Hi
Just what we want, another thread on RGB to SCART cables.
All details in this post and any relevant comments will be used to create an article on my website in due course. I'm a bit busy at the moment so updating the website is a low priority.
This cable has been tested with multiple LCD and CRT TVs.
Before explaining the differences, time for the diagram.
Before creating this cable I undertook a number of measurements on the Amiga video outputs. The first issue I noticed was the voltage on the SYNC signal, it was 4.8V peak to peak.
Modern TVs have video decoder devices, their job is to decode the analogue video waveform into a digital video format, for SDTV this is usually a BT.656 video stream with a YCrCb colur space. Point aside, these devices take the composite video input or sync on green and process the signal to extract sync timing. They expect a 1V peak to peak signal, connecting the CSYNC signal from the Amiga could harm the device, especially as most parts run from 3.3/1.8V supplies. Video decoders are used in CRT and LCD TVs.
As the composite sync input has a 75 ohm terminator, adding 330 ohms, in series with the 47 ohms inside the Amiga has the effect of reducing this signal to less than 1V.
As an option, you can use the CVBS signal from the Amiga. For my testing purposes, I had both connected via a switch.
The BLANK signal (pin 16 of the SCART) is fed from +5V via a 75 ohm resistor, which together with the 75 ohm terminator in the TV, creates a voltage of 2.5V, adequate to RGB input mode.
The AV mode signal (pin 8 of SCART) is fed +12V via a 1K resistor. The resistor is present to limit the current in case of a short circuit.
Pin 18 of the SCART lead (blanking ground) was connected to pin 13 of the Amiga (digital ground).
Each video signal had an appropriate ground signal.
I did not connect the SCART shield, pin 21. My £1 cable, which was cut up to make this, was not fully shielded so there was no point.
So after making the cable it was time to test it. Pluged it into the Amiga and the TV, switched TV, then Amiga. The TV selected AV1, so the SWITCH signal was working but I had a black screen.
Fired up the oscilloscope and looked at the CSYNC signal There were 2 sync signals present. Switched the Amiga off and I saw a nice composite video signal!
In my cheap SCART cable, pins 19 (Video Out) and 20 (Video In) had been swapped over in the cable. So connecting pin 20 at the TV end to the CSYNC signal of the Amiga and it worked.
The picture was nice and clear, here is Directory opus
Colour bars
Oh and my test setup
The only issue I did notice, was that the LCD showed up the anti-aliased fonts of the Amiga. This is to be expected and I have seen similar effects when using old equipment with LCDs.
Now that I had a stable setup I set about injecting faults to see what happens.
The first test was to disconnect the blank signal (pin 16 of the SCART)
The display went black.
If I switched the CSYNC source from the TTL CSYNC of the Amiga, to the Composite video output of the Amiga, I had an image, albeit a crap one!
Does the composite video output normally look this crap?
The next test was to change the voltage on the SWITCH (pin 8 of SCART)
9-12V = RGB, 4:3 aspect ratio
3-7V = RGB 16:9 aspect ratio
0-2V = TV
So when making a cable, ensure you have 9-12V at pin 8 of the SCART lead and that the blank signal is 1-3V.
To test the interchangeability of my cable, I tried another LCD TV, this time a cheap Akura set. This system had trouble synchronising to the composite video signal from the Amiga but worked fine with the reduced TTL CSYNC signal.
It incorrectly detected the video as NTSC when fed from the composite video output!
Loss of CSYNC, from any source, lead to a black screen.
The final test was to use a Ferguson CRT TV.
This worked fine when fed a CSYNC signal from the composite output or the modified TTL CSYNC signal.
If the CSYNC signal was removed, the picture rolled diagonally.
In all tests the colours were well defined and I used colour bars, multiburst and convergence test patterns to confirm the display quality.
When I write this up for my website, I will add more test picture some oscilloscope plots and the rationale behind it. Before I do this, some feedback is required from you, the reader.
1) Have you tried this cable and did it work?
2) Did you try it with CRT and LCD displays?
Purchasing cables
I do not make these cables for sale but the great guys at AmigaKit do.
http://amigakit.leamancomputing.com/...roducts_id=226 and
http://amigakit.leamancomputing.com/...roducts_id=778 are the best cables on the market and have been tested with a wide variety of LCD TVs. Beware of cheaper cables that only work with CRT TVs.
That's all folks
Ian