Sync detection

Moderators: Dave Moll, Steve Anderson

Sync detection

Postby Harry Dalek » Wed Mar 01, 2017 5:18 pm

I am up for ideas on the sync detection ...i recall what i did on my first sstv monitor i used 2 monostables one for horizontal one for vertical timing cap on both with a trimmer to adjust ...it triggered to the correct pulse ..like tuning it in ..at the time it was just trial and error trying different ideas and that one worked .
Klaas I had a look at your monitors circuit it uses an ic i am not familiar with .
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: Sync detection

Postby Klaas Robers » Wed Mar 01, 2017 6:44 pm

How is it named?

On the other hand, you are using the circuits of ROBOT, that Steve improoved. He has also circuits for sync detection. Keep that line.

In my electronics there is a tuned cicuit (inductor and switchable capacitors), resonating on 1200 Hz (1000, 1100, 1200, 1300, 1400 Hz). That wave form is rectified (diodes) and low pass filtered (C-L-C low pass filter). This analogue signal is limited (FCL101 = Schmidt trigger level detector), which converts it to a binary (0 or 1) sync signal. The circuits of Steve will also give a binary sync signal. That is more or less interchangable with this sync signal.

The horizontal saw tooth oscillator should sync directly on this sync, the vertical needs an extra R-C low pass, such that it does not react on the short line pulses, but only on the long frame pulse.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Sync detection

Postby Harry Dalek » Wed Mar 01, 2017 7:34 pm

Klaas Robers wrote:How is it named?

On the other hand, you are using the circuits of ROBOT, that Steve improoved. He has also circuits for sync detection. Keep that line.
n my electronics there is a tuned cicuit (inductor and switchable capacitors), resonating on 1200 Hz (1000, 1100, 1200, 1300, 1400 Hz). That wave form is rectified (diodes) and low pass filtered (C-L-C low pass filter). This analogue signal is limited (FCL101 = Schmidt trigger level detector), which converts it to a binary (0 or 1) sync signal. The circuits of Steve will also give a binary sync signal. That is more or less interchangable with this sync signal.

The horizontal saw tooth oscillator should sync directly on this sync, the vertical needs an extra R-C low pass, such that it does not react on the short line pulses, but only on the long frame pulse.


OH the FCL101 is a schmidt trigger !
I forgot Steves 1200hz band pass filer in hes demodulator circuit that would filer the sync for me ! might use the monostable idea for the vertical triggering well the horizontal is half the battle so i will work on that next .
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: Sync detection

Postby Klaas Robers » Thu Mar 02, 2017 12:10 am

Harry, in SSTV there are two ways of sync detection.

1. called "the TV-way".This is the method of the clippped video signal. Everything above a level of half way the sync pulse is considered as a "0", not belonging to the sync signal, and everything below that level as a "1", belonging to the sync signal. This is how all TVs work and all NBTV monitors.

2. called "the RADAR-way". In this method not the video signal is the input, but the FM-signal that carries the video signal. Here a resonator (passive L-C circuit or active opamp-circuit) resonates at the frequency of the synchonisation signal (1200 Hz). The amplitude of the resonating signal triggers a sync signal.

I don't know why this is called the RADAR-way, although some forms of radar use a sweeping frequency. Many SSTV monitors in the past used it. I think that when there are lots of interference on short wave, the 2nd method is more insensitive to that interference.

But with clean signals, coming from .wav files, there is no reason to use this more difficult method. The first method needs only a level detector, which you feed with the demodulated video signal. This can be a single opamp, where the - input is connected to the demodulated video signal (that goes to the base of the BD115) and the + input to a potentiometer that adjusts the level voltage. I think this is the better way for you to go.

In NBTV the 2nd method was never used, because NBTV only has a video signal. It is never FM modulated.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Sync detection

Postby Steve Anderson » Thu Mar 02, 2017 2:56 pm

In the SSTV-625 converter sync is detected in software. The analogue output of the Robot 70 demodulator is fed into the write processor where it is converted into a digital value. Anything below a certain number count is declared as sync, anything above is presumed to be video. As the internal sampling rate is higher than strictly necessary it allows the micro to check that several samples in a row are sync, thus the odd glitch is ignored.

