I flipped a coin about three weeks ago and ordered an S-A-A-2 (NanoVNA-V2) analyzer for our ham club. We already had one the other three variations (-H_v3.4, -H4 and -F) and have been very satisfied with their price to performance ratio. Adding the -V2 for testing above 2 GHz was an easy decision. The -V2 arrived today - about the same time as some other members have reported receiving their units.
The NanoVNA-V2 was $65 usd from Tindie, including shipping. Tindie filled the order promptly after receiving my PayPal payment, but then the shipment was stuck in transit at the Shanghai postal depot for over a week - most likely due to COVID-19 delays. Given the circumstances, 3-weeks for delivery was understandable. As of this writing Tindie is out of stock. There are two eBay sellers of the NanoVNA-V2, but they both are asking more than double the Tindie store price (no thanks!).
The NanoVNA-V2 was developed by OwOComm under the design leadership of Gabriel Tena-White. Gabriel is well known and respected in the on-line technical community. OwOComm originally developed a VNA, let's call it the S-A-A-1, prior to the release of the NanoVNA without a display and a top end frequency of 300 MHz. Before it could be manufactured, hugen's NanoVNA burst on the retail scene with display, a top end frequency of 900 MHz, and priced below $100. OwOComm met with its client and they decided not to manufacture the S-A-A-1 and instead started on a new development called the S-A-A-2. OwOComm does not sell or manufacture the NanoVNA-V2. Sales and support are by the HCXQS group. HCXQS does not have a technical support forum or GitHub page that I could locate. Their initial production run of NanoVNA-V2's appears to have been about 150 units (Tindie reported 144 orders before posting their out of stock notice).
Un-wrapping:
The NanoVNA-V2 came in a small cardboard box and was packed in bubble wrap, no accessories included. The packing was sufficient to survive the 3-week land-air-land journey.
The contents of the storage box included:
NanoVNA-V2 (with no battery) Note: developer recommends 1000-2000 mAh Li-On batt, 6 x 40 x 60 mm, connector: JST-XH 2.54mm.
No USB cable (micro USB data cable required for connecting to PC).
No SMA RF cables (RG316, SS405, or RG405 cables recommended by designer).
No SMA calibration kit.
No SMA female to female connector for through calibration.
If you don't already have any of the above accessories you will need to account for their cost in the overall purchase price of the NanoVNA-V2. My unit arrived un-calibrated and would not have been useable if I didn't already have a calibration kit.
Immediate visual impression compared to the original NanoVNA
The original 2.8" NanoVNA did not have a case and its power switch and multi-function thumb-wheel switch felt a bit flimsy. Subsequent -H and -H4 models have an ABS case that provide guards for the side mounted port connectors; additionally, the -H4 has better quality switches.
The appearance of the NanoVNA-V2 harkens back to the original 2.8" NanoVNA. It has two PCB's sandwiched between standoffs with a 2.8" display on top. The multi-function thumb-wheel switch has been replaced by three small push button switches. The push-button switches were placed between the SMA connectors so users designing a protective case will need to be imaginative. The PCB version is marked as SAA V2_2.
The NanoVNA-V2's 2.8" display will be a disappointment to user's hoping for a larger display. My understanding from past posts by Gabriel (the board designer) is that a drop-in 3.2" display is available for user upgrade. The on-screen layout of the display will be immediately familiar to anyone who has used a previous NanoVNA variant. The display firmware seems to be based on edy555's ver 6.0 firmware.
PC Software:
None of the current PC software (NanoVNA-saver, NanoVNA-Sharp, TAPR, WebApp) work with the NanoVNA-V2. Rune has ordered a NanoVNA-V2 and has indicated that he intends to support the device in NanoVNA-saver. Erik has also indicated that TAPR may support the NanoVNA-V2. The NanoVNA-V2 has a native PC program called NanoVNA-QT Software. It is a port from an open source VNA program and some functions are not relevant to the NanoVNA-V2.
