Searching \ for '[EE] Extracting the video signal from a 38.9MHz TV' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/displays.htm?key=video
Search entire site for: 'Extracting the video signal from a 38.9MHz TV'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Extracting the video signal from a 38.9MHz TV'
2009\07\30@054122 by Alan B. Pearce

face picon face
Arghh, Philip put a colon on the [EE] ...

Subject: Re: Extracting the video signal from a 38.9MHz TV IF


{Quote hidden}

> --

2009\07\30@102945 by Philip Pemberton

face
flavicon
face
Alan B. Pearce wrote:
> The 'usual' method was to use a standard diode detector after the video IF.
> This then provides the baseband video plus 6MHz audio carrier. The standard
> trick was to use a parallel tuned circuit as a stop filter in series with
> the video siganl, and have a secondary winding on this to feed the audio IF.

Sounds reasonable.

> I wouldn't use a SAW filter. You are not wanting the proper video, so I
> beleive that a bandpass tuned circuit ith a few stages (probably a single IC
> worth) of gain should get you enough.

I was wondering if I could do this with an L/C filter, but figured that
SAW filters were being used for a reason (they seem to have a much
steeper falloff than most L/C filters).

> Essentially you are looking for the
> sync pulses and the data stream in the horizontal retrace periods. I cannot
> remember the data rate, but it is fairly slow, and has lots of ECC info as
> well.

True. I don't care if the video looks like crap, as long as I can find
the sync pulses, 66%-white TTX ones and black-level TTX zeroes.

At 25fps with two fields, each line is 64us in length, consisting of (in
time order, left-to-right as you'd see on a 'scope):
  1.65us  front-porch
  4.7us   horizontal sync
  5.7us   back-porch
  51.95us visible area

I'm going to round the visible area up to 52us to make the math easier.

One teletext packet contains 45 bytes of data: a 2-byte clock run-in, a
framing byte, then a 2-byte type and address and 40 bytes of text.

45 bytes * 8 bits = 360 bits

52us per line / 360 bits = 0.1444(recurring) us/bit

1/0.1444... = 6.92308MHz bit rate

I've managed to find a copy of the BBC/IBA/BREMA spec (on
<http://www.doc.ic.ac.uk/~ih/teaching/teletext/>) which is a bit more
concise and to the point than the current ETSI standard. In the BBC
spec, the bit rate is specified as 6.9375MHz +/- 25ppm (or 444 times the
TV line rate).

I also spotted the closed-caption decoder on Eric Smith's website
(<http://www.brouhaha.com/~eric/pic/caption/>), which looks like it'll
be a nice starting point for the white level extraction/data slicing
hardware.

Recovering the clock back might be a little more "interesting"... My
first idea was to use a PLL but I suspect that it either won't lock in
fast enough, or will lock in and then lose lock once the data starts
streaming in. I'm probably missing something obvious, but there doesn't
seem to be a way to make a 4046 type chip hold the current VCO state,
i.e. "ignore the reference, stay at the current frequency and hold".

The other option would be to make use of some of the varicaps in my
junkbox and make a "pullable" crystal oscillator, and only adjust the
frequency when the run-in is running.

> <VBG> See if your local library has issues of Wireless World. They had a
> good technical article or two when the service first started. I have a
> feeling they also had a project to build a decoder from discrete chips.

My local library has pulped all the electronics books, and the Central
Library no longer allows access to the magazine collection... :(

My university library on the other hand has a near complete set of E&WW
from 1970 or so onwards. It's just not indexed in any sane fashion (i.e.
no copies of the tables of contents, or anything useful like that).

Cheers,
--
Phil.
spam_OUTpiclistTakeThisOuTspamphilpem.me.uk
http://www.philpem.me.uk/

2009\07\30@110517 by Alan B. Pearce

face picon face

> Recovering the clock back might be a little more "interesting"... My
> first idea was to use a PLL but I suspect that it either won't lock in
> fast enough, or will lock in and then lose lock once the data starts
> streaming in.

