As of 2023 there are many badly performing clones on the market. V2/3GHz NanoVNA uses parts like ADF4350 and AD8342 which are costly and clones have been cutting costs by using salvaged or reject parts.
See official store and look for V2 Plus4/V2 Plus4 Pro versions only to avoid getting a bad clone. We have stopped selling V2.2 versions since October 2020, so all V2 hardware that are not Plus or Plus4 are not made by us and we can not guarantee performance.
Hello. I have used my nano VNA on mobile duplexers and it works great. I bought an external duplexer that isolates up to 90 DB. I see that the specs only show the nano VNA going to 60 to 70 negative Db gain. When I hook the external duplexer up I can only get it down to just above -30 DB. Is that adding the -60 max +30? so does that mean I have 90 DB of isolation? How does that work? Thank you
John,
Assuming you have calibrated your NanoVNA (short, open, load, and through), It probably means you only have about 30 dB of isolation in your setup. Beware that you need to be very careful in setting up the test of a duplexer. Poor cables and connectors can easily ruin otherwise good isolation.
The NanoVNA can't directly provide a measurement to -90dB, but it should be able to at show the response down to -60dB or better.
Do a search on "NanoVNA duplexer" to get some ideas on how to do your setup.
--John Gord
On Tue, Dec 21, 2021 at 07:57 PM, <johnjgormley@yahoo.com> wrote:
Hello. Thanks for your reply. No, I don’t believe the -30db is accurate. I can hook up a mobile duplexer and it gets down to -51. Also, since I sent this out yesterday, I tested another external duplexer and it shows the approximate same results. Right around -30. I doubt that both external duplexes are only getting down to 30 DB of isolation. It must be adding the max (60) plus what is displayed.
thanks
john
No, John, you do not add the -60 dB to your reading of -30 dB to get -90 dB
of isolation. You have -30 dB of isolation. If your noise floor is -60
dB, that is the bottom of your dynamic range. You can not "squeeze"
another 30 dB below that.
Are you measuring in the transmission mode: Port 00 to duplexer to Port
01?
After a complete cal and storage of that cal (properly executed and stored
cal), what do you measure on the through configuration without the duplexer
using the same cables you use to measure the duplexer?
Disconnect the through cable. What do you measure for the noise floor in
this configuration?
Dave - WØLEV
On Wed, Dec 22, 2021 at 5:35 PM johnjgormley via groups.io <johnjgormley=
yahoo.com@groups.io> wrote:
> Hello. Thanks for your reply. No, I don’t believe the -30db is accurate. I
> can hook up a mobile duplexer and it gets down to -51. Also, since I sent
> this out yesterday, I tested another external duplexer and it shows the
> approximate same results. Right around -30. I doubt that both external
> duplexes are only getting down to 30 DB of isolation. It must be adding the
> max (60) plus what is displayed.
> thanks
> john
>
>
>
--
*Dave - WØLEV*
*Just Let Darwin Work*
To say in short.. You can not tune a duplexer to 90 or 100db isolation with
that device.. Impossible!!
A kc901 can be used for that but that is way more expensive...
If you use the software satsagen
And TWO plutos.. Both mounted in its own metal case... And with good shielded
cables... You maybe get enough isolation... But the dynamic range is limited..
Means its not easy to setup right..
Getting to over 100db isolation and similar high dynamic range is tricky
Dg9bfc sigi
Am 22.12.2021 18:51 schrieb W0LEV <davearea51a@gmail.com>:
> No, John, you do not add the -60 dB to your reading of -30 dB to get -90 dB
of isolation. You have -30 dB of isolation. If your noise floor is -60 dB,
that is the bottom of your dynamic range. You can not "squeeze" another 30 dB
below that.
>
>
>
>
>
> Are you measuring in the transmission mode: Port 00 to duplexer to Port 01?
>
>
>
>
>
> After a complete cal and storage of that cal (properly executed and stored
cal), what do you measure on the through configuration without the duplexer
using the same cables you use to measure the duplexer?
>
>
>
>
>
> Disconnect the through cable. What do you measure for the noise floor in
this configuration?
>
>
>
>
> Dave - WØLEV
>
>
>
>
>
> On Wed, Dec 22, 2021 at 5:35 PM johnjgormley via
[groups.io](http://groups.io)
<johnjgormley=[yahoo.com@groups.io](mailto:yahoo.com@groups.io)> wrote:
>
>
>> Hello. Thanks for your reply. No, I don’t believe the -30db is accurate. I
can hook up a mobile duplexer and it gets down to -51. Also, since I sent this
out yesterday, I tested another external duplexer and it shows the approximate
same results. Right around -30. I doubt that both external duplexes are only
getting down to 30 DB of isolation. It must be adding the max (60) plus what
is displayed.
> thanks
> john
>
>
>
> \--
>
>
> **Dave - WØLEV**
>
>
> _ **Just Let Darwin Work**_
>
>
>
_._,_._,_
* * *
V2 Plus4 can tune duplexers to -90dB with the appropriate settings.
________________________________
Hi, did you manage to figure this out? I have two duplexers — one factory-tuned that works perfectly, and another one that I tuned myself using the NanoVNA. It’s strange that when tuning my duplexer, I achieved 75 dB of isolation, while the factory-tuned one only shows about 30 dB, even though it performs far better. How is that possible? I assume the factory-tuned unit should have at least 100 dB of isolation, yet the NanoVNA doesn’t show the isolation floor — just a weird 30 dB reading…
How was the factory tuned duplexer tuned? The actual impedance of the
equipment and antenna is not exactly 50 +j0 ohms so the final adjustment could
have been done with the actual equipment.
Mike N2MS
> On 10/28/2025 7:46 PM EDT revdns via groups.io <revdns=gmail.com@groups.io>
wrote:
>
>
>
>
>
> Hi, did you manage to figure this out? I have two duplexers — one factory-
tuned that works perfectly, and another one that I tuned myself using the
NanoVNA. It’s strange that when tuning my duplexer, I achieved 75 dB of
isolation, while the factory-tuned one only shows about 30 dB, even though it
performs far better. How is that possible? I assume the factory-tuned unit
should have at least 100 dB of isolation, yet the NanoVNA doesn’t show the
isolation floor — just a weird 30 dB reading…
_._,_._,_
* * *
I'm not sure I can help, but what frequencies is/are they now tuned to? What frequencies were the duplexers originally made for and tuned for?
Are all of the connectors & cables original or any changes made???
Arnold, KQ6DI
Hi, thanks for your interest in this topic.
Both duplexers are designed for 430 MHz with a 10 MHz offset. I’m measuring both using the same cables.
What puzzles me most is that the duplexer that supposedly has only 30 dB of attenuation (factory-tuned) performs a million times better than the one I tuned myself,
even though the NanoVNA shows 75 dB of attenuation on mine.
That’s just a paradox to me.
In both duplexers, the insertion loss does not exceed 0.4 dB. The SWR is better on the factory-tuned duplexer (close to 1.1), while mine shows around 1.4. But what really bothers me is the attenuation. How is it possible that the better-performing unit shows only 30 dB of attenuation, when it definitely has more, and the NanoVNA should show at least 70 dB (which is the limit for the NanoVNA H4)?
When I tuned duplexers, I added a 2db (or close, 1 to 3db works) attenuators to the end of the cables (at the duplexer) . Calibrate out the pads like you normally would. This “keeps” the effects of the cables out of the measurements
Frank
Totally agree. When I worked in the RF semi industry, we always used 3dB
pads at cable ends.
On Fri, Oct 31, 2025 at 9:00 PM cocopuppy via groups.io <ka2fwc=
bellsouth.net@groups.io> wrote:
Re: duplexer tuning - you didn’t say what your application was but in general, check the bandwidth of your tuned duplexer compared to the factory tuned duplexer. My guess is that the factory tuned one is tuned for a wider bandwidth than you did for yours. If you widen yours to the same as theirs they will probably behave similarly. The wider bandwidth tuning makes it less sensitive to load variations and temperature changes.
When a duplexer is tuned on the bench for maximum rejection and minimum attenuation performance it becomes even more essential to fine tune the duplexer in each of the two paths in service but this may not be essential in your application if the factory tuned one performed sufficiently well - just reduce the bandwidth of yours with the VNA to match it and see if that works for you.
HTH …John, EI1EM
===============
Hi
Is it that your not comparing eggs with eggs. Is your diplexer a band-pass structure (75dB measured) and the factory diplexer is LP/HP network (30dB measure)? I think, as you've declared that your using an H4 Nano, that you've forgotten that above 300MHz your using a third-harmonic mode.
If the factory diplexer is LP/HP, your probably measuring the fundamental frequency leakage (~ 143MHz). If your design is band-pass, your probable seeing a combination of BP fundamental leakage and the third harmonic signal. Neither of these results would be anywhere near correct.
Achieving any sensible measurements on multiplexer configuration is not trivial usually involving additional filtering on the VNA ports to clean the source signal - a V2, which only uses fuundamental signals, would be a good start...hi
Gary
There may be insights you can glean from this posting ?
<https://groups.io/g/tinysa/message/12644>
73
Dave, G1OGY
On Sat, 1 Nov 2025 at 01:00, revdns via groups.io
<revdns=gmail.com@groups.io> wrote:
Only attenuators will eliminate the effect of cable mismatch and make the readings more reliable, which doesn’t change the fact that 30 dB of maximum attenuation is probably not realistic for a duplexer that’s working perfectly. I tuned an identical one using a NanoVNA at the same frequencies and got 70 dB of attenuation, yet the factory-made one still performs much better but the readings suggest it should be performing worse — there’s no sign of attenuation greater than 30 dB.
There are a few problems measuring high-Q or narrow notch type things using a sweep other than raw dynamic range for a wide attenuation bandwidth path.
You will be better off using a fixed frequency generator and not trying to sweep.
When sweeping, the VNA receiver has to reject the strong source pass signal when the transmission frequency is outside the notch. This is different than just measuring something like an attenuator pad, where the source is never strong and only broadband loss is being measured. This makes the source broadband noise and receiver overload dynamic range come into play.
Also, the speed of the sweep can matter. Try narrowing the sweep range as much as possible and sweeping slow.
I pretty much never set cavities using my ENA. I use a fixed generator and a tunable selective level receiver to set notches.
if your measure device has 70 to 90 db max dynamic range (from 0 db to
its own noisefloor) .... its a bit difficult to find a 100 to 120 db
notch :-)
a kc901 can do that ... but not a nanovna v1 or v2
dg9bfc sigi
you can not suck an elefant through a garden hose
Am 02.11.2025 um 00:18 schrieb revdns via groups.io:
On Mon, Nov 3, 2025 at 06:29 PM, Siegfried Jackstien wrote:
>
>
>
> if your measure device has 70 to 90 db max dynamic range (from 0 db to its
> own noisefloor) .... its a bit difficult to find a 100 to 120 db notch :-)
>
>
>
>
> a kc901 can do that ... but not a nanovna v1 or v2
>
>
>
> dg9bfc sigi
>
>
>
> you can not suck an elefant through a garden hose
>
>
Yes, but if the NanoVNA H4 device shows a maximum attenuation of 70 dB and then the noise floor, then even at 100,000 dB attenuation it should still show its noise floor so 70dB (the maxiumum of device), right?
To reply to this topic, join https://groups.io/g/NanoVNAV2