The SSTV PD modes and vertical sync?

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

The SSTV PD modes and vertical sync?

Postby Steve Anderson » Tue Jan 04, 2022 3:45 pm

The PD modes are well-established hi-res colour modes for SSTV.

In digesting the specifications for these I notice there is no mention of vertical sync. There is mention of VIS codes, I cannot see why these couldn't replace the usual vertical sync. That assumes the VIS code is transmitted at the start of every frame, no problem there. It does possibly mean a somewhat different line (or insert shall we say) at the start of each frame, again no problem.

But so far I haven't been able to find any documented example of how this is done. I have quite a bit of data regarding this, but it's not very explicit. e.g. "VIS Codes Dayton Paper", particularly the PD section (attached).

Having downloaded a few PD mode files I was going to 'reverse engineer' these via a SSTV demodulator and 'scope and see what I might find. Though it's not the best way to get data.

Alternatively I was going to contact Paul Turner again and ask. But I'd rather not take up his time unless I really have to.

Can anyone here enlighten me? Has there been any updates to the version attached here. (Dayton SSTV forum, 20 May 2000).

Steve A.

VIS Codes Dayton Paper.pdf
(108.19 KiB) Downloaded 140 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4863
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: The SSTV PD modes and vertical sync?

Postby Klaas Robers » Wed Jan 05, 2022 5:30 am

As far as I understood this: there is no V-pulse. There is always just one frame transmitted and then there is silence.
So it starts with the VIS-code,
which ends with a 30 ms 1200 Hz,
you may see this as a V-pulse.
Then there come the specified number of lines,
and when they are sent, there is no H-pulse at the end,
there is just silence.

So it is always a one frame mode, just like .jpg. Does this help you?
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1611
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: The SSTV PD modes and vertical sync?

Postby Steve Anderson » Wed Jan 05, 2022 4:03 pm

Klaas Robers wrote:Does this help you?

Yes it does! I hadn't looked at it as a single-frame mode before. It does make a lot of sense when you think of it that way.

The 30-ms/1200Hz pulse at the end of the VIS code could be used as a frame trigger/vertical pulse. There is also a total of 300ms of VIS code which is either 1100, 1200 or 1300Hz from the VIS start bit to the VIS stop bit inclusive. That could be used to reduce spurious frame triggering as the line sync pulse at 20ms is not far different in length to the VIS bit-length, 30ms.

Though one should be able to differentiate between 20 and 30ms quite easily.

Thanks for the idea/concept Klaas!

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4863
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: The SSTV PD modes and vertical sync?

Postby Steve Anderson » Fri Jan 07, 2022 1:16 pm

An observation...a large percentage of PD-mode files seem to be recorded as a 48kHz sample-rate WAV file. I don't see why. An 8kHz sample-rate should be perfectly adequate, it's still standard SSTV as far as the sub-carrier is concerned. In most cases it will have been through a receiver with a baseband audio bandwidth of about 3kHz making a sample-rate of 8kHz adequate.

If you wanted a bit of 'breathing space' due to SSB mistuning surely 12 or 16kHz sampling should cover it. Though the output is still subject to the filtering in the receiver. So that reasoning is dubious.

Although storage (disc) space isn't as precious as it used to be, using 48kHz sampling means the files are 4-6 times larger than they need to be...

...or have I missed something?

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4863
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: The SSTV PD modes and vertical sync?

Postby kd2bd » Sun Jan 09, 2022 4:37 am

Modern versions of QSSTV default to 48 kHz sampling. I don't recall what it was in the past, but I do know that it used to be lower.

qsstv_config.png

The ISS SSTV samples from Dec. 2020 I posted last week were recorded using some homebrew recording software. I selected 48 kHz to be compatible with QSSTV. (FLAC compression could reduce the file size without introducing loss, but this site doesn't accept FLAC files... :( )

Modern sound systems have become amazingly complex, with inputs taken from multiple sources, and rate conversions being performed on-the-fly, working into a sound card that may be operating at some fixed, standard sampling rate (48 kHz?):

Pulseaudio-diagram.png

73 de John, KD2BD
User avatar
kd2bd
Research Scientist
 
Posts: 41
Joined: Wed Sep 28, 2016 5:47 am
Location: New Jersey

Re: The SSTV PD modes and vertical sync?

Postby Klaas Robers » Sun Jan 09, 2022 8:40 am

John, SSTV in the past was never sampled at whatever rate. SSTV was recorded on cassette tape. A;lso that was invented and introduced rather recently in 1973 when I built my 7BP7 monitor.

We did rather recently some experiments in recording SSTV digitally in 8 bits resolution and a sampling rate of 8 kHz. There was almost no degradation visible compared to 48 kHz an 16-bits resolution. That was a reduction of a factor of 12 (and 24 if you record in mono). For hard disc space there is no reason any more to do that, but if you want to attach an SSTV .wav file to an e-mail it makes things much easier.

However upsampling from 8 kHz to 48 kHz for reproduction should be done in a proper way, otherwise you get quite some added noise.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1611
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: The SSTV PD modes and vertical sync?

Postby Steve Anderson » Sun Jan 09, 2022 1:25 pm

kd2bd wrote:...but this site doesn't accept FLAC files... :( )

73 de John, KD2BD


Fixed...if anyone wants to upload a file to this forum with a currently unaccepted extension, just let Andrew, Dave or I know and we'll add it to the 'accepted' list...within limits...

FLAC is a lossless audio compression system, unlike mp3 etc. which is lossy. FLAC files are typically half the size of WAV files all other things being equal...

