I contacted KS support and they suggested I post my questions here.
I am designing a multi-stage composite amplifier based around a FDA opamp, to be used in a low noise/distortion converter front end. Despite my best efforts rolling through the RC parts bin, I have been unable to get a design to reach stability. It turns out it is quite unintuitive and a much trickier task than I was anticipating. I figure I need a less experimental/plug-pray approach. So I am looking for a VNA to help me quantify what is going on. I haven't owned or used one before, so I am trying to figure out what I need.
I mostly find myself working in audible and lower ultrasonic bandwidths. So I am looking for a VNA that will reach to 20Hz or so. I would also like to use it for PDN design and general PCB optimisation. And occasionally component Z testing too. The ENA-LF looks like it would suit me best. I have a few questions:
* Is it possible to use its 2 ports to capture 4 port measurements of my diff amp stages? I note that baluns are typically used in such circumstances, but I read that doing this can return inaccurate results. Also, in transforming the amp ports to S-E, I lose the ability to quantify the bal+common parts separately, yes? I.e. for CMRR measurements? It looks like I need more accuracy in the amp models than supplied by the manufacturers. But I believe I need all 16 mixed mode S-parameters if I want it to sim correctly? Is there a simple means to create/compute the 16 terms from multiple 2 port measurements? Does the ENA-LF include the tools to help me do this, i.e. calibrate for the fixtures, etc? Or are there 3rd party solns? Obviously, a native 4 port analyser would suit me best. But it seems the e5061b is the singular unit with 5Hz option. How to get around this?
* I am trying to learn what differentiates the various VNA hardware on the market. From what I understand, the bidirectional couplers at an analyser's S-parameter test ports are typically limited in their LF reach, and this is why most devices stop in the KHz or MHz. So, how does the ENA-LF achieve 5Hz on its 50R ports? What sort of bridges are used? Are they transformer coupled? I.e. are there limits in the performance because of the low roll-off or otherwise? I found a simplified block diagram (below). It shows the high-Z test ports sample the inputs directly, so I can understand that there are no constraints in LF extension there. So, what makes the 50R ports different/special, compared to other analysers? How does it compare to the competition, i.e. the Omicron Bode100, which is also a direct sampling unit, and which I have also been advised to check out?
I have access to a 3577A (and 35677A), through a colleague. Even though it is almost as old as I am and is in an unknown state of cal, I see it also, uniquely, goes to 5Hz too. I am looking at borrowing it to experiment and learn, until I can purchase my own VNA for the lab. I figure I will learn lots getting the old CRT barge to work for me anyway. It seems its S param test set is limited at 100KHz also. Am I able to use the 3577 standalone and just make multiple forwards/backwards connections to my DUT, to get those parameters without the box?
Thanks for reading through the waffle