Narrow-Bandwidth Television Association

Hey, been lurking around a for a little while, made an account and continued to lurk a little more...until now!

So here we go for a first post, a question!


I have been working on my own little televisor projekt. It's nothing too complicated, just something to play around with really.

Everything on it seems to work great except I have an issue with the picture on the disk coming out split up for some reason.

Here are some pictures to better explain this problem...

#1: I'll get the picture to hold in place for the most part. I don't have any special stuff to keep it from rolling up and down other than my variable power supply and a finger on the disk, but it works well enough for now.



#2: But then, it'll split the picture;

Sometimes it'll split it like this:
<--(this is the left side of the smiley appearing on the right side of the view area)
Or like this:
<--(same as above)

And also a few variation in between. The image just cuts into 2 portions, with the leftmost part of the main picture appearing over on the right side.

Sometimes the picture will even start out split.

It does this with both Static and moving images.

I'm using the video2nbtv program I found here to make the pictures: http://users.tpg.com.au/users/gmillard/nbtv/nbtv.htm

Is it just my lack of sync stuff that's doing this, is my homemade nipkow disk just made of fail, or is it something totally different all together?

Any help would be appreciated...

Firstly, welcome to this forum

Fear not, there is nothing wrong with your monitor and what you describe is completely normal.

To get a correctly framed picture, i.e. not split, you have to brake the disk a little with your finger so that the images start rolling. As the pictures roll, the split point will shift one line each time. Eventually, a picture will come around that is correctly framed at which point you stop braking the disk.

It's difficult at first but eventually you learn the skill of 'syncing your disk'. It is something that people in the mechanical television era had to learn, so you are reliving history by doing this!

Using your finger all the while might result in a blister. A common alternative for braking the disk was to secure the end of a length of string on one side of the monitor and to loop it around the motor shaft. The other end is held in your hand and pulled so that it tightens around the motor shaft and slows it down slightly.

It is nice to see someone enjoying this fascinating hobby!

Sadly imageshack is blocked by my ISP/government here due to the amount of pornographic content it hosts. So can't see the pictures. Attached is what I get instead....registering makes no difference...

Steve A.

First: Welcome to the forum.

As Kareno said: http://youtu.be/8CtjhWhw2I8

Maybe a solution: For the frame-sync the 4046 "is looking for" the missing sync-hole at the first line aperture. Sometimes it helps if you close the first two or three sync-holes to get a faster automatic frame-lock.

Steve Anderson wrote:Sadly imageshack is blocked by my ISP/government here due to the amount of pornographic content it hosts. So can't see the pictures.
Steve A.

Couldn't let you miss out entirely...

Steve Anderson wrote:Sadly imageshack is blocked by my ISP/government here due to the amount of pornographic content it hosts. So can't see the pictures. Attached is what I get instead....registering makes no difference...

Steve A.

Steve, try accessing the page through an anonymiser service. Those sorts of services serve you a page that *they* retrieve from the URL that you give. So you basically bypass any country internet filtering. Of course, you have to find an anonymiser service they don't filter.

Cheers
A

The dreaded frame sync problem. I've done a fair bit of head-scratching over the years regarding this. Is it a detection problem? A motor/disc problem? Or something else? Or a combination of the above?

From my perspective I've always viewed the 'missing-sync' arrangement somewhat lacking, I'm aware of its origins, but what to replace it with? And/or derive a compatible waveform. I nominally make my syncs 125us, 5% of line-time. With the bandwidth restrictions imposed by using audio medium there's not a lot one can do in 125us.

Tone bursts are fine, but you'll have to kiss a whole line goodbye (or the best part thereof) as I did on the 48 and 72-line displays in the newsletter and somewhere else here on this forum.

With virtually all audio gear these days being stereo a high-frequency burst mixed into the audio channel is a possible, most of us old farts wouldn't hear it anyway (say 16kHz). But I feel that's cheating. Anyway a notch filter would remove it for the teenagers...erm...OK, next paragraph...

With the dynamic range afforded by the audio CD and other digital media a 'blacker-than-line-syncs' frame reference is entirely possible...but it's not compatible with existing equipment. Alternatively a 'whiter than white' pulse is compatible thereby retaining the 'missing pulse' compatibility. This has been mooted before but I'm not aware of the reaction to this (very informal) proposal.