FLAC - Free Lossless Audio Compression.

https://en.wikipedia.org/wiki/FLAC

https://www.whathifi.com/advice/mp3-aac ... -explained

FLAC software here under the 'downloads' tab:- https://xiph.org/flac/ versions for Win, Mac and Linux...

VLC media player also supports FLAC playback. 'Audacity' also supports the creation, importing, editing, conversion and exporting of FLAC files.

Steve A.

The FLAC file extension should be 'filename.flac', not 'filename.fla' or similar on this forum.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4863
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: The SSTV PD modes and vertical sync?

Postby kd2bd » Mon Jan 10, 2022 12:20 pm

Thanks, Steve, for allowing FLAC audio attachments! :D

And Klaas, yes, I have many cassette tapes containing classic (120/128 line) SSTV signals that I recorded off-the-air from around 1979 to 1982 when I was just a teenager (and SWL). In fact, it was my desire to revisit those old tapes that provided the incentive for me to create my scan converter. I'm still going through them as time permits, and reliving old memories.

Frequency response was never really an issue with tape recordings, but poor tape speed regulation (wow and flutter) would often cause horizontal jitter that grew progressively worse as the horizontal scan moved towards the right side of the image. With each line video line came a new horizontal sync pulse that brought the horizontal sweep back into alignment, even if it was only briefly:

frame03424.jpg
frame03424.jpg (25.21 KiB) Viewed 2423 times

The image above was transmitted by AC5D of Stigler, Oklahoma (USA) on 14.230 MHz more than 40 years ago. Not only did I capture it on tape, but the original transmission made by AC5D was probably a tape playback of his own.

I seem to remember he used an analog sampling vidicon camera, and produced many clever and artistic "programs" by concatenating individual SSTV frames together on tape. Then he would play his various "programs" during a QSO by selecting the appropriate tape.

Today's non-interactive SSTV is incredibly boring and impersonal by comparison. Sadly, AC5D is probably a silent key, today. His callsign has been re-issued to someone else.

73 de John, KD2BD
User avatar
kd2bd
Research Scientist
 
Posts: 41
Joined: Wed Sep 28, 2016 5:47 am
Location: New Jersey

Re: The SSTV PD modes and vertical sync?

Postby Steve Anderson » Mon Jan 10, 2022 2:19 pm

kd2bd wrote:...but poor tape speed regulation (wow and flutter) would often cause horizontal jitter that grew progressively worse as the horizontal scan moved towards the right side of the image. 73 de John, KD2BD

I wonder if it's worth thinking about a 'Timebase Corrector' for signals like these? Or does the 'jitter' somehow convey the state of the art at the time and best left as is? If the wow-and-flutter is compound from two different tape machines it might be impossible to make a significant improvement.
To a degree it also depends on the frequency of the wow/flutter.
I guess there's only one way to find out - and that's to try it. In principle this could also be applied to NBTV.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4863
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: The SSTV PD modes and vertical sync?

Postby Klaas Robers » Tue Jan 11, 2022 3:13 am

Steve, this depends on the frequency spectrum of the irregularities of cassette tape move. Of course you can rely on the precision of line pulse detection, read each line into a memory and then stretch of shrink the line during read out, such that the line duration is again 60 ms. However velocity changes higher than say 15 Hz cannot be corrected. You see on the picture above that on the left side of the picture the vertical lines are almost straight and at the left you see the accumulated timing errors. You could eliminate the irregularities at the right side by this type of time base correction, but you cannot do this for picture contents in the middle.

But it would be a good idea to try how far you can go. May be it is a useful improvement.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1611
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: The SSTV PD modes and vertical sync?

Postby Steve Anderson » Tue Jan 11, 2022 2:25 pm

Not having a cassette machine of any type in the house I cannot make any measurements. I have an open-reel machine, but its wow & flutter figures are already very low, measured at around 0.04%. I recently replaced all the rubber-type components, belts, pinch-roller etc.. I'll see if I can beg, borrow or steal a cassette deck for a while.

As a guess the lowest wow component is likely to be the capstan/flywheel/pinch-roller assembly as it generally has the lowest RPM in the tape transport. From memory a cassette capstan is around 2.5mm in diameter, maybe less. So at a tape speed of 1.875cm/s or 18.75mm/s with a capstan circumference of say 7.85mm is 2.39 revs/s, a wow frequency of 2.39Hz, or a period of 420ms. (All very approximate). But it gives us a ball-park figure, it could be between double or half that.

That very approximate 420ms is seven lines of 60ms/line SSTV. Yet looking at most screen-shots of SSTV displaying what appears to be wow/flutter artifacts it seems to be a shorter period. But that's better than longer - less memory in the TBC (timebase corrector) required.

More investigation is necessary...

In a similar manner this could be applied to 400Hz line frequency NBTV. Not only for recorded material, but perhaps taking out some of the errors in disc-drilling.

I assume few use tape or cassettes these days to record NBTV or SSTV, preferring to use a PC which shouldn't have any evidence of wow & flutter. This is more for trying to recover old recordings on such media.

Steve A.

This might require a separate thread of its own...

Didn't Karen Orton do something regarding this at some point? I have a vague recollection of such, but I might be mistaken...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4863
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: The SSTV PD modes and vertical sync?

Postby Steve Anderson » Thu Jan 13, 2022 2:55 pm

This is an ideal application for some software run on a PC. Over the past day or so I've been thinking of making this a separate hardware/software unit, but really that isn't necessary.

So if there's someone willing to have a go I think it will be of interest to many. Unfortunately writing code for PCs isn't within my remit.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4863
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand


Return to SSTV

Who is online

Users browsing this forum: No registered users and 2 guests