Z7020/AD936x RF performance data

[yramgu]

Hi all

I recently acquired a LibreSDR and was frustrated that I could not find any test / RF performance data about the platform. So, I decided to test it myself (I have access to a professional lab) and created a website to make my data available to whoever wants it:

https://yramgu.github.io/Z93-Lab/master.html

I hope it can be of use, and I'm happy to receive any feedback!

[F5OEO]

Awesome ! Very interesting ...I will defintively link to it on the readme ! Thanks also for this CC contribution .
Hope we could have such valuable documentation on other clone board.

[yramgu]

Yes my plan in the long run is to extend to other boards. Since I was not quite satisfied with LibreSDR I will purchase a PlutoSky next. It has a PA, and we can find some "test" data (Tx power) so I can see if they're lying or not :)

Now obviously I won't buy them all so if someone wants data for a specific platform and can provide a sample, I can definitely run it and add to the website

[PA3AEF]

Very interesting. Thanks for great work! You mentioned S11 is ok, but -5dB isn't much imho.

[yramgu]

I agree with you, by "OK" I mean it's "At least not catastrophic" (not very scientific I know). You will definitely not get the best out of your device with that kind of RL but it will still work reasonably well

[Exteec]

yramgu - do you have an access to other boards mentioned? It's valuable data to verify what can be expected from different clones.

[yramgu]

Hi Exteec, no as I mentioned above I currently have only a LibreSDR. I plan to buy a plutoSky so I will test this one next but I can't buy them all, so for other models I might need help from people to lend me samples for testing (which I will of course return after testing). They would have to be "unaltered" samples to make them representative of a standard unit

[itaniepitas]

For the price range in which someone can acquire such a device for experimentation, its technical specifications are perfectly fine. If someone is looking for high-end devices, they should check out Ettus and similar companies.

[Exteec]

You are right, however I'm curious what real world advantages on RF path we could expect from NI devices. Unfortunately I don't have an access for such anymore. On the other hand, I'm not convinced that there would be night and day difference when we look at S11, input linearity and power output. What I would expect would be bit better image rejection, LO leakage and maybe timings.

[itaniepitas]

I completely understand what is happening. Knowledge gained from performance measurements is good, but on the other hand, we have to consider what kind of measurement results you can get when you know that a genuine AD9361 chip costs around 400 USD on its own, and the entire device with components made from exotic materials costs half as much as the chip.
Also, I believe that even identical models in the lab will have different measurements due to the production line and the lack of serious QC. And finally, I am not against measurements so that there’s no misunderstanding.
I just don’t place absolute importance on them, since all the SDRs I have tested generally did what they were supposed to in order to meet the needs of the hobby. We are no professionals.
At last all these devices are lacking in firmware. Whatever exists of it is a meager copy of Analog Device. There is nothing written entirely for these devices and for their design in each model, with small or large differences in the hardware. A clopy paste situation. :-)