[ntp:questions] Performance?

David Woolley david at ex.djwhome.demon.co.uk.invalid
Wed Dec 31 00:15:03 UTC 2008


dhavey wrote:

> "Voltage levels
> Diagrammatic oscilloscope trace of voltage levels for an uppercase
> ASCII "K" character (0x4b) with 1 start bit, 8 data bits, 1 stop bit
> 
> The RS-232 standard defines the voltage levels that correspond to
> logical one and logical zero levels. Valid signals are plus or minus 3
> to 15 volts. The range near zero volts is not a valid RS-232 level; "
> 
> Wikipedia.

Never use Wikipedia as a source.  A correctly constructed Wikipedia 
article will always identify an original source for every significant 
fact, and you should only use Wikipedia to trace that source.  In this 
case, I think they are over-simplifying because the author is going by 
popular understanding, rather than the actual spec.

Paragraph 2.5 of RS232-C requires that RTS, DSR, DTR, SecRTS (CA, CC, 
CD, and SCA) interpret any voltage between -2 and +2 volts, with a 
source impedance of at least 300 ohms as being OFF, and any interface 
point voltage of more than +3V be treated as on.  As the input 
resistance can be as high as 7k, this seems to imply that the zero going 
threshold must be at least +1.92V, at the interface point.

Specifically, the last paragraph of the Voltage Levels section of edit 
260896780 of the Wikipedia article is misleading, as it omits the power 
off detection case.  Also the oscilloscope diagram (as of 
2007-12-04T03:01:53) is incorrect, because it takes more than 4% of the 
signalling unit to cross the transition region.

Driver open circuit voltage can be as high as 25 volts, but in a steady 
state must be in the range 5 to 15V when driving a resistance between 3k 
and 7k.

I practice, though, any RS232 interface will be emulating the 1489A line 
receiver chip, which doesn't quite meet the 2 volt power down source 
voltage requirement.  It has a configurable threshold, but the 
unconfigured negative going threshold is at least +0.75V and the 
unconfigured positive going threshold is at most +2.25V, which seem to 
me to be very TTL compatible.

The receiver has Schmidt type characteristics.  Although it isn't 
formally characterised, it looks like the typical hysteresis ia just 
over 1V.

There is an alternative version, with a smaller typical hysteresis and a 
maximum positive going threshold of 1.5V.

Chip data from a, rather old, Fairchild Linear Interface Data Book.
> 
> I thought it was plus or minus 3-12 volts with the range between -3
> and +3 volts undefined.

Upper limit is 15 volts, although the receiver must tolerate 25V.  The 
driver should produce at least 5 V at the interface point, in the steady 
state.




More information about the questions mailing list