Slow Scan Television Experiments

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: Slow Scan Television Experiment

Postby Harry Dalek » Sat Nov 22, 2014 8:21 pm

Heres 2 quick videos more a demonstration than any thing one for those who can not be bothered downloading ,i am a little more careful and slower scanning...having a camera in one hand and having to do it in the other its a bit hard need to get my tripod out ! :roll:

forgive the tape it was a prototype :wink:
Attachments
MVI_0014_001.mpg
i wish the sound came out better
(2.57 MiB) Downloaded 617 times
MVI_0013 (1).gif
posted this up as a gif as you see i was having sound problems the curse of using nature light and it changes
MVI_0013 (1).gif (2.44 MiB) Viewed 41400 times
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Slow Scan Television Experiment

Postby Harry Dalek » Sat Nov 22, 2014 8:37 pm

I am pretty sure i know what i did wrong as in the picture definition.

I only used a picture the size of one mirror face . now that will do it, no wonder 15hz was to slow ! :roll:

I am counting about 30 lines in the image i posted 4 of those are 120 lines have i invented 30 line sstv by mistake ! :lol:

Just thinking i need to repeat the experiment by having 4 of the same images in place over the 4 mirrors do it in square instead of a cylinder as i am trying to do with the mirrors, i could also then try the frame mirror .

Doing the cylinder test i now know the distance needed to the sensor and its just an optical problem now .

I love how one thing leads to another without knowing ! you could start and end a project as you wanted but it is fun when it takes you in another direction you were not expecting.
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Slow Scan Television Experiment

Postby M3DVQ » Sun Nov 23, 2014 5:08 am

Steve Anderson wrote:I've had a fiddle around with MMSSTV with you comments foremost, but I cannot find the correct calibration 'knob'. This is a live analogue input generated electronically fed straight into the PCs line input, it's not recorded or processed in any manner.

The clock accuracy you are referring to, is it the soundcard clock? The PC clock? Or the clock the generates the SSTV signal itself? I have been assuming the 'knob' for this is within MMSSTV, maybe I'm wrong there.

When the sample rate is set to for example 11025, but the sound card clock is actually 48000 there is a resampling of the audio which good enough for audio screws up the frequencies and timings enough to upset things like SSTV and PSK etc.
The setting for the sample rate in mmsstv is on the misc tab of the options window. With this set to 48000 and the auto slant turned off both TX and RX should be at 48000Hz and two computers should be able to send and receive pictures to each other closed circuit with no slant error.


Steve Anderson wrote:As per your suggestion I recorded 60 seconds of the generated SSTV signal at 12000Hz sample rate using 'Audacity' on a Windows XP machine. If you can quickly play it and let me/us know what you find. Please don't spend much time on this. This .wav file has been nowhere near MMSSTV at all.


It's all very strange. I'm getting exactly the same slant you've been experiencing. Everything here is going through a software stereo loopback running at 48000Hz on windows 7.

If I play the file in with the clock rate set to 12000 then adjust the slant in MMSSTV until the bars are vertical MMSSTV reckons to have resampled it (if my understanding of how slant correction works is correct!) to 11962.8Hz.
M3DVQ
Just nod and pretend you understand me
 
Posts: 338
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Re: Slow Scan Television Experiment

Postby Steve Anderson » Sun Nov 23, 2014 3:39 pm

M3DVQ wrote:If I play the file in with the clock rate set to 12000 then adjust the slant in MMSSTV until the bars are vertical MMSSTV reckons to have resampled it (if my understanding of how slant correction works is correct!) to 11962.8Hz.

Yes, very strange indeed! That's an error of -3100ppm in the sample rate which far greater than any crystal frequency tolerance. Even the cheapest ones are spec'd at +/-100ppm, yer average common-or-garden variety as used in PCs are around +/-30ppm. So where that error is coming from is beyond me.

More interestingly (to me) is did it sync up? Both line and frame?

It is possible that I am using a defective PC, or the driver of the AC97 codec is slightly corrupted (Realtek). I did tweak the clock frequency as per your instructions, the adjustment required was in the opposite direcion...but it refused to frame sync.

The reason I initially thought of using MMSSTV was simply to confirm that the modulator is producing a valid SSTV signal from the baseband video input (in the absence of any other way of checking it). So far it would appear it is. I had no intention of using MMSSTV on a long-term basis.

I might quickly try my nieces laptop tomorrow as a 'sanity check'. a Gateway Windows 7 machine. Being a laptop it probably only has a Mic In not a Line In - a couple of resistors will fix that.

Thanks for your help...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Slow Scan Television Experiment

Postby M3DVQ » Sun Nov 23, 2014 10:55 pm

Steve Anderson wrote:Yes, very strange indeed! That's an error of -3100ppm in the sample rate which far greater than any crystal frequency tolerance. Even the cheapest ones are spec'd at +/-100ppm, yer average common-or-garden variety as used in PCs are around +/-30ppm. So where that error is coming from is beyond me.

Indeed, it's a long way out.

Steve Anderson wrote:More interestingly (to me) is did it sync up? Both line and frame?

I got just the same as Harry's screenshot on Friday.[/quote]

There is no "frame sync" without VIS etc. MMSSTV just starts receiving as soon as signal opens the squelch. As for line sync it would sync if there was no slant, or if auto slant correction is turned on the auto sync gets the horizontal alignment fine.
M3DVQ
Just nod and pretend you understand me
 
Posts: 338
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Re: Slow Scan Television Experiment

Postby M3DVQ » Sun Nov 23, 2014 11:24 pm

Steve Anderson wrote:Yes, very strange indeed! That's an error of -3100ppm in the sample rate which far greater than any crystal frequency tolerance. Even the cheapest ones are spec'd at +/-100ppm, yer average common-or-garden variety as used in PCs are around +/-30ppm. So where that error is coming from is beyond me.

Indeed, it's a long way out.

Steve Anderson wrote:More interestingly (to me) is did it sync up? Both line and frame?

I got just the same as Harry's screenshot on Friday.

There is no "frame sync" without VIS etc. MMSSTV just starts receiving as soon as signal opens the squelch. As for line sync it would sync if there was no slant, or if auto slant correction is turned on the auto sync gets the horizontal alignment fine.
M3DVQ
Just nod and pretend you understand me
 
Posts: 338
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Re: Slow Scan Television Experiment

Postby M3DVQ » Sun Nov 23, 2014 11:56 pm

AHA! I've finally worked out what's going on. Curse me for not reading through the whole thread properly, and only now picking up Ralph's book.
Now I understand the vertical sync tones and everything in your recording. :oops:

The 8 second mode MMSSTV implements is the the Robot Research model 1200 variation with the 4:3 aspect ratio, longer line syncs, and no vertical sync tone, not the original "Macdonald" 1:1 format. Doh
M3DVQ
Just nod and pretend you understand me
 
Posts: 338
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Re: Slow Scan Television Experiment

Postby Steve Anderson » Mon Nov 24, 2014 11:21 am

M3DVQ wrote:AHA! I've finally worked out what's going on...The 8 second mode MMSSTV implements is the the Robot Research model 1200 variation with the 4:3 aspect ratio, longer line syncs, and no vertical sync tone, not the original "Macdonald" 1:1 format. Doh

Right. Now we're cooking with gas...

As I suspected a few days ago MMSSTV basically requires VIS, without them it will not sync. That's OK, for me it has done what I required of it, basically confirmed the modulator is doing its job, slant or no slant. Sync or no sync. It would have been nice if it could work at this standard, but if it doesn't, no big deal. That's one of the reasons I mentioned much older DOS-based software that might have worked VIS-less.

Time to move on...genuine thanks for your time and effort.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Slow Scan Television Experiment

Postby Harry Dalek » Mon Nov 24, 2014 4:13 pm

Reading the posts i think its time to find something that plays the original SSTV system .
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Slow Scan Television Experiment

Postby Steve Anderson » Mon Nov 24, 2014 4:54 pm

Harry Dalek wrote:...I think its time to find something that plays the original SSTV system .

Agreed. But I suspect that it's not going to be anything recent, hence my wondering about DOS based software which as yet I haven't tried. There may also be hardware issues too. In the days of DOS sound-cards were a luxury and not many had them so the I/O was often via either the RS232 port or the LPT port. A lot of modern machines have neither of these, but you can get USB-RS232 or LPT 'cables', but often these create more problems than they solve. Thankfully the two PCs I use have both legacy ports plus USB 2.0.

Ignoring PC-based software solutions for the moment, I can go either of two ways (or even both), build a 1970s-style P7 CRT monitor but with modern components (I have both 3" and 5" P7 CRTs), and/or an upconverter to 625. The second alternative from a hardware point of view is much easier, no big metal chassis, no bulky transformers and no risk from kV of DC. It would also probably be quicker and far less expensive.

So for a while I'm goint to apply the brakes to this whilst I consider what to do next.

But whatever option I choose I am going to need a demodulator. There are plenty around from the designs of the 70s, but they generally don't do particuarly good job of it. So that's probably where my focus will be next.

Harry, sorry for 'hijacking' your thread...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Slow Scan Television Experiment

Postby Harry Dalek » Mon Nov 24, 2014 5:44 pm

Hi Steve

This only receives but it does 8 sec 12 sec B/w but input is via sound card live from what i am finding out so far .

http://users.belgacom.net/hamradio/rxsstv.htm

http://www.w4cn.org/Field-Day-2013/sstv

above a set up of the sound card for that software


I will see if theres any thing else out there

http://www.navymars.org/central/reg4/al/SSTV.htm

OH i found this on MMSSTV have you seen this ?

No thats fine this is for slow scan experiments let the experiments run free i say,my interest is doing a mechanical camera and optics so on my side of things it will be on that ...i like guests so no problems posting here all good
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Slow Scan Television Experiment

Postby Steve Anderson » Mon Nov 24, 2014 9:28 pm

Thanks for the links Harry, but all are based on the MMSSTV engine, i.e. it's gonna require the VIS signal at the start of each frame.

There is one site I found a few weeks ago that looked promising at first - forget the complexity for the moment, but I have my reservations as to the claims made. I hope that I'm proven wrong...

http://www.qsl.net/kd2bd/sstv.html

..and the site specific to his converter...

http://www.qsl.net/kd2bd/TriplePIC.html

...but at least we have a.n.other who's also interested.

I have been in contact via e-mail with him, but he's giving nothing away, perhaps as a result of his agreement with 'ATV Quarterly' which is understandable.

Steve A.

You'll also note that all is under 'perfect' signal conditions, no noise, fading etc...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Slow Scan Television Experiment

Postby Harry Dalek » Tue Nov 25, 2014 7:04 pm

Steve Anderson wrote:Thanks for the links Harry, but all are based on the MMSSTV engine, i.e. it's gonna require the VIS signal at the start of each frame.


This is a pain i did hope that program was new not a rehash ,i did see theres stuff on Linux and mac i also have the one and only sstv app for iphone that does 8 sec as well might have to pinch one the app is 6 dollars i think .

As you think the Dos ones more than likely are original system but needs that little opamp circuit for the communication port, now i have tried jvfax 7.1 in windows 7 and it seems to work but i have not tried it with the circuit in anger not of late any way i recall doing this in xp many times.

There is one site I found a few weeks ago that looked promising at first - forget the complexity for the moment, but I have my reservations as to the claims made. I hope that I'm proven wrong...


Yes its a bit of work not really knowing ,you would rather go an easier path with more choice such as Dos programs even if it is a pain these days .

http://www.qsl.net/kd2bd/sstv.html

..and the site specific to his converter...

http://www.qsl.net/kd2bd/TriplePIC.html


MMM well i think using jvfax is less work try it on your pc worth a shot does not seem to have a problem in windows .

i don't think it would be clean of that robot system.

...but at least we have a.n.other who's also interested.


Yes this is good more that better may be one day it could be a topic on its own :wink: SSTV is just as interesting as NBTV and some ways easier .....for me i learn for one to use in the other .

I have been in contact via e-mail with him, but he's giving nothing away, perhaps as a result of his agreement with 'ATV Quarterly' which is understandable.


Oh ok but he is welcome to join say hello :)

[code]Steve A.

You'll also note that all is under 'perfect' signal conditions, no noise, fading etc...[/quote]

I like the weather satellite fax system you just lock onto the one AM tone line frequency 2400hz of the system the gray scale white to black is just higher or lower volume of that tone, great system for a PLL,you can lock on to the signal in noise levels we would have no hope of detecting .
And its a mechanical camera with a rotating mirror drum for the line and the movement of the space craft for a never ending frame number of lines,well you would have to follow the signal around the earth and it does end when the space craft falls from the sky end of it s useful life.

