SpyVerter Upconverter now for sale

The team behind the Airspy software defined radio (as well has the popular SDR# software package) have just released the SpyVerter upconverter for sale. Upconverters shift HF frequencies (0 – 30 MHz) “up” by a fixed amount, giving receivers that can’t tune that low like the RTL-SDR and the Airspy the ability to receive HF signals.

The SpyVerter extends reception all the way down to DC and has a 60 MHz low pass filter. Its main selling point is its H-Mode architecture which provides excellent IIP3 performance. This basically means that strong HF signals are unlikely to cause overloading in the up-conversion stage. The good IIP3 performance should improve HF reception when compared to other upconverters even with lower end SDR’s like the RTL-SDR. The reason is that when hit by strong HF signals many other upconverters will overload in the upconversion mixing stage, before even reaching the SDR, thus requiring the need for attenuators or antennas with less gain.

Another selling point is its good performance down to DC, making it ideal for VLF reception.

SpyVerter is designed for optimal performance with the Airspy and can be powered directly by the Airspy’s bias tee. However, RTL-SDR users can also use the SpyVerter by powering it through the micro USB connector, or by using it with one of our RTL-SDR Blog units with the activatable bias tee. 

The SpyVerter sells for $59 USD and comes in a metal enclosure with three bonus SMA adapters. There is a $9 USD discount for Airspy owners.

At these prices combined with its claimed performance and metal enclosure we now generally recommend the SpyVerter over any other upconverter. The designers of the SpyVerter have sent us a sample unit and we will review it after testing it out over the next few weeks, but our initial tests already show good performance.

The SpyVerter upconverter.
The SpyVerter upconverter.
Notify of

Inline Feedbacks
View all comments
Douglas Almquist

Who does this and every other article state that the frequency range of the RTL-SDR.com dongle is limited to 24 – 1766 MHz. It’s simply not true. The R820T2 RTL-SDR v3 can reach frequencies below 24… far below.


I bought a new RTL-SDR Blog R820T2 RTL2832U 1PPM TCXO SMA and I’m pretty sure it only goes down to 24MHz. What am I doing wrong then because I’d really love to tune into AM and SW?


Terrific! It works. I found the Q Branch Direct sampling as per your instructions, then, added a long aerial which I just clipped to the rabbit ears aerial linked to the RTL-SDR and there is reception! Thank you very much for that. Cheers.


It also only gets under 24 MHz using Direct Sampling, which is like cheating. =D


Since this is “H-Mode”, that mean that it can, at least in theory, support TX as well ? Probably not +33 dBm, but maybe +10 dBm ?


If it could handle around +15dBm that would mean that you could use it with a HackRF One as well ( https://github.com/mossmann/hackrf/wiki/HackRF-One#transmit-power ). Well with offset tuning and a frontend BPF, to remove the DC spike and keep spectrum clean.


If this thing is really built around an H-Mode design, then RIP SDR-Play, Ham-It-Up, SV1AFN, etc. How can they build such a mixer for cheap?
For the novices/ignorants who commented and who, I believe, don’t see any difference between poop and cake:


It looks like this HF front end is the best thing that happened to the hobby since years ago. I hope they are kind enough to publish the schematics.

– O


Yes a great deal, Airspy $200 Spyverter $60 plus extra connectors all to do what SDRPlay does for $100 less


Or you could spend $106 less than the cost of a sdrplay, if you already have a RTL-SDR, and just want to see what is on HF. Is HF something that would be of interest to you.


Eat poop. 50,000,000 flies can’t be wrong.
I suspect RTLSDR + SPYVERTER outperforms SDR-Play hands down. Someone to confirm please?

– O


Just $100 dollar less and you can receive all stations at the same time on the same frequency…
Nobody needs dynamic range J


“Nobody needs dynamic range” <= You win. LOL


Internal reference clock: 25MHz 0.5 ppm high precision, low phase noise clock (TCXO)
External reference Clock: 10MHz (configurable up to 100MHz)

Internal reference clock: 24MHz +/-10ppm (crystal)
External reference Clock: none. But you could de-solder the crystal, a few SMD capacitors and connect an external 24MHz reference. Extra shielding would also be required to avoid a major degradation in performance.

The real question should be why is the airspy so cheap, or why is sdrplay so expensive. If you try and compare the two products you have failed to understand that they are in totally different categories.


That’s just great, name the one spec. that’s better than the competitor’s and make the product sound superior.
Does the ‘Do the Math’ commercial from Atari’s first 64 bit game console ring a bell?


A high spec clock in a SDR is a not fundamental component ? The foundation on which everything else sits ?


Yep. But, the SDRFail lacks good software. Its users have to conform with old or slow drip developed software. Sad story for those dumped money on it.


Gotta love the advertising ‘DC to Daylight’, what a load of BS, why can’t the vendors be honest anymore??


Where does it say Daylight ? I looked on the airspy site and I looked on the itead site and I can’t see it mentioned anywhere ?


pigtail+ham-it-up+Aluminum Enclosure is almost $75, the spyverter looks like a very good deal on price alone. Looking forward to see how it compares in performance.


I’m curious what the conversion loss would be, don’t see it in the specs


The typical performance of such a mixer is around 5dB conversion loss and +40dBm IIP3.


The conversion loss should be close to NF declared, so 8dB. On the other hand, declared performance indicate clearly that the measurements of the IP3 were not performed nor the NF and the conversion losses. Declaring some values like better then some value or lower then some value is marketing trick. Reaching the IIP3 of +40dBm with the H-mode mixers using the FST switches is possible but not on such a high IF. Such a performance is possible up to 50MHz.In spyverter the LO used is 120Mhz thus the FST switch is running on 120MHz. To reach the declared the switch should be separately biased and proper transformers used to match properly the input and decrease the losses. Of course, the 120MHz signal should be as squared as possible to switch the FST switches properly. When you comply to all this requirements the OIP3 on such a high IF hardly reach the OIP3 of +30dBm. What I have seen from the available photo, this requirements are not followed so the declared IP3 values = hard to meet. On the other hand, even if this converter is high IP3 unit, what is the purpose when you RTL dongle is poor IP3 performer. You should know that IP3 chain is working opposite than NF chain, so the last part in the chain should have the best IP3 figures, not the first. Another marketing trick is declaring the high IIP3 of the Airspy. Such a high IIP3 can be reach (if reached at all) only when you decrease the gain of the tuner to 0 (zero). You know what is the reception quality when you set the gain of the R820 to zero? Not to mention the NF when the gain is reduced.


Mouthing competition without testing is easy. We all know about the upper performance limit of ADE-1 converters (including yours) : 5 dB CL, +15dBm IP3.
Can someone do the testing to confirm?


Testing the units is the best way. But, this should be done by vendor, to support the figures declared.
You can find the measurements for mine (LNAs) products published. We already discuss the NF of the R820T tuners where real measurements never reach declared 3.5db but we can see that figure as declared. ADE-1 is not a big IP3 performer, this is just a Level 7 mixer and the point of using this mixer was value for the money. Descent performance for the price. Two or even three tone test should be done on the unit to get the idea how much this unit is better from the others.
If the job was done “by the book”, properly, the Airspy should be better regarding the IP3 than the rest of the upconverters on the market, but this should be proved by measurements.