Constant Current Source for PT100 circuit

**sameerk** · 19th November 2009 04:23 PM

Hi,

I am trying to develop a PT100 (RTD) based circuit. I have a PT100 connected in a Wheatstone bridge.
The bridge is excited with +5v supply and has
4k7-RTD and 4k7-100e resistance arms. (1% resistors)
I wish to apply a constant current source to this bridge and pass a near 1mA current through each bridge arm.

Not sure with the change in resistance of RTD(PT100) how the current can still remain constant at @ 1mA! .... Please throw some light.

Can I implement the current source using LM134/234/334?

Please share your ideas/experience about the exact circuit that I can try out.

**MikeMl** · 19th November 2009 05:03 PM

Yes, an external (to the bridge) constant current source will do it. Google "constant current source circuit". Lots of possibilities, including a simple LM317.

**crutschow** · 19th November 2009 07:04 PM

Originally Posted by MikeMl

Yes, an external (to the bridge) constant current source will do it. Google "constant current source circuit". Lots of possibilities, including a simple LM317.

The LM317 has a minimum load current of 10mA so would not work for a 2ma constant current source.

**MikeMl** · 19th November 2009 07:38 PM

Originally Posted by crutschow

The LM317 has a minimum load current of 10mA so would not work for a 2ma constant current source.

So shunt the other 8mA to ground through a parallel resistor. If I was doing this, I would use the LM317 that I have on hand rather than drive across town to buy something more elegant, or wait several days (and pay shipping) to get something via mail order.

**Roff** · 20th November 2009 02:20 AM

Originally Posted by MikeMl

So shunt the other 8mA to ground through a parallel resistor. If I was doing this, I would use the LM317 that I have on hand rather than drive across town to buy something more elegant, or wait several days (and pay shipping) to get something via mail order.

The current through the bridge won't be constant if the resistance of the bridge changes and it has a fixed resistor in parallel with it. The current change might be insignificant, but the effective output impedance of the "current source" driving the bridge will be ≈1.2k instead of infinity.

**MikeMl** · 20th November 2009 03:29 AM

Actually, 600Ω.

I'm not sure that it matters a hill of beans, anyway. Here are three identical bridges excited with 2mA, 10ma shunted by 600Ω, and 4.8V. The RTD resistance is varied from 75 to 125Ω.

The top plot pane shows the differential output of circuit C (the other two are the same on this scale).

The middle plot pane shows the difference between the output of C and B, and the difference between C and V. Note that the largest difference is a measly 66uV (out of 25mV) of this range.

Finally, the last pane shows the actual current through the RTD in the three versions of the circuit.

Driving a four-arm bridge with a constant-current source vs a voltage source has benefit when the tempco of the four arms is matched. By definition, in this circuit, the four arms do not have the same tempco, so why bother driving it with a constant current source, anyway?

**sameerk** · 20th November 2009 04:02 AM

Guys

Thank you so much for such quick responses.
Mike, thank you for your well directed efforts.

So, does that mean I should use a CV source instead?
What may be a better (component) choice for that?

**ke5frf** · 20th November 2009 04:23 AM

You do not need a constant current source with a 2 or 3 lead RTD wheatstone circuit. In fact, with a balanced bridge you don't even have to have a stable voltage supply. Equal currents will be provided by the precision of your resistors and balancing of the RTD with the variable pot. The only concern might be temperature stability on the circuit board. the change in potential created by the RTD (and consequently change in current in that leg) is what causes current to flow through the ammeter and deflect the needle. It seems like regulating the current might be counter productive in the balanced design.

Are you building a balanced bridge with an ammeter or an unbalanced with a voltmeter? The unbalanced voltmeter circuit is a little more complicated, but I still see no justification in regulating the current for typical applications.

Constant current is typically used for 4-wire RTD applications, which isn't being accounted for in the responses here. Is your RTD a 2,3, or 4 wire sensor? 4 wire circuits are for nullifying lead resistance, but is typically not neaded unless your talking about a long distance. Are you making a long cable run between circuit and sensor?

