Constant current and pt100

[gojkosisa]

Hello,

as current source for pt100Ohm sensor I'm using LM317 to get 1mA. But after some time, maybe two or three month pt100Ohm get damaged and I have to change the sensor. I'm using attached schematic as current source (in my schematic R1=1k2).

What could be the problem, I'll appreciate if someone suggest some better IC for this purpose.

Regards.

 

[crutschow]

The LM317 has a minimum worst-case output current of 10mA so it will not reliably deliver only 1mA as a constant-current source.

Don't know of a better IC.

One common approach is to use a voltage reference, an op amp and a MOSFET connected as a constant-current source (see below). The op amp can be any rail-to-rail type and the transistor can be just about any logic-level type P-MOSFET. The reference is a two-terminal 1.2V shunt-type regulator.

The op amp feedback adjusts the transistor gate voltage to drop 1.2V across the source resistor R1, thus generating a constant current at the drain output through the PT resistance, Rpt.

Edit: The LT1009 actually has a drop of about 1.235V nominal so R1 should be 1.235k ohm (1.24K is the closest 1% value) to give 1.0mA current.

 

[Ice-Tea]

You can use a accurate voltage reference and put your PT100 in a resistor devider with an accurate resistor. A bit less straight forward to "read" the temperature, but it should work well

 

[FvM]

You can use a accurate voltage reference and put your PT100 in a resistor devider with an accurate resistor. A bit less straight forward to "read" the temperature, but it should work well

Furthermore, if the resistor is fed from the same reference as the ADC, it doesn't need to be accurate at all. We call it a ratiometric measurement.

 

[gojkosisa]

The LM317 has a minimum worst-case output current of 10mA so it will not reliably deliver only 1mA as a constant-current source.

Don't know of a better IC.

One common approach is to use a voltage reference, an op amp and a MOSFET connected as a constant-current source (see below). The op amp can be any rail-to-rail type and the transistor can be just about any logic-level type P-MOSFET. The reference is a two-terminal 1.2V shunt-type regulator.

The op amp feedback adjusts the transistor gate voltage to drop 1.2V across the source resistor R1, thus generating a constant current at the drain output through the PT resistance, Rpt.

Edit: The LT1009 actually has a drop of about 1.235V nominal so R1 should be 1.235k ohm (1.24K is the closest 1% value) to give 1.0mA current.

View attachment 89051

Thank you for this schematic. It is working but now I have new issue, temperature reading is oscillating. Ambient temperature is the same but sensors reading are oscillating in the interval +/- 2'C.

 

[FvM]

Ambient temperature is the same but sensors reading are oscillating in the interval +/- 2'C.

I won't be surprized if the current source is oscillating due to the high capacitive OP output load. Usually current sources with large MOSFETs need some compensation.

 

[crutschow]

Thank you for this schematic. It is working but now I have new issue, temperature reading is oscillating. Ambient temperature is the same but sensors reading are oscillating in the interval +/- 2'C.

Is "It" your original circuit?

What is the frequency of the oscillation?