NTC Thermistors - I''m baffled!
Archiver
Posts: 46,084
I've since figured out the datasheet for the thermistor...
If I look through the many tables, at the row that defines the resistance at
25 degrees C, I found the value 1000, which must be 1K ohms. So, I've
developed (on paper) a circuit that will develop a voltage out from .1V to
about 4.6V to represent about 54 degrees F to about 104F. It uses an op amp
with an NPN transistor as a constant current source supplying 2 mA to the
thermistor, and the resulting voltage drop is input into another op amp
stage to give me my final output. I've, since then, figured out that a lower
current source will allow me to sense lower temperatures (more voltage drop
across the thermistor at the lower temps), but I'll have to recalculate my
amplifier output.
I don't yet know if my circuit will work (living in Navy barracks limits my
workshop!), but when I get over to my friend's garage I will test my design.
I've learned a lot about op amps during my design process. I've also decided
after modeling my design using Excel, of all things, of the curvature of the
A/D converter output versus temp, that I can neglect the curvature of the
thermistor output, and interpolate the points between the 5 degrees C that
the datasheet gives me, to get a good enough sense of temperature.
I'd like to calibrate my sensor once I get it built, by my math is assuming
ideal parameters of the components. For example, my constant current source
is assuming a voltage drop of 1V for VCE - That is the ideal drop from the
base to the emitter of the transistor, but I think the Vce of the transistor
can be lower than this.The lowest temp that I can sense is limited by the
voltage drop of the resistor I use to limit the current and the Vce of the
transistor.
Does anyone have a good (and cheap [noparse][[/noparse]I'm overseas, and I have limited
resources]) method of calibrating this? I've read that using a water bath,
and another thermometer, will allow me to figure out my error, by measuring
a few different temps. Will an oral thermomter work for this? Or should I
use an average outdoor mercury thermometer (I know that reading this device
accurately is subjective)?
I am also going to initially use 5% carbon resistors for setting the gain of
my op amp. I've read that using 1 or 2% metal film may be better. When I get
this finished, I plan on putting my design on the web for others to
hopefully learn from and critique.
If you're wondering what this is for, I plan on using the thermistor to
sense the temp of a qater tank and turn a heater on/off using a BS, to grow
triops. Triops are pre-historic creatures that should grow from barely
visible eggs to 2 inches in a few weeks. I bought a kit from the Victoria,
BC bug zoo while on leave. These remind me of sea monkeys, and the actual
application may be overkill, but I need some use for my Basic Stamps! [noparse]:)[/noparse]
Wow, this message turned out to be long... sorry 'bout that.
Steve
Original Message
From: "Luis" <lullamas@y...>
To: <basicstamps@yahoogroups.com>
Sent: Monday, June 16, 2003 8:28 PM
Subject: RE: [noparse][[/noparse]basicstamps] NTC Thermistors - I'm baffled!
All the log functions, are linear terms?
Perhaps, you have missing pow() functions in the "t" equation.
Luis.
Mensaje original
De: Steve Ziuchkovski [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=TS2r39RNIjxbboile2EzjbHwqEP-99v4Vy5K1X8y-la1SxYNQR2Dyl2DiUpLJlQnedAdiSfwSmJrF-QW]zman97211@y...[/url
Enviado el: s
If I look through the many tables, at the row that defines the resistance at
25 degrees C, I found the value 1000, which must be 1K ohms. So, I've
developed (on paper) a circuit that will develop a voltage out from .1V to
about 4.6V to represent about 54 degrees F to about 104F. It uses an op amp
with an NPN transistor as a constant current source supplying 2 mA to the
thermistor, and the resulting voltage drop is input into another op amp
stage to give me my final output. I've, since then, figured out that a lower
current source will allow me to sense lower temperatures (more voltage drop
across the thermistor at the lower temps), but I'll have to recalculate my
amplifier output.
I don't yet know if my circuit will work (living in Navy barracks limits my
workshop!), but when I get over to my friend's garage I will test my design.
I've learned a lot about op amps during my design process. I've also decided
after modeling my design using Excel, of all things, of the curvature of the
A/D converter output versus temp, that I can neglect the curvature of the
thermistor output, and interpolate the points between the 5 degrees C that
the datasheet gives me, to get a good enough sense of temperature.
I'd like to calibrate my sensor once I get it built, by my math is assuming
ideal parameters of the components. For example, my constant current source
is assuming a voltage drop of 1V for VCE - That is the ideal drop from the
base to the emitter of the transistor, but I think the Vce of the transistor
can be lower than this.The lowest temp that I can sense is limited by the
voltage drop of the resistor I use to limit the current and the Vce of the
transistor.
Does anyone have a good (and cheap [noparse][[/noparse]I'm overseas, and I have limited
resources]) method of calibrating this? I've read that using a water bath,
and another thermometer, will allow me to figure out my error, by measuring
a few different temps. Will an oral thermomter work for this? Or should I
use an average outdoor mercury thermometer (I know that reading this device
accurately is subjective)?
I am also going to initially use 5% carbon resistors for setting the gain of
my op amp. I've read that using 1 or 2% metal film may be better. When I get
this finished, I plan on putting my design on the web for others to
hopefully learn from and critique.
If you're wondering what this is for, I plan on using the thermistor to
sense the temp of a qater tank and turn a heater on/off using a BS, to grow
triops. Triops are pre-historic creatures that should grow from barely
visible eggs to 2 inches in a few weeks. I bought a kit from the Victoria,
BC bug zoo while on leave. These remind me of sea monkeys, and the actual
application may be overkill, but I need some use for my Basic Stamps! [noparse]:)[/noparse]
Wow, this message turned out to be long... sorry 'bout that.
Steve
Original Message
From: "Luis" <lullamas@y...>
To: <basicstamps@yahoogroups.com>
Sent: Monday, June 16, 2003 8:28 PM
Subject: RE: [noparse][[/noparse]basicstamps] NTC Thermistors - I'm baffled!
All the log functions, are linear terms?
Perhaps, you have missing pow() functions in the "t" equation.
Luis.
Mensaje original
De: Steve Ziuchkovski [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=TS2r39RNIjxbboile2EzjbHwqEP-99v4Vy5K1X8y-la1SxYNQR2Dyl2DiUpLJlQnedAdiSfwSmJrF-QW]zman97211@y...[/url
Enviado el: s