[Coco] Re: Help using the Coco3 with a 1084S-D2 Monitor

[jdaggett at gate.net]

Torten

I seriously doubt that there is timing issues. The NTSC HSYNC signal is essentially 
the xtal frequency divided by 1820 and the VSYNC is the xtal divided by 477272 +/-
. To loose sync, the crystal will really have to be way out of whack. To move the 
horizontal frequency 1 Hertz requires the xtal to move 1820 Hertz. About the most 
you can reliable warp a third overtone crystal is +5KHz/-15KHz.  Also the xtal will 
stop oscillating before one can shift it enough to effect the vertical frequency. 

Now drive levels from the GIME chip is a potential cause. The GIME chip drives an 
inverter and the PIA chip U5. For the PAL versions, the inverters drive a composite 
sync circuit and the RGB port. My suspect is that VR2 is not adjusted properly 
and/or that the circuit to create a composite sync signal for the PAL encoder is 
loading down the inverters that drives the RGB port. As a last resort the inverters 
may provide marginal drive to the Comodore monitor. 

I also seem to remember that LS TTL have poorer noise immunity than the old TTL 
devices. That maybe due in part to the fact that the input impeadances of the two 
are remarkably different. In an LS TTL device you drive the base of a transistor 
through a Schotky diode. In TTL you are driving the emitter of a common base 
configured bipolar transistor. Also in the LS TTL the input transistor is seldom driven 
to full saturation!


regards
james


On 12 Nov 2003 at 12:13, Torsten Dittel wrote:

> Hi Hank,
> 
> > You guys have any ideas on this one?
> 
> I checked your pin out on all 3 components and it looks OK (just to
> complete it, pins 6,7,8 on the NEC 3D are the GND lines for R,G,B).
> 
> I noticed "glitches" in vertical sync on my 1084S-P1 (AFAIR the only
> difference between the P1 and the D2 is a head phone jack) too. The
> funny thing is that it's only with one of my two (PAL) CoCo3s. Both
> work fine with the NEC 3D. One of the CoCo's is US-made, the other
> Korean. The difference is, that the (newer) US model has a new version
> of the GIME chip and I guess this is causing the problems because of
> slightly different signal timing and/or amplitudes. I never
> cross-changed the GIME chips nor checked it out with an oscilloscope.
> 
> Hope this helps.
> 
> Regards,
> Torsten from Germany
> 
> 
> -- 
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco