Other Parts Discussed in Thread: TINA-TI, OPA192
I created a design based on TIPD102 https://www.ti.com/tool/TIPD102#3. For the most part it is working great.
Unlike the original design, which has current between 0-100mA. My design requires much lower current with high precision 0-1mA, with emphasis on 100nA, 300nA, 1uA, 3uA, 10uA
In the TINA-TI simulation (attached), the design seems to be working as expected. But in practice at 100nA and 300nA levels, I am seeing fluctuations in current bigger than what is desired.
For example, at 100nA I see the current fluctuating between 98nA and 102nA.
Design changes:
The relevant schematic of my design is shown below.
- I removed the gain between the 2 stages, not sure if that is a good idea but seems to be working for all the other ranges. RS2 and RS3 both set at 499R.
- I used OPA192 as the opamp.
- RS1 is controlled by a switch TMUX1112RSVR (not shown in the simulation) and one of the following values can be selected: 1K, 10K, 100K, 1M. For the 100nA range I am using the 1M.
I am not completely sure where to start the debugging process and how to reduce this fluctuation, any feedback highly appreciated. I did some voltage measurements and the output from DAC seems stable.
Thank you,