Got my RSP2 today! First impressions:

General discussions
13dka
Posts: 136
Joined: Sun Feb 14, 2016 8:40 am

Re: Got my RSP2 today! First impressions: Images II

Post by 13dka » Sun Dec 11, 2016 6:00 am

Short excursion to much shorter waves:

Before I continue yapping about HF on the RSP2 again, I wanted to show you the solid improvement in the area around 400 MHz. The ranges between 327-350MHz and 374-395MHz are particularly infested with images from the entire FM band and above on the RSP1. I chose the upper end of that range because on the right you can even see an image from a ridiculously weak DAB station on 178MHz (the only one that can be received around here so far):
RSP1-vs-RSP2-394MHz.jpg
RSP1-vs-RSP2-394MHz.jpg (119.35 KiB) Viewed 10651 times
The FM bandstop filter of the RSP2 was not turned on, the improved bandfilters reliably remove the FM muck and let only a lonely local TETRA station show up on the screen. The antenna used for that shot is still my HF dipole, which works pretty well (for a 2x10m wire antenna) on FM and up to 200MHz, but not so good on 400MHz anymore. Here's what's to see on that frequency range when I use an LPDA antenna actually covering that frequency band:
RSP1-vs-RSP2-394MHz-LPDA.jpg
RSP1-vs-RSP2-394MHz-LPDA.jpg (107.41 KiB) Viewed 10651 times
The antenna rolls off below 200 MHz, so it's suppressing the FMBC band nicely but the RSP1 is still adding a little muck. Not sure why the data bursts at 396MHz show up much clearer on the RSP2.

Back down to HF...

... and the question whether or not the RSP2 is also moar betterer on HF. What I can say for (almost) sure is that the 50 Ohms ports on the RSP2 are not working identically to the RSP1. The RSP2 seems to be more finicky about levels behind the LNA in general, and behaves differently on the 50 Ohms ports than the RSP1, most noticable when the LNA is off. Take a look at this composite screenshot:
RSP1-vs-RSP2-InputComp-16MHz-8MHzBW.jpg
RSP1-vs-RSP2-InputComp-16MHz-8MHzBW.jpg (105.25 KiB) Viewed 10651 times
All parts of that screenshot were acquired with roughly the same overall GR. What I meant to show there is:

- At the bottom you see the RSP1 with the LNA off, for comparison.

- In the section above you should notice that the RSP2's 50 Ohms port can be pretty deaf in comparison, even in the upper half of the SW band, when the LNA is off.

- If I add a little gain from the LNA (and reduce IF gain if necessary) I still can't match the performance of the RSP1 with its LNA off for some reason, also it looks a little noisier in the center.

- Only the Hi-Z input on top shows a proper match to the RSP1's performance at the bottom. That's why I marked the CODAR (coastal radar) signal on the left side, it had a nice constant level and reflects the differences pretty well. Obviously, the Hi-Z setup changes my dipole's sensitivity and noise distribution a bit, but still the crisper and less noisy looks of that section reflect what I hear.

So it looks as though the answer to the question above is that the RSP2 has at least the same (if not better) HF performance as the RSP1, but only with the Hi-Z port. However if you tweak IF and LNA gain for the optimum, the difference might be rather marginal, even when you use the SMA ports.

I wish I could repeat these comparisons with a bunch of other antennas, since my 2x10m dipole is quite an oddball ad-hoc construction. It should be a very high-ohmic full-wave antenna over a stretch around 13-14MHz, and a low impedance 1/2 Lambda dipole around 40m (everyone, correct me if I'm wrong!). It has one of those NooElec baluns in the middle because I wanted to compensate for the high impedance around 20m, and hoped it would deteriorate the performance around 6-7 MHz because that's the range that brings up the strongest signals the RSP's input has to tackle. Now the RSP2 has a Hi-Z input and when I connected the dipole directly to that input, it did not really behave differently than on the 50 Ohms inputs. But when I used a 2nd of those baluns in reverse to "undo" the first balun's impedance transformations, the picture changed a bit, but only on the Hi-Z port.

Reason: No reason

g1hbe
Posts: 494
Joined: Sat Jan 17, 2015 3:28 pm
Location: Cheshire, UK

Re: Got my RSP2 today! First impressions:

Post by g1hbe » Sun Dec 11, 2016 9:33 am

All agreed. I find the HF performance via the Hi-z port a big improvement over the RSP1, especially the lack of FM images. It's actually very good and comares well with my other (far more expensive) receivers. I am particularly impressed with the LF performance, which is very clean for such a low cost RX. I still use my homebrew filter bank (0-400KHz LPF//0-1.7MHz LPF//160m amateur band BPF// HF >2MHz HPF), but even without it the performance is commendably clean.
On VHF UHF, things are improved. There is still the DAB image problem around 250 MHz+ which clobbers the weak satellite signals. Moving the converter LO plan to another setting helps get around some of the mess, but not all. I find this part of the spectrum to be a bit poorly.
Re the AGC and LNA settings - there is a bug in SDRUno which causes the LNA gain to be set at zero (or GR max) following a LO plan change or a sampling rate change. You can't tell it's happened by looking at the GAIN slider, as this graphic is left untouched - all you know is that the band has become worryingly quiet... All but the strongest signals to fall into the noise and it caused me a lot of grief at first. I have informed support of this and they have confirmed it.
Overall, I'm mucho impressed.
HF antenna here is a Wellbrook 1530 loop on the Hi-z input for all freq's below 30 MHz. VHF/UHF a selection of homebrew dipoles and yagis fed to SMA1.

Reason: No reason
Andy

Post Reply