Restart after band change (RTL-SDR) ?!

[M*I*B]

Hello forum community,

first of all please excuse my ugly english...

First of all, I would like to say that I have been using SWL for decades, but exclusively analog. For this I use devices from my collection, mostly FRG-7, FRG-7700, Jennen 9R-59 or 9R-4J, ... A MiniWhip (DIY modified according to DL4ZAO) or an active double loop (2 x 3 .8m diameter at a 90° angle, pure copper 12x12mm) with Reuter RLA4 electronics.

Now I wanted to deal with the topic of SDR and got myself an original RTL-SDR V3 to get started with the topic.
Initially I ran it with SDRSharp. But since SDRSharp causes big problems on my Ryzen7 machine (the GUI freezes), I switched to SDR radio yesterday; I like that a lot better too!

But now I have a problem with it, where due to lack of experience I can't judge whether it's my fault (too stupid), whether it's the stick or the software and/or in combination with my computer (WIN10, Ryzen7, 64gb ). That's where you come in

For example, if I'm currently on MW and then switch to LW (Favorites/Broadcast::HF), many fragments of the previous band remain and I'm receiving stations that aren't broadcasting on LW. If I press STOP once and then START again, everything is fine.
This behavior can be reproduced when switching from any band. Sometimes I have to press stop/start while scrolling with the mouse wheel to get back on track.

Can you do anything about it? And if so, how?

Greetz
Micha

PS: I've attached three files to illustrate this. I just did that.
So I hear a station in the 49m band (01_listen), click on 120m and hear/see something completely different there (02_change). After a restart I get what I expect (03_restart)

[jdow]

I'd guess it is a feature of the dongle. The dongles are 8 bit devices. They are VERY easy to overload. Perhaps you could reduce the signal level into the dongle and get improved results. Personally I'd reserve the RTLSDR dongle for 50 MHz and up. An AirSpy Discovery is moderately expensive; however, it's performance over the frequency range you apparently want is stellar.

{^_^}

[Simon G4ELI]

Look at the Ribbon Bar, Home, Radio and reduce the RF Gain. As jdow says, it's easy to overload.

[RadioSDR1]

I listen mostly to SWL myself and first dabbled with SDR with RTL-SDR too and there are some things you can do to dramatically improve SWL on those things. One is to get a filter to block AM broadcast radio. However, this filter will also take out 160m ham band. Another thing you can do is to get an adjustable attenuator...something for TV would be fine. Once you give the radio's front end some protection, then you have to set SDR Console (or any SDR software just right). I would shut off the IF/RF AGC but still use the audio AGC.

Then what you want to do next is adjust the IF/RF gain up to the point of the noise floor of the radio or maybe a little higher. Because raising the gain beyond that point buys you nothing in terms of signal-to-noise ratio on any radio but because it's easy to overload the RTL-SDR, all the more important here. IMHO the RTL-SDR will perform great for SWL so long as the front end is protected and the gain is set just right.

The main advantage I found with the SDR Play RSP1A is it handled artifacts, images, spurs, much much better. You could also add an upconverter and those help a lot with that but I like to listen to the hams and I could never get the frequency calibration precise enough so I was constantly having to fine tune single sideband transmissions. Drove me nuts. Not a problem with the RTL-SDR by itself or the RSP1A by itself but when an upconverter is used, I had trouble. However I did not find the upconverter a problem for SWL on broadcast stations, the very slight frequency calibration issue is not a problem there at all.

[M*I*B]

... that's a lot of answers on the subject; many thanks everyone! I'll have to muddle through first.

But what I can't understand now is the statement about the reduction of the RF gain... that only refers to VHF and upwards, but not to the direct sampling range. And it is precisely the HF range from 100kHz to 30MHz that interests me; there the amplification of the stick cannot be changed in this way.

Yesterday evening I tried another software called SDR++. This effect does not appear there, regardless of whether I have RTL AGC and/or Tuner AGC on or off there or whether I also manually turn the gain up to the stop.

I'm still getting a Nooelec v5 these days. Let's see if he behaves like that too; I will report...

Greetz
Micha

[jdow]

Exactly. On R820T(2) dongles the tuner has an RF path into one of the two input channels. Since the Realtek chip can work using either of the two inputs, I or Q, together or separately the modern dongles seem to hook a balun up to the alternate input. For HF you switch to that channel. It sort of works. You have no real control over the gain with that path since it bypasses everything tuner-ish.

There is another trick that was used before dongles with baluns came around. It involves detuning the VHF "stuff" very thoroughly and relying on HF leakage through the dongle. "Obviously" this suffers potential analog nonlinear performance as well as reduced sensitivity. I'm not sure if this is the mode Simon has used or if he swaps from I to Q in the dongle. Both options might be educational.

