Flying spot scanner mark 2 the Resurrection

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

Postby Harry Dalek » Wed Aug 21, 2013 10:40 pm

gary wrote:As long as both pulse trains are identical in frequency there is no *technical* reason you shouldn't be able to achieve speed lock. If you had a half size picture it implies one or other of the pulse trains was twice the speed of the other (it is always possible to lock at harmonic intervals).


I have to have a look if i made a wave file of the test i recall the picture looked a little washed out and half picture size i was thinking it might be because i couldn't adjust the pulse width of the crystal clock ...i suppose the disk speed is still the problem sync might be cleaner but the nikpow speed was still varying .

If you have a different number of pulses between the two pulse trains then lock is unlikely to be achieved.


mmmm still have a mechanical clock if i tried to cheat using the crystals

If you had a free running clock at EXACTLY 400 Hz, and your disc was running at EXACTLY 750 RPM and you used a single frame pulse from the disc to reset the free running clock then yes, the free running clock pulses would be the same as that created from an encoder on the disc.

However the two *EXACTLY*'s are somewhat difficult to achieve methinks.


Yes still have the motor speed problem which all comes down to again the motor and disk motor control electronics how well you made it all LUCK :roll:

If on the other hand you generated a single pulse, 1 out of 32, from the free running clock and used THAT to phase lock with the disc frame pulse (which is what a single pulse from the 32nd line of the disc is) then you might be in business - I am just not sure that one pulse per revolution is enough to overcome windage, temp changes etc. But from a purely theoretical point of view it could.



Thats good to know what you think i wonder if a PLL locking on 12.5 hz
on one mechanical pulse is easier better? than a 32 mechanical pulses
......is slow easier or harder for a PLL for a bistable it wouldn't matter a pulse is a pulse so long as its there .

How about that one mechanical 12.5hz pulse triggering a bistable having a free running 400hz clock on the other then you would have 32 pulses which should match what would of been there if you had used a 32 line encoder ...i am again only thinking of cleaner electronic sync pulse .
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: 4379
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby gary » Wed Aug 21, 2013 10:45 pm

.....is slow easier or harder for a PLL for a bistable it wouldn't matter a pulse is a pulse so long as its there .


3200 best, 32 good, 1 worst possible...

and most importantly:

number of mechanical pulses MUST equal clock pulses otherwise chaos!
Perfecting an NBTV system is like trying to slam a revolving door...
User avatar
gary
"Fester! Don't do that to 'Thing'"
 
Posts: 2303
Joined: Sat Jan 27, 2007 11:29 am
Location: Bundanoon, Australia

Postby Klaas Robers » Wed Aug 21, 2013 11:36 pm

One pulse per evolution is the easiest, as you get frame lock at the same time. However the stability of the picture is worse and the locking will become rather slow: information is coming in slower.

I suggest to start with one pulse per revolution. If that works you have got feeling on how to work with it, and then you may switch to 32 pulses per revolution. then you have lost the frame sync, but you gained a much better stability.

When that works try the missing pulse to obtain frame lock again. Always go step by step.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1521
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Harry Dalek » Wed Aug 21, 2013 11:40 pm

gary wrote:
.....is slow easier or harder for a PLL for a bistable it wouldn't matter a pulse is a pulse so long as its there .


3200 best, 32 good, 1 worst possible...

and most importantly:

number of mechanical pulses MUST equal clock pulses otherwise chaos!


Sounds like a 60 liner encoder would be easier for a PLL than a 32 liner .

Chaos sounds like a lot of my work Yakes :!:

BTW off subject i find your first nbtv viewer the little one very good for the camera projects i find it easier to adjust things using that as the live viewer than big picture till i have things working better .
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: 4379
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby gary » Thu Aug 22, 2013 10:23 am

Harry Dalek wrote:Sounds like a 60 liner encoder would be easier for a PLL than a 32 liner .


I just want to make sure you understand that for speed control you don't need to match the number of apertures on the disk.

When I said "3200 best" I wasn't referring to a 3200 line disc! :shock:

That number could just have easily been 322 or 6100 - I was merely indicating that a higher number of pulses is going to be better than a lower number of pulses because the subsequent time constant (RC) of the speed controller low pass filter is shorter and, I think, more sensitive, especially if the duty cycle of the input pulse train is low (i.e. narrow pulse width).
Perfecting an NBTV system is like trying to slam a revolving door...
User avatar
gary
"Fester! Don't do that to 'Thing'"
 
Posts: 2303
Joined: Sat Jan 27, 2007 11:29 am
Location: Bundanoon, Australia

Postby Harry Dalek » Thu Aug 22, 2013 1:13 pm

gary wrote:
Harry Dalek wrote:Sounds like a 60 liner encoder would be easier for a PLL than a 32 liner .


I just want to make sure you understand that for speed control you don't need to match the number of apertures on the disk.


This i am not sure about so you can sync a disk of 32 line to a encoder if it has more slots double so on ?

When I said "3200 best" I wasn't referring to a 3200 line disc! :shock:


So if used a 120 slot encoder it would sync the thing better than a 32 line one ? has any one tried that suppose PLL controlling is faster .

That number could just have easily been 322 or 6100 - I was merely indicating that a higher number of pulses is going to be better than a lower number of pulses because the subsequent time constant (RC) of the speed controller low pass filter is shorter and, I think, more sensitive, especially if the duty cycle of the input pulse train is low (i.e. narrow pulse width).


I was thinking the reason a 32 line nipkow and a 32 slot encoder had to match for speed control because of the sync pulses had to match....

Is there a reason a higher number encoder isn't used then if it could be ?


Thanks for the advice to Klaas i can see that all these ideas have been tried and tested !
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: 4379
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby gary » Thu Aug 22, 2013 3:46 pm

Well it would speed control it better rather than sync it better. Sync implies frame or line lock. I suppose as long as your pulses line up with your apertures (or an integral multiple do) then line lock is inherent - but that is not necessary just to obtain speed lock (750 rpm). OTOH the only time I can think of where you WOULDN'T select a multiple of the disk number is if you had a precision commercial encoder that wasn't an integer mutiple. In that case you would STILL need to create your NBTV standard sync pulses somehow if you want to mix them into your outgoing video signal.

The reason it is 32 for a 32 line *monitor* is because there are 32 max* in the signal (including missing frame pulse) in an NBTV compatible video signal. For a camera it is different because you normally don't have an incoming signal to lock onto so what you want is good speed control - and note - that is often done from the mains 50 hz using a synchronous motor, so there is another number for you.

* you could have more of course but then there would be sync bars in the middle of your lines! ;-)

If you placed the sync pulses on a separate channel you could have as many as you want and no black bar at the bottom! :-)
Perfecting an NBTV system is like trying to slam a revolving door...
User avatar
gary
"Fester! Don't do that to 'Thing'"
 
Posts: 2303
Joined: Sat Jan 27, 2007 11:29 am
Location: Bundanoon, Australia

Postby Harry Dalek » Thu Aug 22, 2013 6:46 pm

gary wrote:Well it would speed control it better rather than sync it better. Sync implies frame or line lock. I suppose as long as your pulses line up with your apertures (or an integral multiple do) then line lock is inherent - but that is not necessary just to obtain speed lock (750 rpm). OTOH the only time I can think of where you WOULDN'T select a multiple of the disk number is if you had a precision commercial encoder that wasn't an integer mutiple. In that case you would STILL need to create your NBTV standard sync pulses somehow if you want to mix them into your outgoing video signal.


OH okay learn some thing new every day ! i wasn't sure that was possible ,if it does speed PLL controlling there must be a best NBTV control encoder where a to few slot encoder is the same as one with to many so one in the middle ?

The reason it is 32 for a 32 line *monitor* is because there are 32 max* in the signal (including missing frame pulse) in an NBTV compatible video signal. For a camera it is different because you normally don't have an incoming signal to lock onto so what you want is good speed control - and note - that is often done from the mains 50 hz using a synchronous motor, so there is another number for you.


I think Graham uses the mains a bit for this ....i can see i need to concentrate on speed control more than anything .

* you could have more of course but then there would be sync bars in the middle of your lines! ;-)


That would bother me :shock:

If you placed the sync pulses on a separate channel you could have as many as you want and no black bar at the bottom! :-)



How does that system work again Gary i have forgotten, lost my old mail about it ......least i can come back to this and review .i can only think you mix the wave files of the video and sync later as a mono file and then use it with you nbtv viewers ?

BTW is the Baird setting on the viewers video with no sync ?
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: 4379
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby gary » Thu Aug 22, 2013 6:56 pm

Harry Dalek wrote:How does that system work again Gary i have forgotten, lost my old mail about it ......least i can come back to this and review .i can only think you mix the wave files of the video and sync later as a mono file and then use it with you nbtv viewers ?

BTW is the Baird setting on the viewers video with no sync ?


Oh there is nothing special about that Harry - the sync can always be put on the right channel of a stereo (instead of audio). I always do that for my cameras as it is far easier than mixing the syncs in at the camera. I rarely use audio but if I did I can just record it on another sound card (I have several). To make an NBTV compatible file it is just then a matter of mixing the components together in software (much easier than in hardware).

There is another advantage for keeping the syncs separate for FSS and the like in that I then don't have to have a fancy speed control - I can get there abouts and then use my TBC software to correct it.

The Baird format has sync yes, but being the Baird format it, of course, has no frame sync. Baird format is 30 line don't forget.
Perfecting an NBTV system is like trying to slam a revolving door...
User avatar
gary
"Fester! Don't do that to 'Thing'"
 
Posts: 2303
Joined: Sat Jan 27, 2007 11:29 am
Location: Bundanoon, Australia

Postby Harry Dalek » Thu Aug 22, 2013 9:16 pm

gary wrote:Oh there is nothing special about that Harry - the sync can always be put on the right channel of a stereo (instead of audio). I always do that for my cameras as it is far easier than mixing the syncs in at the camera. I rarely use audio but if I did I can just record it on another sound card (I have several). To make an NBTV compatible file it is just then a matter of mixing the components together in software (much easier than in hardware).


So head amp to the left channel sync to the right (might be a silly question are you using the encoders signal) or lining up a software 400 hz with the video if thats possible .
So is the draw back to this no being able to see it live ?

There is another advantage for keeping the syncs separate for FSS and the like in that I then don't have to have a fancy speed control - I can get there abouts and then use my TBC software to correct it.


Oh what do use sounds like you can do the FSS with very minimum electronics .

The Baird format has sync yes, but being the Baird format it, of course, has no frame sync. Baird format is 30 line don't forget.


No worries i wasn't sure ...again of topic and info might be wrong but i read that the Baird record recordings had no sync ? so no way to view them at the time ?
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: 4379
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby gary » Thu Aug 22, 2013 9:30 pm

Harry Dalek wrote:So head amp to the left channel sync to the right (might be a silly question are you using the encoders signal) or lining up a software 400 hz with the video if thats possible .


Very definitely from the disk pulses otherwise useless.

Harry Dalek wrote:
So is the draw back to this no being able to see it live ?


Certainly not! I have software that *expects* the sync to be on the right channel - I have all sorts of things that you plebs don't... ;-)

Oh what do use sounds like you can do the FSS with very minimum electronics .


Yes, electronics can always be added later once you have proven the concept - alas for me proving the concept is often enough.

info might be wrong but i read that the Baird record recordings had no sync ? so no way to view them at the time ?


TOTALLY wrong! BUT!!!! Baird's sync was NOT a negative going pulse - it was BLACK (zero value) - therefore the scene being televised could not have pure black.

