Page 1 of 2

Project: Field sequential 2 colour camera/monitor system

PostPosted: Fri Sep 14, 2012 3:03 pm
by aussie_bloke
G'day all.

I have embarked on another colour wheel television related project, this one being a 2 colour field sequential camera/monitor system.

This involves a 4 segmented perspex wheel with red/green or aqua/orange colour segments, a 1974 Sony AVC-3250CE B&W vidicon tube camera and its viewfinder being the monitor and a fan motor to drive the colour wheel.

I chose 2 colours to start with as our 50 fields/sec system is easily divided by 2 opposed to 3, so each primary colour will hit each field on the camera's image pickup and the monitor's scan and at rapid speed I will get a 1920s style 2 colour picture!:) The viewfinder and camera lens will face the same way so I can just use one colour wheel in which each primary colour identical on diagonal opposite will hit both camera lens and viewfinder at the same time.

I don't know how well this experiment will work out but am already in the swing of constructing it and will soon find out!:)

Anyhow I have to go out so don't have time to upload pictures and videos at present but will do so this evening when I'm back. So stay tuned!;)

PostPosted: Fri Sep 14, 2012 5:48 pm
by AncientBrit
Depending on how 'laggy' the vidicon is you may have a form of colour crosstalk between the 2 coloured frames.

But well worth a try, good luck.



PostPosted: Sat Sep 15, 2012 1:17 am
by aussie_bloke
AncientBrit wrote:Depending on how 'laggy' the vidicon is you may have a form of colour crosstalk between the 2 coloured frames.

But well worth a try, good luck.



The lag with my 3250 is most prominent in low lighting like most vidicon cameras are but pretty much non-existent in outdoor lighting and good indoor lighting. I actually like the idea of colour crosstalk happening as it will make great for a picture effect! :wink:

PostPosted: Sat Sep 15, 2012 2:05 am
by aussie_bloke
Okay now that I'm back at my computer, as promised here below are some photos of the project starting to take shape along with a sketch design plan for this colour system. As seen the viewfinder screen and camera lense are both facing in the same direction so only one colour wheel is needed, the good thing about this setup is I can watch myself live in living 2 tone colour! :wink:

I have also done an experiment on computer with frame sequential two colour using a still photo of a lorikeet and splitting the channels and colour tinting each channel with it's correct primary colour and making a video of each channel in sequence which on fast scan gives you a flickery two colour image of the lorikeet. I have attached a couple of videos of this experiment, one using red and green primaries and the other using aqua and orange.

Anyhow I believe I will get this whole apparatus set up within the next couple of days and ready for testing! :)

PostPosted: Sat Sep 15, 2012 5:16 pm
by Harry Dalek
Hi troy good luck ! looks good never tried colour interested to see the results .
I was looking up the 2 colour system as i don't know much about it ,i came across this ..
Mechanical 2.5-color TV is possible, too. For instance, imagine a color wheel with red and green wedges. This wheel scans your TV. Viewing the red and green picture through a blue transparency adds the third color. Instead of adjusting a DC level, you swap transparency densities until you find the best one.

It might be an easy add on to see later what that blue filter does...should work better on an electronic system as yours .

BTW i am interested in how you make the colour wheels.

PostPosted: Sat Sep 15, 2012 7:35 pm
by aussie_bloke
Thanks for the info Harry will definitely look into that!:D

How I'm making the wheel, I have a sheet of transparent perspex which I am going to stick red and green cellophane segments onto as the colour filters, of course I prefer to find actual dye plastic filters but cellophane is what's easy available and it worked with my colour video still experiments with vidicon cameras.

PostPosted: Mon Sep 17, 2012 12:55 am
by aussie_bloke
G'day all.

Have managed to assemble the 2 colour field sequential colour camera/TV system to testing stage and have been testing it today.

There is good news and bad news with the results of the test.

Starting with the bad news. I have been unable to get a red/green colour picture with my vidicon cameras I tested with which are the Sony AVC-3250CE and AVC-3200CE. Reason behind that is as AncientBrit mentioned there was indeed colour crosstalk between two frames due to after image persistent lag of the vidicon camera so hence the succeeding frame/field has a ghosting of the previous frame/field hence the colour response is not instant enough for the field rate of the camera so the colours kind of melt together resulting in a yellow monochrome picture once the wheel hits a certain speed. So that was a bit of a bummer :(.

Now for the good news. Taking into account that vidicons lag and CCDs don't I decided to try using a 90s CCD camcorder to see if that makes a difference, whilst it's a colour camera, it will respond to colours in the same fashion as a B&W camera except in colour. Anyhow having my 90s Panasonic camcorder and a little portable B&W TV rigged up behind the colour wheel, I switched on the motor and at a surprisingly slow speed I saw on the TV a red/green composite colour picture!!!:D I was seeing shades of with in the vicinity from red to orange to yellow to green and as expected the colour was flickery. Just like with the mechanical TVs I made, there is no auto sync circuitry for the AC motor so I just finger brake the wheel to slow it to correct speed. So it's working!!!:D

Thinking about vidicon lag resulting in poor colour response from the filters, makes me wonder with NASA's field sequential colour vidicon camera system, did they built a custom vidicon tube that has pretty much minimal to no lag for that system?

