RCTIme with low level resistance?
Archiver
Posts: 46,084
Guys,
If you don't mind I want to bounce off some ideas in low resistance
measurment by RCTime.
I need to measure the resistance of a Platinum RTD sensor (100Ohm at
0 Celcius, about 140 Ohm at 100 Celcius. Important interval in my
application 125-140 Ohm)
Actually 100Ohm RTDs are so common in industry that this would
benefit many industrial engineers.
All I read about RCTime lead me to believe that a low value of my RTD
poses a problem because the pin resistor that limits the current
during initial phase has to be some 100-200 Ohm not to fry the pin
and at the same time it must be "negligable" to the measured
resistance for things to be nice and linear.
Since my resistor is only 100-140 Ohm....
I came up with an idea - what if I use a 220Ohm protected pin to
initial phase and a different pin with no resistor for RCTime phase?
If I tie them both together and manipulare them in the following way:
1. Protected pin does its thing for needed period. That takes care of
the initial state of the capacitor. Since the pin is protected, no
problem.
2 At the end of that time unprotected pin goes the same way and its
safe, because the capacitor is almost charged, so there is no high
current drawn.
3. Protected pin goes INPUT (disconnects as a source of voltage)
4. Unprotected pin goes RCTIme, etc., etc.
May be a diode on an unprotected pin will be also helpful just in
case? In that case timing of the 1-2 is not that important...
Any comments? Any other suggestions how to measure the low
resistances with RCTime?
All and any responces, ideas or suggestions will be humbly and
greatly appreciated. Parallax hardware gurus, whatdayathink?
Cordially,
Alex Chaihorsky
If you don't mind I want to bounce off some ideas in low resistance
measurment by RCTime.
I need to measure the resistance of a Platinum RTD sensor (100Ohm at
0 Celcius, about 140 Ohm at 100 Celcius. Important interval in my
application 125-140 Ohm)
Actually 100Ohm RTDs are so common in industry that this would
benefit many industrial engineers.
All I read about RCTime lead me to believe that a low value of my RTD
poses a problem because the pin resistor that limits the current
during initial phase has to be some 100-200 Ohm not to fry the pin
and at the same time it must be "negligable" to the measured
resistance for things to be nice and linear.
Since my resistor is only 100-140 Ohm....
I came up with an idea - what if I use a 220Ohm protected pin to
initial phase and a different pin with no resistor for RCTime phase?
If I tie them both together and manipulare them in the following way:
1. Protected pin does its thing for needed period. That takes care of
the initial state of the capacitor. Since the pin is protected, no
problem.
2 At the end of that time unprotected pin goes the same way and its
safe, because the capacitor is almost charged, so there is no high
current drawn.
3. Protected pin goes INPUT (disconnects as a source of voltage)
4. Unprotected pin goes RCTIme, etc., etc.
May be a diode on an unprotected pin will be also helpful just in
case? In that case timing of the 1-2 is not that important...
Any comments? Any other suggestions how to measure the low
resistances with RCTime?
All and any responces, ideas or suggestions will be humbly and
greatly appreciated. Parallax hardware gurus, whatdayathink?
Cordially,
Alex Chaihorsky
Comments
I think there are a couple of problems with using RCTime with a 100 ohm RTD.
Choice of capacitor. To get a 2 millisecond time constant
(count=1000) you would need something on the order of 15 microfarad
capacitor. (tm = 1.347 * R * C)
<http://www.emesystems.com/BS2rct.htm#RCtime>
Stability is an issue with such a large capacitor, and the surge
current through the RTD would quite high, and the pin would
definitely need discharge protection.
The 100 ohms of the RTD is very low with respect to the internal
resistance of the Stamp pins themselves. In the standard rctime
circuit (and also I think in the alternative circuit you proposed)
one end of the 100 ohm rtd is tied to common. The stamp pin has to
supply current not only to charge the capacitor, but it also has to
drive wasted current through the rtd. The protection resistor and
the internal pin resistance act like a voltage divider with the rtd,
so the capacitor only gets charged part way to the level you are
expecting. That is why in the standard circuit, it is recommended to
use 1 kohms or 10k or higher for the resistance, with as low a value
as possible for the capacitor and the protection resistor.
I would go with an op-amp to buffer the RTD. If we are determined to
use rctime, a circuit similar to those at
<http://www.emesystems.com/BS2rct.htm#RCtime> should work to make a
resistance to current converter. I would probably use an LM10 to do
it, because the LM10 contains a built-in 200 millivolt reference.
-- regards,
Tracy Allen
electronically monitored ecosystems
mailto:tracy@e...
http://www.emesystems.com
>Guys,
>
>If you don't mind I want to bounce off some ideas in low resistance
>measurment by RCTime.
>I need to measure the resistance of a Platinum RTD sensor (100Ohm at
>0 Celcius, about 140 Ohm at 100 Celcius. Important interval in my
>application 125-140 Ohm)
>
>Actually 100Ohm RTDs are so common in industry that this would
>benefit many industrial engineers.
>
>All I read about RCTime lead me to believe that a low value of my RTD
>poses a problem because the pin resistor that limits the current
>during initial phase has to be some 100-200 Ohm not to fry the pin
>and at the same time it must be "negligable" to the measured
>resistance for things to be nice and linear.
>Since my resistor is only 100-140 Ohm....
>I came up with an idea - what if I use a 220Ohm protected pin to
>initial phase and a different pin with no resistor for RCTime phase?
>If I tie them both together and manipulare them in the following way:
>
>1. Protected pin does its thing for needed period. That takes care of
>the initial state of the capacitor. Since the pin is protected, no
>problem.
>2 At the end of that time unprotected pin goes the same way and its
>safe, because the capacitor is almost charged, so there is no high
>current drawn.
>3. Protected pin goes INPUT (disconnects as a source of voltage)
>4. Unprotected pin goes RCTIme, etc., etc.
>
>May be a diode on an unprotected pin will be also helpful just in
>case? In that case timing of the 1-2 is not that important...
>
>Any comments? Any other suggestions how to measure the low
>resistances with RCTime?
>All and any responces, ideas or suggestions will be humbly and
>greatly appreciated. Parallax hardware gurus, whatdayathink?
>
>
>Cordially,
>
>Alex Chaihorsky