[time-nuts] GPDSO is working

[MailLists]

Historically U(S)ARTs had the data signals in positive logic, and the 
control ones active low - after inversion in the level translators, 
signals are in "mark" (-12V) condition for an "inactive" interface.
The threshold being slightly positive in "modern" EIA-232 receivers, is 
a "fail-safe" measure to unambiguously interpret the signals as "mark" 
in case the line is interrupted. If the standard thresholds would be 
used, in case of a line failure the line receiver would "memorize" the 
last state.
The 75154 is one of the line receivers which can be configured in both 
modes (even so the thresholds are not symmetrically around GND, but the 
hysteresis is much larger). Also the ubiquitous 1489 could have the 
thresholds adjusted using the response-control pin.

For experimenting, an USB-serial adapter like the FTDI MM232R could be 
used. It has the possibility to invert the logic signals (configurable 
on-chip), and to adjust the logic levels 1.8-5V (5V TTL or 3.3V LVTTL 
with the internal regulator), to adapt it to most logic level circuits. 
Even a lower power load up to ~400mA (eg. GPSRx) could be powered 
directly (5V), or through an external (adjustable) regulator, through 
the USB bus (available on most computers, unlike a "real" COM interface)...


On 7/14/2013 12:25 AM, Chris Albertson wrote:
> On Sat, Jul 13, 2013 at 6:52 AM, Chuck Harris<cfharris at erols.com>  wrote:
>
>> If a TTL signal does "just work" with your RS232 receiver, you have
>> a faulty receiver.  The receiver is supposed to have a dead zone from
>> +3V to -3V.
>>
>
> You are 100% correct, almost all modern RS-232 receivers are faultily as
> you describe and will work with TTL level signals
>