3-wire RTD with Lead Compensation

[Nitra]

This question has been posted before with unsatisfactory replies.

A 3-wire RTD wired for Remote Sensing will transmit two analogues from the SCC-RTD01 into the DAQ, one proportional to 1 leg of the lead + RTD, and the other proportional to one leg. Then for a well balanced system (i.e. all legs of the leads are the same resistance) compensation can be carried out by processing (i.e. adding/subtacting the two analogues) and then calculating the temperature.

Discussion of other configurations i.e. 4-wire is not an option for this project. 

Question 1  - Is this mode supported by the NI-DAQ Test Software (Only three options appear available 2, 3 and 4 wire, not 3 wire with compensation).

Question 2 - Is there example or toolbox code available to support this mode.

It should be noted that I am the supplier of the hardware and need to confirm correct operation prior to delivery. I do not have access to any software other than that supplied with the hardware. Although a reference to appropriate code would be useful for the Customer.

[gt3000]

Hello Nitra,

Thank you for being very specific with your description.  The direct answer is that 3 wire RTDs can be connected to the SCC-RTD01, however there is no facility where compensation is calculated based on your description.  The software that directly interfaces with all the hardware plugged into a computer is Measurement and Automation Explorer (MAX), which will provide the intelligent driver for allowing user control of testing and scaling of measurements, before any data is collected.

As you may know, one of the key points of RTDs is that although measurements can be affected by lead resistance and self-heating, they are linear throughout their temperature range, and the error caused by its physical factors are stable over a large range.  Their linearity is dependent on the quality of their production, which I am sure you are aware of as a supplier.  Therefore, within MAX, you and your customer will be able adjust all data points by your choice of compensation in degrees C, or similar.

The National Instruments Developer Zone has much documentation that yourself and customers can read, regarding RTDs, and how to make adjustments within the software.  An example is linked below, for the benefit of those needing more information on RTDs

RTD: http://zone.ni.com/devzone/cda/ph/p/id/208

Developer Zone: http://www.ni.com/zone

Because of the MAX driver software, users of the hardware will need to setup one DAQmx Task to read from the RTD, if they are using the latest version of our software, such as LabVIEW.  Over 2000 examples exist in LabVIEW's example finder which ships with the software.  To find premade DAQmx code for RTDs the following steps can be used:

1. Run the LabVIEW program

2. On the main menu, navigate to 'Help'->'Find Examples...'

3. In the 'Browse' tab, Browse by 'Task'

4. Click the folders through to 'Hardware Input and Output'->'DAQmx'->'Analog Measurements'->'Temperature'

5. Open 'Acq RTD Sample.vi'

Many more examples exist within the help files, online, and through support, for getting the most accurate measurements using NI hardware for data acquisition and signal conditioning.  Please reply back with any further questions on this matter, or if you thought this information was what you were looking for.

Regards,

[Nitra]

1. In our application the cable connection to the RTD is long and the temperature along its route will vary 20-150degC, therefore the lead resistance will change significantly.

2. We are wiring the RTD to the SCC-RTD01 using 3-wire with Remote Sensing in accordance with NI document User Guide 371072C-01, figure 1.

If my interpretation (refer previous) of the operation of this configuration is incorrect can someone please advise me otherwise?

[gt3000]

Hello Nitra,

My previous post described a possible software solution, focused around your questions regarding the NI drivers and software.

From a hardware approach, lead resistance can be ignored using a wire and resistor configuration.  Taken from the RTD section of the article on http://zone.ni.com/devzone/cda/tut/p/id/4084, 'Signal Conditioning Fundamentals for Computer-Based Data Acquisition Systems':

"To keep costs down, RTDs are also available in 3-wire configurations. The 3-wire RTD is most effective in a Wheatstone bridge configuration (see the following Strain Gauges section). In this configuration, the lead resistances are located in opposite arms of the bridge, so their errors cancel each other out."

To prepare the Wheatstone Bridge, you can place the known resistors with the wires straight into the required pins of the SCC-RTD01.  If you require further help with the Wheatstone Bridge configuration at this point, please reply back.  Further information regarding Wheatstone Bridge theory is available in textbooks, and online, however I can try and find some links if you require them.

Regards,

[cebailey]

It would be VERY surprising for any hardware system with a 3 wire RTD input to NOT do the lead compensation automatically. I don't have the hardware, but, are we sure there is any additional treatment necessary? I do a lot of work with RTDs and have never seen equipment that said it used 3 wire RTDs but didn't do lead compensation.

Historically, RTDs connected to Wheatstone bridges did take advantage of the 3 wire circuit to reduce wiring costs and have an approximate correction (a friend worked out the math carefully and found that the correction is not quite right). For this to work, all the leads must have the same resistance. Modern systems, though, don't use a Wheatstone bridge. They produce an estimate of lead resistance and then correct for it in software, generally.

[gt3000]

Hello,

The Wheatstone bridge is a typical approach for canceling the effect of lead resistance.  As Nitra said, the wiring configuration for '3 wire RTD with Remote Sense' is being used.  The purpose of using hardware with a Remote Sense terminal is that the input now has an incredibly high input impedance.  The input impedance is so great, that the lead resistance becomes negligible.  This is a characteristic of all hardware that allow Remote Sensing. 

Regards,

[Nick_Alexeev]

George T. wrote:

The input impedance is so great, that the lead resistance becomes negligible.

In a 4-wire configuration, the voltage drops across the sense leads is negligible, because current is small, because input impedance is high.

In a 3-wire configuration, the lower lead carries excitation current, and the voltage drop across it is non-negligible.  It also acts a reference lead for measurement.  The assumption behind the 3-wire RTD probe is that the resistance of the upper lead is the same (or close, albeit still non-negligible) as that of the lower lead.  Then DAQ system can estimate the voltage drop across the upper lead and subtract it from the voltage drop across RTD.  Perhaps, the best way is to carry out the subtraction is to do it in analog, because that helps remove 50/60Hz noise that RTD leads may pick up.