NBTV without any moving parts. THE NIPTRIX

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

NBTV without any moving parts. THE NIPTRIX

Postby Viewmaster » Thu Oct 04, 2012 11:44 pm

Having dabbled with a tiny NBTV (The Nipper), I am considering the complete opposite. A project based on three items I have seen.........

1. Some years ago at NBTV convention was a flat array of 1563 LEDs
forming a large NBTV picture. But the LEDs were switched by a large rotor, so intermittent connections were the problem.

2. In recent newsletter Jeremy outlined his single line scanner using two CMOS 4017 as switches to light up a straight row of 60 LEDs in sequence.

3. On utube was a 3D 8x8x8 LED cube controlled from a Raspberrypi.

So if No1 were combined with an extended version of No2 but with an additional analogue NBTV input to control the light o/p of 32 rows of 48 LED's one would have a Non Mechanical flat, bright, NBTV display.
(All synced from the line sync pulses)

Using 5mm LEDs it would give a picture about 10" x 6"
LEDs can be bulk bought at about 2 p each
.
No noise at all, as it hangs on the dinning room wall. Mmmmm.

Maybe a Raspberrypi would help in its control as in No3, but this is not really necessary and would only add to the cost.

The 4046 pll could be put in the dustbin. Imagine that! :wink:

I estimate about 3200 soldered joints required, hence my consideration only at this point. :lol:
Last edited by Viewmaster on Fri Oct 12, 2012 10:33 pm, edited 2 times in total.
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1289
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby Harry Dalek » Fri Oct 05, 2012 1:12 am

Hi Albert
Sounds like a great project a matrix NBTV ?
This is a simple scope and i am not saying this could be used but just thinking about it ..if you feed this sawtooth as in a scope at the right frequency or adjust as in a crt scope you would get the lines and in this case look i think like all the leds a lighting up same time ..matter of modulating mmmm
i wonder if you could get a simple on off tv led matrix display by modulating that sawtooth on and off to the LM3914..
Perhaps wishful thinking but i like the idea of your project even silly ideas like mine might help.
Attachments
LED Oscope.gif
LED Oscope.gif (63.5 KiB) Viewed 10120 times
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 DrZarkov » Fri Oct 05, 2012 2:25 am

To save some work: There is this kit available in Germany: http://www.elo-web.de/elo/mikrocontroll ... s-pingpong

The kit contains an LED array with 120 LEDs and an ATMega8 Microcontroller. The price of that kit was at www.conrad.de about 10 EUR, I bought two of those kits (and unused until now) for 4,99 EUR.
The official price was 29,99 EUR (at Amazon they sell it now for 65 EUR, used), but sometimes you'll find them in bookstores like Wohlthat (a kind of German "the Works") for the 5 EUR I've paid.
User avatar
DrZarkov
I think I've had a cranial implosion.
 
Posts: 1037
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

Postby Viewmaster » Fri Oct 05, 2012 4:44 am

Thanks guys for your interesting ideas, both the circuit and the kit idea.

One of the many interesting problems is that there is only the NBTV 400 Hz sync pulses to lock on to. But the 48 vertical lines of LEDs needs to run at a switching rate of 400 x 48 = 19.2 kHz.

So some kinda, line sync x48 timing needs generating.....maybe.
I am also looking at the 4017's control pins too, to see what inter connections are required to achieve the sync locked vert/hor LED timing too.

Years ago there was a 16 way IC with 4 bit addressing. Unlike the 4017 this could switch analogue signals through instead of just on/off. That would have been useful.
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1289
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby M3DVQ » Fri Oct 05, 2012 7:58 am

If you only want to use it with crystal controlled sources (e.g. a computer) then you can run the counters from a crystal clock divided down and re-synchronise it to the line pulses to stop it drifting enough to cause a picture tear. This is what I did with my oscilloscope driver circuit and it works very well.

If you want to be able to feed it with an unreliable source like a mechanical camera then you would probably have to have a voltage controlled oscillator generating your master clock and a phase locked loop comparing the sync pulses with a suitably divided version to generate the feedback. You would have to deal with the missing sync issue in such a way as to not affect the PLL though there is a moderately easy way to do this.
M3DVQ
Just nod and pretend you understand me
 
Posts: 331
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Postby Steve Anderson » Fri Oct 05, 2012 4:51 pm

This is something I have pondered before, how do you address 1536 LEDs? Assuming minimal pixel resolution.

The answer is in multiplexing them, 32 columns, 48 rows. Right, great, easy. BUT only one LED will be on at a time, 1/1536 = 0.065%. It's gonna be kinda dim.

So you could break it down into smaller chunks, say 8x8 in a 6x4 matrix...and parallel address them...the number of connections is rising!!

There is no easy answer to this...well, except perhaps the good old CRT!...no moving parts...except the electrons....avitar (left) is not a bad example methinks...

Steve A.

The original (cropped) screen-shot follows...
Attachments
Nong Jeap 1.jpg
Nong Jeap 1.jpg (56.81 KiB) Viewed 10083 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4143
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Viewmaster » Fri Oct 05, 2012 6:37 pm

Steve Anderson wrote:This is something I have pondered before, how do you address 1536 LEDs? Assuming minimal pixel resolution.

The answer is in multiplexing them, 32 columns, 48 rows. Right, great, easy. BUT only one LED will be on at a time, 1/1536 = 0.065%. It's gonna be kinda dim.
Steve A.

Your CRT has a beautiful, stable display, Steve, but I would like to try my own wall mounted NBTV in a suitable frame. :)

5 years ago a spinning LED carousel was demonstrated at NBTV convention.
I cannot find any links/photos of it, but I recall that it was colour and the image was quite brightish.

I might experiment with switching an ultra bright LED pulsed at your .065%
just to see.
Thanks M3 too for suggestions.

That 19.2kHz I mentioned could be generated simply on Audacity and put on the 2nd track of the CD, so this and the 400hZ sync would both be available.
The 400 clocks the 32 rows and the 19.2 clocks the 48 columns.

All the 8 Cmos 4017's would be zeroed at the 32 blank sync pulse in some way, ( cutting off the supply volts? :shock: ), but the very precise timing of this zeroing point is problematical as it's just at a frame ending.....maybe a very slight timing +- adjustment required on the front panel.
I shall put all the ideas given me and put 'em in the melting pot. :)


.
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1289
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby gary » Fri Oct 05, 2012 7:14 pm

Baird's matrix used incandescent bulbs - in this case the slowness of them helped as they added "persistence" - perhaps this could be emulated with LEDs using capacitance in some way or another?

The write up for Kevin Hadfield's carousel, at least the monochrome version, in volume 30 - number 4. He used 74574's with the common cathode of the LEDs tied to the drain of the video driver MOSFET.

Of course he only had to drive one line at a time....


PS: The colour version was in 33-2 - sample principle though.

PPS: Grant Dixon's 32x48 matrix was in 20-3
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 AncientBrit » Fri Oct 05, 2012 8:03 pm

Albert,

There is an alternative to using a PLL locked to 400Hz to derive the 19KHz pixel clock.

I tend to use gated clocks rather than PLLs

The clock is free running at slightly faster than the target freq.

You derive a 'needle' pulse from reference line syncs (extra circuitry need to produce the missing line sync, see Steve A's PIC cct. in a recent Newsletter)) and use this needle pulse to reset the pixel clock.

If the reference syncs are above or below target freq then the displayed pix will lose a few pixels or wrap around.

Cheers,

Graham
AncientBrit
Green padded cells are quite homely.
 
Posts: 858
Joined: Mon Mar 26, 2007 10:15 pm
Location: Billericay, UK

Postby Steve Anderson » Fri Oct 05, 2012 8:49 pm

Viewmaster wrote:5 years ago a spinning LED carousel was demonstrated at NBTV convention....


