Post by thread:Sync restoration (how) ?

[ Ok, this is *not* a pure PIC question, but I really need some help...So don't shoot please ...] Dear All, I have a composite video signal (PAL) with *no* sync pulses at all. I would like to build a simple circuit to put the sync back (a sync restorer). Ideally, vertical & horizonal phase (and frequency -- slight changes) would be adjustable. - Can you suggest me any special IC's for that purpose ? (I know about National's LM1882 but it is too complicated for my purpose). I was thinking to use a PIC for that purpose , but this must be an overkill... - Any other suggestions or ideas ? Thank you very much in advance, Argiris A. Kranidiotis -- ____________________________ __________________________________ / /\ / /\ / Argiris A. Kranidiotis _/ /\ / E-mail (InterNet): _/ /\ / University Of Athens / \/ / / \/ / Informatics Department /\ / spam_OUTakraTakeThisOuTdi.uoa.gr /\ /___________________________/ / /_________________________________/ / \___________________________\/ \_________________________________\/ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \

In message <.....m0u6eod-0000nfCKILLspam@spam@dc.cis.okstate.edu>, "Argiris A. Kranidiotis" write s: > I have a composite video signal (PAL) with *no* sync pulses at all. There needs to be some sort of timing signal available which is synced to the video signal. The actual sync generation is simply the addition of a pulse of the correct width of maximum black level sent at the start of each new video frame which in the case of European PAL is every 625 lines. One could divide the color burst frequency down enough to provide the right suitable frequency, but there needs to be some sort of time reference in order to know where the exact start of the new frame is. Some subscription video systems send no sync, but transmit a pilot carrier over the audio so that the decoder can rebuild a sync pulse. The short answer is that there is no magic way to restore sync where there is none. Martin McCormick 405 744-7572 Stillwater, OK OSU Center for Computing and Information services Data Communications Group

On Tue, 9 Apr 1996 17:45:45 +0300, you wrote: > I have a composite video signal (PAL) with *no* sync pulses at all. >I would like to build a simple circuit to put the sync back (a sync >restorer). Ideally, vertical & horizonal phase (and frequency -- slight >changes) would be adjustable. > >- Can you suggest me any special IC's for that purpose ? (I know about >National's LM1882 but it is too complicated for my purpose). I was >thinking to use a PIC for that purpose , but this must be an overkill... There is a sync generator chip available that provides all the sync signals you'll probably need. It is the CD22402E LSI Sync Generator. Works with 525 or 625 line systems. It cost $7.95 from DC Electronics. Don't know if 800 numbers work from Athens but their number is 1-800 467-7736 or 1-800 423-0070. Good luck. Terry

> I have a composite video signal (PAL) with *no* sync pulses at all. >I would like to build a simple circuit to put the sync back (a sync >restorer). Ideally, vertical & horizonal phase (and frequency -- slight >changes) would be adjustable. Although there're a variety of timebase chips that would do this cheaply, you're still going to need a reference to lock on to... eg you can't just re-insert syncs from a free running timebase - (you could try locking to colour subcarrier and dividing down - but it's not too easy in PAL.. I had been thinking that a PIC would be good for doing an intelligent sync capture (look for absence of picture etc) and locking a pll to the output of the Pic - haven't done the sums tho'.... Jim

>> I have a composite video signal (PAL) with *no* sync pulses at all. >>I would like to build a simple circuit to put the sync back (a sync >>restorer). Ideally, vertical & horizonal phase (and frequency -- slight >>changes) would be adjustable. > >Although there're a variety of timebase chips that would do this cheaply, you're >still going to need a reference to lock on to... eg you can't just re-insert >syncs from a free running timebase - (you could try locking to colour subcarrier >and dividing down - but it's not too easy in PAL.. If the colour burst (or even a remnant of it) is present, it should be possible to detect this & then at least generate sync pulses for the following lines, meaning you'd miss one line sync. Once this has been locked on to, you'd then need to look for a missing pulse at the vertical interval and derive the vertical serration sequence. around line 8 or 9 the colour burst starts again. This again could be detected and your insertion sequence initiated. (Perhaps this is where the missing line could be tolerated, as it's already in the vertical area.) Kenneth The lion and the calf shall lie down together, but the calf won't get much sleep....