Sunday, June 13, 2010

A600: Apollo 630 fix

OK my board arrived 2 days ago from Sweden, where it have been modded by PG to have a CPU socket instead of CPU soldered to PCB.
Since I had some instabilities with the replaced 030@40MHz with a 030@50MHz... I thought about removing the crystal and putting the former 40MHz one to test...

That was something I didn't expect since desoldering the crystal also removed the pin socket (eye-let) from the PCB along with some track
Ok that was something not for the faint hearted to realize...





I made some tracking through the PCB to find the signals from other places getting them with wire but haven't managed to make it work...
I can assure you that I had a really bad sleep that night
Next morning, I thought of asking Zetr0 "teh fluxmasta" about these signals since he has experience with the board!
Keith was very happy to help and provided a nice picture with the signals clearly:

CystalOscillator: 4pin (14 Dill package)
Pin # Desc
--------------------------------------
1 --> NC (not connected) *sharpe edge*
7 --> GND
8 --> Output
14 --> +5VDC
After trying it... OMG IT WORKED LIKE A CHARM!!! I was so happy about it that made a clean hack for it in order to be able to change whenever I wanted the oscillator.




In case you ask WHY I putted an external mini-PCB with a socket and not mounting it at the designated spot wiring the signals there... it's because YOU CAN'T put an oscillator socket along with the Indivision ECS since the height forbids it!

Well it's not bad after all aye?




One last picture from the 50MHz version working now semi-stable is this:



Thanks tons Zetr0 my friend! You're the ambassador of Kwan!