Temperaturmessung: Info Dallas DS 18S20

[Stefan <Hegi> Hegner]

Hier noch weitere Infos ... aus www.brianlane.com/digitemp.php

> Das wuerde bedeuten das man entweder pro fuehler eine
> Spannungsversorgung macht (Unpraktikabel) oder aber eine seperate
> leitung mitfuehrt - Aus 2 mach 3 ...
> 

Temperature Range
The temperature range of the DS1820 is actually -55 C to +125 C. But if 
you are using the passive adapter, which is powered from the serial port, most 
systems will only be able to read temperatures up to about +70 C. I have 
two  machines here, one which reads up to about +120 C and the other that 
only reads up to +75 C. If you require temperature readings in the +70 C 
to  +125 C range then you will need to supply +5v to the power pin of the  
DS1820. DS9097-U adapter has a strong pull-up option which may enable 
reading  higher temperatures, but it is still limited by the power 
supplied  from the serial port. The bottom line is that if you need reliable
temperature reading above +70C you should supply power to the 3rd pin of 
the temperature sensor. 

> Das problem mit der original schaltung mit einem Fuehler ist schon 
> das Temperaturen oberhalb 70 C nicht messbar sind weil die
> Spannungsversorgung nicht ausreicht. Ich denke das weitere Fuehler
> den Spannungsbereich weiter runter ziehen. Lieber was bauen um mit
> der Datenleitung mehr Fuehler zu versorgen.

Long Wire runs (glitches)
If you have long wire runs and are having problems with glitches you 
should add a Schottky diode to the end of your network. I used a 1N5818 diode 
reverse biased across the DS18S20 sensor in the attic. I was getting 
multiple glitches per hour before adding it and now I have seen 1 glitch 
in several weeks. 

See this TechBrief from Dallas Semiconductor for more info on using the  
diode fix, cable length issues  and lots of technical information about 
one wire networking issues. (http://www.maxim-ic.com/1st_pages/tb1.htm)

bis denne

STefan.