4:3 PAL colour test card

[See_Mos]

Starting in the late 70's and for around 20 years my main income was from repairing video recorders.

Around the mid 1980's UK television stations stopped transmitting engineering test cards and started transmitting daytime programmes instead.  I needed a stable and repeatable test signal, something more than just colour bars and cross hatch.  I could not afford several hundred pounds for a commercial test card pattern generator so I built a project from the May 1979 of the UK magazine Television Servicing.  The logic card used 45 TTL logic IC's and a very expensive sync generator IC.  The colour encoder used a couple of television IC's and other components.  A copy of the original magazine cover is shown in picture 1.  It was taken from the only TV they had handy at the time.

Picture 2 shows the results of the recently adapted AD724 colour encoder on a Daewoo 4:3 LCD TV.

I had thought for a long time that it should be possible using a PIC so I decided to give it a try using an 18F25k22 and Positron.  The result so far is shown in picture 3.   Note the glitch just below the colour bars, this does not appear on a cheap Techwood LCD TV but as it is not present with the original logic card there has to be a problem with the code even though everything appears to be OK when checked with a Picoscope. Note the extra white line below and above the castellations

Picture 4 shows the test setup. Only PortB is needed to drive the AD724, PortC is used for various test signals during development and testing with a previous TEA1002 encoder which needs more logic signals.

There is some camera distortion of the pictures and the shading is due to LCD viewing angles.







   

[keytapper]

Please move this post to the Wiki.

[TimB]

See_Mos

I love your history. If others have any history stories I for one would love to hear them.

Tim

[Mapo]

See_Mos
you have created a great generator.
One question, is the resolution standard or HD?

[John Drew]

That's a great project. Am I right in assuming the Pic is outputting RGB. That's remarkable.
More details if you're willing to share. A friend is attempting to replace a failed CRO tube with a 7"TFT display. Maybe your project could provide some possible solutions.
John

[top204]

A wonderful project. :-)

In 1979, I was an  apprentice TV engineer in a local shop, and the "new fangled" video recorders were extremely expensive to buy and were laden with fancy components. :-). They had not made there mind up whether it would be the Philips, Betamax or VHS that would finally win the race, and unfortunately, it was the VHS. I say unfortunately, because the Betamax was far superior in both mechanical and video quality, but they cost more to build than VHS. And the philips was like all philips stuff back then, and absolutely dreadful. Over complex and very unreliable. Remember the G6, G7 and G8 and the terrible, G11 chassis? Then they went on to single board TV chassis that would totally die because the EEPROM packed in with too many sound volume changes stored in it! A lot of people do not realise that I2C was first developed for communications in television and video recorder circuits. :-)

Me and dad used to buy the "Practical Television" magazine every month from him being a young man to me being a young boy onwards. :-) Then it turned into the "Television" magazine and it was an invaluable aid and an excellent read for us nerds. LOL.

The magazine you are talking about is here:
https://worldradiohistory.com/UK/Television-Servicing-UK-1979-05.pdf

and part two of the pattern generator here:
Television-Servicing-UK-1979-06.pdf

I used to use the test card on the early colour televisions for aligning the colour guns and coils around them with a panel full of pots and movable coils they had in them. It was a time consuming task and was soooooo boring. Then they aligned the electron guns inside the tube itself, which made the job a lot easier.

I could never afford a colour generator, so I used a crosshatch for the standard repairs, and recorded the test card for alignments, then played it back when setups were required. :-)

Looking back at the mastery of digital logic gates to generate timed pulses, it was a work of dedicated art. I did it once when I made a pong game from 74 series TTL devices and it took months of work using the scope and experimentations. :-) Then they brought out the single 40-pin chip that did it all for you, but it was not as much fun.

I made a simpler colour and crosshatch device back in the late 1990s using a PIC16F84 device, but it was nothing special, but saved on a large amount of CMOS 40 devices.

Question.... What was your favourite video to repair in the 1980s? Mine was the lovely Ferguson with the piano keys and the top loading tape. It rarely went wrong and all that usually needed replacing was the belts/pinch wheel, tension strap, and sometimes the head. All the mechanics were metal and lasted forever! Then "Philips" brought out the dreadful plastic cogs and others followed. Then videos became so unreliable.

I look forward to seeing the project, and "bloody well done" see_mos. :-)

[See_Mos]

Perhaps we could start a new thread for personal history / background stories.

I think Les and I lived in parallel worlds.  I learned everything I knew about video recorders from Television Servicing magazine and a couple of Hitachi training sessions. I used to go to the Harrogate Hi-Fi shows and one time got a sneak preview of a  Victor NTSC machine before it was released in the UK as the Ferguson 3V00 / JVC HR3000.  My favorite was the 3V29 and 3V30 or the JVC equivalent HR7200 and HR7300 and I must not forget the Mitsubishi machines with the yellow solder resist.  you could replace an IC, knock off the flux residue and no one could tell any work had been done.  I was once accused by another engineer of not replacing a part because of this.

The generator output is standard 625 line 50Hz driving an AD724 colour encoder module in PAL mode. Only one port is needed to provide R.G.B. mixed sync and three or four bits connected to a modified R2R network for the grey areas.

I solved the glitch, a couple of clock cycles had crept into some of the blocks of line code, it's surprising how much 125nS can affect the display.  If every line was extended by the same amount I think I might have got away with it.

I will release the code once I have tidied it up a little, meanwhile, here is a picture of the original modified to give colour bars and R.G.B only screens.

[JonW]

WOW, that world radio history website  has so much content, many thanks for sharing!

[See_Mos]

Earlier this year when I decided to try the AD724 colour encoder with the old logic board there were several sellers on EBAY offering the board shown in the test setup picture.  A couple of months ago when I wanted to buy another there was only one seller of the board and now there are none.

There are quite a few sellers offering the IC alone, many of them from China.  The board is quite simple so I don't think it would be difficult to do the CAD and have some made.