NBTV without any moving parts. THE NIPTRIX

[Steve Anderson]

This is the big issue with missing sync recovery, you only know it's a new frame too late. There are two ways around this....

I would like to say there's a third, but this may be viewed as a modification of #2 above. It's in V35 #4, also recently brought up here...scroll down to near the bottom of the page...

http://www.taswegian.com/NBTV/forum/vie ... php?t=1474

If the source is precise in it's timing then a simple monostable arrangement triggered on every sync pulse to produce a sync pulse 2.5ms later, then OR-gated with the original...you have continuous 400Hz. But mechanical sources, although doing an amazing job, are not quite precise enough.

I too wish that the missing pulse method was never instigated, but under the circumstance when it was introduced many years ago, there was little else to choose from.

Steve A.

Gary, as an afterthought, you've mentioned that you've done some form of TBC for NBTV. Did it get published or is it still in the works?

[Viewmaster]

This is the big issue with missing sync recovery, you only know it's a new frame too late. There are two ways around this....

If the source is precise in it's timing then a simple monostable arrangement triggered on every sync pulse to produce a sync pulse 2.5ms later, then OR-gated with the original...you have continuous 400Hz.

So I believe it also follows than an AND gate also connected to sync pulses and mono will show false at this missing sync point which is also a frame pulse.

[Viewmaster]

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

Going back to this.........
If the video came from a CD player with various picture tracks (like the test cards for example), then each time a track were changed there would be a break in sync pulses leading to the above if the LED's were greatly over run.
Or even if someone pulled out the CD player's jack o/p
By mistake of course

So some detection of both row and column clock pulses is required and the modulator turned off if either or both clocks stopped.

As previously mentioned, if I have to over run the LEDs I would do it the way I understand, by using two Freq to Volt converters to continually monitor the row and column clocks.

[M3DVQ]

If the video came from a CD player with various picture tracks (like the test cards for example), then each time a track were changed there would be a break in sync pulses leading to the above if the LED's were greatly over run.
Or even if someone pulled out the CD player's jack o/p

Except that you would be outputting "black" if there is no input shirley

[rommert]

The contraption on Dr Zarkov his photo's is made by me.
Although it was not working yet as desired it was put on display at the
NBTV meeting in Overpelt as a piece for discussion.
I see it provoked some discussion on the forum. I hope to give some
clarity on some comments.
1: Flatcable. The leds are in a 32 by 48 matrix. Each connection to the
circuitboards is made via two wires to avoid the need for cutting the
tracks between two successive pins from the flatcable connectors.
2: Red leds are used for cheapness. Less then 1 eurocent a piece on
a radio flea market. Probably the had been stored in a damp envirement
for they were soldering badly. No specifications known.
3: Indeed the leds are soldered in rows and each led tested before
soldering the next row. Iused an aluminium strip as an alignment tool
for each row in order to have all leds on the same distance to the vero-
board mounting board,and all aimed in the same direction.
4: Three circuit boards ( each two 100 X 160 mm veroboards side to
side ) are used. a) horizontal deflection b) vertical deflection
c) video circuit, overload indication and sync. separation.
5: A screen between each led as indicated by Vieuwmaster would be
a usefull improvement, because there is some optical interference
between adjacent leds.
6: The leds are overdriven. I found when the current is higher than
160 mAmp they change color. (from red to orange) I also have the
impression that the relation between current and light output is not
lineair in this mode of operation. The adjustment from contrast and
blacklevel etc. is very critical to get a recognizable picture. Also
the switchable gamma correction does not work as desired.
So there is room for improvement in the video circuit.
7: frame sync. is manual.

[Steve Anderson]

Hello Rommert and welcome to the forum. We were all interested in your display as Albert (by coincidence) is building something very similar and your experience will be invaluable.

It certainly looks well-made and one can see the amount of time you put into it.

It would surely be a well received item for the NBTVA newsletter, as will Albert's version in due course. If you have more information on this display (schematics etc..) we would like to study them. And perhaps some might make suggestions to the problems you're having with the contrast and black-levels.

Of course sync is always a headache in NBTV but the problems can be negated.

We hope to see more in time...

Steve A.

