GM Coolant sensor
Darrell Norquay
dnorquay at awinc.com
Thu Sep 19 05:08:11 GMT 1996
At 11:44 AM 9/18/96, Dan wrote:
> Does anyone know if there's a positive curve temp sensor around, ie.
> increasing voltage for increasing temp. Would make life a lot easier
> for analogue circuitry.
>
There are a couple of solid state sensors which come to mind. Analog
Devices has the AD590, which is a current output device. It outputs a
current of 1uA per degree Kelvin. Thus, at 0C it delivers 273uA, and at
100C it delivers 373uA. Run this through a resistor and you have whatever
voltage you want, but you still have to deal with the offset. The other
device worth looking into is the National Semiconductor LM34/35 series.
This device puts out 10 mV per degree C (LM35) or degree F (LM34), but in
most cases requires a negative supply voltage to get temp readings below
zero. Unfortunately, neither is going to eliminate the opamp, since both
signal levels are too low for most A/D converters, and both will need some
offset to allow readings of temps below zero. There are positive temp
coefficient thermistors available, but thermistors are very nonlinear.
BTW, just because a thermistor has a negative tempco, doesn't mean you can
only get a negative slope voltage out of it. Consider a voltage divider
with a resistor tied to the supply, in series with the thermistor to ground:
v+ v+
| |
R1 Th1
| |
+----out +----out
| |
Th1 R1
| |
--- ---
Fig 1 Fig 2
The circuit of Fig 1 will give a decreasing voltage with increasing temp.
However, if you reverse the positions of the resistor and the thermistor, as
in Fig 2, you will also reverse the slope of the output voltage, and thus
have a positive slope output even though the sensor itself has a negative slope.
regards
dn
dnorquay at awinc.com
More information about the Diy_efi
mailing list