Are you AC coupling the LM1881?You shouldn't connect it directly to the
camera, you should have a 0.01 uF cap inbetween.

Sean


On Fri, 26 Jan 2001, Chris Carr wrote:

> As you have an oscilloscope first check the signal from the camera is to
> specification i.e. 1 volt peak to peak comprising 0.3 volt sync pulses and 0
> to 0.7 volt video.
>
> Have you terminated the video camera ?
>
> Regards
> Chris
>
> ----- Original Message -----
> From: "Stephen B Webb" <sbwebb0@SAC.UKY.EDU>
> To: <PICLIST@MITVMA.MIT.EDU>
> Sent: Friday, January 26, 2001 6:35 AM
> Subject: [EE]: Having trouble with LM1881 / NTSC sync separator
>
>
> > Breif background:
> >
> > I have a dual CCD camera module that outputs 2 NTSC composite video
> > signls.  The two signals are in sync.
> >
> > In order that I can view this stereo video feed on a head mounted
> > dual-display device, I need to take the two signals and FRAME interlace
> > them into one NTSC composite signal.
> >
> >
> > The plan goes something like this:
> >
> > By using a national semi. LM1881, pull the ODD/EVEN field sync signal off
> > of one of the composite signals.  Use this information to switch between
> > the two video signals during the vblank period, thus getting a FRAME
> > interlaced composite output.
> >
> >
> > The problem goes something like this:
> >
> > To start with I have only a basic EE background, so I don't necessarily
> > understand what happens to video signals as they go through wires, traces,
> > capacitors and resistors, etc.  I'm  a software type, so if you find
> > yourself thinking "Man, this guy doesn't know what he's doing", you are
> > probably right.
> >
> > So, I start by trying to get the LM1881 to give me a nice steady odd/even
> > field signal.  I connect the various components to and let it rip.
> >
> > First problem:  Nothing happens.  So, I add the optional lowpass filter on
> > the signal input, and still nothign happens.  I pick up the camera board,
> > and inadvertantly touch the RCA connector; my finger bridges the two
> > leads.  The circuit starts working.  So, I add a 1 meg resistor between
> > the signal and ground leads on the incoming composite feed.  I get a
> > somewhat consistent odd/even signal.
> >
> > Second problem:  If I expose the camera to more or less light, it tends to
> > cause the odd/even output to stop working (for instance if I cover the
> > lens of the camera, the odd/even line stays high or low until I uncover
> > it)  I'm sure this is related to #1 and my 1 meg resistor hack.
> >
> > Third problem:  When I have the two capacitors and resistor on the input
> > signal, it drastically changes when viewed on my [very crappy] scope.
> > (as compared to measuring the signal not connected to anything)
> > Since I want to use that signal as an output, I am concerned that it
> > doesn't look the same as it used to (I'm not even sure if it contains much
> > video information at all any more).  Do I need to run the initial signal
> > through some sort of amplifier, or something first?
> >
> >
> >
> > Note:  I first put this on a breadboard, and when I ran in to trouble I
> > moved it to a Rad Shack style prototype board.  Haven't been able to make
> > it do much on the proto board yet. :(
> >
> >
> > Any insight would be appreciated.
> >
> > -Steve
> >
> > --
> > http://www.piclist.com hint: To leave the PICList
> > mailto:piclist-unsubscribe-request@mitvma.mit.edu
> >
> >
> >
>
> --
> http://www.piclist.com hint: To leave the PICList
> mailto:piclist-unsubscribe-request@mitvma.mit.edu
>
>
>

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu