Page 1 of 1

Spectrum Display Noise Floor Randomly Lowers

Posted: Tue May 29, 2018 8:35 pm
by Jen
Not sure why this happens. Maybe someone could help out.

The spectrum display noise floor randomly lowers and then rises up. Then it repeats randomly.
It could go from -85dBm down to -100 dBm. It will at times go much lower, well below -110dBm.

At the same time….

The waterfall display will go from a light to dark color in sync with the spectrum display.

I have messed around with AGC, gain and other settings…no luck.

It happens in both SDR Sharp and SDR Console V3 Beta 2.

I have both a NooElec NESDR SMArt and A RTL-SDR v3 dongle. Happens with both of them.

Computer is as follows:

Processor Intel(R) Core(TM) i5-3570K CPU @ 3.40GHz, 3401 Mhz, 4 Core(s), 4 Logical Processor(s)
Installed Physical Memory (RAM) 16.0 GB
Total Physical Memory 15.9 GB
Available Physical Memory 11.2 GB
Total Virtual Memory 18.3 GB
Available Virtual Memory 13.1 GB
Page File Space 2.38 GB
Page File C:\pagefile.sys
Kernel DMA Protection Off
Virtualization-based security Not enabled

Video card Name NVIDIA GeForce GTS 450

Windows 10 Pro 64 bit

Thanks for all constructive advice in advance.

Re: Spectrum Display Noise Floor Randomly Lowers

Posted: Wed May 30, 2018 4:13 pm
by hotpaw2
With or without the antenna plugged in? If the noise floor only cycles when the antenna is plugged in, then it could be from some periodic RF noise source in your room, building, or neighborhood.

Re: Spectrum Display Noise Floor Randomly Lowers

Posted: Wed May 30, 2018 9:27 pm
by Jen
It does not appear to happen with the antenna unplugged.
Funny thing is that it could happen anywhere in the frequency spectrum.
I'll have to note times and frequencies and see if I can get some kind of correlation.

Re: Spectrum Display Noise Floor Randomly Lowers

Posted: Thu May 31, 2018 1:14 pm
by rtlsdrblog
Sounds like there might be a very strong signal turning on and off, causing the dongle to go from normal operation, to overload, then back to normal.

Maybe a pager?

Re: Spectrum Display Noise Floor Randomly Lowers

Posted: Thu May 31, 2018 8:25 pm
by Jen
Weird thing is I get no indication of any frequency activity popping up in spectrum display.
Happens at HF/VHF and UHF frequencies.

I'm starting to think it is computer related. Will have to test dongles on another system.

Re: Spectrum Display Noise Floor Randomly Lowers

Posted: Fri Jun 01, 2018 3:21 pm
by hotpaw2
Computer fans, power supplies, voltage regulators, plus USB, network, and video cables (especially cheap ones) can radiate a ton of broadband RF noise into a nearby antenna or unshielded SDR.

Motors and LED lightbulbs in the room can be another problematic RF noise source. Nearby radio stations, as well as aviation, weather, and military radar installations (often pulsed) could also cause front-end overload problems in a radio. I would be tempted to make some sort of directional mini broadband antenna for my RTL-SDR to hunt down these type of sources.

Re: Spectrum Display Noise Floor Randomly Lowers

Posted: Sat Jul 07, 2018 4:04 am
by W1ABA
You are seeing proof, right before your eyes that the RTL-SDR is overloading. It can happen in the tuner as the result of marginal mixer operation or in the A to D converter (when it's maximum total power input is exceeded).

With the RTL-SDR's 40 db dynamic range and lack of tuned input filters, it is always in a state of overload.

This is why a signals don't show up on the waterfall-the signals contributing to the overload can be 100's of MHz away from the frequency the waterfall is displaying.

Overload, regardless of whether it's in the tuner or the A to D causes a non linear response, so broadband hash is created. It's inescapable and no amount of filtering will render the overload moot.

With 40 db of dynamic range before an antenna is attached, it's easy to see why the RTL-SDR's overload so easily.

Another indication of overload is the fuzzy 'clouds' that appear in the time domain of the waterfall....those are artifacts of signals that aren't really on the frequency that the RTL-SDR is tuned to.

Once any signal encounters a non linear element, it is distorted (likely by clipping). The abrupt change in the waveform creates wideband spurious responses, which in theory are infinitely wide band. This is how and why massively strong signals in the AM broadcast band can destroy reception at vhf and uhf frequencies.


Just between me and the group, I am working on an extremely high dynamic range super regenerative preselector which will work as an active filter to limit the input frequencies to the RTL-SDR to only those frequencies that are of interest. I have a vacuum tube version running now and it seems perfect-I bought my RTL-SDR to use as a test vehicle to evaluate the super regen preselector's performance...the regenerative design is straight out of a 1930's circa publication. Obviously, conversion to solid state is needed to make it practical.

When the super regen preselector is complete, my next project will be an extremely high dynamic range mixer, with the ability to handle 100's of milliwatts of wideband rf inputs without producing a non linear response.