Though it finally works, the video output is not perfect. On my LCD screen it looks like all characters move sideways one or two pixels, about every 2 seconds. Very annoying. But it could be a LCD monitor issue, just not coping with this non-perfect AV signal. So I connected it to my TRS-80 monitor just to see if it would work better with a device that originates from the same era.
Not really. Apart from the image being shifted to the left, all lines move slightly left and right like a wave is passing through the text.
So, 50.0 Hz is important !
Probing around with the oscilloscope I find the screen update frequency on pin 10 of U11. This is exactly 50.5 Hz, which might be the problem since that is exactly 0.5 Hz off from the required 50Hz, which could explain the 2 seconds period in the wave. I actually did not think this would be a problem on a modern monitor, but it looks like it is.
The reason for this frequency to be off is the fact that I used a 6.5536 MHz crystal instead of the original 6.50 MHz, since that was the closest I could get from Reichelt. Maybe not such a good idea after all. A quick search shows that 6.50 MHz crystals are indeed not very common, but DigiKey does actually sell them.
So I ordered 10 pcs. (they are not expensive). And indeed, after swapping the crystal the video image is rock steady and super clear.
( I just noted that @Ken Yap actually warned me for this when I started the project. And at the time I also answered to him that I could get the right crystal at Mouser or DigiKey. I just forgot about that later..)
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.