ADA4522-2
Recommended for New Designs
The ADA4522-1 / ADA4522-2 / ADA4522-4 are single/dual/quad
channel, zero drift op amps with low noise and power, ground
sensing inputs, and rail-to-rail...
Datasheet
ADA4522-2 on Analog.com
LT6657
Production
The LT6657 is a precision voltage reference that combines robust operating characteristics with extremely low drift and low noise. With advanced curvature...
Datasheet
LT6657 on Analog.com
I have been using the attached constant 1mA current source circuit for quite some time to drive 4-wire RTD sensors.
The circuit uses an LT6657-2.5 reference and an ADA4522-2 op amp to ensure 2.5V is driven across a source-side 2.5k set resistor.
This sources 1mA to the RTD.
This circuit has worked well for for my original sensor resistance values, but I recently tried new sensors that were higher resistances. Here are some example resistance comparisons between old and new sensor:
| Old/working | New | |
| Rsns | 2.1 kohms | 2.8 kohms |
| R_exc_lead+, R_exc_lead- | 178 ohms | 309 ohms |
| R_sns_lead+, R_sns_lead- | 2.1 kohms | 2.9 kohms |
The new sensor causes the voltage measurement across Rsns to be much noisier--about an order of magnitude worse, from ~20uV p-p to >100uV p-p.
My initial thoughts are that the higher series resistance of the excitation leads and sensor resistance are causing the drive circuit to go out of compliance, but I'm not sure why the circuit couldn't drive these higher resistances. The highest output / input voltage U11 would see with the new sensor would be just under 6V, which seems within the (implied) input voltage range of the ADA4522-2 of 0 to ~10.5V.
It also doesn't seem like the reference would run out of headroom.
Does anyone have any suggestions for why the circuit might be performing worse for higher resistances?
Was wondering if there are op amp or reference specs that I overlooked, or chopping artifacts of the ADA4522-2 that might factor in differently for different load resistances.
Thanks
