Current status and HF quality


I am looking for buy a SDR device (currently I just have rtl-sdr dongle).

The device will be used for:

  • Receive radio (pmr, dmr …)
  • Lab usage: receive and transmit modulated signals (FSK, ASK …) for RF developments

But the main usage will be on RX.

I see lot of topics with potential problems, are them HW or SW problems ?

What about the reception quality, how is the LimeSDR R1.4 vs SDRPlay RSP2 and AIRSPY R2 ?

The device will be mainly used for RX, that’s why I want to compare with other SDR device full RX.

What about the bandwidth, In the paper the LimeSDR is able to use 60Mhz bandwidth. Is it really true ?

Should we see 60 Mhz bandwidth with GQRX ?

In practice is it possible ? Does the computer specification limit the bandwidth ?

If you set the bandwidth higher does it reduce something else ? (example the refresh interval)


Like owner of one RTL-SDR v2 stick, two SDRPlay RSP1 and one Lime SDR my best
advice for you is to take look on schematics of all of them. They are not in the same
class of SDR equipement.
Take in consideration that SDRPlay units are ready for RX as is - while LimeSDR is just
last critical block of SDR unit ready for TX or RX on HF or above freq…
On my Lime RX1_W is modified and for HF have one LNA4HF (HF LPF) in front with some overvoltage
protection on signal divider in front of all…


If you are interested in HF RX, and can wait, me personally I’d hold out for the AirspyHF+. 12-bit vs 18-bit SDR is on paper up to a maximum of 64x the distance, if everything else is equal (FSPL equation). They just uploaded the source code to github last night, Support has been in SDR# since January, so I can imagine the hardware landing very soon. Oh and it comes with a rugged metal case by default, granted that is probably a prerequisite for 110dB of dynamic range at HF frequencies. On the downside it is only 660 kHz of bandwidth, but I can’t think of a single signal in the HF range that 660 kHz is too small to be able to receive. There are limits to the physical laws of nature which are hard to work around.

660khz of BW is too narrow for dvb-s2…
LimeSDR-USB was not slated as a ham SDR. Its 71db ADC is mid rage for my interests but not worthless … 130db would be great but 130db @ 60mhz BW is more throughput that even USB3 can handle (i think)

Anyhow who exactly uses dvb-s2 in the HF band ?
The problem with 60 MHz BW with a DR of 130 dB is that thermal noise from the circuit, if at room temperature, will start at around -96 dB. So unless the nearby signals are extremely powerful, part of your range will be underneath the thermal noise floor. Which is great for generation of random numbers.

The frequency range of the airspy HF+ is out of range for that i need and the bandwidth is too small, i want to use the device for receive ham radio, it’s more easy to find a station when you have a biggest bandwidth.

I need to receive 400 - 470 Mhz, 800 - 900 Mhz and 100 - 200 Mhz

About the libraries. Did limesdr provide something for modulate/demodulate (FSK, ASK, OOK, 2-FSK, PSK…) ?

Thanks all for your responses

100mhz of spectrum is outside LimeSDR-USB capabilities. That would be 200mhz sampled at 12bits (100mhz I + 100mhz Q) …

Its why u can only rx 30mhz of spectrum… 30x2 , as i understand it …

Part of the problem is that LimeSDR-USB does not do I/Q image rejection on board. Its done in software on your computer so that limits the amount of data through the USB bus – more so if your stuck with USB2.0 like me.

Any one can use dvb-S2 on HF … its not legal but that dose not stop the recorded conversations being rebroadcasted on 40m either. Ill stick to dvb on the microwave bands where the LimeSDR-USB can do that mode due to the 30mhz bandwidth … thank goodness phase4 only requires 200khz (last i read)

Psk quam etc, isnt realy part of lime — that stuff is done with gnuradio or even fldigi/wsjt through a data pipe into/out-of the lime layers …

Example would be quisk (a linux SDR program) transmitting digital voice in codec2 compression or psk31 through quisk digital sink/source…


I just see in the schematic that the limesdr is not able to rx on frequency <700mhz…

Is it correct ?

If yes the limesdr will not be my next sdr board…

You just need to remove the MN18 inductor. After that it should run on HF with no problems. You can ask Crowd Supply for a modified board. I believe that they still have modified USB Type-A boards but you should order your’s quickly.

Removal of MN18 – an smd inductor-- brought rx of rx1w close to my UHFSDR using the same high Q bandpass filter and antenna.


I checked the schematic and i’m not sure to understand why removing mn18 from rx1_l will help Rx1_W to receive less than 700mhz

Is it possible to do the same with rx2 ?

What is the lowest frequency accessible after the modification ?
Did we loose in “quality” doing this ?


Yes, it is possible to do the same on RX2
On RX1_W with coil removed, from approx 1,5 MHz and above up to 30+MHz when used with LNA4HF

I didnt need a addon LNA … I wonder if the manufacturing tolerances for the chip isn’t as tight as one would hope … and i wish that the first LNA’s DC blocking capacitor was 1000X its value … all of these discussions about HF would not be here i think

It is LNA + LPF for freq bellow 30 MHz so FM and GSM signals are filtered out by LNA4HF
My experience: LNA is necessary for 3,5 and 1,8 MHz

you can “pop off” 4 inductors … 2 on rx1 and 2 on rx2

on RX1L and on RX1W … and the same on rx2

that is the “simple fix” for shortwave … changing the caps (or piggy pack soldering ontop of the original caps) is the next step for better low end performance … yes the coupling caps should have been bigger … but only if you want higher sensitivity on longwave or below … for shortwave the sensitivity with just deleting those inductors (coils) is high enough (comparable to my other receivers)