[Coco] TC-9

[jdaggett at gate.net]

On 7 Jul 2008 at 23:46, Chuck Youse wrote:

> On Mon, 2008-07-07 at 23:35 -0400, Chuck Youse wrote:
> > On Mon, 2008-07-07 at 23:29 -0400, jdaggett at gate.net wrote:
> > 
> > > The MC6809E, no internal oscillator,  the E and Q clocks are
> > > inputs and sink very little current.   
> > 
> > I believe you are mistaken.  The E clock on the 6809E is not a
> > simple TTL signal, it has different voltage and current requirements
> > because it directly drives the internal circuitry.  Look at the data
> > sheet.
> > 
> > C.
> 
> Hmm, so I just looked at the data sheet myself; it lists Vih for the E
> clock (separately) as VCC-0.75V (nominally 4.25V).  I can't see any
> specific information on the current draw, but the sample clock circuit
> shows the E to the CPU separated from the E to the system and pulled
> up with 680ohms.  I don't know if they're just trying to keep the rise
> times short or if they're trying to ensure a lower-impedance supply to
> the pin.  
> 
> ANYWAY, point still is that you can't drive those with a standard TTL
> output and certainly not a 3.3V FPGA IO pad.  I'm pretty sure both the
> GIME and the SAM were designed to meet the E clock requirements
> without any external assistance.
> 

The E Clock needs the higher voltage due to the larger internal buffer that it drives 
on the MC6809E. In order to operate to the 2+MHz range and overcome the larger 
input capacitance. Internally the E Clock is the dominant clock source. 

True the output of the IOB do not reach the voltage level requirements. That would 
require an externa buffer of some sort. 

I never trried to omply that an FPGA was a cure all. The approach to an FPGA 
based system would require that all the 5 volt logic be converted to 3 volt logic. 
That initself can present extra circuitry for level translation. Then again 5 volt logic 
is also becoming a bit more scarce lately. That led me to wonder if the HD63C09 
could operate at 1 to 1.5 MHz at 3.6 VDC.

 
> But now we're just drowning in details.. :)
> 

True.