SSTV Demodulation.

Forum for discussion of SSTV topics. Slow Scan television (SSTV) is a picture transmission method used mainly by amateur radio operators, to transmit and receive static pictures via radio in monochrome or colour.

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Re: SSTV Demodulation.

Postby Steve Anderson » Tue Aug 03, 2021 5:05 pm

kd2bd wrote:Klass,The test SSTV images I am currently using follow the 30 ms vertical sync pulse with 0.031666 seconds of black, followed by a 5 ms horizontal sync pulse. That is a new change for me. In the past, I immediately followed the 30 ms vertical sync with a new line of video, followed by a 5 ms pulse, and so on. 73 de John, KD2BD

Hmm, interesting, I've always interpreted Cop Macs timings as below...I guess in those early days they weren't carved in stone...so different people will see it in different ways...all should work fine, so really there's no 'right' or 'wrong'...this is probably one of many permutations...though it seems the same or similar as what John is saying above...

There's no particular advantage with this arrangement, perhaps with the exception that during the vertical interval the horizontal timebase keeps the same frequency throughout...more advantageous with direct magnetic deflection rather than electrostatic as a guess...

Steve A.
SSTV Basic Timings 1l.gif
SSTV Basic Timings 1l.gif (24.96 KiB) Viewed 535 times


The reference point for timings could be changed to the negative-going edge of the sync pulse, it really makes no difference, but might help with envisioning how to do this in practical terms...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4694
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: SSTV Demodulation.

Postby Klaas Robers » Wed Aug 04, 2021 2:19 am

Yes Steve, this is also the way I made my SSTV-video. However with those ugly VIS-codes the continuity of the horizontal sync is any way interrupted. I never used that code.

It is that I started my carreer in TV technology at Philips Research, so I don't know better than H-sync that continues running during the V-sync interval. That is also the reason that I intend to program a H-sync flywheel to get away with the sync instability because there is no pre- and post pedestal. In the beginning of the VCR-time we had the same problem with the sync at the top of the picture. To get away with that the time constant of the H-sync flywheel had to be changed. I still know that this was only on the last "program" of the preset channels. It was never the idea that the H-sync had larger timing instabilities.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1568
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: SSTV Demodulation.

Postby kd2bd » Wed Aug 04, 2021 3:22 am

Thanks for posting your timing diagram, Steve.

For decades, the only SSTV video standards I ever saw in U.S. publications came directly from an article written by Copthorne MacDonald for the January 1961 issue of QST magazine:

sstv_standards.gif
sstv_standards.gif (131.9 KiB) Viewed 523 times

"Item D" from your illustration was completely new to me until very recently when I saw it mentioned in an on-line SSTV document. I only recently made that addition to my SSTV generation software to keep the phase of the horizontal sync consistant from frame-to-frame while performing my current sync detection experiments.

My TriplePIC follows a "triggered sweep" philosophy, and so it doesn't care about the phase of the horizontal sync between video frames. My original SSTV Oscilloscope Adapter had a horizontal sweep that was triggered by each individual sync pulse as well.

But a year or so later when I built a Robot 70 (with a PLL-triggered horizontal sweep), I often saw on-air signals where the first few lines of the image were out-of-alignment as the PLL was forced to re-acquire lock at the start of every new frame of video. I thought (at the time) that the problem was due to the PLL being thrown out of lock by the vertical sync pulse, but I now have a better understanding of what was going on. I felt that the poor sync handling of the Robot 70 was its biggest downfall.

In the early 1970s, Copthorne had an SSTV column in "CQ" magazine (published in the U.S.). In the August 1974 issue, he wrote a product review of the Robot 70 where he stated:
"This reviewer's only real complaint about sync performance is the line-to-line jitter which starts to appear when the signal drops to the point where "snow" is visible in the picture".

In the September issue, he offered his own solutions to the jitter problem that was prevalent in probably the majority of SSTV monitors and scan converters at the time. In fact, the cover photo of that very same issue shows the output of W0LMD's SSTV keyboard after being scan converted and displayed on an NTSC (fast-scan) television set, complete with very noticeable line-to-line jitter! :oops:

W0LMD-jitter.jpg
W0LMD-jitter.jpg (140.64 KiB) Viewed 523 times

In Cop's opinion, the biggest cause for jitter was what Robot (and others) had employed in their P7 monitors at the time: A sync separator that stripped sync from the demodulated video, followed by a PLL-triggered horizontal sweep. He even cited his original P7 monitor as having instability problems as it employed a 1200 Hz bandpass detector followed only by half-wave rectification and no post detection filtering at all.