EDIT: but I must also point out that the NBTVA Baird format, whilst contemporaneous in most aspects, has a negative going sync pulse.
Perfecting an NBTV system is like trying to slam a revolving door...
User avatar
gary
"Fester! Don't do that to 'Thing'"
 
Posts: 2303
Joined: Sat Jan 27, 2007 11:29 am
Location: Bundanoon, Australia

Postby Klaas Robers » Fri Aug 23, 2013 4:57 am

For my disc monitor I used a DC motor from a video disc player. Nice and strong. It has a magnetic speed pick up, with a small magnetic ring having 18 pulses per revolution. So I had to invent something difficult to match the 18 pulses per frame with the 32 (31) from the CD signal. It took me more than one year to come to a reliable and still simple enough system.

It is easiest to start with two identical sync systems, e.g. 32 per revolution with or without one missing. Then the 'only" thing you have to do is see that the pulses come in at the same rate and at the same moments.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1521
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Harry Dalek » Fri Aug 23, 2013 12:04 pm

gary wrote:
Certainly not! I have software that *expects* the sync to be on the right channel - I have all sorts of things that you plebs don't... ;-)


Well can't blame you for that you did all the hard work.
I again use the first small nbtv viewer of yours a lot its very helpful for me .

Yes, electronics can always be added later once you have proven the concept - alas for me proving the concept is often enough.


Thats again a useful tool i,d still like making it but just to see call me crazy but i enjoy soldering.


TOTALLY wrong! BUT!!!! Baird's sync was NOT a negative going pulse - it was BLACK (zero value) - therefore the scene being televised could not have pure black.


Mmmm that was a bad site by the sounds of it ,bit like a show i watched on foxtel last week on inventions of last century they did Baird but a dreadful job of the lens Nipkow and it did erk me they made out he invented the Nipkow .

EDIT: but I must also point out that the NBTVA Baird format, whilst contemporaneous in most aspects, has a negative going sync pulse.


What did he use to make it ? have to have a look at old photos to see if theres some sort of hole encoder idea .
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: 4379
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby Harry Dalek » Fri Aug 23, 2013 12:22 pm

Klaas Robers wrote:For my disc monitor I used a DC motor from a video disc player. Nice and strong. It has a magnetic speed pick up, with a small magnetic ring having 18 pulses per revolution. So I had to invent something difficult to match the 18 pulses per frame with the 32 (31) from the CD signal. It took me more than one year to come to a reliable and still simple enough system.


Yes i to used one ,that one on the drum monitor but never used the Ac pulse i have another in my shed i might scrap to use in the future and i have 2 other small ones that do the same thing ...pulses per turn i need to test if i ever use them .
Did you make use of the large neon laser ?

It is easiest to start with two identical sync systems, e.g. 32 per revolution with or without one missing. Then the 'only" thing you have to do is see that the pulses come in at the same rate and at the same moments.


Thats true thats easiest even for me to work out!
I am going to take Garys advice and buy the laser printer for the next encoder (my wife had trouble with the new printer last night ended up she hadn't had the usb plugged in but gave me an excuse to mention the laser printer idea which she liked :wink: ) cheaper than buying the ink cartridges for this HP/
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: 4379
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Postby gary » Fri Aug 23, 2013 12:40 pm

Harry Dalek wrote:Well can't blame you for that you did all the hard work.


Oh don't get the impression that I am holding them back from general use - It's just that they are custom jobs for a specific project that are unlikely to be generally useful to others. Also they tend to be "lash up" jobs that aren't particularly aesthetically pleasing.

Harry Dalek wrote:Mmmm that was a bad site by the sounds of it.


Could you post or send me a link to that site please?

They may have just been talking about the fact that a phonic wheel can be made to work with just the video signal because there is a strong peak in the video signal at the frame x lines frequency.

I think that Baird realised fairly quickly that that peak could be made more reliable by having a black bar at the beginning of each line.

Harry Dalek wrote:What did he use to make it ? have to have a look at old photos to see if theres some sort of hole encoder idea .


I think he just placed a mask across the bottom of the camera window - there are some examples of NBTVA constructors doing that in the old newsletters.
Perfecting an NBTV system is like trying to slam a revolving door...
User avatar
gary
"Fester! Don't do that to 'Thing'"
 
Posts: 2303
Joined: Sat Jan 27, 2007 11:29 am
Location: Bundanoon, Australia

PreviousNext

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 6 guests

cron