My assumption here is that to correctly test, I should be using the same antenna on a pair of devices, with no other devices playing part of that mesh (i.e. cheating by providing a better hop). This is to make sure I get a truer reflection of both the transmitting and receiving effectiveness.
Is this a correct or good approach, or is the a simpler/better one?
The easiest way to do that is to flash your LoRa-Board with an RSSI software. I did this to ceck for places to put repeater nodes. 
Can you point me to any?
This is the T-Beam repository created by LilyGO.
I’ve done some modifications so two T-Beams will output the RSSI to the display when connected.
Edit: You the code is now public and you can use it to test your connection and antennas in the field
Get you a VNA and learn how to use it, Thank me later
VNA gives you only S parameters so you know how your antenna looks electrically. What is its impedance and reflection but you know nothing about efficiency and antenna gain.
If you want measure antenna you can use VNA. But if you want know how your antenna is effective you need at least field test. Or better test in chamber on rotator and proper software with generator and reciever / spectral analyzator.
I wish… but, what are some practical ways I can test different antennas for meshtastic use, in the field.
Go to Google and search for the best Antennas that humans have built in the history of Radio Communication 
There are many photos or models of antennas that must not be invented but built for personal use and with the desired frequencies.
You just need to have interest and creativity 
Edit: However the VNA gives you a clear idea of ​​the quality of the antenna. You don’t have to use thousands and thousands dollar equipment to know if the antenna is actually good or not.
As a start you could start with the Moxon Dipole antennas, there are the exact measurements for various frequencies on Google
Download also this App (https://play.google.com/store/apps/details?id=com.daveyhollenberg.amateurradiotoolkit)
Good app find. Words words words
Communication is made by words
However, if someone here wants to build antennas they have to do a little research to understand their needs and what they need.
Words could not help so much but experience and manual skills built over time.
However someone above seems to have problems measuring the effectiveness of the antenna with the NanoVNA, I remind you that the NanoVNA measures both the swr but also the decibels of the antenna. No nasa laboratories are needed to test them.
A deep hole in the spectrum of the NanoVNA means that the radiated power is almost all translated into radiation with no return back into the radio amplifier, that is the effectiveness of the antenna.
However, if we want to give the narrow or wide projection, the shapes of the antenna and the elements that can be added come into play.
Your NanoVNA and Google is the best friend in all the situations
@TitanTronics
“the decibels of the antenna”
“A deep hole in the spectrum of the NanoVNA means that the radiated power is almost all translated into radiation with no return back into the radio amplifier, that is the effectiveness of the antenna.”
What do you see? Almost perfect impedance match and SWR -13.13 dB / 1:1.56 and do you think it will work?
Your deep hole doesnt mean that your antenna readiated that power because you dont know its efficiency.
If you put a 50 ohm Dummy Load you can have an even better antenna than that resistor you put in! 
In fact I’ll show you a perfect antenna too “In reality it’s not perfect at all” it sucks “but measured on purpose to make it look so”.
I think you have a different concept of how you see the VNA working and how the Smith chart works that I totally disagree.
This however is my opinion what I think, since I have a lot of experience, but it does not mean that I am right or not. Maybe you’re right, maybe I’m wrong who knows
Maybe you watch this video? https://youtu.be/6cVYsHCLKq8
Hi.
Are you andreas?
Yes, you are right. I am Andreas…
It is an honour to meet you, your videos taught me so much about LORA and all things RF, Thanks for telling all makers to get a VNA even if it is a Nano, Yet to get my T beams, they should be here by next week while I wait, practising how to make 433mhz antennas of varying types and figuring out how to tune them, So far I have made 1 433mhz Folded end Dipole, please look at my readings and tell me what you think. Please see Finally got me a VNA (noobie)
Really good to know its you.
Really appreciate your help and support and advice…and…
Waiting for your next video
Great to see you here, Andreas! I’ve been a follower of your youtube channel for a while now.
If you haven’t been here in a while and you have some time, you should take a look at this thread, since I know some of the RAK boards have come up in your recent videos:
Meshtastic now has official builds for the RAK 4630 Wisblock board, albeit in an early stage. That board promises to be a very nice low power option at a reasonable price though.
Best way with minimal equipment is to use one good antenna on one end and either find a second known good antenna or evaluate an antenna against the known good which will give you a comparison against your “reference”. Subsequent antennas will be either better or worse, or the same, as your reference. If you are testing a lot of antennas you can eventually find the highest performer and then start using it as your new reference.
You will also need some kind of metric to use to evaluate. BER is only good if you are near the fringe of its range. Because you won’t see anything better than 0 BER. Someone mentioned RSSI but that can fluctuate a lot so it would take some really long term averaging in order to get meaningful results.
Since antennas are reciprocal, you could get a rough gauge of an antennas efficiency operating at close range by using RSSI. Get it up out of the noise and just measure the TX power of the “known good” with the DUT and do the whole reference eval I described above. This will give you a relative measure of the antennas gain. Best scenario is you can get a hold of two antennas that have been characterized to calibrate the link. Then everything is relative to it.
You can use cheap SDR stick and measure power of recieved signal in fiexed bandwidth. It will not be precise but it gives you some results.