I adopted his suggested techniques in my TriplePIC scan converter, using modern-day components, and the results have been very good. But now I realize it may be possible to do even better. :D

73 de John, KD2BD
User avatar
kd2bd
Laboratory Assistant
 
Posts: 21
Joined: Wed Sep 28, 2016 5:47 am
Location: New Jersey

Re: SSTV Demodulation.

Postby Steve Anderson » Wed Aug 04, 2021 12:08 pm

Klaas Robers wrote:Yes Steve, this is also the way I made my SSTV-video. However with those ugly VIS-codes the continuity of the horizontal sync is any way interrupted. I never used that code.

Agreed, VIS codes interrupt the smooth continuous flow of the line-syncs and would probably/definitely upset monitors from those early times. Such is progress.
These 120/128-line 50/60Hz derived standards do have an assigned VIS code, and likewise I've never implemented it. For more recent SSTV modes, OK, but here I am sticking to the origin of the format, so no VIS codes.

If ever you came across this 'Robot 8' mode on-air (most unlikely*) it may be possible to remove the VIS code, but that would be a moderately complex device with very little application. Quite a bit of effort for little return.

Steve A.

*..or tell me otherwise!
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4694
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: SSTV Demodulation.

Postby Steve Anderson » Wed Aug 04, 2021 3:56 pm

kd2bd wrote:Thanks for posting your timing diagram, Steve. 73 de John, KD2BD

Thanks, but it is my interpretation of what I've seen and read over the past 5-odd decades, it would seem to be the most logical way to do it, and it seems many others have drawn the same or similar conclusions. We'll never know if it's what Cop Mac intended, but there's my tuppence for what it's worth...

Even if you 'reverse engineered' Cops' articles and designs, I doubt you'd reach a definitive conclusion...unless there are others I haven't seen...

Steve A.

It's interesting to note that the vast majority of Amateur Radio and general electronics publications carried items r.e. SSTV. But although covering virtually all other aspects of Amateur Radio, "Wireless World" published nothing at all about SSTV...and I have gone through the issues thoroughly from the mid-50s until the magazine's demise. Someone tell me I'm wrong!

The BATC (British Amateur Television Club) did and still do publish occasional SSTV items, more often these days either colour/hi-res (PD modes etc.) or digital. Though you can raid their on-line back issues (decades) for free...they also have a SSTV handbook, but it's well and truly dated...the regular publication is called CQ-TV...though very little pertains to NBTV or SSTV...or even the analogue 625 PAL system, now largely forgotten it seems..

Will 525/625 attain the same status and cult following of 405?...at least in the UK? I'm not sure...somehow I doubt it...though maybe it's still early days...

https://batc.org.uk/cq-tv/cq-tv-archive/

https://batc.org.uk/wp-content/uploads/ ... panion.pdf

Not been updated since 2002...The website with absolutely nothing new on it... https://www.bvws.org.uk/405alive/
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4694
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: SSTV Demodulation.

Postby Steve Anderson » Sat Aug 07, 2021 8:27 pm

This is maybe where I got the idea for the vertical interval from, though this is for the original AM subcarrier system, I guess Cop carried over the basic timing into the scfm arrangement...

They were comparing the two modes on-air, so it would seem sensible to keep the baseband parameters the same to eliminate those factors from the tests, also the same images to remove any subjective bias...

QST Sep. 1958, p34.

Steve A.

AM SSTV Vertical 1.gif
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4694
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: SSTV Demodulation.

Postby kd2bd » Sun Aug 08, 2021 2:28 am

So, is that vertical sync pulse interval a "bug" or a "feature"? :)

The Complete Handbook of Slow-Scan TV (1977 by Dave Ingram, K4TWJ, SK) shows a similar timing diagram for an SCFM signal in the first chapter of the book:

timing1.gif

while a later chapter on the subject of digital scan conversion illustrates a demodulated waveform where video is not in any way interrupted by vertical sync:

timing2.gif

The original specifications for slow-scan television indicate 120 video scan lines. However, some of the early timing diagrams show the vertical sync pulse "clobbering" 40.5% of the first video scan line, thereby producing only 119.6 video scan lines.

My guess is that they had little choice but to do this in the early days because the simple timing circuits used were based on the frequency of the local line voltage, whose phase could not be controlled or modified as needed to produce 120 lines of video in their entirety. As SSTV video generation grew more sophisticated, it finally became possible to produce video where all lines were of equal length, and actually contained valid picture information.