I downloaded the USB-Serial port driver and NanoVNA-QT software from, . There are installations for Windows (7+) and Linux. The Mac OS is not supported - yet. On Windows you have to load the driver package first. This went smoothly for me on Windows 10. The archive for NanoVNA-QT should be un-zipped to a desired directory and then the executable in the directory can be run without any install required.
Using NanoVNA-QT is different than NanoVNA-Saver and takes some getting used to. It seems to have most of the features you would expect from this type of software and connected to the -V2 without issue - once the correct COM port was selected. From within the program you can perform up to 1024 point sweeps, which eclipses the 201 point maximum sweep on the device. Overall the program gives a good first impression, and for updating the firmware of the NanoVNA-V2 is the only available option.
NanoVNA-V2 does not support terminal commands. I find it nice when using the NanoVNA to load up a terminal program and type in console commands to remotely program the device. The design of the NanoVNA-V2 does not support doing this. Gabriel indicated that a python class might possibly be ready when the V-2 was launched, but I was unable to use the one at , at least not in the way I use the class for the NanoVNA.
While the V-2 is connected to NanoVNA-QT, the display is blanked with a version message. When I closed the NanoVNA-QT program, I couldn't find a way of restoring the display for manual use without cycling power.
In Use:
Turning on the NanoVNA-V2 presented a version screen of NanoVNA V2_2 Jan 18 2020. I checked the NanoVNA-V2 GitHub site for newer firmware releases but none were posted. You'll want to take care of what operations you perform until back-up firmware becomes available. Updating the firmware, as previously noted requires NanoVNA-QT. STM DfuSeDemo utility is a thing of the past.
The specified NanoVNA-V2 frequency range is 50kHz - 3GHz, but as reported previously by Wolfgang, the top measurement frequency can be set to 4.5 GHz- with good performance to 3.5 GHz. The V-2 operates only using fundamentals and does not use harmonics for higher ranges as the other NanoVNA variants do. The waveforms from the two V-2 synthesizers are not sinusoidal so there are still plenty of harmonics in the output spectrum, but unlike the NanoVNA, you don't have base frequencies higher in amplitude than the signal you are generating.
The menu structure for the NanoVNA-V2 mimics that of the NanoVNA. Calibration, Trace, Sweep and Marker settings all work the same. There are a couple of new options for setting the number of display points, and flipping the display orientation.
Under the hood:
The NanoVNA-V2 is powered by a GD32F303CCT6 processor. It has 48KB of SRAM and 256KB of flash memory which is comparable to the NanoVNA-H4.
Published specifications are:
Frequency range: 50kHz - 3GHz
System dynamic range (calibrated): 70dB (up to 1.5GHz), 60dB (up to 3GHz)
S11 noise floor (calibrated): -50dB (up to 1.5GHz), -40dB (up to 3GHz)
Sweep rate: 100 points/s
Display: 2.8'', 320 x 240
USB interface: Micro USB
Power: USB, 300mA
Battery: not included. Includes charging circuitry.
Maximum sweep points (on device): 201
Maximum sweep points (USB): 1024
VNA-QT software supported platforms: Linux, Windows (7+), Mac OS planned
Since the device arrived un-calibrated, the first order of business was to perform an OSLT cal. There were a few differences noted from the normal NanoVNA OSLT calibration. The first was that there is no isolation step. The second is the save step does not change the display notation to the saved channel (i.e. saving to location 0 did not change the notation to C0). The third was that cycling power did not automatically recall save location 0. The display always comes up in the un-corrected state and you have to manually recall 0 from the "RECALL" menu, at that point the C0 notation is displayed.
I purchased a 1 - 2GHz BPF that I have been meaning to test and it seemed like a good candidate for the NanoVNA-2. The attached photo shows the filter measured from 20 MHz to 3 GHz. The filter skirts and band pass are as specified by the manufacturer.
The NanoVNA-V2 is a welcome addition to my club's test instrumentation. Judging by comments from some NanoVNA-H owners, the 2.8" display is going to be a disappointment. If you can get by the display size and judge the NanoVNA-V2 on its performance and cost, you will be hard pressed to find a comparable 3 GHz device.
- Herb