I haven't looked at the links you provided, but I would be tempted to lock
the VCO to the line rate, and then when it needs to lock to the data it
"just" needs trimming slightly ... as you point out, the data rate is an
integer multiple of the line rate, so a small phase error should lock in
quickly.

2009\07\30@135554 by Philip Pemberton

face
flavicon
face
Alan B. Pearce wrote:
> I haven't looked at the links you provided, but I would be tempted to lock
> the VCO to the line rate, and then when it needs to lock to the data it
> "just" needs trimming slightly ... as you point out, the data rate is an
> integer multiple of the line rate, so a small phase error should lock in
> quickly.

The thing that's putting me off doing that is in section 1.1.3 of the
BBC/IBA/BREMA teletext spec:

  The data timing reference point is the peak of the penultimate '1' of
the Clock Run-In sequence (see figure 3). This point has been selected
to reduce the effect of any transient distortions at the start of the
Data-Line.
  The line time reference is the half-amplitude point of the leading
edge of the line synchronising pulse.
  The data timing reference in the signal as transmitted shall be 12.0
(+0.4/-1.0)us after the line time reference.
  The data timing may vary from Data-Line to Data-Line.

So AIUI, the start of the clock run-in on the scanline can be between
11.6us and 13us from the leading edge of the line sync pulse. 12us is
the length of the horizontal blanking period (frontporch, sync, and
backporch). If that is the case, then locking a VCO to the line rate
wouldn't help if, say:

  * The broadcaster is providing a signal that's "just a bit" fast or
slow -- still within the (+0.4/-1.0)us spec, but not perfect.
  * Their teletext inserter is waiting for the sync, then just "brute
forcing" it -- waiting 12us and spitting out a line of data.

Say the broadcaster's signal is 0.5us fast (that is, the Hblank is
12.5us). Their inserter waits for the sync, waits 12us exactly, then
inserts the clock run-in, data and so on. Problem is, if the receiver is
locking off the line rate, then it's not going to see the clock run and
framing sequence, or the clock is going to skew as it goes along the line.

AIUI, the receiver is *supposed* to wait for the clock run-in (101010...
repeated to a length of 16 bits), and lock its own clock to it. Problem
is, as I said, a PLL would need a loop time of at least 1x the line
rate, otherwise by the end of the line (depending on the data) it'll
have skewed off. Effectively what I want is an oscillator that can be
pulled in sync with the clock run-in, then left to stay where it is
until the end of the line.

In other words, it's got ((1/6.9375MHz)*16) = 2.31us to lock, then needs
to hold its phase and frequency at where it's been set for at least 60us...

Hm.

--
Phil.
.....piclistKILLspamspam@spam@philpem.me.uk
http://www.philpem.me.uk/

2009\07\30@160919 by Richard Prosser

picon face
>
> My local library has pulped all the electronics books, and the Central
> Library no longer allows access to the magazine collection... :(
>


One would wonder why they bother to retain it then.

RP

2009\07\30@192107 by Philip Pemberton

face
flavicon
face
Richard Prosser wrote:
>> My local library has pulped all the electronics books, and the Central
>> Library no longer allows access to the magazine collection... :(
>>
> One would wonder why they bother to retain it then.

The library or the magazine collection?

There are a couple of good things about the library system itself.
Access to BSI standards free-of-charge is good when you're meddling
around creating CAD libraries -- if memory serves, I grabbed a couple of
schematic symbol standards from their site to use as a baseline for my
Kicad and EAGLE libraries. Which still aren't done. Woo. (Though to be
fair, the EAGLE symbols are pretty good, it's the PCB footprints that
are b0rken)
If I'm being honest though... my university's ScienceDirect and ACM
Digital Library subscriptions have been far more useful...

As for the collection itself... I seem to recall you can still get
copies of individual articles for a fiver a go (there's a page limit but
I can't remember what it is), but there's no refund if for some reason
they either can't find it or can't copy it (which seems to be about 90%
of the time). Takes about three weeks to do it, too. Also, they *do*
have every Wireless World from 1983 onwards, but again none of them are
catalogued to TOC level, so unless you've got an actual reference to an
article (including title, author, issue, year and page number), you're
up the creek.

