SDR DGPS Shootout: AirSpyHF+ Discovery vs AFE822x vs netSDR

[ChrisSmolinski]

Which SDR is best for DGPS decoding?  For my nightly recording sessions I use my AFE822x, it has always provided higher decode rates than my netSDR. But wait, isn't the netSDR a "better" receiver since it has a 16 bit A/D?  Well yes and no. Or maybe "it depends".

The DGPS band is 285-325 kHz, so these results may be applicable to NDB enthusiasts as well.

In the case of DGPS decoding in particular, you want to maximize the dynamic range, use as many of those A/D bits as possible. The AFE822x has a programmable gain amplifier on the front end. I can crank up the gain to maximum, because I feed my antenna through a LW low pass filter which kills the MW and SW bands, and prevents overloading. I always get better results (more decodes) with the AFE822x vs the netSDR. For general purpose listening, I don't notice much of a difference. The netSDR may be slightly better, or it could be my imagination. But there is no real objective way to tell which one "sounds better".

For a fair comparison, I fed the antenna through the LW low pass filter and then a splitter to each receiver, and recorded the DGPS band over the same time period. Since I had three receivers to test but only a two port splitter, this necessitated running two sets of tests.

For the antenna I used my 950 ft LoG (Loop on Ground) antenna, which is fed via one of my Cyclops antenna transformers and then coax to the shack
https://www.blackcatsystems.com/rf-products/cyclops_rf_ham_shortwave_radio_matching_transformer.html

These tests were done in the late morning and early afternoon. Not prime longwave/DGPS reception time, but it was convenient. Normally I use my 500 ft northeast terminated beverage/longwire, but I did some quick tests with the LoG this morning and it seemed to be producing stronger signals. That is something else to check, is this only the case daytime, or also at night?

All decoding done with Amalgamated DGPS of course, the premier DGPS decoding software. https://www.blackcatsystems.com/software/dgps_decoding_software_sdr.html

AirSpyHF+ Discovery vs AFE822x

AirSpyHF+ Discovery:
Count    ID   ref1 ref2 kHz   Baud City                           Country              Lat      Lon      km     Deg
20       799  044  045  290.0 200  Penobscot, ME                  United States        44.453   -68.776  858    49 
882      778  192  193  292.0 100  Kensington, SC                 United States        33.491   -79.349  723    198
348      929  312  313  296.0 200  St Jean Richelieu, QC          Canada               45.324   -73.317  693    24 
57       838  116  117  319.0 200  Detroit, MI                    United States        42.306   -83.103  590    301
2780     772  198  199  306.0 200  Acushnet, MA                   United States        41.749   -70.889  561    64 
2684     844  094  095  324.0 200  Hudson Falls, NY               United States        43.272   -73.542  489    35 
1034     839  118  119  322.0 100  Youngstown, NY                 United States        43.239   -78.972  427    338
2099     803  006  007  293.0 100  Moriches, NY                   United States        40.794   -72.756  378    70 
1856     806  012  013  289.0 100  Driver, VA                     United States        36.963   -76.562  307    173
3219     804  008  009  286.0 200  Sandy Hook, NJ                 United States        40.475   -74.02   266    70 


AFE822x:
Count    ID   ref1 ref2 kHz   Baud City                           Country              Lat      Lon      km     Deg
396      778  192  193  292.0 100  Kensington, SC                 United States        33.491   -79.349  723    198
88       929  312  313  296.0 200  St Jean Richelieu, QC          Canada               45.324   -73.317  693    24 
2711     772  198  199  306.0 200  Acushnet, MA                   United States        41.749   -70.889  561    64 
2520     844  094  095  324.0 200  Hudson Falls, NY               United States        43.272   -73.542  489    35 
998      839  118  119  322.0 100  Youngstown, NY                 United States        43.239   -78.972  427    338
1998     803  006  007  293.0 100  Moriches, NY                   United States        40.794   -72.756  378    70 
1745     806  012  013  289.0 100  Driver, VA                     United States        36.963   -76.562  307    173
3051     804  008  009  286.0 200  Sandy Hook, NJ                 United States        40.475   -74.02   266    70 


The Discovery wins, with more decodes from each station, as well as picking up several weaker stations completely missed by the AFE822x.