More correctly, it looks at the past 16 samples if 13 or more are below the sync binary value then it says, "This is sync." and acts appropriately. The sync value is derived from a front-panel pot again converted into digits so you still have a sync-slice control, the same applies to black level, gain/contrast and a few others. The control sampling is done during the remainder of the sync period so they sample at around 15-16Hz which is frequent enough. With no input (no sync) the micro defaults to a 10Hz control-scan rate.

The same idea will be used for NBTV (when I get around to it) but it will scan the controls at frame rate, 12.5Hz.

For simplicity you could simply add a comparator to output of the Robot 70 demodulator, anything below a certain voltage is sync, otherwise it's not. Then it's a case of separating the vertical from the horizontal - quite easy.

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: Sync detection

Postby Harry Dalek » Thu Mar 02, 2017 5:44 pm

Klaas Robers wrote:Harry, in SSTV there are two ways of sync detection.

1. called "the TV-way".This is the method of the clippped video signal. Everything above a level of half way the sync pulse is considered as a "0", not belonging to the sync signal, and everything below that level as a "1", belonging to the sync signal. This is how all TVs work and all NBTV monitors.

2. called "the RADAR-way". In this method not the video signal is the input, but the FM-signal that carries the video signal. Here a resonator (passive L-C circuit or active opamp-circuit) resonates at the frequency of the synchonisation signal (1200 Hz). The amplitude of the resonating signal triggers a sync signal.

I don't know why this is called the RADAR-way, although some forms of radar use a sweeping frequency. Many SSTV monitors in the past used it. I think that when there are lots of interference on short wave, the 2nd method is more insensitive to that interference.

But with clean signals, coming from .wav files, there is no reason to use this more difficult method. The first method needs only a level detector, which you feed with the demodulated video signal. This can be a single opamp, where the - input is connected to the demodulated video signal (that goes to the base of the BD115) and the + input to a potentiometer that adjusts the level voltage. I think this is the better way for you to go.

In NBTV the 2nd method was never used, because NBTV only has a video signal. It is never FM modulated.


Hi Klass Steve couldn't do to much today had to get rid of some junk around the house boring work stuff .
i was thinking there was only the radar way this wording is new to me i have not come across this before i will pay more attention to what was used in the different circuits i will review ...
I have tried to draw up something to your suggestion do you think this might work or need changing ? the op amp could be any thing really 741 tl072 so on ?
On Steve's Advice he would use a comparator i am thinking this way would be similar to what is used in the NBTV encoder circuit as in the comparator wiring ? if so either way seems very simple to get the horizontal and vertical pulses if so ...i would have to look at what comes out on the scope if its clean enough perhaps it is enough to sync the horizontal oscillator at least ...i will wait to hear what you think i know i am not great at circuit design .
Attachments
IMG_1192.JPG
IMG_1192.JPG (255.56 KiB) Viewed 11992 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: Sync detection

Postby Klaas Robers » Thu Mar 02, 2017 8:02 pm

Very good Harry. This is more or less what I wrote in words. Althoug I would have done the set circuit slightly different:

- potentiometer 10k
- bottom to "ground".
- top with 47k in series to +12V
- middle via 10k to the + input of the opamp.

You will get sync pulses "upside down", i.e. positive pulses.

It is a good idea to add a resistor of 1M from the output of the opamp directly to the + input of the opamp. This makes it a Schmidt trigger. Without the 1M it is a comparator. Works also.

I don't know the polarity of the sync pulses that your sawtooth oscillator needs. I don't know any more how the BRY39 works. Do you know Steve?
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Sync detection

Postby Harry Dalek » Thu Mar 02, 2017 8:27 pm

Klaas Robers wrote:Very good Harry. This is more or less what I wrote in words. Althoug I would have done the set circuit slightly different:

- potentiometer 10k
- bottom to "ground".
- top with 47k in series to +12V
- middle via 10k to the + input of the opamp.

You will get sync pulses "upside down", i.e. positive pulses.

It is a good idea to add a resistor of 1M from the output of the opamp directly to the + input of the opamp. This makes it a Schmidt trigger. Without the 1M it is a comparator. Works also.

I don't know the polarity of the sync pulses that your sawtooth oscillator needs. I don't know any more how the BRY39 works. Do you know Steve?


