Problem with Instrumentation amplifier Gain INA826

[eengr]

I have built a circuit to measure temperature using 2 wire RTD interface as shown below:


I am effectively using one half of the dual current source (100uA) as 2nd half is connected to 0V-GND (via R41 & R81 0 ohm links). The idea is to use this circuit with Pt-100 & Pt-1000 type devices. The output of amp is going to an ADC. The voltage divider at output (R44 & R45) is to cater for broad range of temperatures using Pt-100 & Pt-1000 with one circuit. I started observing some non-linearity so decided to do some test by connecting a variable resistor (1K) between RT_Pos & RTD_Neg termianls (RTD_Neg is effectively GND)
Using a DVM I took measurements for voltage drops and got the following results:

Resistance connected (ohms)	Voltage drop across resistance (Vin)	Voltage at INA826 Amplifier Output (Vout)	Gain (Vout / Vin)
45.96 ohm	4.596mV	88.11mV	19.17
323.2 ohm	32.32mV	619.4mV	19.16
551.8 ohm	55.18mV	1.057 V	19.15
677.5 ohm	67.75mV	1.2983 V	19.16
771.8 ohm	77.18mV	1.4789 V	19.16
812.5 ohm	81.25mV	1.556 V	19.15
830.3 ohm	83.03mV	1.576V	18.98
886.4 ohm	88.64 mV	1.6052 V	18.109

As you can see than somewhere between 812.5ohms & 830.3 ohms, the gain of amplifier started to drop and it carries on like this up till 1K+ input resistance . I am struggling to understand the reason for this behaviour of the amplifier. Am I overlooking any parameter in datasheet of the amplifier here? It is rated for Single Supply operation from 3V to 36V DC. I am using 5V DC as Rail. The amp has 1uF decoupling cap across its Rail (not shown on schematic) Any help would be greatly appreciated

 

[FvM]

Please review datasheet Figure 12. It shows that the common mode range doesn't include the negative rail for output voltages above 1.5V and Vref = 0.

 

[eengr]

Thanks for the feedback. Learned something new today . I have found an app note explaining these graphs,
https://www.ti.com/lit/an/slyt647/slyt647.pdf


will start reading it now. I might have to ask some questions on this topic if I struggle to understand.

 

[eengr]

Hi there, thanks for the feedback. What I understand from the graphs is that my graph would be somewhere between the graph of Gain = 0 & graph for Gain = 100; as Gain for my circuit is apprx 20.
In any case my differential mode input signal is sitting very close to 0V, which means with Vref = 0V, the maximum voltage I could get at the output is apprx 1V (for linear operation of this Instrumentation amplifier)
The way I understand my circuit and from definition of COMMON Mode voltage, the common mode voltage of my circuit is apprx 0V. so I am sitting on the bottom blue line of the graph.

Is there a way to raise this common mode voltage to say 1.5V (with Vref still connected to 0V) so that I could get wider range on Vout. Would this be a reasonable approach to get wider linear operation at Vout? It would be greatly appreciated if you could suggest a way to achieve this on the existing circuit.