image on the right-hand side.

Running "scanimage -d 'escl:https://192.168.1.73:443' --format=pdf --mode Gray --source ADF -x 210 -y 297 > /tmp/scanimage.out" gives the same result, so the apparent confusion in simple-scan when there appear to be multiple devices is probably not the cause of the output problem...

Plugging a usb stick into the Epson device and scanning from the device itself gives a perfect result,
but it would be good to be able to scan correctly from Ubuntu.

tags: added: libsane
Revision history for this message
Bernard Moreton (bernard-moreton-1) wrote :

Epson WF7710 supports A3 scanning, and this may be the cause of the problem.

scanimage --help gives
Options specific to device `escl:https://192.168.1.73:443':
  Scan mode:
    --mode Lineart|Gray|Color [Lineart]
        Selects the scan mode (e.g., lineart, monochrome, or color).
    --resolution 100|200|300|600|1200dpi [100]
        Sets the resolution of the scanned image.
    --source Flatbed|ADF|ADF Duplex [Flatbed]
        Selects the scan source (such as a document-feeder).
  Geometry:
    -l 0..289.052mm [0]
        Top-left x position of scan area.
    -t 0..423.672mm [0]
        Top-left y position of scan area.
    -x 8.12799..297.18mm [297.18]
        Width of scan-area.
    -y 8.12799..431.8mm [431.8]
        Height of scan-area.
  Enhancement:
    --preview[=(yes|no)] [no]
        Request a preview-quality scan.
    --preview-in-gray[=(yes|no)] [no]
        Request that all previews are done in monochrome mode. On a
        three-pass scanner this cuts down the number of passes to one and on a
        one-pass scanner, it reduces the memory requirements and scan-time of
        the preview.
    --brightness -100..100 (in steps of 1) [inactive]
        Controls the brightness of the acquired image.
    --contrast -100..100 (in steps of 1) [inactive]
        Controls the contrast of the acquired image.
    --sharpen 0..3 (in steps of 1) [inactive]
        Set sharpen value.
    --threshold 0..255 (in steps of 1) [-1226889112]
        Select minimum-brightness to get a white point

Type ``scanimage --help -d DEVICE'' to get list of all options for DEVICE.

List of available devices:
    escl:https://192.168.1.73:443 airscan:e0:EPSON WF-7710 Series

scanimage -d 'escl:https://192.168.1.73:443' --format=pdf --mode Gray --source ADF gives an A3 scan
scanimage -d 'escl:https://192.168.1.73:443' --format=pdf --mode Gray --source ADF -x 277 -y 297 gives an A4 scan, with left-offset again
scanimage -d 'escl:https://192.168.1.73:443' --format=pdf --mode Gray --source ADF -l 0 -x 277 -y 297 does the same
BUT
scanimage -d 'escl:https://192.168.1.73:443' --format=pdf --mode Gray --source ADF -l 5 -x 277 -y 297 gives:
scanimage: rounded value of tl-x from 5 to 5
scanimage: rounded value of br-x from 215 to 215
scanimage: rounded value of br-y from 297 to 297
Segmentation fault (core dumped)
and the same is true for any value other than 0 given to "-l N".
The scan is physically started before the segmentation fault occurs.

Giving "-l N" seems to be the way to fix the left-offset, and autochanges br-x. A value of about 67 would seem right - but we can't get to that point.

I have got the Epson proprietary linux scan utility to work; and (apart from single-sided scan seeming to scan only the reverse side of the paper!) it scans correctly.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.