Flat panel NBTV

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

Flat panel NBTV

Postby Andrew Davie » Thu Jun 15, 2017 12:46 pm

Been mulling this concept for a while, and thought I'd try to describe it. It's an idea for an arbitrary resolution/size mechanical flat-panel display.
Consider a band of material on pulleys, such that a flat face is presented to the viewer, and the material is moving (let's say vertically) with respect to that face - by means of the pulleys rotating. Now put a second band of material behind the first, with pulleys at right angles. Now we have two "flat faces" one behind the other, moving at right angles. Put a light source behind the material.

Now let's consider how to put holes in the material to allow a "scanning action" of the view area. Should work with arbitrary # scanlines, but for argument's sake, 32 lines scanned vertically from bottom right. First, put a hole in the top material layer at the bottom right - that is, start of the first scanline. The movement of the material will "scan" this upward but we won't see any light because the second layer of material is blocking. When the hole gets to the top, we want another hole to start scanning the next scanline, so put a hole at that position - that is, with the first scanline hole having just disappeared off the top, put a hole in the second scanline position at the bottom. And again, for the third scanline. At this point we have three equally spaced holes in the "band", separated by one scanline each. Let's assume that band is just three scanlines long - so now as the third hole is disappearing off the top of frame, the very first hole is now appearing bottom right at the first scanline position. That's OK - we can't see ANYTHING yet because of the second band blocking all the light. But now, put a new hole in the fourth scanline position at the bottom. Equal in position with the first scanline. And the fifth scanline hole will be equal to the second... etc. So now we have multiple holes at any time scanning the "frame" and a reasonably short band - just three times the frame height in its total length. Moving fairly rapidly, I grant - but let's assume that we can find a material that can do that safely.

Now let's move on to the second band of material (behind the first) that is currently blocking all light. As the frame scanning starts, we want to see the first scanline hole (at bottom right). So let's put a hole in the second band to make that hole a real "double hole" through both materials. As the first band is moved to scan the first band hole upwards, the second band is also moving leftwards - so the hole we had is now offset and we can't see anything anymore. So put another hole under the first band hole so we can see the "double hole" again. Repeat. We end up with a diagonal stripe like a "/" in the second band. As the second band shifts leftward, and the first band shifts upward, the "double hole" is kept in synch so that we see the light through the first band hole that's scanning and also through the second band because the "/" is effectively tracking that first hole. We won't see any other of the pixel/holes because of two reasons. Firstly, all holes are on the same horizontal line (one of two - the odd and even scanlines). Secondly, the diagonal masking "hole" in the second band is moving in synchronisation with the hole position for only the scanline currently being displayed. It is vertically offset from other scanline holes, so there will be no "seeing through".

Now consider after the first scanline is finished. As stated, the first line hole disappears off top of frame and we immediately see the second scanline hole at bottom of frame. We again put a "hole" in the second band at this point. And as the second band shifts left and the first band shifts up, we keep adding holes. We get another diagonal "hole" or stripe in the second band which again tracks the vertcial position of the first band hole. We don't see any other pixels, as the diagonal stripe is not overlapping with any other diagonal stripes, so there's only the possibility of "one masking stripe" per scanline. We need each new diagonal stripe to be offset leftwards by one frame and one scanline, so that it scans correctly. If the second band was 32 "diagonal stripes" then it would, I think, scan a frame correctly.

Consider a "diagonal stripe" which is effectively a mask to show/hide holes in the first layer/belt. When the first belt hole is at bottom, we also want the second belt stripe to have a hole in the same position. When both "get to the top" of frame we want the next hole to be one scanline left (which it is), and we also want the next stripe to be one scanline left. So the "next stripe" is actually offset in position just one scanline left from the end of the previous stripe. I suspect/believe that it will be possible to have the diagonal stripes fairly close together so that the secondary belt does not need to be very long, either. I have to work a little bit on the geometry/concept to confirm. But essentially the top layer has 32 scanlines, the bottom has 32 masking bands.

Well, that's the basic concept. I guess the belts will be whizzing by pretty quickly. If we're scanning a scanline that's 15cm long, then that happens 32 times per frame, 12.5 frames per seconds so it's 4 m/s. Fast, but not unthinkable. The secondary belt speed would depend on the slope of the masking hole/channel which in turn would depend on the aspect ratio and the slope of the mask. Gonna come back to that.

Well, that's the basic concept. Just thought I'd jot it down and see if anyone cares to comment on what I've overlooked.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1511
Joined: Wed Jan 24, 2007 4:42 pm
Location: Hobart, Tasmania, Australia

Re: Flat panel NBTV

Postby Klaas Robers » Thu Jun 15, 2017 6:09 pm

Although your idea is more complicated, it resemples more or less to the old idea of a (35mm) film loop which runs over rollers and has black frames, which have one small transparent square (hole). I have seen a nice drawing of this idea, but can't find that picture back.

Of course it worked, but was rather noisy due to the rollers and the fast speed of the film. More difficult was that the film wears out very fast. Within one hour the film loop had to be replaced.

As a principle this looked close to a drum monitor, which had not the problem of the noisy rollers and the heavy wear.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1465
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Flat panel NBTV

Postby gary » Fri Jun 16, 2017 11:11 am

Klaas Robers wrote:Although your idea is more complicated, it resemples more or less to the old idea of a (35mm) film loop which runs over rollers and has black frames, which have one small transparent square (hole). I have seen a nice drawing of this idea, but can't find that picture back.

Of course it worked, but was rather noisy due to the rollers and the fast speed of the film. More difficult was that the film wears out very fast. Within one hour the film loop had to be replaced.

As a principle this looked close to a drum monitor, which had not the problem of the noisy rollers and the heavy wear.


Yes it seems to be a combination of that, and the variable scanning systems (2 disks or 2 cylinders, the latter due to Haskell 1931): details in Newsletter 9 volume 3.

BAND SCANNING.jpg
BAND SCANNING.jpg (15.89 KiB) Viewed 2834 times
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

Re: Flat panel NBTV

Postby Klaas Robers » Fri Jun 16, 2017 6:58 pm

Yes, very good Gary! That is indeed the drawing that I remembered. And I remembered the disappoining story behind this idea.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1465
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Flat panel NBTV

Postby Harry Dalek » Sun Jun 18, 2017 9:20 pm

Hi Andrew i love the different ideas of mechanical scanning and have tried a few over the years i am all for those to try some thing different .
I have to say i did very much enjoy the 2 2 drum projects i had a go at here on the forum ,the line slit drums give you a small CRT size picture and a multi system display also .
Any thing above 32 line i would use pulley system to increase speed for the higher line rates.
The main drum is critical to line speed as with a nipkow but the framing drum is not so as far as seeing a picture only in correct frame rate and aspect ratio .
So this idea looks flat to the viewer but not true flat as your wanting.
I had a look at 35mm film once but its not light proof the thicker colored paper i used in the Drums is so may be making a strip out of this stuff would fix the light proof problem ,if you could keep it on a set of rollers with guides perhaps less wear and do the higher speeds .
With your 3d printing skills you could make some thing that would look pretty neat .
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: 4117
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia


Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 4 guests

cron