Sstv is sort of fax in a way all early image systems should not be forgotten !
Attachments
Screen 00042.jpg
noaa weather fax
Screen 00042.jpg (85.14 KiB) Viewed 41338 times
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Slow Scan Television Experiment

Postby Steve Anderson » Tue Nov 25, 2014 8:50 pm

Interesting the weather-fax system, I wonder if that AM system is still used? I would have thought it all digital these days.

As for your question regarding syncs within a recent PM...I think I understand what you're thinking of. A fixed frequency (1200Hz) oscillator that is switched to when you require some sync tone then back to the video sub-carrier signal after the appropriate length of time. Is that correct?

If it is I don't recommend that method as at the switching point it will introduce a discontinuity in the FM waveform. 'Traditional' demodulators using tuned circuits and the like would probably cope with this gracefully, but a software-based version might get a bit upset, it depends on how the software filtering is done (if any).

If you plan to use a VCO, say a 4046, an arrangement as shown below will do the trick. If for example the VCO produces 1kHz per volt of input the sync pulses disconnect the video amp via a 4066 during sync-time and and the 1.2V input provides the sync tone. At other times the 4066 is on and the vieo amp with an output impenance of much less than 100k dives the VCO. The video amp needs to be configured to produce +1.5V on black, and 2.3V on white.

Of course you can use any sensible voltage, this was just an example.

I don't know what you are using/planning for the video amp but it should be possible to make it fit in with this method.

Steve A.
Attachments
SSTV Mod 1.gif
SSTV Mod 1.gif (4.42 KiB) Viewed 41333 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Slow Scan Television Experiment

Postby Harry Dalek » Wed Nov 26, 2014 12:25 am

Steve Anderson wrote:Interesting the weather-fax system, I wonder if that AM system is still used? I would have thought it all digital these days.


http://en.wikipedia.org/wiki/Automatic_ ... ansmission

Checking here 3 noaas and a meteor Russian one is still going i think they are still in this old analog system so its easy for any one any where to decode the images .
i was thinking it would be nice to repeat the systems idea for fun same sort of idea for a little mechanical camera one day .

As for your question regarding syncs within a recent PM...I think I understand what you're thinking of. A fixed frequency (1200Hz) oscillator that is switched to when you require some sync tone then back to the video sub-carrier signal after the appropriate length of time. Is that correct?


Yes Steve i was wondering which was correct ,sync inserting part isn't really explained clearly in what i have read .

Inserting the sync i was thinking the VCO had to drop back to 1200hz but with the head amp on it would be any thing depending on the light levels between the mirrors opto switch for the sync or not ,i suppose its because i am doing a mechanical system and not a full electronic as in the flying spot scanners i have been reading about .

Sort of why i was thinking about a stand alone 1200 hz oscillator and sync in its own system any case i reading on .

If it is I don't recommend that method as at the switching point it will introduce a discontinuity in the FM waveform. 'Traditional' demodulators using tuned circuits and the like would probably cope with this gracefully, but a software-based version might get a bit upset, it depends on how the software filtering is done (if any).


OK well software is what i am going to use to view so i will have to get it to work with that .

If you plan to use a VCO, say a 4046, an arrangement as shown below will do the trick. If for example the VCO produces 1kHz per volt of input the sync pulses disconnect the video amp via a 4066 during sync-time and and the 1.2V input provides the sync tone. At other times the 4066 is on and the vieo amp with an output impenance of much less than 100k dives the VCO. The video amp needs to be configured to produce +1.5V on black, and 2.3V on white.


Thanks for the plan there Steve thats makes sense ! when i get around to it i will try this circuit idea for sure i have worked with the 4066 in the past and i have a few handy .
Great idea disconnecting the head amp this way !

Of course you can use any sensible voltage, this was just an example.
I don't know what you are using/planning for the video amp but it should be possible to make it fit in with this method.

Steve A.


MY head amps a Ne5534 my system is very simple at the moment just that the light sensor and a 4046 test VCO


I think with just the ujt it could be tiny ,i am playing around a bit with it to see what i can get away see what works .

I will be doing some more tests in a day or 2 the time off i have thought of a test with the line mirror which might get me step closer the last test i have an idea now what to do .
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

PreviousNext

Return to SSTV

Who is online

Users browsing this forum: No registered users and 8 guests