Hello,
I'm having issues trying to measure the through response of crystal ladder filters with the nanoVNA_V2_Plus4

Bandwidth is at 0.8kHz for first nanoVNA pic and at 3.1kHz in the second.  Changing the bandwidth causes artifacts and the peaks start looking like they are separating away from the center into multiple copies.

My suspect is that the sweep is not allowing the crystal to stabilize. Any guidance or clarification would be helpful.

Reference VNA is an HP8752C.

-Tyler

Tyler,
Try increasing the AVG setting.  This will then give the high-Q resonances time to settle.
--John Gord

Software averaging isn't going to fix anything; even 80x gives the same result.

You have to change firmware.
I had the same problem.
My VNA now is not working anymore so I can't check directly, so what I
understand is that probably
the firmware I've uploaded is in my folder named "oji_experimental_XTAL" of
several years ago,
so you can search for it in the web and see if it is still available or
there is some other release that allows
a XTAL measurement.

Fabio - I5LHY

Il giorno mar 16 dic 2025 alle ore 11:29 septr via groups.io <septr=
live.com@groups.io> ha scritto:

> Software averaging isn't going to fix anything; even 80x gives the same
> result.
>
>
>

Try reducing span and enlarging no of points. The frequency steps are
too large to obtain a smooth curve. Reducing sweepspeed may also help
but I don't think that's available on the nanovna
Regards Rien

Op 15-12-2025 om 08:11 schreef septr via groups.io:

> Software averaging isn't going to fix anything; even 80x gives the
> same result.
>

I think you are trying to use the wrong tool to measure a crystal filter.  Because the nanoVNAv2-plus4 is constantly doing background self-calibration to it's own test port, above and beyond any calibration correction the user may have engaged for a measurement,   the "cw" spectrum of its stimulus has significant spectral width.    This means that the energy from it doesn't make it through narrow devices.  So, yes, it's a problem that the group delay of its IF won't let all the signal through because it doesn't 'wait' long enough, but that's not a setting problem, it's built into the architecture.   Here's the spectrum I measure from "CW" at 10 MHz

This energy simply doesn't all fit within the BW of your DUT.
I think that unless you disable the self-cal feature of the nanoVNA you will not be able to measure devices with BWs small compared to this.
Isn't there advice to this effect somewhere in the documentation?    I seem to remember it.

Give a look at this thread:

INFORMATION about Japanese-origin Hacker Software for NANOVNA V2 PLUS4 for
measuring quartz and quartz filters

Il giorno mar 16 dic 2025 alle ore 11:37 Fabio Morsani <
fabio.morsani@gmail.com> ha scritto:

> You have to change firmware.
> I had the same problem.
> My VNA now is not working anymore so I can't check directly, so what I
> understand is that probably
> the firmware I've uploaded is in my folder named "oji_experimental_XTAL"
> of several years ago,
> so you can search for it in the web and see if it is still available or
> there is some other release that allows
> a XTAL measurement.
>
> Fabio - I5LHY
>
> Il giorno mar 16 dic 2025 alle ore 11:29 septr via groups.io <septr=
> live.com@groups.io> ha scritto:
>
>> Software averaging isn't going to fix anything; even 80x gives the same
>> result.
>>
>>
>>

It could be an impedance mismatch problem. Try matching the impedance with BALUN and UNUN transformers and small 3- or 6-dB attenuators.
It works perfectly for me, and all the intermodulation due to reflected waves has disappeared!
Congratulations!!! Checking quartz filters has never been easy, not even with $12,000 equipment!!!
73 de IW4AJR Loris

The nanoVNA V2plus4 is not suited for measuring crystal filters due to
the analog switch directly at the input. This 4-port switch is used  to
compensate for drift by switching to known standards. I think there is a
software version out somewhere that fixes this behaviour for
measurements on crystals.