Why not go with a simpler 2 or 3 wire design?

**sameerk** · 20th November 2009 05:13 AM

Hi

Mine is a 3-wire RTD wheatstone bridge.
I shall amplify the ΔV of the bridge.
I am not sure whether to answer this as a 'balanced bridge with an ammeter or an unbalanced with a voltmeter'.

My intention is to have the stable arm of the bridge (4k7-100e) produce a fixed Voltage and only the ΔR of the RTD should produce a proportional ΔV in the design.

**ke5frf** · 20th November 2009 10:56 AM

Originally Posted by sameerk

Hi

Mine is a 3-wire RTD wheatstone bridge.
I shall amplify the ΔV of the bridge.
I am not sure whether to answer this as a 'balanced bridge with an ammeter or an unbalanced with a voltmeter'.

My intention is to have the stable arm of the bridge (4k7-100e) produce a fixed Voltage and only the ΔR of the RTD should produce a proportional ΔV in the design.

ΔV to an amplifier would essentially be an unbalanced voltmeter type bridge.
And three wire? Explain why constant current is needed in your design? Can you post your schematic? Constant current is typically used when lead resistance is an issue. 4 wire designs overcome that problem. But your using 3 wire, so please explain the need for constant current?

**ke5frf** · 20th November 2009 12:56 PM

Your circuit resembles this, right?

Do you see that the bridge itself is your source of regulation?

In fact, a constant current source eliminates the need for a bridge altogether, because the voltage drop across the RTD in series with a single resistor would suffice as long as no current variability occurs.

**sameerk** · 20th November 2009 01:39 PM

Hi ke5frf

Thank you for your interest.
I understand that constant current is not needed in my bridge design.

However, do you think I need a constant voltage source OR just a stable voltage output of a +5v SMPS would be fine?

The schematic you posted is mostly similar to my circuit.
I am saying mostly as I have connected the 3-wire RTD up side down and one of the two wires goes to ground and other goes to the other arm's 100e resistor

to ^ 4k7 GND 100e (bottom - other arm)
^ ^ ^ (CONNECTION Details)
I_lRTDl__I_____I

Hope I am clear!

**ke5frf** · 20th November 2009 02:03 PM

I think a regulated SMPS output would typically suffice for a hobbyist circuit, but I wouldn't trust a wal-wart (for example) to regulate a critical temperature controller on the Space Shuttle

It depends on the purpose of your final product. Are you building a digital thermometer for a scientific test apparatus, or are you going to be monitoring the soil temperature in your flower garden?

**Boncuk** · 20th November 2009 02:34 PM

Originally Posted by MikeMl

Yes, an external (to the bridge) constant current source will do it. Google "constant current source circuit". Lots of possibilities, including a simple LM317.

Using an LM317L he can achieve a constant current of 1mA without problem.

Boncuk

**Roff** · 20th November 2009 03:07 PM

Originally Posted by sameerk

Hi ke5frf

Thank you for your interest.
I understand that constant current is not needed in my bridge design.

However, do you think I need a constant voltage source OR just a stable voltage output of a +5v SMPS would be fine?

The schematic you posted is mostly similar to my circuit.
I am saying mostly as I have connected the 3-wire RTD up side down and one of the two wires goes to ground and other goes to the other arm's 100e resistor

to ^ 4k7 GND 100e (bottom - other arm)
^ ^ ^ (CONNECTION Details)
I_lRTDl__I_____I

Hope I am clear!

It's certainly not clear to me! If you don't have a schematic capture program, use something like Microsoft Paint.

Electronic Circuits Projects Diagrams Free

Thread: Constant Current Source for PT100 circuit

Thread Tools

Display

Constant Current Source for PT100 circuit

Similar Threads

Constant DC current source

constant current source

constant current source

Constant Current Source

Constant Current Source

Tags for this Thread