OK that sounds promising Klass if i have time tomorrow i will put it together i would use your changes rather go with some one who knows what their doing ,glad i was close but . :wink:
That would be a great step forward if i can get this done ..Thanks Klass ~! BTW last time i synced the scr oscillator i had to have a diode between the circuits .
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: Sync detection

Postby Klaas Robers » Thu Mar 02, 2017 9:10 pm

I have been reading the circuit diagram and I think that you indeed need positive (upside down) sync pulses. And indeed, you need a diode in between, because the opamp goes to -12 volt. Anode to the opamp output, cathode (ring) to the oscillator sync input.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Sync detection

Postby Harry Dalek » Fri Mar 03, 2017 2:01 pm

Klaas Robers wrote:I have been reading the circuit diagram and I think that you indeed need positive (upside down) sync pulses. And indeed, you need a diode in between, because the opamp goes to -12 volt. Anode to the opamp output, cathode (ring) to the oscillator sync input.


Hi Klass happy to report the opamp sync circuit works . :D
I ended at the moment ac coupling the sync to the sawtooth circuit i will see how the diode gos again as i tried but i also needed to adjust the trimmer on the sync circuit by then i was trying a capacitor any case it works ..i still have free running vertical so the start position is different each scan line but the line position is the same .
I did do a test card video but uploading that to youtube and post later today .
Below the sync output of the op amp

youtu.be/Z6FzeGgvAvM
Attachments
IMG_1225.JPG
IMG_1225.JPG (116.95 KiB) Viewed 11979 times
IMG_1222.JPG
IMG_1222.JPG (118.02 KiB) Viewed 11979 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: Sync detection

Postby Harry Dalek » Fri Mar 03, 2017 6:58 pm

OK Klass heres what i made from your description i think i got it right as it works i forgot to put in the led indicator in my drawing .
As i said in the last post i only have Ac coupling at the moment should look into the diode coupling , i find the 10k trimmer pretty much needs to be adjusted right as in this video it can distort the lines a touch if i am off..seems give better results on pictures might be the audio levels too...look into this more tomorrow .
Any case it syncs ...Thanks Klass


youtu.be/5ezRIYTaYTU
Attachments
IMG_1232.JPG
IMG_1232.JPG (291.32 KiB) Viewed 11970 times
IMG_1227.JPG
IMG_1227.JPG (276.5 KiB) Viewed 11970 times
Last edited by Harry Dalek on Fri Mar 03, 2017 7:11 pm, edited 1 time in total.
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: Sync detection

Postby Klaas Robers » Fri Mar 03, 2017 7:07 pm

Very good Harry!

And now you may adjust the vertical scan to about the correct timing. I see that it runs about twice the optimal speed. The circle is a usefull pattern to see that at once. Also the squares around the circle, they have to be real squares, not vertically standing rectangles. That should be done anyway before the vertical sync can work properly.

The distortion in the "top" of the picture is because the vertical sync is a longer pulse. Then the line oscillator is "held". I think that you need a diode and a resistor in series. A diode only might give this same result. The resistor about 47k.

Do you have a circuit for vertical sync separation? I just "invented" one for you:
- From the output of the sync separator opamp a resistor of 100k to a "point"
- From this point a capacitor of 150nF to ground
- The base of a PNP-transistor connected to this point.
- Collector of this transistor to -12V
- Emitter of this transistor via a resistor of 10k to +12V
- Diode from emitter to sync-input of V-oscillator.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Sync detection

Postby Harry Dalek » Fri Mar 03, 2017 7:35 pm

Klaas Robers wrote:Very good Harry!

And now you may adjust the vertical scan to about the correct timing. I see that it runs about twice the optimal speed. The circle is a usefull pattern to see that at once. Also the squares around the circle, they have to be real squares, not vertically standing rectangles. That should be done anyway before the vertical sync can work properly.


Thanks Klaas i will do that ...i am just noticing the circle i know i need to use this to adjust a correct picture .I again didn't have to much time this afternoon to play with this just enough to make it a do a test run ...But yes adjusting the vertical will be next .

he distortion in the "top" of the picture is because the vertical sync is a longer pulse. Then the line oscillator is "held". I think that you need a diode and a resistor in series. A diode only might give this same result. The resistor about 47k.


