Sync Width and Position Modulation.

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

Sync Width and Position Modulation.

Postby Steve Anderson » Mon Feb 03, 2014 8:43 pm

This is a spin-off of the "Hellcat MkI" 625-32 converter. The thread is called "Video to NBTV converter"

In the past it has been advocated that 'porches' be included in the NBTV video signal to eliminate modulation of the sync pulse width and position. It's quite a subtle thing, but it's there all the same. Graham Lewis has often introduced at least a front porch into some of his designs saying that it does help with reliable sync extraction.

In developing the "Hellcat" the timing is such I will end up with with some spare time during the line so I decided to add porches, both front and back.

In the traces below the cyan trace is the TTL output composite syncs and the yellow trace is the filtered NBTV PWM output. The left trace is the reference for the other two, a totally black frame, it's the same with or without porches - as you'd expect. The sync width is 129us.

The centre frame is an all-white frame, note there is no change in width or position of the sync pulse when porches are present.

The right trace is with no porches as is common. Note the delay in position relative to the TTL syncs, which by inference also means a shift relative to the video. The pulse is also narrower at 116us at the same sync-slicing point.

As I said, it is subtle, and most of the time will not be of any great concern. But it will happen wherever there are filters, in and out of PC sound cards for example.

I'm NOT suggesting that this be a standard, but just raised the point which may be of interest to some.

Steve A.

Sorry if you have to download the image, it's wider than my screen allows here...
Attachments
Sync Width & Position Modulation 1.gif
(13.56 KiB) Downloaded 50 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby AncientBrit » Mon Feb 03, 2014 11:11 pm

Hi Steve,

As you say the effect of adding porches is subtle but I think worthwhile.

The NBTV spec for sync duration is pretty loose anyway, I seem to remember it was expressed as a percentage of total line duration.

The porches can of course 'steal' time from either syncs or from active line time.

Kind regards,

Graham
AncientBrit
Green padded cells are quite homely.
 
Posts: 858
Joined: Mon Mar 26, 2007 10:15 pm
Location: Billericay, UK

Postby Steve Anderson » Tue Feb 04, 2014 3:26 pm

The spec. for sync width as per www.nbtv.org is 0.1 to 0.25ms, 100 to 250us. I generally aim for 125us, 5% of line-time.

The output of the 'Hellcat' will have porches simply because it's dictated by the timing during the line. Together they take up a mere 2.7% of the line-time so I doubt anyone would have noticed them.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby AncientBrit » Wed Feb 05, 2014 6:16 pm

Thanks Steve,

I tend to use 150-160uS but I'm not that conscientious.
With a 'mechanical' camera I've achieved porches by masking the scanned area.
The porches are useful if you can add them.

Regards,

Graham
AncientBrit
Green padded cells are quite homely.
 
Posts: 858
Joined: Mon Mar 26, 2007 10:15 pm
Location: Billericay, UK

Postby Steve Anderson » Wed Feb 05, 2014 8:50 pm

Well, it is indeed fortunate that porches were put into the original monochrome standards, 405/525/625/819. The original reason was simply the sluggish performance of early timebases, both vertically on horizontally. Without the back porch where would we have put the colour burst? Without the vertical blanking where would we have put teletext, ITS, VITC and a whole load of other stuff.

But the roughly 1.5us front porch at the end of active video was put there for precisely this reason. No data or anything was placed there otherwise its whole reason for existing would have been negated. In a bandwidth of only a few MHz you couldn't have put much there anyway.

Steve A.

For those interested I attach a few pdf's of the four standards mentioned above, these show the porches in detail and the vertical blanking interval (VBI) with great clarity. Perhaps a bit OT for the NBTV section, but I think it still applies here. i.e. There are valid reasons this was done...

I would post a link to the site I got them from, but for the time being it's down...my thanks to Alan Pemberton all the same...
Attachments
FSTV Horiz Blanking etc.pdf
(24.82 KiB) Downloaded 311 times
405 VBI.pdf
(13.81 KiB) Downloaded 274 times
525 VBI.pdf
(43.51 KiB) Downloaded 368 times
625 VBI.pdf
(40.19 KiB) Downloaded 269 times
819 VBI.pdf
(51.66 KiB) Downloaded 269 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Klaas Robers » Thu Feb 06, 2014 9:01 am

Steve,

The new CCNC test-CD has a sync of 119 usec and there is a front porch of 40 usec. This is mainly done to "clean" the video low pass filter before the sync is coming. We assumed that in almost all cases the (negative going) front edge of the sync is used and not the (positive going) back edge and also not the centre of gravity (mean value of the position in time) of the sync. So in total 159 usec are taken away from the total available line time.

I assume that the black and white CDs have the same width of the sync and the same porch. Why should I have done it differently?
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1453
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Steve Anderson » Thu Feb 06, 2014 4:04 pm

Klaas, agreed. As per the pdf's above I take the timing reference from the leading edge (falling, negative-going) of the sync pulse. The pulse itself is in line n as is the video information that follows up until the next sync pulse where it is line n+1.

I quite agree, the width of the sync pulse isn't that important within limits but that leading/falling edge is.

Doug Pitt used to go on about something connected with this, I never quite understood what he was trying to get across.

I'm sorry Klaas, I don't quite understand your last sentence above.

Steve A.

Not being 'funny', but does anyone still use CDs or DVDs any more? In the resort towns here there used to be guys walking around selling pirated copies of moves for about two quid each. Demand has dropped such that you don't see them anymore. Also in the UK witness the closure of so many high-street shops like HMV etc. for the same reason.

http://www.independent.co.uk/news/busin ... 56780.html

http://www.mirror.co.uk/money/city-news ... ng-1719908

In fact the last CD/DVD I bought was the NBTVA's back-issue newsletter CD and that was about three years ago! Neither of the two PC's here in my workshop/office have DVD/CD drives.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby M3DVQ » Fri Feb 07, 2014 2:40 am

Steve Anderson wrote:Not being 'funny', but does anyone still use CDs or DVDs any more? In the resort towns here there used to be guys walking around selling pirated copies of moves for about two quid each. Demand has dropped such that you don't see them anymore. Also in the UK witness the closure of so many high-street shops like HMV etc. for the same reason.


Yes, but they buy them mail order from amazon for half the price of the high street retailers (before they all went broke)

As for pirate DVDs, I suspect everyone has noticed that if you want dodgy copies of films free is an even better price.
M3DVQ
Just nod and pretend you understand me
 
Posts: 326
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Postby Klaas Robers » Fri Feb 07, 2014 8:17 am

Steve,

I was writing about the NBTV-CDs of the NBTVA. For the CCNC test disc I could read back how long I made the sync and the porch. This is less easier for me for the Black and White NBTV CDs No 1, No 2 and No 3.

However I wrote that I expect that those lengths are the same. I made those discs using Pascal routines and I have to find back how to start that up again, Borland Turbo Pascal.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1453
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands


Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 3 guests

cron