Op 13-12-2025 om 06:05 schreef septr via groups.io:

> Hello,
> I'm having issues trying to measure the through response of crystal
> ladder filters with the nanoVNA_V2_Plus4
>
> Bandwidth is at 0.8kHz for first nanoVNA pic and at 3.1kHz in the
> second.  Changing the bandwidth causes artifacts and the peaks start
> looking like they are separating away from the center into multiple
> copies.
>
> My suspect is that the sweep is not allowing the crystal to stabilize.
> Any guidance or clarification would be helpful.
>
> Reference VNA is an HP8752C.
>
> -Tyler
>
>

IIRC Joe Smith on EEVBlog noted that the NanoVNA H4 is not suitable for
crystals or crystal filters. The original NanoVNA does not have a problem
with this. I don't remember the reason off the top of my head or even if
there is a fix. Search the forum on EEVBlog.com.

73,
Sam
W3OHM




On Sun, Dec 14, 2025 at 2:59 AM septr via groups.io <septr=
live.com@groups.io> wrote:

> Hello,
> I'm having issues trying to measure the through response of crystal ladder
> filters with the nanoVNA_V2_Plus4
>
> Bandwidth is at 0.8kHz for first nanoVNA pic and at 3.1kHz in the second.
> Changing the bandwidth causes artifacts and the peaks start looking like
> they are separating away from the center into multiple copies.
>
> My suspect is that the sweep is not allowing the crystal to stabilize. Any
> guidance or clarification would be helpful.
>
> Reference VNA is an HP8752C.
>
> -Tyler
>
>
>
>

Hi, can you try this firmware:
https://nanorfe.com/downloads/20251215/nanorfe-20251215-v2plus45.bin
And select the 0.3kHz bandwidth setting?

Thanks,
NanoRFE Support

Where can I find information on the v2plus45 version? I could not find it on the versions page:

https://nanorfe.com/nanovna-versions.html

Can I use this code on my NanoVNA V2 Plus4

Thanks,

Mike N2MS

> On 12/18/2025 6:24 AM EST nanov2support via groups.io <edelswartz123=gmail.com@groups.io> wrote:
>
>
> Hi, can you try this firmware:
> https://nanorfe.com/downloads/20251215/nanorfe-20251215-v2plus45.bin
> And select the 0.3kHz bandwidth setting?
>
> Thanks,
> NanoRFE Support

On Thu, Dec 18, 2025 at 05:24 AM, nanov2support wrote:

>
> Hi, can you try this firmware:
> https://nanorfe.com/downloads/20251215/nanorfe-20251215-v2plus45.bin
> And select the 0.3kHz bandwidth setting?
>
> Thanks,
> NanoRFE Support
>
>

Thank you for the reply. I think I forgot to type out that my unit is actually the 'Pro' version of the Plus4. So would this firmware you'd suggest that I test going to be the same for the Plus4 and Plus4 Pro?

-Tyler

To everyone else, thank you for trying your best. All the a lot of those suggestions are something accounted for. The impedance is perfectly matched (please look at my HP8752C reference VNA for demo of that).

Hi, the firmware is for both V2 Plus4 and V2 Plus5 (pro version).
Thanks

I loaded the 20251215 into my nanoVNA Plus4.

Where do I find the setting for the 0.3kHz bandwidth setting? I select the BANDWIDTH tab and it only displays 6.2kHz 400 P/s.

Mike N2MS

It looks a little better, but still not close unfortunately.

Can't something just be added to allow a longer settle time (or whatever the fundamental roadblock is)? I would happily wait a longer sweep time to get a good measurement. It's just demotivating that the OG $30 NanoVNA has no issues with crystal filters while the unit I choose for $300+ can't :(. Forgive me if I'm ignorant to the challenges of fixing it.

Hi, it's because the newer designs switch between DUT and reference path each measurement to correct for temperature drift, but it can be worked around by slowing down the sweep. We will add more lower bandwidth options in the firmware.
Thanks and sorry for the inconvenience