Construction Diary -- Part 2, Spinning the Nipkow Disc

Original build of a televisor by a complete novice.

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Postby Andrew Davie » Mon May 14, 2007 12:48 pm

Klaas Robers wrote:On this photo you can see that the square holes have a little overlap. So it is sharp. Also the difference in horizontal and vertical sharpness is very well visible. Look at both lines of the T.


I think this is very misleading, because that "T" has been so-designed so that the vertical line coincides with a whole and single scanline only. Were it in a different position (say, 1/2 a scanline to the right) or thickness, it would not look nearly as sharp. It's entirely possible that the vertical resolution is better than the horizontal, for objects not aligned exactly on scanlines.

This is one of the reasons I think test patterns aligned exactly to scanline positions are fairly useless as a guideline to resolution.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby DrZarkov » Mon May 14, 2007 4:20 pm

The test patterns are indeed good, if you are measuring or testing contrasts, brightness, etc, but they are not a "real life" testcard. That's why the BBC later in the 405 line systems included the photo of that girl in the testcard, I think. The "Reichspost" in Germany used instead of a testcard a short film with two young girls working at the "Reichspost" singing "Horch, was kommt von draußen rein?" ("Listen what is coming in") and a photo of a ship passing through a bridge. Maybe I can make scans of both and put it here. I think the copyrights are public domain, but I will look up that first.
User avatar
DrZarkov
I think I've had a cranial implosion.
 
Posts: 1041
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

Test cards.

Postby Steve Anderson » Mon May 14, 2007 5:22 pm

Although at the time we all thought that transmitting test cards was boring, now they bring back nostalgic memories to those that are old enough to remember them.

I attach a few here, test05 is one I created from scratch for the 405-Alive society but without the infill picture. It could be modified to suit 32 lines I guess.

I have a book about television that has test cards (BBC) from just after WWII, I'll have a hunt around for it if anyone is interested.

The young girl in test card F was chosen to eliminate any colouring by the use of make-up, she's alive and well, and now around 50 years of age!

Also the same picture was used in test card W, a widescreen format.

The colour accuracy and contrast of these files is questionable, can't recall where I got them.

Steve A.
Attachments
377.jpg
377.jpg (41.08 KiB) Viewed 16993 times
TCD.jpg
TCD.jpg (52.03 KiB) Viewed 16993 times
tcf.gif
tcf.gif (99.73 KiB) Viewed 16993 times
testd05.gif
testd05.gif (9.81 KiB) Viewed 16993 times
TCW.jpg
TCW.jpg (35.34 KiB) Viewed 16988 times
Last edited by Steve Anderson on Mon May 14, 2007 6:31 pm, edited 1 time in total.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby DrZarkov » Mon May 14, 2007 6:09 pm

How about turning testd05 at 90 degrees and putting a picture of Baird or Nipkow in it? It would be a nice new testcard.
User avatar
DrZarkov
I think I've had a cranial implosion.
 
Posts: 1041
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

A good idea.

Postby Steve Anderson » Mon May 14, 2007 6:37 pm

How about turning testd05 at 90 degrees and putting a picture of Baird or Nipkow in it? It would be a nice new testcard.


Yes, that's a very good idea, anyone have a jpg or gif either gentleman?

Steve A.

Just done a bit of 'googling' and came up with this one...I just wish he wasn't so good looking...

Also one of Nipkow...
Attachments
JLB.jpg
JLB.jpg (3.68 KiB) Viewed 16979 times
Nipkow.jpg
Nipkow.jpg (2.19 KiB) Viewed 16977 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Klaas Robers » Tue May 15, 2007 2:38 am

The problem of the T is this:

In the top bar of the T the video voltage goes from one volt to zero and back to one again. So in the video there is a real black for a short moment. The video is band width limited, so the voltage varies gradually, not like a square wave. Bot it becomes zero, I can see that on the oscilloscope.

Now, due to the square holes in the disc, the light is smeared more or less in vertical direction. That is the reason that the horizontal line is nowhere really black.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby DrZarkov » Tue May 15, 2007 6:51 am

What do you think about that:
Attachments
nipkowtest.gif
nipkowtest.gif (25.21 KiB) Viewed 16958 times
User avatar
DrZarkov
I think I've had a cranial implosion.
 
Posts: 1041
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

Nipkov test pattern.