You're thinking of Kevin Hadfield's 'carousel' displays. initially monochrome, later full colour...

Without looking up the back-issues of the newsletter, it used a batch of shift-registers to achieve the result. It was/is very clever and I wish I had thought of it first!!

I have to own up and point out that the above picture is 48 lines (horizontal..count 'em) and 64 pixels per line...it used the full bandwidth of 48kHz sampling...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4143
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Viewmaster » Fri Oct 05, 2012 10:01 pm

Thanks Graham for your suggestions.....more for my melting pot.
I may build a single column of 48 LED and a single top row of 32 LEDs.
In this way any mock up in control and clocking can be seen at the first line and at the top of a test card signal. Just as a guide without full commitment.

Gary, I didn't know of Grant Dixons write up. I must have a look to see if anyone else has a framed mounted NBTV display up on the wall.
If I ever, ever make one it will be called, "The Nipkow Frame up!" :lol:


Steve, re LEDs not being bright enough.
If you read Jeremy's artical in latest newsletter, page 13 at the very end of the last paragraph he says that the LED light o/p is quite competitive with a white CRT.
Now Jeremy is clocking his LED's at 28kHz whilst my NBTV version would only clock at 19.2kHz. So NBTV version would be even brighter when full on, for max white in the NBTV video
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1289
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby Viewmaster » Sat Oct 06, 2012 12:04 am

gary wrote:PPS: Grant Dixon's 32x48 matrix was in 20-3


Thanks for that Gary.

So it has all been done before as this old convention photo shows. Sadly, Grant Dixon passed away. I wonder what happened to his LED
48x32 display?
It was before my time in the NBTVA but wish I had been there to see it.
I found that 20-3 artical by him of great interest.
Attachments
ledmtrx.jpg
Grant Dixon with his LED matrix display
ledmtrx.jpg (27.9 KiB) Viewed 10038 times
“One small step for a man,"......because he has Arthritis.
Albert.
User avatar
Viewmaster
Frankenstein was my uncle.
 
Posts: 1289
Joined: Fri Apr 06, 2007 4:50 am
Location: UK Midlands

Postby M3DVQ » Sat Oct 06, 2012 3:55 am

You could always drive the LEDs with a higher current making them much brighter but that would clearly be outside of the recommended parameters.

Alternatively you could add persistence to each LED by adding enough parallel capacitance to let the LED fade down from on each time.
M3DVQ
Just nod and pretend you understand me
 
Posts: 331
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Postby gary » Sat Oct 06, 2012 10:15 am

You could always drive the LEDs with a higher current making them much brighter but that would clearly be outside of the recommended parameters.


The duty cycle would be so low that you should be able to exceed the parameters by a considerable amount without damage.

-----

It is amazing to me that this form of matrix screen formed the basis of Baird's FIRST patent.

http://www.taswegian.com/NBTV/forum/download.php?id=720

He was later to implement it as a 5ft x 2ft screen (will you match that Albert? ;-))

Edit:
Hmm I wonder if we could use this patent as a basis for a claim that Baird invented the Plasma TV? If only he had used neon bulbs...

LOL actually I think the idea of a matrix like this was published at least 50 years before Baird, but it's always a good story.
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 M3DVQ » Sat Oct 06, 2012 3:59 pm

gary wrote:The duty cycle would be so low that you should be able to exceed the parameters by a considerable amount without damage.


That's exactly what I was thinking. A datasheet should give some idea of how many milliwatts the chip can dissipate but there would still be some trial and error involved to discover how rapidly the junction heats up :)

Also very important to make sure that whatever is driving your matrix can't ever stop clocking with the led driver still active! It would be much like burning a dot into a CRT only there might be a bit more smoke :shock:
M3DVQ
Just nod and pretend you understand me
 
Posts: 331
Joined: Sat Feb 10, 2007 10:52 am
Location: Lincolnshire

Next

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 10 guests

cron