Talk:10338: Difference between revisions

From ElphelWiki
Jump to navigation Jump to search
Ekratzer (talk | contribs)
No edit summary
Ekratzer (talk | contribs)
No edit summary
 
(One intermediate revision by one other user not shown)
Line 1: Line 1:
[[Image:10338-purple-image.jpg]]
--[[User:Andrey.filippov|Andrey.filippov]] 16:49, 3 December 2007 (CST)[[Image:10338-purple-image.jpg]]


Several of our 10338 sensors are producing images of this nature (single purple-ish color.)
Several of our 10338 sensors are producing images of this nature (single purple-ish color.)
Line 12: Line 12:


--[[User:Ekratzer|Ekratzer]] 11:18, 3 December 2007 (CST)
--[[User:Ekratzer|Ekratzer]] 11:18, 3 December 2007 (CST)
What cable length are you using? It looks like the signal phase is wrong. You may try &byr=0..3 in ccam.cgi parameters, but that is just to verify the problem, to fix it you'll need to adjust that phase (see [[Adjusting_sensor_clock_phase]])--[[User:Andrey.filippov|Andrey.filippov]] 16:49, 3 December 2007 (CST)
We are using a 4inch cable.  I have tried adjusting the phase.. sometimes we have success, but most of the time not.  I think it is similar to the problem where the 5Mpix sensor gets confused when setting the clock frequency as described in [[Adjusting_sensor_clock_phase]].  Has there been any more discovery as to why this happens?  --[[User:Ekratzer|Ekratzer]] 11:43, 11 December 2007 (CST)

Latest revision as of 17:43, 11 December 2007

--Andrey.filippov 16:49, 3 December 2007 (CST)

Several of our 10338 sensors are producing images of this nature (single purple-ish color.)

Some of the sensors produce this type of image 100% of the time, others 50% of the time (after reboots), and still others work fine.

I have tested with 7.1.5 from sourceforge and (as of today 2007-12-03) the latest from CVS with the same results.

Is this a bad sensor? Has anyone seen this behaviour?

(note: the lens is a fisheye.)

--Ekratzer 11:18, 3 December 2007 (CST)

What cable length are you using? It looks like the signal phase is wrong. You may try &byr=0..3 in ccam.cgi parameters, but that is just to verify the problem, to fix it you'll need to adjust that phase (see Adjusting_sensor_clock_phase)--Andrey.filippov 16:49, 3 December 2007 (CST)

We are using a 4inch cable. I have tried adjusting the phase.. sometimes we have success, but most of the time not. I think it is similar to the problem where the 5Mpix sensor gets confused when setting the clock frequency as described in Adjusting_sensor_clock_phase. Has there been any more discovery as to why this happens? --Ekratzer 11:43, 11 December 2007 (CST)