A Review of the SpyVerter R2

The SpyVerter is a high performance upconverter that enables HF reception on SDR’s that aren’t able to tune directly to HF frequencies. Like any upconverter it works by converting those lower HF frequencies ‘up’ into a higher frequency range that is actually receivable by the SDR.

Back in December 2015 when the SpyVerter first came out we reviewed the unit and found that it was probably the best and highest value upconverter on the market. It was priced at a similar or cheaper price to competitors, came in a metal enclosure and had excellent performance. The main reason for its high performance is due to the architecture. While most upconverters on the market like the ham-it-up use an ADE-1 double balanced mixer component, the SpyVerter instead uses an H-mode mixer design. This design is harder to engineer, but it provides better dynamic range meaning that strong signals are less likely to overload the upconverter.

The SpyVerter was recently given a refresh, and the SpyVerter R2 is now available. The changes are small and are mostly centered around the clock. The oscillator is now a 24 MHz 0.5 PPM TCXO, run through a SI5351 clock generator to produce the 120 MHz upconversion frequency. A new onboard microcontroller programs the SI5351 on power up.

This change in clock design also now allows you to connect a 10 MHz reference frequency if ultra stable, or phase coherent frequency operation is required. A u.FL connector is provided next to the output SMA connector on the PCB for connecting a 10 MHz reference. Unfortunately there is no breakout hole in the metal enclosure, meaning that you’ll need to drill your own hole in the enclosure to get the u.FL clock cable out. Few people will need this feature however, as thanks to the 0.5 PPM TCXO stock frequency stability is now excellent.

The new design also uses less power, only drawing 10 mA of current compared to 47 mA in the SpyVerter R1. It also has 12 dB lower local oscillator leakage meaning that the gains might be able to be pushed slightly higher without overload. Once again, just like with the SpyVerter R1 the R2 is also powered via the bias tee on the Airspy, and so is compatible with the bias tee on our RTL-SDR V3 dongles.

svr2_1
svr2_2

There’s also an interesting mod that can be performed with the SpyVerter R2. The LO frequency can be modded to run at 58 MHz instead of 120 MHz. 58 MHz is just low enough to avoid the broadcast FM band, and the lower frequency allows the switches used in the H-mode design to run at a lower frequency. This results in an insertion loss better by about 3 dB’s and less LO leakage meaning that the RF gains can be pushed higher. The main disadvantage to this mod is that the lowest input frequency will only be 28 MHz.  The mod details don’t seem to be published yet, but we’ll update this post once they are.

The cost of the SpyVerter R2 remains the same as before at $49 USD. Compared to the Ham-It-Up v1.3 which costs $41.95 USD and does not come with an enclosure or TCXO, the SpyVerter still seems to be the best value. Currently you can buy one internationally from iTead who ship from China, at Airspy.us for US customers, and there are several European distributors linked on the Airspy website.

Disclaimer: The SpyVerter R2 was sent by the Airspy team to us for free in exchange for an honest review.

21 comments

  1. unixpunk

    Has anyone has issues powering the Spyverter R2 with the HackRF bias-T? I can power it from a rtlsdr v3 dongle and usb fine, but with the HackRF, I get some perfectly spaced imaging at 32db gain @ 120mhz. If I power it from USB and turn off the bias-T, I see all the AM stations I’d expect.

  2. Hans Albertsson

    Can the Spyverter r2 upconversion be disabled and any dongle connected to it be used for 24MHz-2GHz directly, without actually physically disconnecting the spyverter r2 box?

    Like what happens if you disable the bias-t feed? Will the antenna signal pass through the r2 box?

    • Beer It

      Short answer: No
      Long answer: No
      As far as I know there is a 60MHz LPF on the frontend (top of PCB) of the spyverter and something like a 120MHz-180MHz bandpass filter on the backend/IF side of the spyverter (bottom of the PCB). And when the power is removed those two filters will block any signal from passing.

      This does sound like a problem until you realise than any antenna that optimal for LF/MF/HF frequencies is going to be pretty much dire at higher frequencies, so you need to swap antennas anyhow.