AirSpyHF+ Discovery vs netSDR


AirSpyHF+ Discovery:
68       799  044  045  290.0 200  Penobscot, ME                  United States        44.453   -68.776  858    49 
606      778  192  193  292.0 100  Kensington, SC                 United States        33.491   -79.349  723    198
52       929  312  313  296.0 200  St Jean Richelieu, QC          Canada               45.324   -73.317  693    24 
34       838  116  117  319.0 200  Detroit, MI                    United States        42.306   -83.103  590    301
1895     772  198  199  306.0 200  Acushnet, MA                   United States        41.749   -70.889  561    64 
1912     844  094  095  324.0 200  Hudson Falls, NY               United States        43.272   -73.542  489    35 
624      839  118  119  322.0 100  Youngstown, NY                 United States        43.239   -78.972  427    338
2011     803  006  007  293.0 100  Moriches, NY                   United States        40.794   -72.756  378    70 
1954     806  012  013  289.0 100  Driver, VA                     United States        36.963   -76.562  307    173
3040     804  008  009  286.0 200  Sandy Hook, NJ                 United States        40.475   -74.02   266    70 


netSDR:
Count    ID   ref1 ref2 kHz   Baud City                           Country              Lat      Lon      km     Deg
1695     803  006  007  293.0 100  Moriches, NY                   United States        40.794   -72.756  370    75 
1533     806  012  013  289.0 100  Driver, VA                     United States        36.963   -76.562  340    173
2508     804  008  009  286.0 200  Sandy Hook, NJ                 United States        40.475   -74.02   258    77 


A clear winner here, the netSDR is pathetic on the DGPS band.



The next step is to run the Discovery overnight, recording DGPS. Perhaps sharing the same antenna as the AFE822x, and then another night with different antennas to each receiver.

[skeezix]

Since you used a splitter, was there any isolation between the two SDRs?

Definitely do night tests with the splitter. One when things are quiet (i.e. no stupid t-storms) and even another when the stupid t-storms are around. See how they both react to that interference.

[ChrisSmolinski]

Since you used a splitter, was there any isolation between the two SDRs?

They shared a common ground.  I'll do some more tests today with other SDRs. I am tempted to throw together a 4 port splitter so I can test a bunch of them at once. Could be useful for HF and MW comparisons also. I am a big fan of testing radios against each other with the same antenna at the same time. The tests you usually see where someone switches back and forth between radio A and B are problematic, as propagation can change in just a few seconds.

[RST111]

I just got my Airspy Discovery this week; started playing around with it and comparing it to my Afedri.  I run 2 copies of SDR-Console, and the radios are on the same 40m dipole through a stidsberg multicoupler.  The one difference is the Afedri is using 50 ohm coax and the Discovery 75 ohm (RG-6 quad shield because that's what I had sitting here).

So far, they seem to pull in the same signals.  The main non-scientific subjective difference I've noticed is that the Discovery has a better SNR.  It's a lot quieter.  One thing to note is that I live in a fairly rural area with low noise and no strong signals that overload things.  So I'm the wrong person to ask if anyone is interested in how the radios compare in overloaded situations.

I'll do some more comparisons and post results; I'll post a youtube video with the consoles running side-by-side if I can figure out how to record my screen.

Bottom line - I'm definitely impressed with the Discovery and have no regrets getting it.  It's performing well and actually comes out a bit cheaper than the Afedri.

[Josh]

Was gonna ask about the snr for each rig. If they're not the same, that will effect the outcome as far as comparing apples to apples.

[RST111]

Was gonna ask about the snr for each rig. If they're not the same, that will effect the outcome as far as comparing apples to apples.

Yeah, I'm trying to figure out the right baseline.  The gain options are very different for each, at least how SDR Console implements them.  And the response when one switches gain is different as well.  I'm going to try to do a video this weekend.  I'll show the different gain options and what they do, and those who have more experience can help figure it out.