[sane-devel] LiDE 90 half ccd

Pierre Willenbrock pierre at pirsoft.dnsalias.org
Sat Apr 26 00:01:07 UTC 2008


Guillaume Gastebois schrieb:
> Hello,
> 
> No progress on 2400dpi. DPIHW is 2400 and dpiset 600 for a 300dpi scan.
> Don't know where to look yet for that problem !

I'd try to get a log from a 2400dpi scan from windows and compare.

> about byte nibbles modifying reg_0x79 from 0x3f to 0x3e (very few 
> contrasted image) or 0x40(black image) gives me bad images !!!
> 
> What is effect of this register on clk3 ?? It's very sensible.

register 0x77 to 0x79 describe a bitmask, that determines when the clk3 
signal is high or low. For 0x0003f(lsb is register 0x79), this would give:

time===>>                          ! next pixel
bit of 0x79    ! 0x78          !0x77! 0x79
0 1 2 3 4 5 6 7!0 1 2 3 4 5 6 7!0 1!0 1 2 3 4 5 6...
1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 ....
___________                         ___________
            |_______________________|           |____

0x40 is this:

bit of 0x79    ! 0x78          !0x77! 0x79
0 1 2 3 4 5 6 7!0 1 2 3 4 5 6 7!0 1!0 1 2 3 4 5 6...
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 ....
             _                                   _
___________| |_________________________________| |____

As you can see, this gives a very short pulse, which will probably not 
be recognized by the ccd, while 0x3e just shortens the clock pulse at 
the beginning. The datasheet mentions that the scanner does not need to 
use the full 18 bits of the bitmask, but uses the first 6, 12 or 18 bits 
depending on the scanning mode. 12 seems to be correct for the mode the 
backend programs for cis scanners.

Please try if modifying bitmask for clk4(0x7a to 0x7c iirc) shows any 
effects. Also, other values you may want to try on clk3: 0xff03f, 
0xff801f, 0xffc00f. Another thing to try may be to send two clock 
pulses, for example 0x001c7. If clk3 is connected to the ccd, this may 
give interesting results(like, reducing the resolution by half), but for 
the 4-bit latch/D-FlipFlop, only the last clock pulse before the gl841 
samples the data bits is relevant.

> regards,
> Guillaume

Regards,
   Pierre




More information about the sane-devel mailing list