That is why I suggested a smaller antenna. A broadcast band rejection filter may also provide a better experience below about 350 kHz to 400 kHz. While the Nooelec dongle is excellent, it is not materially going to change anything. Saving the money and skipping a few packs of cigarettes, if you smoke, for an AirSpy Discovery would be a seriously good idea, especially if you are interested in frequencies below 400 kHz. (I'm not a shill for Youssef. But, I do recognize something that performs well when I play with it.)

{^_^}

[M*I*B]

... learned something new again; ty

The dongles are just the beginning. The NooElec is already on its way. If they both behave the same, then I'll keep the one that offers better reception. I will then test both of them with my notebook and with the smartphone, with a mobile loop and a mobile MiniWhip... It will take a while...

If the whole SDR stuff can convince me, I would like to come to a final, larger solution in the end. I currently have the RSPdx in mind (if there is a matching Linux-based software application for an RPi or similar), or a Malacit. The only problem with the latter is that I don't trust the Chinese cracker and unfortunately contact with the developers seems impossible at the moment; at least they don't reply to my inquiries.

I realize I can't have everything, but a stand alone solution ala Malacit is where I want to go.

Well. Now it's time for the end of the day.

Thank you again for the valuable information. I'll get back to you at the latest when the NooElec has arrived and been tested...

Greetz
Micha

PS_1: Yes, I'm a smoker since round about 45 years... but only self-rolled with natural tobacco (Schwarzer Krauser) without any additives. That's probably why I'm still alive and still have my full lung volume.

PS_2: I use a active splitter between my Loop and my RX which I can use to attenuate or amplify any output. Together with the overlapping control of the two loops at a 90° angle, I find this to be optimal...

[Max]

I currently have the RSPdx in mind (if there is a matching Linux-based software application for an RPi or similar), or a Malacit.

If you are mostly interested in HF listening, do yourself a favour and avoid the SDR Play products in favour of the Airspy HF+. It's superior in every way in the frequency ranges it covers:

https://airspy.com/airspy-hf-discovery/

SDR Play are a better choice only if you also want higher frequency coverage (above 260 MHz). Only real drawback to HF+ is a gap at 6m ham band (50MHz). Also the maximum sampled bandwidth is only 768kHz but that is plenty for serious HF work. Then just use your dongle for the VHF/UHF stuff.

SDR Play use a tuner chip that as I recall was originally designed for UHF broadcast TV receivers and has inherently high gain that then needs taming with attenuation. That will never be the optimum configuration for HF work where excess gain is more of a problem than a help. It does have some good points but it's somewhat of a compromise at HF. Airspy HF+ was designed specifically with LF/MF/HF DXer requirements in mind and performance/cost is extremely good. Many reviews online if you Google Airpsy HF+.

Max

[M*I*B]

Thank you for pointing out the HF+. I'll have to think hard about that.
However, I have already had bad experiences with their software SDR#. It's knitted with a hot needle. Also many links on the page https://airspy.com/download/ (most at the end of this page) also lead to nowhere. Perhaps they can do better in hardware than in software or take care for of the Internet presence?!

They also talk a lot about what great filters they have; the examples sound great... The fact that there is special server software for this is definitely a plus... No questions...

But where are the filters located? Are they in the hardware or in the software (SDR#)?
In the first case the software must be able to address the filters, so I doubt that other SDR software can't address them; As already mentioned, I can't really use SDR#.
In the second case filters are implemented on the software side, then I would be bound to the SDR# (which doesn't work well here) and also the advertising for the great filters on the hardware side is completely out of place...

... all this doesn't leave a good impression ...

At the moment I'm a bit overwhelmed by the many pros and cons ^^
On the one hand, it is correct that my SWL is mainly in the range up to 30MHz, but I also use it for 2m and 70cm as a kind of scanner (currently on an X200U).
Another point I don't understand is why the HF+ has a deadband between 30 and 60MHz. Why this is the case does not seem to be explained anywhere; At least I didn't find it...

... that will be a difficult decision^^

Greetz
Micha

[jdow]

You can avoid aomw confusion with SDRSharp if you run it bare rather than with all the community plugins. Get used to it. THEN add ONE plugin, learn it or discard it, and iterate over all the plugins. Youssef does the SDRSharp tool. He does not involve himself in the plugins with very rare and light exceptions.

In your case as I understand it, I'd get an AirSpy HF+. If you had a serious antenna and were not interested in below 500 kHz use an RX888 has a certain allure; but, below 500 kHz it is pretty numb. So later on with some HF+ experience to prove the concepts you could go to the rtlsdr at VHF on up where it is more or less designed to work natively. Once familiar with that operation (with it's AGC off if at all possible) move up to an AirSpy R2. That's not the best thing around. It is probably the best bang for the buck you can get. It is a little fussy about its computer hookup, though, when used with the 10 MHz sample rate. For plug and go an AirSpy mini might be a better choice. USB2 likes 6 Msps better than 10 Msps.

{^_^}