Now, their catalogue software.. that's a different kettle of fish. It's
an ancient version of TalisPrism that appears to date back to the early
1990s, and is WELL overdue for replacement. Here's a lovely example...

Search for "Author: Horowitz, Paul"... zero results
Try again... "Author: Horowitz Paul"... zero results
And again... "Author: Horowitz P"... zero results
One last go: "Author: HOROWITZ, P"... one result,
  "Art of Electronics, The; Horowitz P., and Hill, W."

Well, that sounds like it... let's expand that and find out some more
about it...

"SHELFMARK: 621.38 UNKNOWN SECTION      NO COPIES AVAILABLE"

Or if you're playing with the amber-screen terminals at the Central Library:

"FURTHER INFORMATION NOT AVAILABLE AT THIS TIME"

Or the best one:

SEGMENTATION FAULT.
TASK TERMINATED. >>>

*

Yep. Dies and drops you to a command prompt. At this point you have to
get the library staff to restart the system. This involves a phone call
to the council offices, and usually takes about 15 minutes. If there
aren't any other (working or not-in-use) terminals left... um... well, I
hope you've got something to read while you wait.


How about something else. Let's see if there's anything else interesting
in that shelfmark (sometimes it's fun to browse).

Search: "Shelfmark: 621.38"  gets 41 results.
  (this is "Electronics")
Search: "Shelfmark: 621.382" gets 93 results. None of which were in the
first search.
  (this is either ham radio or data transmission technologies)

So you basically have to be more specific, to be less specific... It
doesn't understand wildcards, so you can't just do "Shelfmark: 621.38*".
That gets you a "Security Alert: Invalid SQL detected. This has been
logged." warning on the website, or zero results on the terminals.


The infernal contraption does word-stemming on author queries, so if you
enter "Graf, R" as an author search, it'll do an OR search and give you
such delights as "Grafton, Sue" and "Richards, Steven". The fun part?
You can't actually make it do an AND search without serious effort.


For bonus points, there are Bob Pease books (actually different copies
of the same book) listed under "Pease, Bob", "Pease, Robert" and "Pease,
R". Consistency? What's that, then?


Lastly, the real clincher... The search times out after a couple of
minutes (seems to be about two). So you get half-way through the search
results and....

  SEARCH HAS TIMED OUT. Please restart your search.

It's usually easier to get a shelfmark chart from the librarian,
scribble down the shelf codes and go look on the shelf...

--
Phil.
piclistspamKILLspamphilpem.me.uk
http://www.philpem.me.uk/

2009\07\31@040809 by Alan B. Pearce

face picon face
>For bonus points, there are Bob Pease books (actually different
>copies of the same book) listed under "Pease, Bob", "Pease,
>Robert" and "Pease, R". Consistency? What's that, then?

Oh, so it is like trying to search the Farnell website for something then
... that seems to be an amazing set of descriptions that don't quite meet
expectations for a component, and the same component under about 4 different
variations of a description.

2009\07\31@050437 by Ruben Jönsson

flavicon
face
> >For bonus points, there are Bob Pease books (actually different
> >copies of the same book) listed under "Pease, Bob", "Pease,
> >Robert" and "Pease, R". Consistency? What's that, then?
>
> Oh, so it is like trying to search the Farnell website for something then
> ... that seems to be an amazing set of descriptions that don't quite meet
> expectations for a component, and the same component under about 4 different
> variations of a description.
>

For just that reason I have kept an old catalog where I can see what they have
called things. It's a real pain searching on the web compared to searching in
the catalog. You never quite know if you have found all options or there might
be more components of the same type but under other names.

I can't say that I have found any major electronics distributor that have a
better on line catalog than the real thing.

/Ruben
==============================
Ruben Jönsson
AB Liros Electronic
Box 9124, 200 39 Malmö, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
.....rubenKILLspamspam.....pp.sbbs.se
==============================

More... (looser matching)
- Last day of these posts
- In 2009 , 2010 only
- Today
- New search...