The circuit below is basically taken from CN-0099 and is part of a weight cell :
Our problem is that on one of our weight cell units has failed because the ADA4096-2 op amp was broken.
( We use ADA4096-2 instead of the op amp AD8603 suggested in CN-0099 to allow +24V circuit operation).
Measurements with the broken ADA4096-2 (with a 500ohm Rload attached between 4-20mA and 0V):
+IN @ U3.3 was measured to 496mV
SENSE @ U3.5 was measured to 5.996V
REF @ U3.1 was measured to 5.5V
Also the 4-20mA output @ U7.3 was measured to 5.5V across 500ohm Rload.
So, the output 4-20mA output current was actually as much as 11mA (5.5V / 500ohm) even though the current through the 124ohm shunt resistor was only 4mA. Calculation: (5.996-5.5V) / 124ohm. This meant that U7.3 was sourcing 11mA - 4mA = 7mA to the 500ohm Rload.
Measurements after replacing the broken ADA4096-2:
SENSE @ U3.5 was measured to 2.496V
REF @ U3.1 was measured to 2V
The 4-20mA output @ U7.3 was now measured to 2V across 500 ohm which is correct for 4mA.
We need help to understand why the ADA4096-2 op amp in this case have failed in our weight cell application.