Searching \ for '[EE]: Datarecording on a VCR-audiotrack' 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/io/audio.htm?key=audio
Search entire site for: 'Datarecording on a VCR-audiotrack'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Datarecording on a VCR-audiotrack'
2000\06\20@065601 by Jan Lund

flavicon
face
Hello List.

I want to record some serial data on a VCR audiotrack. The idea is to record
video from a surveillance camera and add information on the audiotrack.
Then, at playback, route the data from the audiotrack through an OSD chip
(STV5730) and display the data on screen.

I want to keep the original video signal, without any overlay, thats why i
want ot record the data on the audio track.

I guess a data rate around 1200bps would be sufficient.

Does anyone have any ideas of how to make this ?

(i thought i could use an old modem chip, but i cant find any)

Rgds

Jan

2000\06\20@083119 by M. Adam Davis

flavicon
face
Old modem chips are going the way of the dinosaur, unfortunately.

Fortunately, Scenix has a 1200 bps software modem for their devices, so you may
just mosey on over to http://www.scenix.com/virtual/vp/vplibrary.html .  It
appears as though it is half duplex (ie, either receiving or sending, but not
both at the same time) which is ideal for your situation.

Furthermore, I imagine scenix will be around for a lot longer than nearly any
modem chip you might find.

-Adam

Jan Lund wrote:
{Quote hidden}

2000\06\20@084418 by mike

flavicon
face
On Tue, 20 Jun 2000 12:55:05 +0200, you wrote:

>Hello List.
>
>I want to record some serial data on a VCR audiotrack. The idea is to record
>video from a surveillance camera and add information on the audiotrack.
>Then, at playback, route the data from the audiotrack through an OSD chip
>(STV5730) and display the data on screen.
>
>I want to keep the original video signal, without any overlay, thats why i
>want ot record the data on the audio track.
>
>I guess a data rate around 1200bps would be sufficient.
>
>Does anyone have any ideas of how to make this ?
>
>(i thought i could use an old modem chip, but i cant find any)
>
>Rgds
>
>Jan
Depending on the baudrate you need, you may be able to use a really
crude 2-tone system, with encoding and decoding mostly in software.

2000\06\20@090123 by Phil

flavicon
face
Hello List.

I want to record some serial data on a VCR audiotrack. The idea is to record
video from a surveillance camera and add information on the audiotrack.
Then, at playback, route the data from the audiotrack through an OSD chip
(STV5730) and display the data on screen.

I want to keep the original video signal, without any overlay, thats why i
want ot record the data on the audio track.

I guess a data rate around 1200bps would be sufficient.

Does anyone have any ideas of how to make this ?

(i thought i could use an old modem chip, but i cant find any)

Rgds

Jan

I hope you have a better response to your question than I did (Re 16F84 retrieving code, what a weak response!)

You can encode 1200 BPS data onto an audio track using an AM7910 chip, this will also decode the data at the other end.. I use these chips for remote data collection via radio links and it works very well..

2000\06\20@094024 by James Paul
flavicon
face
At the risk of sounding like a smart aleck, try a new modem chip.

                                       Regards,

                                         Jim



On Tue, 20 June 2000, Jan Lund wrote:

{Quote hidden}

spam_OUTjimTakeThisOuTspamjpes.com

2000\06\20@094437 by aipi Wijnbergen

picon face
Jan,

OK if you are already set on this direction, but, you might also consider
another alternative.
What you want to do sound similar to TeleText broadcasting, which uses the
first few lines of the video signal in order to encode digital data, some
of the data into each of the frames. In your case it would be synchronized
with the frames that come from the camera.

Chaipi

At 12:55 20/06/00 +0200, Jan Lund wrote:
{Quote hidden}

2000\06\20@095518 by Alan B. Pearce

face picon face
>I guess a data rate around 1200bps would be sufficient.

>Does anyone have any ideas of how to make this ?

>(i thought i could use an old modem chip, but i cant find any)

Look at using a phase encoded or NRZ system? I am thinking of something like RCA
described in an application note when they first brought out the CD4046 phase
locked loop. It gave a clock recovery off the encoded data stream as well. Your
audio bandwidth should be wide enough to have a reasonable data rate using this
approach, without worrying about edge shifting due to rise and fall (phase shift
problems) due to limited bandwidth. The encoding end should be doable totally in
software (the receiving end as well if you really wanted).

