search

c0pperdragon / C64-Video-Enhancement

Component video modification for the C64 8-bit computer
MIT License
250 stars 36 forks source link

Troubleshooting with testpattern #14

Closed bwack closed 5 years ago

bwack commented 5 years ago

Hi. Thanks @desaster for recommending to use the test pattern here: https://github.com/c0pperdragon/A-VideoBoard/releases I can see a pattern on my scope, but not on the TV. It seems to be very slow. Maybe that is it. The sawtooth sweeps looks fine, so I guess the resistor ladder is ok.

Looking at Y. 100ms/div, 200mV/div bilde

Maybe the fpga needs configuring.. ?

desaster commented 5 years ago

when powering the board, notice that there are two groundplanes, GND and GNDA. GND1 and GND2 are connected to GNDA. I initially made the mistake of trying to power the board with GNDA instead of GND.

bwack commented 5 years ago

Ok. I did that mistake the first time. Now the board is in the C64. I see I have soldered four grounds except GND5. Not sure what to do. Maybe the crystal is wrong.

bwack commented 5 years ago

Ouf, I have mounted the crystal wrong.. bilde

19061520371910494982

bwack commented 5 years ago

It was the crystal. Now there is picture :) Thanks for help !

c0pperdragon commented 5 years ago

Funny thing that the FPGA actually generates a signal without any oscillator at all. It seems that the PLL inside falls back to generating an ultra-low-speed clock (more than 1000 times slower as planned) when not being driven by an external signal.

bwack commented 5 years ago

Maybe the VCO in the PLL continues without the lock.. or there is a fallback. I tried to figure out by datasheet, but without luck. Just curious anyway :)