[Viewmaster]

5: A screen between each led as indicated by Vieuwmaster would be
a usefull improvement, because there is some optical interference
between adjacent leds.
6: The leds are overdriven. I found when the current is higher than
160 mAmp they change color. (from red to orange) I also have the
impression that the relation between current and light output is not
lineair in this mode of operation. The adjustment from contrast and
blacklevel etc. is very critical to get a recognizable picture. Also
the switchable gamma correction does not work as desired.
So there is room for improvement in the video circuit.
7: frame sync. is manual.

Hello Rommert,
Welcome to this forum and many thanks for posting some details of your display.
Re optical interferance between LEDs....That's very useful information for me. I wouldn't have thought of that.
Will now do a mock up with a small cluster of the ultra bright white LED's I am going to use to see what the inference is....maybe a solution will be to paint matt black each LED around their side surface.

Re over driven... I wonder if you have put in any extra circuitry as I have suggested in order to protect the LEDs if any of the clocking stops?
You only mentioned overload indication.

To M3DVQ.....Yes you are right. No signal so black.....like my mind at times....... goes black! So it's only an internal problem in that if the CD is outputing white and the clock stops then we are in trouble.

[Harry Dalek]

The comment on the interference between Leds is one useful bit of knowledge perhaps the led bricks is a better way to go...
Also perhaps instead of painting leds for the unwanted light between them place a thin black paper barrier along the rows..the black paper i used in the slit 2 drum monitors is light proof well to 6 and 8 watts of luxeons any way so should be no problem for a row of leds .
So if they are already soldered in as our new friend has done it might be an easy fix if you have a stanley knife a ruler ...

[Viewmaster]

Harry, one might need to mask the LEDs in both directions. The rows and the columns, so single strips would have to be augmented with cross pieces too.
My 2,000 LED s have not yet arrived from Hong Kong so all options are still open. My cluster experiment should tell me which way to go.
I'm glad that I choose ultra bright white LEDs, as this should give me the brightest picture for a given LED current without very much over running 'em.
We shall see.
I enclose a picture of the 4017 circuitry. This is NOT my design except for the red diode addition. This circuit is the standard method of cascading many 4017s together. The final display should indicate where the circuit needs modification.......probably the lot !

[gary]

Albert, are you absolutely SURE you don't want to build a 1536 segment commutator?

[Viewmaster]

Albert, are you absolutely SURE you don't want to build a 1536 segment commutator?

O yes, yes, Gary, I'm quite looking forward to getting away from all things rotational for a change...........and probably finish up watching a static silent jumble of pixels on the NIPTRIX

[gary]

Baird twitches lightly in his grave...

[Harry Dalek]

Hi Albert

Thanks for the schematic post its nice to see how its done all nice to see something working and such but to learn i for one need and example !

I have another idea to toss at you if you are by the sounds of it ...making the thing with loose Leds perhaps use these led panel mounts or with out ...drill holes in a thickness wood or plastic panel to length of the led used so you are making one big led brick as in Steves posting of the little ones you can buy yours would be one big mother home made Led Brick ....

[Steve Anderson]

Albert, not wishing to piss on anyone's parade, the HEF/CD series of chips, (even at 12V) may not be up to the task...they more than likely will be, but I wouldn't want to go into production using these devices...

Their output drive capabilities are severely limited such as driving a 4k7 resistor into a b-e junction could be compromised. They probably will work, but I wouldn't want to base my reputation on it. Change technologies, to HC at least...

At the same time consider 74HC154's and the like...fewer packages required...and with 10 diodes as per your diagram you may as well add another DIP package...

4 chips, 3xHC154's, one HC393, pixel decoding is done...line even simpler...and with plenty of drive capability...

Steve A..

There is a slight flaw in the 4017 too, it can produce illegal outputs until it has run through a complete 11-clock pulse sequence...I believe the FBI are on to this...

[Viewmaster]

Albert, not wishing to piss on anyone's parade, the HEF/CD series of chips, (even at 12V) may not be up to the task...they more than likely will be, but I wouldn't want to go into production using these devices...

Steve, thanks for that, but Jeremy drives his LED's at 25ma directly from his 4017s .See recent newsleter.

Also I did a load of googling research and found many examples where the 4017 drives LED's both directly and also by turning on a transistor, as I am going to do, so I was drawn to the conclusion that using the 4017 in some form would be OK.

Circuit wise I am now commited to the 4017, for good or evil now, so we shall see. I do intend however to do a single 4017 mock up to see how it clocks so will soon be able to tell if your warning comes back to haunt me............it probably will: