Mechanical Colour

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Mechanical Colour

Postby Andrew Davie » Thu Jan 25, 2007 12:23 am

Some thoughts on colour...

In the same way that we achieved colour on the video game machine by multiplexing individual scan lines RGB, I wonder if a Nipkow disk with coloured filters over the scan-holes could achieve the same effect? That is, three sets of holes for each disk (yes, the disk would have to be huge, but then maybe it could be a belt instead of a disc) that would scan red for one scanline, green for the next, blue for the next, etc.. and once that has completed the next ‘set’ would scan green first for the first line, and the next set blue. A good colour picture is achievable by ‘interlacing’ and using just two different ‘interleaved’ images, and I thought one could even achieve ‘interlacing’ by slightly offsetting the position of the scanlines on an image. On the game machine, we have a 60Hz frame rate, and we display colour at 1/3 of this (ie: 20Hz). We see ‘shimmer’ but not flicker, and if we use interlaced 2-frame colour, we don’t even see shimmer, as the 30Hz rate is sufficient to fool the eye.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Andrew Davie » Thu Jan 25, 2007 12:24 am

On reflection, a standard Nipkow disk could be used, with a secondary ‘colour filter’ disk spinning behind the Nipkow but in front of the light source (which should be white, of course). The colour filter disk would colour successive scan line ‘holes’ red, green, blue, red, green, blue – and be spinning such that for every revolution of the colour disk, the Nipkow disk would spin three times. Thus the colour filter disk would have three sections – one colouring individual Nipkow holes in the order red/green/blue, the next green/blue/red, and the next blue/red/green. It would consist of very finely spaced colour filters, in three distinct pie-sections. The spacing, of course, the same as that of the apertures in the Nipkow disk.
To be clear, consider the first hole in the Nipkow disk. As the hole scans the image area, the colour filter is red. As the second hole scans the area its colour filter is green. As the third hole… blue. 4th hole… red. 5th hole, green… etc. Now the next time we see the first hole (a complete rotation of the Nipkow disk) the colour filter for the first hole is showing green, and the 2nd showing blue… etc. The third rotation of the Nipkow disk, the first hole colour filter is blue.. etc. So after every three rotations of Nipkow disk, we have scanned red/green/blue for the first hole, green/blue/red for the second, blue/red/green for the third… and the net effect of this (scanning with three colour filters) is that we see a greyscale image (ie: normal NBTV). If, however, we have also changed the actual video data so that for the first frame for the first hole we’re showing the red components, and for the second hole the green components, and for the third hole the blue (4th red, 5th green, 6th blue etc), and then for the second frame for the first hole we’re showing green, for the second hole blue, for the third hole red components, etc… then we get a full-colour image using a system I called ‘interleaved chronocolour’ on the game system.
The above mechanical system would be a reasonable analogue equivalent of what we did on the video game system to get colour display on a B&W system, and my experience is that this does in fact provide quite good colour reproduction. The nice thing is that you don’t have to increase the frame rate of the display signal – you just mix the red/green/blue components on a per-scanline basis. It still gives reasonable black and white images, because the colour sub-components are mixed in a ‘rolling’ fashion and alternate on a three-frame basis.
A mechanical TV would be much better than the game system, in that the game system only had single intensity (on or off) red/green/blue, whereas the mechanical system would have the graduated intensity available to the normal mono display.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Klaas Robers » Fri Jan 26, 2007 12:38 am

Oh Andrew, Nipkow disc colour is done so much simpler. You just make a Black and White Nipkow monitor, but in stead of making an LED cluster of all the same LEDs, Amber or White, you mount a mix of Red Green and Blue LEDs or mount so called RGB LEDs. The R-chain, G-chain and B-chain are driven by a Red, Green and Blue signal simultaneously. So the LED-cluster is giving the right coloured light already and the disc is doing the scanning.

A sequential system that you suggest is not feasible with a scanning frequency of 12,5 Hz. This is already to the bottom, as with monochrome the picture flicker is enormeous. We cannot bear a colour flicker of 4 Hz on top of that.

Besides this it is almost impossible to extract red light from the white LEDs. White LEDs are blue LEDs with a dot of yellow fluorescent material on top of them. So they give a mixture of yellow light and blue light, which our eyes see as a kind of white. There is NO red light and green light available.

You could make white light as well by using Red, Green and Blue LED,s driven by the same driving signal and mixing the light optically in the LED-box. Then you can even define the colour temperature as you like it. But then you are very far on the way of driving the chains independently.
Last edited by Klaas Robers on Fri Jan 26, 2007 2:37 am, edited 4 times in total.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Andrew Davie » Fri Jan 26, 2007 12:52 am

Klaas Robers wrote:Oh Andrew, Nipkow disc colour is done much easier. You just make a Black and White Nipkow monitor, but in stead of making e LED cluster of all the same LEDs, Amber or White, you mount a mix of Red Green and Blue LEDs or mount so called RGB LEDs. The R-chain, G-chain and B-chain are driven by a Red, Green and Blue signal simultaneously. So the LED-cluster is giving the right coloured light already and the disc is doing the scanning.