2000\06\20@095734 by Severson, Rob

flavicon
face
> Old modem chips are going the way of the dinosaur, unfortunately.

TDK has "old modem chips". If you need a quick connection and small amounts
of data, a 1200 bps modem is faster than a 56k. Less handshaking. Thus TDK
is keeping these modem chips around for a long time. (They bought the modems
from, IIRC, Silicon Systems)

The TDK modems should work fine for your application

{Quote hidden}

2000\06\20@140842 by Robert Rolf

picon face
Jan Lund wrote:
> I want to record some serial data on a VCR audiotrack. The idea is to record
> video from a surveillance camera and add information on the audiotrack.
> Then, at playback, route the data from the audiotrack through an OSD chip
> (STV5730) and display the data on screen.
>
> I want to keep the original video signal, without any overlay, thats why i
> want ot record the data on the audio track.
>
> I guess a data rate around 1200bps would be sufficient.
>
> Does anyone have any ideas of how to make this ?

If you want best results, make sure that your VCR audio tracks are
Hi-Fi (20-20Khz vs linear 100-7Khz). With a Hi-Fi audio track you
can store the serial signal directly (with a clamped comparator for
recovery since the path is AC coupled) to 19.2kbaud (it looks like
a sine way at that rate).

With Hi-Fi you have the 2nd track available for location audio.

> (i thought i could use an old modem chip, but i cant find any)

Why bother? Its simple enough to just use a 4046 to encode/decode your
own
FSK since you don't need full duplex. Just use a high carrier frequency
to get the bit bandwidth.

You might also want to look at using the 'user bits' from a VITC
inserter or LTC generator. IMS you get 4 bytes every field.
If your 'information' is time, then the TCG does all the work for you.
The 'user bits' let you ID the camera, etc.

VITC also has the advantage of allowing you to recover the data when
the VCR is in pause mode.

Horita makes a nice series of inexpensive video VITC/LTC TCG's &
readers.
http://www.horita.com/


www.kodak.com/US/en/motion/support/glossary/post.shtml
for a glossary for the above acronyms.

Robert

2000\06\20@143345 by William Chops Westfield

face picon face
Does anybody have a quick rundown on which (if any) modem technology is
suitable for recording on typical audio recording media?

BillW

2000\06\20@144344 by David VanHorn

flavicon
face
At 11:32 AM 6/20/00 -0700, you wrote:
>Does anybody have a quick rundown on which (if any) modem technology is
>suitable for recording on typical audio recording media?


FSK should work pretty well. Bell 202 tones. MX-Com 614 modem does this.
VCRs have a servo loop that keeps the tape speed pretty constant, so you
don't get a lot of warble in the tones.
We use bell 202 for ham radio, which can be a pretty nasty audio path, and
it survives pretty well.


--

http://www.SpamWhack.com  A pre-emptive strike against spam

2000\06\20@150006 by Ralph & Helene

picon face
While, we're on the subject, I have a similar application that would record
a track of audio data on a mag stripe about 3-4 inches long with a
synchronizing digital clock channel so that when the audio is read by a
swiping reader, the audio would come out at the right speed. I suppose the
audio would have to be A/D-ed, be read into memory, and then the reverse
process at the right speed.

Anybody been down this road or have any ideas??

Thx,

Ralph Krongold
Culver City, CA
{Original Message removed}

2000\06\20@151631 by David VanHorn

flavicon
face
At 11:53 AM 6/20/00 -0700, Ralph & Helene wrote:
>While, we're on the subject, I have a similar application that would record
>a track of audio data on a mag stripe about 3-4 inches long with a
>synchronizing digital clock channel so that when the audio is read by a
>swiping reader, the audio would come out at the right speed. I suppose the
>audio would have to be A/D-ed, be read into memory, and then the reverse
>process at the right speed.
>
>Anybody been down this road or have any ideas??

Magstripes, at least the common ones, don't have much density.
A couple hundred bits per inch.

--

http://www.SpamWhack.com  A pre-emptive strike against spam

2000\06\20@153845 by M. Adam Davis

flavicon
face
Ah, but at a slow speed you can record audio on the stipe.  At 1.2cm/s, on a 3"
card you can get 6 seconds of audio.  The one thing you generally don't have
control over is how fast the card goes through the reader head.

If you add an additional stipe that has a reference audio signal, say 40kHz,
then you could use it to determine the speed of the card.  As suggested below
you sample the audio stripe at every peak of the reference signal, and play it
out at 40k samples/s.

