This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Xtr111 4-20mA non linearity at low currents

Other Parts Discussed in Thread: XTR111

Hello,

I am following a TI reference design for a prototype build and have run into an issue where the output deviates from the expectation at low input voltages ~ <1.5V. I did not see this particular problem mentioned in the e2e forum, but if it has already been answered please point me in the right direction.


The circuit is to transmit 4-20mA to an external receiver from a 0-5V input. (valid input signal range is 0.5-4.5V), which matches the TI reference design exactly so that is where i began my design. The schematic is attached.

All resistors are 1% tollerance, the resistors setting Iset and the 4-20 scalling will be changed out to 0.1% parts to reduce static error. The static error, as built, is not a particular problem, it is that the error % changes over input voltage (between 0.5 and 4.5V). I have attached my measurements showing the deviation.
The tollerance of the spescification is ~+/- 0.185mA (25uA Ideally) between the ideal Iout and the actual Iout. As the chart showes Iout deviates from this at VinTx <1.199V.

Desired Voltage Voltage out of Sensor (V) Voltage into Tx (V) Voltage@ Rsns (V) Current @ Rsns (mA) Expected voltage into Tx (Ideal V) Expected Current (Ideal mA) Difference (mA) Difference (V) Deviation   (mA |%|) Deviation   (V |%|)
4.500 4.496 4.595 7.04 18.483 4.5966 18.386 -0.096 0.000 0.523 0.035
N/A 4.229 4.383 6.71 17.616 4.3828 17.531 -0.085 0.000 0.485 0.005
N/A 2.104 2.701 4.118 10.811 2.6811 10.724 -0.087 -0.007 0.810 0.743
N/A 1.803 2.463 3.75 9.845 2.4401 9.760 -0.085 -0.009 0.870 0.940
1.700 1.699 2.381 3.622 9.509 2.3568 9.427 -0.082 -0.010 0.870 1.028
1.600 1.600 2.302 3.502 9.194 2.2775 9.110 -0.084 -0.011 0.923 1.076
1.500 1.498 2.222 3.378 8.868 2.1958 8.783 -0.085 -0.012 0.970 1.193
1.400 1.402 2.146 3.268 8.580 2.1189 8.476 -0.104 -0.013 1.226 1.277
1.300 1.300 2.065 3.157 8.288 2.0373 8.149 -0.139 -0.014 1.709 1.362
1.200 1.199 1.985 3.049 8.005 1.9564 7.825 -0.179 -0.015 2.291 1.463
1.100 1.100 1.906 2.940 7.719 1.8771 7.508 -0.210 -0.015 2.799 1.540
1.000 1.000 1.827 2.836 7.446 1.7970 7.188 -0.257 -0.017 3.582 1.669
N/A 0.945 1.784 2.777 7.291 1.7530 7.012 -0.279 -0.018 3.975 1.770
0.900 0.901 1.749 2.729 7.165 1.7177 6.871 -0.294 -0.018 4.274 1.820
0.800 0.822 1.686 2.645 6.944 1.6545 6.618 -0.326 -0.019 4.929 1.906
0.700 0.700 1.590 2.516 6.605 1.5568 6.227 -0.378 -0.021 6.075 2.134
0.600 0.600 1.510 2.409 6.324 1.4767 5.907 -0.418 -0.023 7.072 2.255
0.500 0.509 1.439 2.312 6.070 1.4038 5.615 -0.455 -0.025 8.095 2.506
 


What would be causing the deviation to slip at low input voltages? Where should I focus my design efforts?
Any assistance in understanding the XTR111 is appreciated

Thank you

  • I think your problem is the input bias current required for current shunt monitor, AD8211.  Consult Figure 11 of the AD8211 datasheet.  These are NOT high input impedance devices. At 24Vdc input common mode voltage the input bias current is about 100uA into each input.  Disconnect the inputs to the AD8211 and check if you are back to desired accuracy.  If you need a high input impedance high common mode (24v) current monitor device that will level down to 100mV (0mA across your 10 ohm resistor) to 4.9V (20mA across your 10 ohm resistor) I will need to design a custom one as I do not know of an off the shelf one that will do this.