73 de John, KD2BD
User avatar
kd2bd
Laboratory Assistant
 
Posts: 21
Joined: Wed Sep 28, 2016 5:47 am
Location: New Jersey

Re: SSTV Demodulation.

Postby Steve Anderson » Sun Aug 08, 2021 2:59 pm

Virtually all analogue TV systems lose (lost?) one or more lines due to the requirement of frame/field synchronization. The notable exception is NBTV, though there may be others. In this case, early SSTV, it is in effect just one in 120, or 0.83%, slightly less if using 128 lines. 0.78%. Contrast that with 625-lines, 50 lines are devoted to frame/field syncs and/or blanking per frame (25 per field) out of 625, or 8%. A similar figure for 525. [1]

So SSTV and NBTV are comparatively efficient in use of bandwidth and available time.

In the case of Dave Ingram's waveforms, well, is this an example of the result of lack of standardization from day one? Not of his doing, a case of haphazard evolution. Hence, thankfully, these vast committees that are set up to standardize matters. The CCIR, NTSC, NAB, IEC and so on. As much as anyone hates bureaucracy, there are times when it's not such a bad thing.

So, where does that leave us r.e. early SSTV? A good question, with not one simple answer...

Steve A.

[1] In later years those blanked lines came in useful for all manner of additional signals, test signals, teletext, 8-field editing (PAL), subtitles, and timecode, among others.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4694
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: SSTV Demodulation.

Postby kd2bd » Mon Aug 09, 2021 12:07 pm

Here in the U.S., our FCC provides guidelines through their rules and regulations for the Amateur Radio Service. They are intentionally broad in nature so as to encourage the experimental nature of the radio service. Basically, the only limitation to what we can do in terms of image communications comes down to bandwidth. This is probably true everywhere on the planet.

If you look back to the system characteristics published for Cop's original subcarrier AM SSTV system, you'll find it had 120 lines, a horizontal scan rate of 20 Hz, and a vertical sweep rate of 6 seconds. The vertical and horizontal sync pulse durations are given as being approximate.

scam-standards.png
scam-standards.png (184.26 KiB) Viewed 382 times

Going further, it is interesting to point out that a 1.5 ms horizontal sync pulse of a 2000 Hz subcarrier should yield 3 subcarrier cycles, yet only 2 appear in Figure 4 above. And a 15 millisecond vertical burst should yield 30 subcarrier cycles, yet only 12 cycles are illustrated in the drawing. I didn't try counting the total number of subcarrier cycles per scan line, but even the ratio between horizontal and vertical sync intervals illustrated in Figure 4 don't match the 10:1 ratio specified in the "standard" published at the time.

It is likely that Copthorne made suggestions for the illustration, but the artwork was probably done at QST / ARRL, and they got it wrong.

That goes for any article published anywhere, of course. But is does appear that even in the earliest days of SSTV, the experimenters played "fast and loose" with setting and closely adhering to standards.

73 de John, KD2BD
User avatar
kd2bd
Laboratory Assistant
 
Posts: 21
Joined: Wed Sep 28, 2016 5:47 am
Location: New Jersey

Re: SSTV Demodulation.

Postby Steve Anderson » Mon Aug 09, 2021 12:37 pm

kd2bd wrote:...it is interesting to point out that a 1.5 ms horizontal sync pulse of a 2000 Hz subcarrier should yield 3 subcarrier cycles, yet only 2 appear in Figure 4 above...

I wouldn't assume that the quantities of subcarrier cycles would/should be correct, it's more of a conceptual overview/sketch. The other waveform sketches show more of an 'envelope' view, or a baseband signal, prior to subcarrier modulation. (Either AM or FM). FM modulation shouldn't look anything like Fig.4., the aim is a constant amplitude subcarrier, though in practice there probably will be an amount of AM due to filters or other bandwidth restrictions.

But that Fig.4 could cause more confusion as you've pointed out.

I must admit I've never looked at the SSTV FM subcarrier level generated here with a view to looking for AM, something to put on the TO DO list...for no particular reason, just out of interest...

Steve A.

I just looked at a few of the FM SSTV subcarrier files generated here, there is an element of AM. around 20% between syncs and peak white. That'll generate a few moderately low-level unwanted sidebands, but the results don't seem to show any ill effects...it could be improved, but having gone through a SSB transmitter/receiver chain it would possibly have a similar result anyway...

That was with 8-kHz sampled files, those that were sampled at 48kHz show much less AM, around 10%, perhaps less. Possibly down to the filtering (or lack of) in the PC. Either way, I don't see it as a big issue...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4694
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Previous

Return to SSTV

Who is online

Users browsing this forum: No registered users and 1 guest