The problem I see with a 'whiter-than-white' pulse is it could show up in the bottom-right and/or the top-left of the display as an artifact.

Your thoughts?

Steve A.

Metallica Man X wrote:

I have been working on my own little televisor projekt. It's nothing too complicated, just something to play around with really.

Everything on it seems to work great except I have an issue with the picture on the disk coming out split up for some reason.
...

As has been said all is OK with your little televisor.
Some have mentioned the 4046.
Don't know if you are using this. I get the impression from your letter that you are not.

So, if not, your next step might be to try to get some form of electronic sync. to lock you picture.


If you look at the NBTVA site you will see a circuit for syncing your motor.
This generally works for most, but is a bit temperamental to get going.

http://www.nbtv.org/

Goto "handbook" link.
Click 'here to enter' and then 'contents' and look at No4 link,"Building a complete NBTV televisor."
Just over half way down that page see, "Automatic speed control circuit."
which uses the 4046 refered to by others.

To use this circuit you would need to have 32 sync holes on your disc, with one blanked off. You could, depending on what you already have, put in a seperate disc with the sync holes in it.

If your Nipkow disc is very heavy or very light the 4046 circuit might need jiggling a bit, as the mechanics of the system and the motor chosen, play a part in its correct operation, most of us have found.

Welcome Metallica man x

Steve Anderson wrote:
The problem I see with a 'whiter-than-white' pulse is it could show up in the bottom-right and/or the top-left of the display as an artifact.

Your thoughts?

Steve A.

I have the problem on the MUTR. Not all the time.
I thought the black level was too close to the blacker-than-black.
So I read the NBTVA standard, and notice the thing about the missing pulse on the first line.
I saw videos on youtube and notice that, at the place of the missing pulse, there is a bit of the picture ( 3 pixels )
So I thought i had to put there something far from black, so white.
And i created the "synch gard" that i put on my videos before transfer them to WAV

edit : i there a way to transmit things like ceefax or something during the synch pulses ?

I have submitted an article on a related matter for the newsletter. It isn't automatic frame sync, but it could be the basis of it. I am at this moment working out how to achieve reliable automatic frame sync and more

Lowtone wrote:edit : i there a way to transmit things like ceefax or something during the synch pulses ?

well the sync pulse lasts for aeons in terms of video frequencies, you could easily just modulate a standard teletext signal at a low amplitude onto all the sync pulses...

Would rather destroy the notion of narrow bandwidth tv though... hmm

I don't think there's anything wrong with a missing pulse sync mechanism, the problem that we have with the nbtva scheme is that you can't detect a missing pulse until... you've missed it!

If rather than missing the sync pulse for line 1 you remove the pulse from line 32 then it is easy to generate a frame sync.
(obviously if you are doing a digitally derived scanning waveform like mine it is easy enough to use a missing line 1 pulse and compensate for this problem, just count the lines as [frame sync]-2-3-4-...-31-32-1[frame sync]. I did this on my cro driver and it is just a case of resetting my staircase generator to binary 1 rather than binary 0 when the frame pulse arrives)

It's nice to see that this problem plagues everyone...and I'm not totally made of fail

I'll have to check out that 4046 circuit and be ready to drill a few more holes in my disk.

...I just wonder if that circuit will work with that video2nbtv program....

For NBTV "Ceefax" (actually it would be a kind of "Videodat" which was used by some television stations during the late 1980s/early 1990s for transmitting computer software in "Basicode 2" we could use a second line, which gives us 32 Bits (or 4 Byte) per frame for information. Per second we could transmit about 54 letters, which is not even enough for subtitles, but would be enough for a callsign and some information about the transmitter. It could indeed be an interesting item for shortwave broadcasts.

Metallica Man X wrote:...I just wonder if that circuit will work with that video2nbtv program....

It will work for a given value of "work"

The video2nbtv program creates an nbtva spec signal
i.e. with a "too late to do anything about it" missing pulse



Further to what I was saying about resetting a counter to line 2 rather than line 1 in a digitally scanned system, on a mechanical monitor you can do the same thing by having your scanning holes rotated 1/32th from the point that you are synchronising the disk to the derived frame sync pulse.