Anyhow here below are pics of further construction of this apparatus and testing of it, photographing the TV making a colour picture was not possible but did some snaps anyway:

PostPosted: Mon Sep 17, 2012 11:35 am
by aussie_bloke
Just had thought of an idea to overcome the vidicon lag issue though at the possible compromise of a field and having to use another colour wheel on the spindle for viewfinder. I was thinking perhaps have the colour wheel with two blanking segments in the clockwise fashion of red, black, green, black. Here below is a picture to visually describe my idea:

PostPosted: Mon Sep 17, 2012 7:13 pm
by Harry Dalek
Hi Troy

Interesting results i can see with no lag your better off with a newer camera or a very old type ghosting on either .

Very well done i don't understand the speed it has to run at i am thinking twice the speed of one frame ? or slower ? i suppose on test card non movement stuff it could be dropped back to a flicker at NBTV speed.

I was reading this page troy

This old CBS colour montior was converted to NTSC which i didn't think was possible for that system to be converted to !
Has me wondering why not mechanical Pal ?

PostPosted: Mon Sep 17, 2012 7:32 pm
by aussie_bloke
G'day Harry.

Thanks for your input. I am most certainly aware of the Gray Research monitor which was used for the CBS field sequential colour system used. Their method of scan was 6 fields per frame which 2 fields were used per primary colour, and the scan speed was a whoppin 144 fields/sec which equates to 24 full frames/sec which is pretty much close to PAL colour speed and so the motion still looks pretty fluid!:) They used 3 inch image orthicon tube cameras which they placed a 6 segment colour wheel in front of the tube inside the camera.

From the B&W programs I've seen the image orthicon tubes seem to be snappy and lagless as like with CCDs when the camera tracks fast motion like a football kicked in the air or a cowboy swinging a lasso the motion looks choppy whereas with vidicon and plumbicon tube cameras (B&W and colour) there is image lag and hence the fast motion tracked is has smooth transition between frames which is in similar fashion to human eye tracking motion as the eye has persistence of vision which when looking at an object for a little while then turning away a ghost of that object is still persistent before it decays.

Anyhow whilst it's good that this experiment works with a CCD camera, it is kind of disappointing that it doesn't work with a vidicon camera as I wanted to see my vidicon cameras produce motion colour, but will try the blanking method as described above and see what that does.

I am also gonna have to make a bigger wheel as the one used barely fits over the Sony cameras viewfinders, so have ordered an A3 sized piece of perspex off Ebay plus am now getting gel filters as cellophane is ripply and bounces bright light around which is undesirable. Also been thinking about the positioning of the camera lens and the viewfinder, camera lens must be positioned so the filter segments just fit and same with the viewfinder so the scanning is proportional which each filter hits precisely at the right time for both camera and monitor!

Anyhow will keep on tinkering away at this experiment and hopefully will eventually get it good enough to make some colour telerecordings.

PostPosted: Mon Sep 17, 2012 8:09 pm
by Harry Dalek
Thanks for the information there Troy ..yes i can see the size of using one colour wheel could be a problem since the monitor is bigger than the camera lens .. they might be out of sync ?

Your lens i suppose fixes this with a zooming it to same size so they start and finish the colour wheels start and end .

The New idea of the colour wheel with blanked out segments is an interesting idea....passive way to do it
I think another way might be to sync the video off and back on to the colour wheel colours every time they came around but i think it would be no different to your idea ..yours no electronics sensors.

Troy where did you by the perspex ? how thick is it ?

PostPosted: Mon Sep 17, 2012 9:21 pm
by AncientBrit

re vidicon lag.
Some commercial vidicon cameras mixed in light bias within the optics to reduce the effect.

Not sure if you could arrange this.



PostPosted: Mon Sep 17, 2012 9:53 pm
by aussie_bloke
harry dalek wrote:
Troy where did you by the perspex ? how thick is it ?

I bought the perspex sheets off Ebay, someone in Queensland is selling them and also quite a few other people too. Here's a couple of links to perspex sheets:
A3 size
A4 size

PostPosted: Mon Sep 17, 2012 10:56 pm
by gary
I managed to get a reasonable quantity of perspex from reverse garbage for a very reasonable price before someone came in and started buying it all up before anyone else got a look in. Problem is I am afraid to to use it in case I never get to see any again. Perspex is one of those growing number of products that are becoming like gold in Australia - try buying some hardwood - where will it end?

Edit: At least the postage for the A4 size perspex is free - that makes a big difference.

PostPosted: Wed Sep 19, 2012 1:48 am
by aussie_bloke
G'day all.

Some real good news. Have done further testing of my red/green 2 colour system and have managed to get it all to work a lot better making some real distinguishable red/green colour pictures!:D

I am still using the Panasonic RX3 CCD camcorder but had decided to use one of the Sony vidicon camera's viewfinders as it just fits over the colour wheel. To make the viewfinder work with the camcorder I made an adapter for it to be powered individually and take any composite video input from any source. I then positioned the camcorder so the lens is center positioned between the filter segment edges of the wheel and the viewfinder I had to do some trial and error to find the best height position due to the nature of how the colour comes up on the screen.

Anyhow I have taken some screen snaps which include a picture of a lorikeet printed with red and green primary colours only and some colourful objects on a stand. I have also made a telerecording which I will later upload to YouTube. Here below are the screen snaps: