Re: Plus/4 RS232 woes

On 05.09.2018 11:25, smf wrote:
> I forgot to mention, later in the datasheet it's clear that a crystal 
> and clock are different things.
>
> "XTAL1, XTAL2 (Crystal Pins)
>
> These pins are normally directly connected to the external crystal 
> (1.8432 MHz) used to derive the various baud rates. Alternatively, an 
> externally  generated clock may be used to drive the XTAL1 pin, in 
> which case the XTAL2 pin must float. XTAL1 is the input pin for the 
> transmit clock."
Hi,

In our (VLSI Solution) IC's XTAL1 is a clock input, so it can also 
receive a logic-level external clock.

XTAL2 is a buffered (inverse of) XTAL1, so XTAL2 and XTAL1 with a 
crystal, load capacitors and possible bias resistor create an oscillator 
amplifier circuit, and thus generating a clock signal to XTAL1.

So the system doesn't really need any actual clock switch and the /16 
mode works with both an external clock and with the crystal oscillator.
> The block diagram shows that the /16 is always active, the baud rate 
> generator takes XTAL1 & XTAL2 and bits 0-4 of the control register. I 
> don't know what happens electrically when you switch the internal baud 
> rate generator off, which the datasheet indicates the XTAL2 must float 
> and it's not floating because it's hooked up to a crystal.
XTAL2 must be unused if an external clock is fed into XTAL1.

Without seeing the datasheet I'm pretty sure the internal baud rate 
generator and /16 will work exactly the same with an external clock and 
crystal oscillator.

One possibility for the wording is that maybe the original design wanted 
to put the XTAL2 driver into powerdown when an external clock is used. 
But it would not save you much and the driver would need to know too 
much about the circuit...

So, it sounds like the baudrate generation just have a separate mode for 
/16, and the clock generation is totally separate and determined by the 
external components.

-Pasi