Airspy HF+ Official Specifications Released

Last month we saw news of the Airspy HF+, which is a yet to be released software defined radio with a focus on high performance reception in the HF bands. Some preliminary specs were unofficially released back then on the Airspy Yahoo forums.

Now over on the Airspy website, the official specifications have been released and they are pasted below. The specs suggest that the Airspy HF+ will have extremely high performance when it comes to strong signal handling. This means that there should be little to no chance of overloading, and thus no intermodulation or spurs.

The goal pricing is to be below $200 USD. If this is true, then it will compete heavily with the $249.95 USD ColibriNANO which is another new HF specialty radio with similar specs.

The Airspy team write:

Airspy HF+ is a paradigm shift in high performance HF radio design. It is a joint effort between Airspy, Itead Studio and some famous chip maker to build a state of the art SDR for HF and VHF bands.

Like most high-end HF receivers, the HF+ uses very high dynamic range ADC’s and front-ends. But unlike the current offerings in the market, it also brings more frequency agility by using high performance passive mixers with an excellent overtone rejection structure.
Both the architecture and level of integration achieved in this design allow us to bring top performance reception at a very affordable price.

HF Tuner

Airspy HF+ achieves excellent HF performance by mean of a low-loss band filter, a high linearity LNA, a high linearity tunable RF filter, an over-tone-rejection (OTR) mixer that rejects up to the 21st harmonic and an IF filter.
The 6 dB-stepped AGC gain is fully controlled by the software running onto the DSP which optimizes the gain distribution in real time for optimal sensitivity and linearity. OTR is a key issue in wide band HF receivers because of the large input signal bandwidth. The output of the IF-filter is then digitalized by the IF ADC for further signal processing.

VHF Tuners

Excellent VHF performance is also achieved by using optimized signal paths composed by band filters, high linearity LNAs with a stepped AGC and an over-tone-rejection mixer and IF filters optimized for their respective bands.
The amplifier gain is switchable in 3 dB-steps and is fully controlled by the AGC processing running onto the DSP. The RF signal is converted to baseband by a high linearity passive mixer with overtone-rejection structure. The low-IF signal is then converted into the digital domain by the IF ADC for further digital signal processing.

IF Digitalization

The IF digital to analog converter has a 4th order multi-bit topology; it features very high dynamic range and linearity. The IF-ADC sampling frequency is determined by a control algorithm running on the DSP. This advanced technique changes the sampling frequency depending on the tuning frequency with the goal of avoiding the disturbances generated by the switching discrete-time sections of the IF-ADC.

Digital Down Converter

Once the IF signal is digitalized, the high sample rate I/Q stream is then frequency translated and processed with cascaded CIC and FIR decimation stages. After every stage, the sample rate is reduced and more the resolution is increased. The final signal at the output has 18bit resolution and the alias rejection performance is 108 dBc. The data is then scaled to 16bit and sent to the Micro-Controller for streaming over USB.

Use it over the network!

Connect as many SDR applications as needed to the HF+, over the Internet or in your own local network with near zero latency thanks to the new SPY Server software.
This setup basically brings all the flexibility of Web based SDRs while still benefiting from the full power of desktop applications. The IQ data is processed in the server with state of the art DSP and only the required chunk of spectrum is sent over the network. What is sent is the actual IQ signal, not compressed audio. This means you can use all your favorite plugins to process the IF, eliminate noise and perform heavy lifting of the signals as you are used to do with locally connected SDR’s.
We have a tradition of building multi-tools, so we made sure the SPY Server runs on 32/64bit Windows and Linux on Intel and ARM processors without any compromises. Low cost Raspberry Pi 3 and Odroid boards are in the party.

Technical specifications

  • HF coverage between DC .. 31 MHz
  • VHF coverage between 60 .. 260 MHz
  • -138 dBm MDS at 500Hz bandwidth in HF
  • -142 dBm MDS at 500Hz bandwidth in VHF
  • +26 dBm IIP3 on HF at maximum gain
  • +13 dBm IIP3 on VHF at maximum gain
  • 110 dB dynamic range in HF
  • 95 dB dynamic range in VHF
  • 120 dB Image Rejection
  • Very low phase noise PLL (-110 dBc/Hz @ 1kHz separation @ 100 MHz)
  • +10 dBm Maximum RF input
  • No Silicon RF switch to introduce IMD in the HF path
  • Routable RF inputs
  • Wide Band RF filter bank
  • Tracking RF filters
  • Sharp IF filters
  • Smart AGC with real time optimization of the gain distribution
  • All RF inputs are matched to 50 ohms
  • 2 x High Dynamic Range Sigma Delta ADCs @ up to 36 MSPS
  • 600 kHz alias and image free output
  • 18 bit Digital Down Converter (DDC)
  • 0.5 ppm high precision, low phase noise clock
  • 4 x Programmable GPIO’s
  • No drivers required! 100% Plug-and-play on Windows Vista, Seven, 8, 8.1 and 10
  • Industrial Operating Temperature: -45°C to 85°C

Typical Applications

  • High Performance Networked HF Radio
  • Ham Radio (HF + 2m)
  • Short Wave Listening (SWL)
  • AM DX
  • FM DX
  • Remote Telemetry Radio Receiver
  • Low Bands IoT

Supported platforms

  • Windows Vista, 7, 8, 8.1 and 10 (For Windows XP, please contact us)
  • Linux
  • *BSD
  • OSX

Minimum hardware requirements

  • 1GHz Pentium or ARM
  • 1GB of RAM (to run your own OS, HF+ barely needs 1MB of memory)
  • High speed USB 2.0 controller
The Airspy HF+ Architecture
The Airspy HF+ Architecture
Notify of

Inline Feedbacks
View all comments

“…over-tone-rejection (OTR) mixer that rejects up to the 21st harmonic” What the heck is an OTR? Is he referring to a “Harmonic Reject Mixer”? If-so then the mixer will reject (to some degree) ALL odd harmonics (and not the even harmonics), not just “up to the 21st harmonic”. If he’s not referring to a harmonic reject mixer, then he should tell us about this new magical mixer.


Is it transmitt capable because it using Si5351 programmable oscillator.


The very first line is “focus on high performance reception”. There is also no mention of a Si5351 being used at all, which is just a configurable clock generator. So you would buy the cheapest high quality mass produced TCXO you find and then use a Si5351 chip to generate the clock signals required by other chips in the design. You may as well be asking “is it transmit capable because it uses a resistor”. Correlation does not imply causation, a lot of RF designs use the Si5351 for clock generation because it is currently in the right price/performance point for most designs.


Open source drivers or I don’t care. My existing airspy is mostly a paperweight to me as I don’t want to run binary blobs on my Linux machines.


So which part of the existing Airspy is a binary blob that is causing it you to use it as a paperweight ? The 100% open source firmware ( ) ? Or the 100% open source host library and tools ( ) ?

I’m also interested what Linux flavour are you running, that has minimum blobs theses days ?
Most Debian/Redhat/SuSE based distributions still have lots of blobs in /lib/firmware
archlinux has lots in /usr/lib/firmware
Maybe gentoo if you can avoid a “emerge linux-firmware” and /lib64/firmware/, /lib/firmware are both empty ?

I personally find myself drawn away from Linux to OpenBSD where blobs are just not allowed and fully rejected, if your hardware will works with blobs it is not supported, but SDR support is poor.


Well, when you use the Airspy, you are tied to the dev’s will.
Example: you are wanting to perform frequency sweeps and .csv export. Well, you can’t do that easily. It’s a shame that spectrum spy doesn’t offer that option. A lot of things using the Airspy seems kind of unfinished. Even the plugin making in SDR# is difficult (the dev often break the backward compatibility).
We are speaking of a ~200$ device here. I’m not a fan of the method, here you go take that open-source stuff and make your own programs.


I’ll make a suggestion that has a utility called called rx_power which if you have soapy installed and configured will probably do what you want and is even supported by

You probably need to change yourself from a mindset of complaining, to one of seeking solutions. And failing that you could always hire a software developer with a specialisation in DSP, and see how long that $200 lasts.


Thanks for your advice. I’m already involved in the qspectrumanalyzer issues reporting and bug fixing. But keep in mind that there are few Airspy users that know Linux, Python and Qt. The Windows instructions for installing the program easily are new (<2 weeks), and it will soon gain more visibility.

I still think it's a shame that the developers are wanting to keep the SDR# so locked, and they are still annoying the plugins devellopers. If their goal is to sell devices, being more open would surely help them to. Perhaps it's their French mentality that prevent them to do so? Btw, it's kind of sad.


They probably just don’t don’t share the same conception of open source you have and only do it as a necessary evil? Or maybe they only give stuff for money?


I’ve yet to see the source code on the binary blob inside a FunCube that runs on the PIC microcontroller or the source code of the binary blob inside a SDRPlay that runs on the 8051 microcontroller at the heart of the MSI2500 chip. You can’t always get access, but you can choose not to buy products from companies who only provide an API. But I can fully understand companies using secret blobs, it means that the hardware is extremely generic, and basically anyone could produce a much much cheaper knockoffs at really low prices with no development time and R&D costs to recuperate.


How does Airspy HF+ support VHF-L TV with only 600 kHz of instantaneous bandwidth? It needs about 6 Mhz of bandwidth.


If it did not mention VHF, I would have assumed a typo and assumed SSTV which is like 3 kHz of .
Initially I thought that may have been a typo and they really meant 3 kHz SSTV (Slow Scan television) which only requires 3 kHz of bandwidth, but most of those are usually at 27MHz and lower frequencies. The only exception that I can think of is the 25 watt SSTV broadcasts at ~145 MHz by the ISS (International Space Station). But none of that is VHF-L (Band I – 47 MHz to 88 MHz), so yea ?


wish i could edit the post ignore the first incomplete line that I meant to delete.

Terry Gaff

Sounds great, I look forward to getting my hands on one. Any chance you could add the ability to select either one of two (or more) AirSpyR2 devices that are connected ? Would be a great help.


Very good, but it should reach at least 500 Mhz to be competitive with ColibriNANO.


ColibriNANO does 10 kHz to 55 MHz direct sampling, and 55MHz to 500 MHz in oversampling mode.
So at a guess they will be switching in a number of fronted filters, lets randomly pick the number 10, and then using Nyquist aliases to their advantage. I’d like to see how it performs, a comparison would be good. But maybe I’m wrong and they are doing something else, if the chips did not have their part numbers ground off you could better guess what way their device actually functions –

DB Gain

The design concentration upon hf performance can only benefit vhf performance, this should be a very nice rig for ambc dxing with the high dynamic range that doesn’t sacrifice sensitivity, as well as for general hf to high vhf pursuits. I find it very interesting it requires no additional driver. Can’t wait to test one.


it only goes up to 260MHz, any hope for 270/280MHz for Ultra Low Noise Floor SATCOM monitoring ?

DB Gain

I suspect a rtl with a turnstyle antenna will do that chore perfectly.