The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: Chuck Simmons
Organization: You jest.
X-Mailer: Mozilla 4.61 [en] (X11; U; Linux 2.0.33 i586)
Subject: Re: FPGA+DAC: decoupling and impedance matching
Date: Tue, 17 Sep 2002 02:26:33 GMT
NNTP-Posting-Date: Mon, 16 Sep 2002 19:26:33 PDT
Dmitri Katchalov wrote:
> I'm trying to get TV output from FPGA dev. board with mixed
> results. My VHDL and digital stuff worked perfectly from the
> first attempt. My problem is the picture is very noisy.
> I understand there are issues with power supply decoupling
> and impedance matching. I know a bit of theory but I'm having
> problems applying it to the problem at hand.
> First, my setup: I'm using Xilinx SpartanIIE board from
> www.burched.com.au The board requires regulated +5V that comes
> from 5V/2A switch-mode pack of unknown quality. The DAC is
> Philips TDA8702. Maybe wrong choice, I picked it because of
> minumum number of external components (and because it was
> available for same-day delivery). DAC is mounted on a small
> proto board connected to the main FPGA board with a short
> ribbon cable. At present analog and digital rails are connected
> together and bypassed with 0.1uF ceramic and 100uF electrolytic
> caps. DAC output goes into RCA socket through another 100uF cap
> (as per datasheet). DAC runs at 15Mhz but I would like to run
> it at 30Mhz later on when it comes to DSS of color subcarrier.
> There are 3 problems with the picture:
> 1) There is a fine moire pattern. As a result black is not
> quite black and white is not quite white. I understand this
> is most likely digital noise getting into analog output through
> power supply rails.
> 2) There are strong overshots on the borders of high contrast.
> Eg. white areas have bright-white edges. I assume this has to
> do with impedance matching. I wonder why, the datasheet seems
> to indicate I can connect 75ohm output directly.
> 3) There are horisontal stripes (about 20 scanlines each)
> of varying intensity. These stripes run up or down the screen
> for some time then they seem to get phase-locked and the
> pattern stays still for some time then starts rolling again.
> I don't have any idea about the source of this interference.
> BTW damping the output with a capacitor to the groung eliminates
> the first 2 problems at the expense of horizontal resolution.
> I understand that I need to decouple analog supply from digital,
> but how? The only power I've got is +5V coming from the main
> FPGA board. Also what do I do about those overshots? Do I need
> to add output driver stage, some kind of filter or both?
> And what can be the cause of those stripes? Unfortunately
> I don't have tools to investigate what's going on.
> I'm really looking for a simple solution that works.
> I don't need broadcast quality but I want to display text
> as clearly as possible.
1) I suspect that the DAC is not perfectly glitchless. Filtering the
video might help but see 2) below.
2) This is likely impedance mismatch. Is the TV monitor really 75 ohm?
Is your cable 75 ohm? Change the cable length and see if there is some
change in this effect.
3) this would seem to be interference by some nearby device. For
example, a close by computer monitor can cause this (I see the problem
occasionally at work when we put a TV monitor too near a computer
monitor). Move the TV to a new location and see if the problem changes.
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons firstname.lastname@example.org
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup