[Coco] [Color Computer] Re: Interfacing question.
j_e_daggett at yahoo.com
Wed Mar 23 12:07:54 EST 2005
--- In ColorComputer at yahoogroups.com, Gene Heskett <gene.heskett at v...>
> On Tuesday 22 March 2005 22:03, j_e_daggett wrote:
> >--- In ColorComputer at yahoogroups.com, Gene Heskett
> > <gene.heskett at v...>
> >> Or switch to a 6309, which can cheerfully source or sink 20 ma per
> >> pin, and its logic one outputs are the supply rail for all
> >> practical purposes.
> >> [...]
> >on a good day and if you hold your breath for a year.
> >Not according to the Hitachi spec I have will the 6309 sink and
> > source 20ma per pin. The max output current is 100ma and that is
> > the sum total of all the address buss, databus and control signals.
> > You run that current to long and I guarentee the part will get warm
> > as nice bed bug.
> Oh not much doubt there. But the individual pin can do that if it has
> to. The one I've spent quite a bit of time looking at with scope
> probes did go rail to rail quite nicely, this with a multipack full
> of cartridges plugged in, a disto SC2 4n1 in the back slot, a wp-rs
> in the next one, and 2 rs232 packs, one of them modded for two ports
> in the front two slots. And it ran cool enough (on an external power
> supply, not that built in toaster the coco's have) that I can turn it
> on, boot up nitros9, bury it under a furniture blanket, come back and
> lift the blanket a day later to read the darkroom thermometer laying
> over the memory, and its 3 degrees above room temp.
Considering the the multi-pak interface adds buffers to the address
lines, no matter if all four slots are filled the 6809 sees only one
extra TTL load. The 6809 address lines have afan out of 5 and with the
multipack interface you achieve a fan out of 5.
The 6309 is a CMOS device. To sink current from a CMOS output all the
current to the load is through the uppper PMOS transistor. T he lower
NMOS transistor is turned off. Derived from the Hitiachi spec and 20
ma current, the on resistance of the drain-source PMOS transistor can
be as high a 45 Ohms. This is not uncommon is MOS ICs. Small geometry
MOS tranistors have large on or saturated resistance. The smaller the
geometry the higher the sorce drain saturation resistance is.
> Even on the more heavily loaded A0-A3 lines, both transitions were
> well under 10ns, the bandwidth limit of my 100 mhz dual trace hitachi
> V-1065 scope. Its an amazing chip IMO. I'd love to take one running
> exclusively on a WP-RS for its video i/o, and crank up the clock to
> see just how fast it could run. With fast enough memory, I'd bet on
> 20+ mhz. Something else would fall over first.
The beuaty of CMOS is that nearly all the current is when the output
is switching from one state to another. During this transistion period
the load capacitance has more influence on the transient current
seen by a CMOS output stack. From the instance the change of state is
initiated the charge in the external load capacitance will rise to a
very high level and then assymtope to a steady state value determined
by the load resistance. Rise and Fall times are the time it takes to
charge or discharge the external load capacitance.
My guess is that the HD6309 will start to fail at around 6 to 10 MHz.
In this processor and most the major determination of speed is by the
propogation of the signal through the many gates that determine things
like address mode decode and instruction decode. In the MC6809, these
are assynchronous processes and have to pass through about 3 to 5 NMOS
gates. Each gate can have up to about 4 nS delay time from input to
output. If Hitachi uses also asynchrounous process to do address mode
and instruction decode, then I would expect delays through these
circuits to be on the order of 25 to 35 nS. This would yield between 7
and 10 Mhz maximum clock speed.
> >The 6309 has a VOH voltage than the 6809. Which would improve his
> >problem. The 6809 is rated VSS+2.4 VDC min while the 6309 is rated
> > 4.1 VDC min.
> That, I'd venture to say, would be under a lot heavier load than the
> coco puts on it to pull it down that far.
The VOH for the 6309 is base on a sink current of 400 microamps. For
the 6809 it is 205 microamps.
Yes the 6309 has a fanout of 10, or twice that of the 6809.
Brought to you by the 6809, the 6803 and their cousins!
Yahoo! Groups Links
<*> To visit your group on the web, go to:
<*> To unsubscribe from this group, send an email to:
ColorComputer-unsubscribe at yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
More information about the Coco