Hi y'all,

I just recently got a NanoVNA V2.2 in an effort to get more into measuring RF circuitry. I strongly believe it's a a clone (I'm sorry, but I only discovered afterwards and a genuine V2plus4 was sadly out of my price range for just starting out...), so I'm no trying to measure whether it's a "good or bad" clone with worse parts (out of spec for some frequencies). Doing some research, I discovered that especially the return loss of Port 2 is usually worse on such clones, which I found was quite interesting since my own measured return loss seemed kinda strange to me (picture from VNA-QT below, calibrated with the included standards and with a through cable from Port 1 to Port 2); especially when comparing to other pictures posted here.
So my question: is seen "ringing" (I don't know how to call it) normal? The return loss seems to be in spec for 3 GHz with about 16dB peak, but not for 1.5GHz with only about peak 15dB).

My version has a metal enclosure so that could be a part of the problem.
Same with the cables, at the moment (I will get better ones) I am using the cables provided with the kit.

Any help would be greatly appreciated.
Also are there any other measurements I can do, to tell exactly if my unit is one of the worse ones (no, I don't have a known-good VNA to compare it against)?

Sorry if my questions are dumb or anything, I'm still quite new to this.

Best regards, l0rdl4nx

The "ringing" is probably due to the test cable between the ports not being exactly 50 ohms.
Try calibrating at the end of the cable, or try another different length or brand of cable to see if the ringing changes.
--John Gord

On Sun, Dec 29, 2024 at 08:00 AM, @l0rdl4nx wrote:

>
> Hi y'all,
>
> I just recently got a NanoVNA V2.2 in an effort to get more into measuring
> RF circuitry. I strongly believe it's a a clone (I'm sorry, but I only
> discovered afterwards and a genuine V2plus4 was sadly out of my price
> range for just starting out...), so I'm no trying to measure whether it's
> a "good or bad" clone with worse parts (out of spec for some frequencies).
> Doing some research, I discovered that especially the return loss of Port
> 2 is usually worse on such clones, which I found was quite interesting
> since my own measured return loss seemed kinda strange to me (picture from
> VNA-QT below, calibrated with the included standards and with a through
> cable from Port 1 to Port 2); especially when comparing to other pictures
> posted here.
> So my question: is seen "ringing" (I don't know how to call it) normal?
> The return loss seems to be in spec for 3 GHz with about 16dB peak, but
> not for 1.5GHz with only about peak 15dB).
>
>
>
> My version has a metal enclosure so that could be a part of the problem.
> Same with the cables, at the moment (I will get better ones) I am using
> the cables provided with the kit.
>
> Any help would be greatly appreciated.
> Also are there any other measurements I can do, to tell exactly if my unit
> is one of the worse ones (no, I don't have a known-good VNA to compare it
> against)?
>
> Sorry if my questions are dumb or anything, I'm still quite new to this.
>
> Best regards, l0rdl4nx
>
>

The ringing comes from mismatch. All parts don't  have  exactly same impedance at every frequency.
Port 2 is  not  so  well matched as other  parts maybe. But I have no figures to  give you about that.

Thanks for the replies!

The problem seems to be just due to bad cables (could be other hardware too, i can't really check that as of now).
Below the curve with calibration done at the end of one cable (still the one's supplied with the unit; I don't have other ones). Thru is now just that same cable connecting both ports, without the female-female adapter and two cables.

(sorry for the bad resolution, groups.io seems to compress images a bit too much or I am doing something wrong...)

What worries me still a bit is that I don't see a "perfect" 50 Ohm Load on the Smith Chart for the same cable. Maybe a bad matched port 2? Or, even more plausible, it is due to me calibrating with an adapter (I dont have a female cal kit) and removing said adapter for the through measurements.

Either way, at least the return loss seems way better now, it should be in spec. I will look out for better cables; Huber & Suhner seems to be a common pick.

Is there any other way, I can check whether my unit is fully functional or defective/bad?

Don't expect a great return loss on the receive port.  Even much more expensive units are not too good.
The maximum uncorrected load match for the Agilent 8753E is specified as 18dB from 300kHz to 1.3GHz, 16dB from 1.3GHz to 3GHz, and 14dB from 3GHz to 6GHz.
Corrected, the match is 55dB, 51dB, and 46dB, but doing the correction requires a full bi-directional system, not just S11 and S21.
If you need better load match, add an attenuator at Port 2.
--John Gord