Fast Feedback Loop with 2657A

Hi everyone,

I would like to implement a feedback loop using two SMUs, ideally a Keithley 2657 and a Keithley 2634. The idea is to utilize the fast ADC of the 2657 to measure current values and then adjust the output voltage of the 2634 accordingly.

Here are my main questions:

1. Is it even possible for the two devices to communicate directly and form a closed control loop?
2. If direct communication is not possible: Could this be implemented with a software solution like LabVIEW or Python? However, I am concerned that such an approach might be too slow for my application.

Ideally, one SMU should control the other to ensure the feedback loop operates as quickly as possible.

Maybe you got some input :)

Cheers

Interesting topic.
A few years ago I had a customer use two SMU channels:  one for sensitive current measure and one to generate a voltage signal for a fiber alignment tool.
If simulating a changing current with few hundred Hz sinewave + resistors, we saw about 200usec of dwell time on each step of the following voltage signal.

In that case, the two SMU channels were within the same physical box, e.g., dual channel 2602B.

Can you use the two channels in the 2634B?
Since 2634B does not have TSP-LINk feature, it is limited for coordination with the 2657A.
If strongly determined to eliminate a PC from the mix, you could try the use of tsp-net and have the 2657A send commands over LAN to the 2634B.

Sounds like you are hoping the 1MHz digitizer in 2657A will give you fast loop rate.  But you are still taking single shot measurements, computing the voltage level and needing to source it.  I’d expect other overhead, specifically the voltage sourcing rate of the other SMU, will be your rate limiting factor.