Oh ok will do.... i know i used a diode many years ago trying every way i could think of to sync the oscillator not sure if i also used a resistor but i will try this tomorrow .

Do you have a circuit for vertical sync separation?


No but i have been thinking about it

I just "invented" one for you:
- From the output of the sync separator opamp a resistor of 100k to a "point"
- From this point a capacitor of 150nF to ground
- The base of a PNP-transistor connected to this point.
- Collector of this transistor to -12V
- Emitter of this transistor via a resistor of 10k to +12V
- Diode from emitter to sync-input of V-oscillator.


OH thats a easy one but not for me to work out :idea: so that will be made tomorrow also ...Big Thanks again Klaas letting me use your brain ! :wink: not much to do after that really ...would of taken me much longer to work out a vertical sync circuit i will be keen to see every thing in sync !


[/quote]
Attachments
IMG_1237.JPG
IMG_1237.JPG (247.38 KiB) Viewed 11966 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: Sync detection

Postby Klaas Robers » Fri Mar 03, 2017 10:31 pm

You got it Harry. Good luck. But first adjust the vertical sync.

Synchronisation works in this way:
- the spot on the tube runs in a certain speed,
- you can adjust that speed.
- It is sensible to have that control on your front.

- Suppose the sync input is floating........
- When the spot is at a certain position the spot returns to the beginning of a line and starts a new line.
- This is called the "fly back".
- If you raise the voltage on the sync input the fly back starts earlier,
- so you get shorter lines.
- The sync pulse should come just before the fly back would start on its own.

- With a resistor in series with the diode you can ensure that the sync input goes just 1 volt up
- or even less.
- Then a sync pulse works only in the last cm of the line,
- so the length of the sync pulse is of no influence.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Sync detection

Postby Harry Dalek » Sat Mar 04, 2017 3:15 pm

Klaas Robers wrote:You got it Harry. Good luck. But first adjust the vertical sync.


Well Klass i know you know what your doing so ....so long as i can follow instructions and i have a bit of soldering skills i can't go wrong .
The vertical detection circuit was built today and works.. :wink: i have used the diodes now connecting both sync's to the sawtooth oscillators ,i didn't have use a resistor for the vertical with the diode but i have 10 k on the horizontal ..both seems to work so i just left it for now and tested .

Synchronisation works in this way:
- the spot on the tube runs in a certain speed,
- you can adjust that speed.
- It is sensible to have that control on your front.


So far on my testing with sync control via your circuit it seems to control every wav file i have tested so far flawlessly ! i am in two minds about a panel control for the sync but you know better than me .

- Suppose the sync input is floating........
- When the spot is at a certain position the spot returns to the beginning of a line and starts a new line.
- This is called the "fly back".
- If you raise the voltage on the sync input the fly back starts earlier,
- so you get shorter lines.
- The sync pulse should come just before the fly back would start on its own.


The vertical being so slow i can see here play with the control that was very useful getting the raster look ....your description there is very much what i am seeing on the vertical as it syncs..I was thinking the flyback line would be seen as before on the blank raster but it doesn't seem to be a problem on my tests of late .
- With a resistor in series with the diode you can ensure that the sync input goes just 1 volt up
- or even less.
- Then a sync pulse works only in the last cm of the line,
- so the length of the sync pulse is of no influence.
[/quote]
As i said the circuits work really well Klass and such small circuits as well i am very impressed !
i still have the vertical with a diode connected direct to the sawtooth oscillator and the horizontal diode and 10 k ...i didn't have to go to the point of changing things so i am not sure if i should bother as it works ?
Below testing i think it works great i am very happy .
In the picture the veritical circuit added and alligator clips white green syncing the oscillators

youtu.be/srGUJgH3P5A

youtu.be/pGV5d-i-v9w

youtu.be/GwyIugWYxWY
Attachments
IMG_1261.JPG
IMG_1261.JPG (213.22 KiB) Viewed 11957 times
IMG_1262.JPG
IMG_1262.JPG (286.21 KiB) Viewed 11957 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

Next

Return to Harry Dalek's 5AHP7A Vintage SSTV The Deep image

Who is online

Users browsing this forum: No registered users and 0 guests

cron