Postby Stephen » Tue May 15, 2007 8:20 am

Nice job, Volker. I like it. I was just thinking though that one would need a high resolution system to make out the patterns except for the outer border! However, by changing the grey squares to a black and white checkerboard pattern and making the diagonal striped bars solid black, the whole pattern should be usable with 30 or 32 lines.
Stephen
User avatar
Stephen
Anyone have a spare straightjacket?
 
Posts: 427
Joined: Fri Jan 26, 2007 7:00 am

NBTV Test card.

Postby Steve Anderson » Tue May 15, 2007 12:59 pm

Nice job, Volker. I like it.


Me too. It would need a lot of simplification to be of any value on NBTV, also the picture would have to be proportionally larger to be able to discern the features of the face.

I might have a go at it in the future or others can have a go at it now. Is there any current NBTV test card? I've seen in a few places the same NBTV picture, but not this one, I presume from the NBTVA CDs.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby DrZarkov » Tue May 15, 2007 5:50 pm

I think this is a job for my good old Amiga, using PPaint, a pixel painting program. I will try this evening to simplify the testcard.
User avatar
DrZarkov
I think I've had a cranial implosion.
 
Posts: 1041
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

Postby Andrew Davie » Wed May 16, 2007 1:28 am

Well I've just surfaced after a very painful and slightly embarassing two days trying to get a simple IR pair working. Since my first attempt at running the motor synch circuit as it is newly installed on my single board TV didn't work, I thought -- well, let's have a look at the IR signal as I hand-spin the disc. Nothing.

So, I took out the IR bits and created a stand-alone version of just that part of the circuit on a breadboard. Just the IR transmitter, the 680 ohm resistor, the IR receiver, a 100K pot (what I had handy) and a 33K resistor in series with that.

Nothing.

I bought some new IR send/receive components. Nothing. I checked and re-checked. Basically, as I'd mixed up the components and they don't really have much of any markings to help out... I had a pile of IR transmitters, none of which I knew worked or not -- and a smaller pile of receivers -- ditto. I had assumed that something I'd done blew them up -- and had no real way to tell (I can't seem to get my multimeter diode function to work -- perhaps it's not putting enough voltage out?).

I thought to check the setup by placing a LED across where the IR transmitter goes, and similarly across the receiver -- just to see that there was a reasonable voltage/current there. In hindsight I could have used the multimeter, but in any case, the LED lit (not too brightly), and so I figured that was OK (I understand the specs of the transmitter and receiver are about 1.3V). I calculated the correct resistance for the transmitter, noting that the 680 ohms was on the large side, but should be OK.

I've been frustrated in this for a couple of days, but tonight I had a bit of inspiration! A long time ago I noticed that my old videocamera when pointed at a TV remote actually showed the remote led brightly in the viewfinder, even though it wasn't visible to the eye. So I'd figured the camera was showing the IR downshifted to a visible wavelength on the viewfinder. Dug up the old camera, pointed it at a transmitter IR LED and *voila* I could confirm that LED worked. From then, things got much simpler. When I plugged in my original IR LED it worked.... and then a minute or two later it did not work. Very strange, I appeared to have blown that one up while testing. I'd done this a while back, too... perhaps adding/removing components to the live circuit is a no-no?

In any case, now that I had a working and visibly working (with the camera) IR transmitter going, I could concentrate on the receiving side of things. Plugged in the receiver. Nothing.

I figured I'd blown that up too, but it's the only one I have left that seems suitable. I have a bunch of others, but they reacted strongly to the lamp I'm using (last I had things working) so I'm avoiding those.

After trying installing the receiving LED leads backwards ... things started working. When I disassembled my original working breadboard circuit, I carefully 'tagged' the negative leads of the wires with black tape. It appears that I did this for the wrong wire with the IR receiver. This is very annoying, because I was trying to be careful -- and this mistake has cost me a few days.

Now I have a working IR pair, running in the small subcircuit. The problems were two -- firstly, because I couldn't SEE if the IR was transmitting, I had no idea if the IR LED was blown or not. I had several IR LEDs which (it turns out) were blown. These have now been thrown away. Secondly, I had the receiver plugged in backwards, because of my stupid tagging mistake.

OK, so now I'm happier... a working IR signal, giving a good 12V when they're in proximity, and 0V when not. Next step is to connect these back to my single-board motor control circuit, and start to trace the rest of that circuit knowing that the IR side of things is working.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Klaas Robers » Wed May 16, 2007 7:34 am

Andrew, there is an even easier way to check if an LED is working: measure the voltage accross it. So connect your multimeter to the terminals of the LED and read a voltage of about 1,8 volt (for IR). If the LED is blown you read the supply voltage, as well as if the LED is connected in reverse.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Postby Andrew Davie » Thu May 17, 2007 12:57 am

After a lot of effort, my single-board NBTV setup is now functioning! I have two rectifiers, the NBTV sync separator and LED driver circuits, and the motor control circuit all together and working -- not perfectly -- but very well. Once I sorted out the issues with the non-working IR transmitter and the wrong-way-around connection for the receiver -- plugged this into my implementation of the motor driver circuit on the single board, and it worked. First time.
There's a bit of an issue with the stability of the image -- it locks, but every few seconds it kind of 'pulls' almost like it's sticking or something, then it goes back to it's locked position. I saw this before I transferred the circuit to the single-board, so I'm suspicious of the motor, the disk friction, perhaps the belt. I put the IR signal on the oscilloscope and it looks excellent -- there's no glitching there.
But for now, I'm pretty much done with the actual electronics 'development'. I must say, my old LED matrix (which I'm using because I blew up the nice new one, as mentioned before) is far inferior. The picture quality I'm getting isn't anywhere near as good. I have 100 new ultra-bright LEDs on the way, so I'm itching for those to arrive so I can build yet another LED array. I might build a few, so if anyone wants one let me know.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: Commonality

Postby Andrew Davie » Thu May 17, 2007 2:09 am

Steve Anderson wrote:..but heck, you're doing a damn fine job!! 10/10. Wanna job as a lab assistant? Genuinely, I would seriously consider someone like you, even with no formal education in electronics. A concept I have being trying to get over to the locals here with little success, where paperwork is everything. Where enthusiasm and a will to learn is dismissed.


Sorry, I didn't notice the above until just now. Thanks for the compliment. Never say die :)
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby Andrew Davie » Sat May 19, 2007 10:11 pm

Well, I've been a bit disappointed with how the motor synchronisation circuit has been working, now that I've transferred it to my single-board. Yes, it synchs, yes, it's just about the same as on the breadboard... but it really suffers from the occasional "pulling" that I mentioned earlier.

Tonight I set about trying to find out why. Firstly, the pulses from the IR seemed to look pretty solid -- a nice sharp curve. Didn't seem to be anything there. I tried reversing the motor (and figure-eighting the band)... worse, if anything (which is what I'd expect with the variable friction on the band).

Finally, I removed the 'frame' stickytape -- the one hole in the disc that was covered to indicate start of frame to the circuits. And... whoa... now my motor control synchronisation is ROCK SOLID. It was that missing 'pulse' -- deliberately masked out -- that was throwing the circuit for a loop. I always wondered how/why the circuit was expected to behave itself when it wasn't really getting 32 pulses/rotation, yet it was expected to synch to 32 pulses/frame. Or is it?

I'll have to chase that up, so that I understand it properly. But in any case, without that stickytape on the disc, I get very very nice solid frame 'lock'. With it, as I said, the locking is very disappointing.

So now, at least, I have something that I'm happy to show off. Speaking of which, I showed the family the converted footage from the GIF files online (those of Don MacLean's restoration efforts). They were most impressed that we were looking at 1930s video pretty much as it might have looked back then. 'Spooky' was the feeling.

Last week I dropped into a local glazier and asked if they had any glass that would diffuse but not attenuate the light. I think I used "too big" words, as I just got blank stares. After explaining what I wanted it for, though, they were only too helpful. They gave me three samples, one of which is a slightly frosted glass, which operates beautifully.

At this point I have come to realise that the diffusion filters you place over your light source have a HUGE difference on the quality of the image -- not just the brightness, which I expected, but also the clarity/contrast, which is NOT what I expected. My images go from average/good to excellent, depending on the filter used.

Interestingly, this is still with my old LED matrix -- remember, I blew up the nice new one. I have 100 LEDs on order via eBay (same seller as before), so I expect with the new new matrix I plan, and the new glass diffusion filter... and of course now I have the locked in synchronisation working nicely... my image should be awesome.

Next thing on my plate, while I wait, is to build the circuit that allows me to play 'negative video' properly.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

PreviousNext

Return to Andrew Davie's Televisor #1 (2007)

Who is online

Users browsing this forum: No registered users and 0 guests