Retro SSTV Anyone? II

Where it all started as far as most are concerned and saw heavy use from the 60s through to the 80s. Colour and Hi-res modes have unfortunately pushed this system into the backwaters of SSTV. Time to resurrect interest in this simple analogue system.

Re: Retro SSTV Anyone? II

Postby Steve Anderson » Fri May 12, 2017 11:00 am

You've looked at it closer than I and with it's condition and the apparent mods done to it the more I am not interested. Shipping to Europe...twice the price of the unit itself. As for not shipping here, you (more correctly I) see it a lot here on e-bay...one of those things.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3642
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Retro SSTV Anyone? II

Postby Klaas Robers » Sat May 13, 2017 6:38 am

It looks as if the SSTV input chain is working correctly. I connected my tone generator to the SSTV input and I see that when the tone is slowly varied between 1500 and 2300 Hz that the information in the memory goes from very dark grey (= black?) in small steps to white. I haven't counted the steps and because of the missing 6 RAM chips it is not easy to observe, but it is not far from 16 steps. So the A to D converter looks to work correctly.

The input starts to show noise for signals als low as 2 mV, that is also not bad.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1342
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Retro SSTV Anyone? II

Postby Klaas Robers » Mon May 29, 2017 12:40 am

Yesterday a small box arrived from Turkey, with the 8 RAM ICs. Today I had time to place them and they worked. After carefully adjusting the contrast and the brightness of the demodulated SSTV video I can show these two screen shots:

Scherm4b.JPG
Scherm4b.JPG (29.44 KiB) Viewed 705 times

This is the ramp signal coming from the generator built inside the "Heathkit" SSTV monitor. If you count carefully you might see 16 grey steps, that is black, white and 14 shades of grey. That is the maximum which the Robot 400 can display (4 biits). At the edges you see the hesitating write action into the memory. The dotting is still switched off. I think that is added to mask the hesitating behaviour.

Scherm5b.JPG
Scherm5b.JPG (70.82 KiB) Viewed 705 times

This is the SSTV test picture. Also the Robot has some problems with the sync detection.

The grey scale at the bottom is the remains of the full height grey scale, that can be flashed digitally into the picture memory by the rotation of a knob. The memory is 128 lines high, but when writing SSTV the vertical addressing jumps back on the V-sync, after 120 lines. If no V-sync is received it continues writing until line 128 and then starts again at line1 (top).
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1342
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Retro SSTV Anyone? II

Postby Steve Anderson » Mon May 29, 2017 2:58 pm

That's interesting as the updated Robot 70 demodulator is very good and stable on syncs, and it's the same as that in the 400 (but without the update). I would hope that they wouldn't have considered the display you're getting as acceptable as I presume the input signal is noise-free. Maybe something downstream isn't quite as good as it should be.

There is a clue as it seems to be sync-width modulated, if possible it would be worthwhile generating your test-card with a short front and back porch, this would point to if that's a factor or not.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3642
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Retro SSTV Anyone? II

Postby Klaas Robers » Mon May 29, 2017 5:47 pm

That is correct. The point is that the black and white blocks at the borders of the picture are there on purpose. This reveals immediately these problems with direct, hard sync. I will not adapt the test picture, as this is to check this behaviour of a monitor / display. A well designed form of flywheel sync mightl also repair this.

But you also can see that there is still something visible of the 1200 Hz resolution lines. Not bad.

On the other hand, if you look in the central horizontal bar, you see the ringing on the vertical white lines. There is also a black overshoot, which is almost invisible, because it is clipped away in the black, but if I raise the brightness of the FS-monitor, it becomes obvious. We also observed this on the output wave forms of your SSTV modulator. It is impossible to prevent this to happen.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1342
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Retro SSTV Anyone? II

Postby Steve Anderson » Mon May 29, 2017 6:12 pm

Klaas Robers wrote:We also observed this on the output wave forms of your SSTV modulator. It is impossible to prevent this to happen.

I assume you mean demodulator, if so, yes the output filter being a Butterworth one inevitably introduces over and undershoot. A bessel filter would be far better in this respect but it wouldn't suppress the remnants of the sub-carrier enough. Hence my interest in digital filtering, but that's a way down the line.

However, the modulator that produces the files I currently generate from a .tiff file also has a filter on the input of the A-D. I've been thinking this over for a while now. There's a whole D-A, filter, than A-D process that I need to bypass. It's just finding the time with all the other stuff going on.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3642
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Retro SSTV Anyone? II

Postby Andrew Davie » Mon May 29, 2017 6:19 pm

Steve Anderson wrote:
Klaas Robers wrote:We also observed this on the output wave forms of your SSTV modulator. It is impossible to prevent this to happen.

I assume you mean demodulator, if so, yes the output filter being a Butterworth one inevitably introduces over and undershoot. A bessel filter would be far better in this respect but it wouldn't suppress the remnants of the sub-carrier enough. Hence my interest in digital filtering, but that's a way down the line.

However, the modulator that produces the files I currently generate from a .tiff file also has a filter on the input of the A-D. I've been thinking this over for a while now. There's a whole D-A, filter, than A-D process that I need to bypass. It's just finding the time with all the other stuff going on.

Steve A.



I have a possibly related question in this post. It's about digital resampling/filtering and I thought I would put a link here because of what I think is similar subject matter...
User avatar
Andrew Davie
Frankenstein was my uncle.
 
Posts: 1379
Joined: Wed Jan 24, 2007 4:42 pm
Location: Hobart, Tasmania, Australia

Re: Retro SSTV Anyone? II

Postby Klaas Robers » Mon May 29, 2017 8:16 pm

Yes, of course I meant "demodulator". For video a Butterworth filter is the best compromise between filter steepness and ringing. This it too for digital filters, also there one needs to define a steepness of the cut off.

However the "ringing" can be spread over before and after the edge in time. An analogue minimum phase filter (only one path for the signal from input to output, giving the lowest number of inductors and capacitors) never can do this, so the "ringing" is always following the edge. The ringing can be shifted in time by an All Pass filter. In fact this filter delays the edge, so some ringing will occur before the edge. I did this in my passive butterworth filter in the "Heathkit"-style SSTV monitor, but an All Pass can also be made using op-amps and capacitors - resistors. Robot didn't do that. I don't know the topology of active all pass filters, nor how to optimize them. An All Pass halfs the ringing amplitude (50% is before, 50% is after) and also the number of rings, if you only counts from the center, the edge itself. Symmetrical ringing gives for video a sensation of extra sharpness.

Another way is making a FIR-filter. If you make a symmetrical FIR-filter the ringing is always symmetrically divided between before and after the edge. In digital audio the ringing of the filter following the DAC is very strong and long. I have seen this on the oscilloscope. But those low pass filters are extremely sharp, which results in more ringing.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1342
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Retro SSTV Anyone? II

Postby Steve Anderson » Tue May 30, 2017 12:41 am

Of course to fully test a demodulator you ideally need a perfect modulator. In this situation I have more faith in the modulator than the demodulator - but not 100%. A chicken and egg situation. I have mused on ways to get around this as mentioned before, it's a case of having the time. I have a queue of things, mostly 'real work' things to attend to.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3642
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Retro SSTV Anyone? II

Postby Steve Anderson » Wed May 31, 2017 12:55 pm

Andrew Davie wrote:I have a possibly related question in this post.

Well I'm not re-sampling i.e. changing the sample-rate, just a straight-forward 10-bit A-D. The over and undershoot (ringing though heavily damped) is the result of using an analogue Butterworth filter at the output of the demodulator. There are better analogue filters in this respect but they're worse in other areas. There's no such thing as a perfect analogue filter.

Digital filtering can often be better than any analogue filter, though they do have short-falls too. Investigating this is on my 'to do' list.

I read the posting you referred to this morning and Gary's response is better than I could have posted.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3642
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Retro SSTV Anyone? II

Postby Steve Anderson » Wed May 31, 2017 2:00 pm

Just to clarify here's how the whole system fits together in block form. It's more complex than it needs to be and simplifying it is also on my 'to do' list.

First off I want to eliminate the two blocks in red, that is the DDS modulator is moved into the TIFF-to-baseband processor which eliminates one filter. It's a Bessel, so that isn't the cause of my over/undershoot. It's a simplification exercise.

Next to go is the blue block. I've already made a start on this as noted in the original up-converter thread. The video signal proper was improved, better resolution and far less over/undershoot. The remaining issue is its currently poor sync performance, more to come...one day.

There aren't two laptops, it's one and the same.

Steve A.

A footnote:- No, I'm not that clever to do USB on a PIC, and that particular chip doesn't have a USB module anyway, though quite a few do. I cheated and used one of those FTDI USB-to-TTL level RS232 cables. See below. They're good up to 3Mb/s - that's more than enough for this.
Attachments
SSTV-625 MkIII Peripherals-Model.gif
SSTV-625 MkIII Peripherals-Model.gif (20.35 KiB) Viewed 671 times
TTL-232R1.jpg
TTL-232R1.jpg (8.13 KiB) Viewed 671 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3642
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Retro SSTV Anyone? II

Postby Klaas Robers » Sat Aug 12, 2017 6:27 am

I became into contact with a radio-ham John Magliacane (USA), who has a website about his "three PIC SSTV scanconverter". I sent him the .wav file of my "testbeeld" and asked him to show me how well his converter works. He promised me to do that in this weekend. This might be a good "land mark".

His website: http://www.qsl.net/kd2bd/TriplePIC.html shows for me remarkable results.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1342
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Retro SSTV Anyone? II

Postby Klaas Robers » Sat Dec 02, 2017 5:44 am

Due to the summer holidays this stopped more or less. However I really got pictures of KD2BD, John Magliacane.
First is here a computer picture "taken" at the output of his 3-PIC circuit:

Testbeeld3.png
Testbeeld3.png (52.88 KiB) Viewed 67 times

You see that his circuit looks to have the well known problems with the horizontal sync stability. But it is nicely demodulated. If you look in close up, you will see the normal ringing of the white to black and black to white transients.

And then he sent me too a picture of the video output connected to a small 60 Hz television set:

KD2BD_3PIC.jpg
KD2BD_3PIC.jpg (47.51 KiB) Viewed 67 times

Then we see that the black-white bar at the top is gone, as well as the first half of the blocked grey bar. But this is a problem that cannot be simply solved for NTSC-standard monitors and TVs. But to my surprise the horizontal stability in the sync is better than in the computer picture on the top. John also could not explain that. The 1200Hz bar is almost 100% modulated. Clear, his video low pass goes to 1200 Hz, see his website.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1342
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Retro SSTV Anyone? II

Postby Harry Dalek » Sat Dec 02, 2017 5:33 pm

Thanks for posting Klass don't want SSTV to be left behind ! ...the image he's converter does is pretty good and yes does look much better on the television monitor . makes me want to make Steve's when we can get around to it .
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: 3539
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Retro SSTV Anyone? II

Postby Steve Anderson » Sat Dec 02, 2017 5:54 pm

Harry Dalek wrote:...makes me want to make Steve's when we can get around to it .

...should be a matter of days when I can resume work on the NBTV/SSTV-to-625 up-converter. First I have to sort out my office/workshop, clear out all the 'stuff' (read junk) cluttering the place up, get rid of all the rubbish and transport the office/workshop to the new location (earlier than thought) and get on with it.

Generator installed, works fine but only under manual control at the moment.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3642
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

PreviousNext

Return to Cop McDonald's FM SSTV System

Who is online

Users browsing this forum: No registered users and 1 guest

cron