Baffled, confused. What happened in my FM receiver?

[Skogsgurra]

I was doing measurements om the bench. A white noise source with settings +/-10 V and 100 MSa/s was going via an unshielded wire from the arb to the test circuit and passing close to my little FM radio receiver. When I switched from sinewave to white noise, the receiver started to output something that sounded a little like SW reception, but not quite. Listen:

http://www.gke.org/pub/files/FM receiver with sine and white noise.WMA

I have no idea what happened. I am weak on radio and signal analysis. Can anyone explain what this is and why it happens? It would be nice if I got a little more than "the radio picked up interference".

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[btrueblood]

Skogs,

What is the bandwidth of your white noise source?

It sounds like the bandwidth is large enough (say from 10Hz to several MHz?) that the blips are the source's attempt to cover the audio range, but with frequency steps so large that they come though as audible random blips instead of well-mixed random "static".

 

[Skogsgurra]

It generates noise from a 100 MHz clock. The BW is 5 MHz (-3 dB).

Does that help? Still not "unconfused".

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[VEBill]

The Voice of SkogsGurra! Hello!!

As you surmised, the 10.7 MHz is getting into the IF strip. So the FM capture effect with no modulation results in silence. The 400Hz tone makes sense. You're asking about the weird noise.

What you've discovered is that your digital "white noise" source is actually a pseudorandom pattern that reveals its deterministic repeating properties in the acoustic patterns output from the FM receiver. In other words, it's an artifact of the source.

True white noise (+/- preemphasis) is already available between the stations.

Is the source made by Fluke?

 

[waross]

Is the source made by Fluke? .... .. .... ..

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[VEBill]

"Fluke"

_ _ _ . _ _ . _ _ _ _ _ _ _ . _ . !!!

 

[Skogsgurra]

Thanks VE1
Spot on.It is a PRG. Not made by Fluke. I have another noise generator that is based on a reverse biased base-emitter diode. It doesn' have the same punch (much lower amplitude) so I don't think it will affect my radio. I shall run an FFT on that PRG to see if it is white or rainbow striped.

Do you happen to know any mathematical derivation that casts light on this?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[MikeHalloran]

I think Donald Knuth devoted an entire chapter to PRNGs.
Not easy reading, as I recall.


Mike Halloran
Pembroke Pines, FL, USA

 

[Skogsgurra]

Good ol' Knuth - yes, but lost the book many years ago. And I think it is easier and more rewarding to put a question here and get it answered by you that breathe and live communication and signal processing.

I took a few time domain and FFT with 50 ns and 50 us/div resolution.

http://www.gke.org/pub/files/PRNG time domain and FFT 1.pdf

There is actually a "rainbow" pattern - a little like what you get from a square wave with limited rise and fall times - in the spectrum.

But, I still don't quite understand why the beeps stay on for approximately 100 ms and then there's a new beep frequency for 100 ms. And so on.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[msquared48]

... --- ... ... --- ...

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[Skogsgurra]

Isn't it "Mayday" these days?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[MacGyverS2000]

The keywords to use here are "linear shift register" and "pseudorandom generator". I have Knuth's 3-volume set on the shelf, and once you get past his slightly off "language" for computer architecture, it's quite straightforward verbiage.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[MacGyverS2000]

Oh, add the keyword "feedback" to that "linear shift register" search.

Dan - Owner

http://www.Hi-TecDesigns.com