[time-nuts] RS 232

Chuck Harris cfharris at erols.com
Fri Jul 26 18:53:13 EDT 2013


The old MC1489's and 1489A's had both positive and negative
thresholds (about +/- 1.5V.. not quite correct, but better
than nothing), and as such wouldn't work with a 0 to 5V signal.
Then along came the IBM-PC, with its comport card that had both
20ma current loop, and RS232 implemented using single transistor
circuits, and the die was cast.

The MAX232 also implemented the standard incorrectly by leaving
out the hysteresis and going with a TTL threshold.

I don't know that there has ever been an IC line receiver that
implemented RS232's thresholds correctly.

-Chuck Harris

Bob Camp wrote:
> Hi
>
> I can't say if they were more than 10 years old or not, but I have seen some ports
> that do not work reliably with a 0-5V signal in the last few years.
>
> Bob
>
> On Jul 26, 2013, at 3:58 PM, Didier Juges <shalimr9 at gmail.com> wrote:
>
>> I believe the trend for the last 10 years has been to use a positive threshold.
>> I mess with serial ports all the time professionally and not and I do not
>> remember the last time I saw one that did not work with a positive threshold.
>>
>> Didier KO4BB



More information about the time-nuts mailing list