On 3/2/21 9:18 PM, Solder Soldier via groups.io wrote:
My approach is to attach the antenna to the NanoVNA to mimic the handie-talkie as best as possible. Hold the VNA as you would a radio and make your measurements. A rubber ducky on a ground plane is not a realistic test and my guess that the engineering is not even optimized for a ground plane.
It kind of depends on what you're trying to measure. If you're confirming S11 for the design, then a ground plane works great - you model it over a ground plane, build it, and see if it matches the model.
If you're making pattern measurements, then you need some sort of "arm and head phantom" to replicate the RF properties of your hand, body, and head. That is difficult. Holding the VNA in your hand is close, but you have a hard time pushing the right button if you're holding the VNA up to your head.
What most people do is model the "hand/head/body" with the antenna to get the design optimized. Then, they model that antenna over a ground plane, which is what they use for manufacturing verification.? If you're real fancy, you have a series of test jigs in various configurations that are easier to replicate and which you can also model. Maybe you have a lossy dielectric clamp that holds the radio/phone next to a dielectric head (or a bag of a mixture of salt water and shredded plastic)? There's a whole literature on making RF phantoms with all sorts of recipes for the material to replicate human body tissue RF properties.
It just depends on what you're trying to do. If you're just curious about the match for various rubber duckies that's sort of different than "optimizing the radiated RF field".? For the latter you'd need good information about the output Z of the radio, which is almost certainly not 50ohms regardless of load. But maybe someone will figure out a way to use the NanoVNA to make a load pull test set.