A sequential system that you suggest is not feasible with a scanning frequency of 12,5 Hz. This is already to the bottom, as with monochrome the picture flicker is enormeous. We cannot bear a colour flicker of 4 Hz on top of that.


Gotcha. How very elegant!

I had assumed that it was a frame of red, a frame of green, a frame of blue. My system interlaces those frames which does dramatically reduce the flicker in colour (it is not frame sequential, but line sequential). But you're right, 4Hz is not workable... and I just missed that very simple colour system that you point out.

So, why aren't there more colour monitors around, I wonder???

Cheers
A
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Klaas Robers » Fri Jan 26, 2007 12:58 am

This idea is not so old, I guess that Vic Brown showed it in 2004 for the first time. In the mean time I saw a few more at the Conventions in the UK. I still have to make one myself or change my existing Black & Amber monitor to colour.

Now we have an idea on how to store colour signals on CD it becomes more attractive.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Improved Baird Sequential Detail Scanning System for Colour

Postby Stephen » Sun Feb 04, 2007 3:25 am

I have a proposal for a compatible monochrome/colour scheme that requires no special electronic circuitry and little additional bandwidth. I submitted it to the NBTVA last year for the newsletter, but since it has not yet appeared in print, I have posted it at http://www.taswegian.com/NBTV/images/SDS.pdf . It is based upon John Logie Baird's sequential detail scanning system as shown in British Patent GB391924A.pdf filed 2 February 1932. I have posted a copy of this patent at http://www.taswegian.com/NBTV/images/GB391924A.pdf .
Last edited by Stephen on Fri Feb 09, 2007 10:44 am, edited 1 time in total.
Stephen
User avatar
Stephen
Anyone have a spare straightjacket?
 
Posts: 427
Joined: Fri Jan 26, 2007 7:00 am

Postby Guest » Mon Feb 05, 2007 8:56 am

Colour NBTV has been of interest to me for some time. I thought a potted history might be of interest. I joined the NBTVA around 1991 and fairly soon afterwards exhibited a large three-colour monitor and a separate camera at their convention. The 45-line monitor used three small RGB filtered fluorescent tubes fed from EL34s. This first attempt at colour wasn't successful though, the camera proved hopelessly insensitive and all one ever saw on the 'screen' was coloured noise. You can see this early rig three pictures down the page here: http://www.radiocraft.co.uk/nbtv1.htm

My next attempt was more successful and was the first working demonstration of colour of some sort at an NBTVA convention. This took place in the mid/late 'nineties (I forget the exact year) and consisted of a 30-line camera-monitor using a pierced LP record as the Nipkow disc. Noddy, rotating on a turntable, was televised in two colours (red and blue) but reproduced in three - the green being produced from a judicious mixture of red and blue! The exact mix was determined by a "tint" control. This was remarkably effective.

Following on from this I built the large colour monitor with 20-inch disc which remains the mainstay of my experiments today. To start with, this displayed patterns and pictures (in full colour) originating from a triple EPROM generator designed by the late Grant Dixon and was demonstrated at an NBTVA convention around the turn of the century.

A little later it was joined by my matching mechanical colour camera. This started as a two-colour unit, just like my earlier camera/monitor, with the results displayed in three-colour via my monitor. However over the past few years, thanks to the suggestions, encouragement and resources of other members I've been able to further enhanced this camera. Today both the camera and monitor incorporate laser-cut stainless-steel disks with square holes. The camera is now a full RGB unit, with each primary handled by a dedicated photomultiplier channel. A dichroic splitter (taken from a professional tv camera) does the colour splitting.

Last year I brought along for the first time two stereoscopic adapters; one for the camera, one for the monitor. These use beam splitters and shutters to switch between the left and right channels, frame by frame. Unfortunately last year the demonstrations were hampered by other problems with the apparatus. Things have now been improved and I hope to exhibit stereo in full colour at this year's convention. Having said that, I consider the stereo to be no more than a novelty - the monoscopic viewing is much easier on the eyes! :wink:
Guest
 

Postby Andrew Davie » Tue Feb 13, 2007 12:06 am

An interesting proposal for mechanical colour NBTV can be found at http://www.televisionexperimenters.com/samscolr.html
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia


Return to Mechanical NBTV

Who is online

Users browsing this forum: Google [Bot] and 20 guests

cron