NHL-10
Baseband Troubleshooting
CCS Technical Documentation
ISSUE 2 01/2004
COMPANY CONFIDENTIAL
41
Copyright © 2003 Nokia. All Rights Reserved.
If no visible defects (items 2-4) can be found, a couple of test images should be taken and
checked. Generally, a well illuminated typical indoor office scene, such as the one in Figure 33:
A good quality picture taken indoors,
can be used as a target.
The main considerations are:
The amount of light, 300 – 600 lux (bright office lighting), is sufficient.
The scene should contain e.g. small objects for checking sharpness and distance to them
should be in order of 1 – 2 meters
If possible, compare the image to another image of the same scene, taken by different device.
The taken images should be analysed on PC screen at 100% scaling simultaneously with ref-
erence image. Pay attention to the computer display settings; at least 65000 colours (16bit)
have to be used. 256(8-bit) colour setting is not sufficient, and true colour (24bit, 16 million col-
ours) or 32 bit (full colour) setting is recommended.
If there appears to be a clearly noticeable difference between the reference image and the test
images, the module might have a misfocused lens. In this case, the module should be changed.
Always re-check the resolution after changing the camera. If a different module produces the
same result, the fault is probably in the camera window. Check that you can see through the
window when replacing the module.
Bit errors
Bit errors are defects in image caused by a data transmission error between camera and phone
baseband. This type of error is expected to be rare since usually missing bits will cause a hard-
ware failure message. Bit errors can be typically seen in images taken of any object, and they
should be most visible in full VGA resolution images.
Viewfinder images may not contain the errors at all due to lower bit rate used in this mode.
A good practice is to use uniform white test target.
The errors will be clearly visible as colourful sharp dots or lines in camera VGA images.
