Over on YouTube Leif SM5BSZ has uploaded two new videos. The first video shows a set up that compares the Airspy and the SDRplay RSP on several lab tests that test for dynamic range performance at various frequency offsets. The Airspy definitely shows better results, but Leif notes that the differences are fairly small. The Airspy and SDRplay are two SDRs that compete in the mid range SDR price bracket.
As lab tests can only approximate real world performance, in the next video Leif does a HF reception comparison on a real world antenna. In this video he compares our RTL-SDR.com V3 in the special direct sampling HF mode, a Funcube Pro+, SDRplay RSP, Airspy+Sypverter, Afedri Net, and an FDM-S1. The test injects an artificial signal and combines signals from a real antenna via an adjustable attenuator. Leif adjusts the attenuator to increase the antenna signals until the test signal strength is degraded by 3dB from reciprocal mixing/overload. That attenuation setting is then recorded.
The results for the daytime and nighttime results results rank the SDR’s in order from best to worst: FSM-S1 ($400 + shipping), Afedri ($259 + shipping), Airspy+Spyverter ($218 + shipping/$149 + shipping (mini)), SDRplay ($129 + shipping), Funcube Pro+ ($155 + shipping), RTL-SDR.com V3 direct sampling ($20 incl shipping). Interestingly the performance seems to correlate nicely with the unit cost. Of course the V3 in direct sampling mode can be significantly improved by using filtering on the front end, or just by using an upconverter and quadrature mode instead.
At the end of the video Leif also shows a final ranking of the HF performance of all radios tested in his previous videos.
RTL-SDR enthusiast and blogger Akos has recently uploaded three new articles. In his first article he discusses what he believes is the differences and advantages of Generic vs Premium branded RTL-SDR dongles.
In his second article he shows how easy it can be to perform the direct sampling mod on newer dongles, as most have the direct sampling break out pads. He shows how it can be as easy as sticking a wire into these holes. Please note that if doing this we would caution you to take decent ESD precautions as these pins are not ESD protected.
In the third article he reviews the recently release Nooelec SMArt dongle. The SMArt is a new RTL-SDR variant which comes in a smaller black case, cooling via thermal pads and with an SMA connector. With these modifications it is very similar to our RTL-SDR.com units, however the one advantage of the SMArt is that it is small enough to fit two side by side on closely spaced USB ports, like on the Raspberry Pi. In the post he shows what is inside the SMArt and discusses various points such as heat generated, included antennas and performance.
Tim Havens is an avid CW operator on the ham bands and primarily uses his Yaesu FTDX-5000 transceiver for this purpose. At the same time he also uses a software defined radio coupled with an upconverter as a panadapter by connecting the SDR to the 9 MHz IF output of the Yaesu.
To get around this Tim decided to use the Airspy in a special configuration. First he used the external clock input of the Airspy to connect to his Jackson Labs “Fury” GPSDO. This device uses GPS satellites to generate a very accurate 10 MHz clock, with almost zero drift. Secondly, to get around the need for an upconverter with it’s own frequency drift he used the ADC1 direct sampling input ports on the Airspy to connect to the 9MHz IF output of his FTDX-5000 through an extra band pass filter and LNA.
Tim writes that he will soon update his post with more images and a video.
So in order to optimize VLF reception, Martin built an external frequency equalisation network which has the following components and functions:
2MHz Low pass Filter – to minimise alias signals originating at 30MHz
20dB Variable attenuator – to set the overall signal level fed into the dongle
Switched LF roll-off – to optimise the performance at frequencies around 10KHz in the presence of strong lightning surges
-10dB notch at 198KHz – to reduce level of BBC R4 broadcast station in the LF band
-10dB notch at 800KHz – to reduce level of local broadcast stations in the MF band
His screenshot results show that his filters work well and significantly reduce the effect of lightning pulse noise at 9 kHz. With the filters in place and properly optimized with the attenuator and various switches, he is able to receive Russian Alpha navigation signals at frequencies around 12 and 14 kHz and the 300 to 500 kHz aeronautical and maritime navigation bands.
Previously we posted about Japanese RTL-SDR experimenter Nobu and his work in prototyping a new 14 MHz low pass filter and galvanic isolator for use with the RTL-SDR. The low pass filter improves reception when using the RTL-SDR in direct sampling mode by reducing out of band interference and the galvanic isolator (isolation transformer) reduces computer and other noise when using the RTL-SDR with an upconverter.
Another product Nobu is working on is an isolation transformer (aka Galvanic Isolator) which can be used together with an upconverter to help reduce noise generated from common ground sources such as the PC. The isolation transformer is inserted between an upconverter and antenna.
In the image below Nobu shows the effect of inserting the LPF . An interfering FM broadcast band signal is removed after inserting the LPF.
The image below shows the effect of the isolation transformer showing a clear decrease in noise floor and increase in signal strength.
The Airspy software defined radio comes with breakout pins that connect directly to the analogue to digital converter (ADC) inputs. By connecting an antenna to these inputs together with experimental software it is possible to receive frequencies between 0 – 10 MHz. This is the same trick used in the RTL-SDR direct sampling mod. However, to get decent reception and to keep the Airspy circuitry safe, a low pass filter and isolation transformer are required.
Over on YouTube W9RAN discusses this mod in his latest YouTube video.
You can get these changes from https://github.com/mutability/rtl-sdr/ (you’ll need to build from source yourself). There should be no application changes needed, just tune as normal. (gqrx needs the “no limits” option turned on) These changes work by limiting the tuner to a range of frequencies that it can reliably tune to, then allowing tuning beyond those bounds by making the 2832’s downconverter do the final bit of tuning. This can add up to 14.4MHz to each end of the range. Also, the tuner is switched to low-side mixing at the top of the range which gives a bit more range there. The practical range is limited by the width of the IF filter and aliasing effects at the extreme edges of the downconverter’s range. I’ve been able to pick up broadcast AM and amateur CW/SSB down to around 15.5MHz without too much trouble. I’d be interested to know how this works for others. Also.. these changes are likely to have broken offset tuning, direct sampling mods, and tuners other than the R820T, as it touches all those areas but I only have an unmodified R820T to test against. If you have different hardware and are willing to spend some time testing then please let me know. I expect that the range of the other tuners can be extended in the same way with not much trouble.
Over on the Reddit RTL-SDR discussion board there has been talk about this patch. Most users are reporting that it works well down to around 15 MHz, but some people are reporting that they have been able to receive signals down to around 4 MHz. Testers also report that this modified driver works much better than the no-hardware direct sampling mod patch released a few months ago. Update: Commenter J.B has been kind enough to upload a ready to go Windows binary for SDR#. Download here https://db.tt/0JuVpWBL. Simply copy the files in the zip into the SDR# folder.