PT100 Scaling in RSlogix 5000

[InstrumentationUK]

Hi Group,
I used the scaling block in RSlogix5K to display PT100 on HMI but the problem is, every time when I measure the resistance across the sensor and do manual calculation it does not match with the scaled value from RSlogix.
( Please find the attached screen shot of RSlogix Scaling block)

Manual calculation:
How I derived at the formula, Temperature= ( 100- measured resistance)/0.384 ?

PT100 data from its data sheet
Transmitter : 4~20mA Output,
Range: -50/250 Deg C
Detector Element Resistance Value :100 ohms at 0°C

According to data given by, the probe is 100 ohms at 0 DegC and 1 DegC change will cause 0.384 change is resistance ( Ref: http://www.picotech.com/applications/pt100.html ).
So
1DegC change = 0.384 ohms
Now for Example, if measured resistance is 115 ohms then
=>15.09 +100 = 115 ohms ( because 100 ohms at 0DC)
=> (100-115)/0.384 = 39.05 Deg C
I have scaled the sensor in Rslogix as per the data provided, which is -50/250 @4/20 mA and I am getting reading as 37.9 DC

After this, I changed the parameters in RSlogix scaling block to match with the manually calculated value and again after a while when I measure the resistance value and do the manual calculation, the calculated value does not match with the value from RSlogix scaling.

Can anyone please tell me where I am making mistake or please tell me what is the best way to do scaling of PT100 in RSlogix5000.

Thanks for the help in advance.

 

[InstrumentationUK]

Any ideas mates !

 

[Ken Roach]

Disclaimer: I'm no calibration expert. But I use a lot of RTDs and thermocouples.

The ControlLogix can only deal with what the 4-20 mA loop gives it. If you keep changing the maximum and minimum limits so that a particular measured point matches your external resistance measurement, you're introducing an offset and you won't get accurate measurements across a span of temperatures.

Can you use an ice bath to at least establish a zero C reference, to see what the 4-20 mA transmitter puts out at that temperature ?

My guess is that you are using an ordinary multimeter to measure the resistance of the PT100 and that doesn't take into account the resistance of the meter leads. As the Pico article mentioned, even a small measurement error in resistance can lead to a large error in temperature indication, which is why four-wire sensors are used to compensate for connection lead resistance.

You're using the Scale block correctly, and it is functioning correctly.

 

[TConnolly]

RTDs are non-linear devices. The data sheet you linked provides the polynomial necessary to properly linearize it. Use the CPT instruction and that polynomial.

Since you asked for a recommendation on the best way: My recommendation is that the best way is to install an RTD input module and wire the RTD directly to it without the transmitter, using three or four wire setupt to null out lead resistance. The module will return the linearized reading to the PLC, no further scaling necessary on your part.

 

[Ken Roach]

Most "transmitters" that output 4-20 mA perform linearization onboard, which is why you configure them for specific types of RTDs or thermocouples.

Is this a configurable transmitter or something that just performs a straight resistance-to-current-loop conversion ?

 

[shooter]

for one you have to measure the resistance in the several lines
you get a calculation for example AB = 115.5
Bb is 0.5 ohms
so real resistance is 115.0 ohms gives 38.6 celsius
calculation gives 39.05
the line is not straight.
mostly the use a XTR105 converter in these heads see datasheet of it.
it has built in compensation.
what type of Pt100 is used a 0.385 or a 392 (jis)
etc.

 

[InstrumentationUK]

Problem Sorted !
Just used Loop calibrator and scaled 4mA and 20mA properly in Rslogix instead of trying to match the value from PT100.