      They could add two solid-state bypass switches, or an electromechanical switches or a mechanical DPDT switch, but they would all add additional attenuation (each would be like ~0.1 to 2dB depending on technology and price) and would also be a source of additional noise, as well as extra cost. And electromechanical switches may interfere with the coil near the middle of the PCB, which I suspect is the core of the device, and one of the reasons for a metal box.

      So in theory they could redesign it to have lower performance by adding that feature ?

  3. Anonymous

    I not see any issues with microUSB since all my smartphones had it past 10 yrs and only the first one had an issue with it.

    What I do see is noise when tuning below 3Mhz with GQRX, though this might be the quality of my USB power supply and using a laptop.

    • Anonymous

      PS
      This SMA connectors are certified for 500 times connecting, now we should want PL259/BNC on the SDR-stick/SpyVerter too?

  4. Freddy

    Thanks for the tips Bertie,
    I’m French and not professionnal in terminal, but this is what it says :
    “$ osmocom_fft -a airspy=0,bias=1 -v
    -bash: osmocom_fft: command not found”…

    For airspy_rx -b 1 -t 4 -r /dev/null :
    “Average speed 19.9996 MSPS Real
    done”

    • Bertie

      I don’t use OSX.
      Was the bias-T enabled with airspy_rx, do you have any way to actually check that the command enabled the bias-T ? (e.g. a voltmeter)

      As for where the osmocom_fft application was installed, maybe it was not, but it was installed on my Linux machine.

      [email protected]:~/src/gr-osmosdr$ ls apps
      CMakeLists.txt osmocom_siggen osmocom_siggen_nogui
      osmocom_fft osmocom_siggen_base.py osmocom_spectrum_sense
      [email protected]:~/src/gr-osmosdr$ which osmocom_fft
      /usr/local/bin/osmocom_fft
      [email protected]:~/src/gr-osmosdr$

      • Freddy

        “$ ls apps
        CMakeLists.txt osmocom_siggen_base.py osmosdr_source_1rtl.grc
        osmocom_fft osmocom_siggen_nogui osmosdr_source_2rtl.grc
        osmocom_siggen osmocom_spectrum_sense”
        but no Bias-T.. and I have no voltmeter.. Forget it,
        and thank you for the help, 73

    • admin

      What else would you like to see? The performance is pretty much exactly the same as the R1 (that review is already done see the menu or link in this post), side by side screenshots show the exact same thing. The addition of the TCXO is the main improvement.

      • Adam

        Hi,

        as the device is advertised as high performance high dynamic range, I would like to see that performance proved using a two tone test. This is the only advantage (advertised) against the other upconverters. All the rest is same or worst comparing the other upconverters. Let us see the two tone tests. I am really interested in that figures. Anybody experiment with the H-mode mixer is well aware of limitations and the problems especially above the HF frequencies. In the presented upconverter the 120MHz oscillator was used. This is rather high frequency for this type of H-mode mixer. If this type of H-mode mixer is superb against the standard DBM mixers, why they are not used in the VHF frequency region?

        Show me the figures 🙂
        Have fun, as always.

  5. NLA

    I wish they would use a USB B connector instead of micro. B should be more rugged for constant plugging/unplugging of a USB cable.

      • Freddy

        Because not all systems allow Bias-T,
        example : GQRX on OSX :
        “airspy=0,bias=1”, (via gr-osmosdr API),
        doesn’t work at all..

        • Bertie

          That sounds like a fault. But you would need to isolate where the fault is to report it.
          If it was me I’d be running rc10 firmware on my airspy, the very latest gr-osmosdr, and the very latest gqrx.
          And test each one individually and see where bias-t was working or not.
          I’d use airspy tools to test that libairspy is working as expected.
          e.g. open a terminal and run.
          airspy_rx -b 1-t 4 -r /dev/null
          Then I’d use osmocom_fft to validate that gr-osmocom was working as expected (no space in “airspy=0,bias=1”).
          osmocom_fft -a airspy=0,bias=1 -v
          And finally I’d test gqrx to narrow down where the fault lies.

      • NLA

        I am talking about plugging/unplugging USB cable from the device. Granted it would be smarter to do it from the computer. Other than the Airspy Mini, all the others use micro USB. Heck they might even consider USB C. I know they use micro due to its higher insertion rating, but that doesn’t change the fact that micro is a weak connector physically.

Post a comment

You may use the following HTML:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>