Surely not high quality audio, and perhaps 40kHz is too high, but the principle
is the same, and you can probably get a manufacturer to make a card with a
better stripe.

-Adam

David VanHorn wrote:
{Quote hidden}

2000\06\20@155134 by David VanHorn

flavicon
face
At 03:37 PM 6/20/00 -0400, M. Adam Davis wrote:
>Ah, but at a slow speed you can record audio on the stipe.  At 1.2cm/s, on
>a 3"
>card you can get 6 seconds of audio.  The one thing you generally don't have
>control over is how fast the card goes through the reader head.
>
>If you add an additional stipe that has a reference audio signal, say 40kHz,
>then you could use it to determine the speed of the card.  As suggested below
>you sample the audio stripe at every peak of the reference signal, and play it
>out at 40k samples/s.

This assumes that the head and tape support that. The gap length, and tape
domain size both have to support it.


>Surely not high quality audio, and perhaps 40kHz is too high, but the
>principle
>is the same, and you can probably get a manufacturer to make a card with a
>better stripe.

This is true, to some degree. Check out MagTek, they have the best
offerings in my experience.
--

http://www.SpamWhack.com  A pre-emptive strike against spam

2000\06\20@161602 by Robert Rolf

picon face
"M. Adam Davis" wrote:
> Ah, but at a slow speed you can record audio on the stipe.  At 1.2cm/s, on a 3"
> card you can get 6 seconds of audio.  The one thing you generally don't
Except that the usual magnetic material for cards is designed to behave
in
a binary way. The B-H curve is squarish, and LOUSY for audio.
I know. I've tried. Get a hold of some old 1/2" magtape and try
to record analog audio on it. You need very high bias to record
anything, and it sounds like crap in playback.

have
> control over is how fast the card goes through the reader head.

Hence his point about using an A/D. Grab varispeed audio using the
data clocks to measure speed, then play back corrected audio,
interpolating
the sample clock rate variance.


> If you add an additional stipe that has a reference audio signal, say 40kHz,

The card standard allows for 3 tracks, with different bit densities
specified for each.

> then you could use it to determine the speed of the card.  As suggested below
> you sample the audio stripe at every peak of the reference signal, and play it
> out at 40k samples/s.
>
> Surely not high quality audio, and perhaps 40kHz is too high, but the principle
> is the same, and you can probably get a manufacturer to make a card with a
> better stripe.

If you have custom cards made, you will probably want high coercivity
for the data tracks, and the lower/linear for the audio.

Wasn't there a 1980's slide projector that played a 2 second audio track
of the edge of the slide? Used 2.25" cubes to store the slides.

{Quote hidden}

But in linear mode that's irrelevant. You can also push the density
if you can live with a higher error rate or shorter life. The density
is low so that a well worn card is still readable.

2000\06\20@162628 by David VanHorn

flavicon
face
>
>Except that the usual magnetic material for cards is designed to behave
>in a binary way. The B-H curve is squarish, and LOUSY for audio.
>I know. I've tried. Get a hold of some old 1/2" magtape and try
>to record analog audio on it. You need very high bias to record
>anything, and it sounds like crap in playback.

I forgot about that part. We only recorded data :)
True though, non-linear as hell, by design.


>The card standard allows for 3 tracks, with different bit densities
>specified for each.

Given that this is non-standard, you could do a custom head that records
say six tracks at track 3 densities.
Being clever about the encoding, you could avoid the need for a clock
track, so maybe you could reach 1800-2000 bpi.
Still, less than 10kbits for the whole thing.  How much audio can you pack
into 10kbits?
Custom heads are pricey too. (Ask me how I know)


>If you have custom cards made, you will probably want high coercivity
>for the data tracks, and the lower/linear for the audio.

Doing this with audio would be really problematic. You get at least a 10-1
variation in wipe speeds from person to person. The good news is that most
people can't exceed 4-1 in speed variation through the wipe, and it's
pretty linear.
Though it seems like you can, when you scope it you'll see. Arms have lots
of mass.


--

http://www.SpamWhack.com  A pre-emptive strike against spam

2000\06\20@215511 by Bob Ammerman

picon face
Instead of searching for 'modem' chips, try FSK.

Bob Ammerman
RAm Systems
(high performance, high function, low-level software)

{Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...