Why does the noise floor change with zoom level?

Sergey R5AU

1. When you switch off yours 6700(may be on 6500 too) pulsed noise available on Lin-out and Phone output. This noise disappeared after switch radio ON 2. waterfall level on panadapter are different with changing panadapter sweep, look into the images: can some body comment it ?

Michael - N5TGL

Sergey, 1. The pulsed noise is from the GPSDO. It's a known issue. 2. Noise level does change as you change the width of the spectrum. Noise is additive, so wider spectrum = higher noise floor. When you zoom all the way in, the noise floor will be at its lowest, which is very helpful for weak signals. On 20m zoomed all the way in I see a noise floor of -130 or so...

Sergey R5AU

Michael, with GPSDO - OK, but in with Noise level -i am changing width of the spectrum and not a effective filter bandwidth of the receiver, may be i am wrong in this case

Dave - WB5NHL

Let me see if my guess is correct when somebody from Flex answers..... I think as you are changing the spectrum the ham band preselector maybe turning on? I'm not sure how the preselector works so we need a Flex input.

Michael - N5TGL

Sergey, I'm probably not using the correct terminology for #2, but I do remember Tim or some other Flex rep covering it. I do remember them saying that behavior is completely normal.

Sergey R5AU

Michael you are right with explanation in general, but efficiency for weak signal i assume should be correlated with effective bandwidth of the receiver - in our case after DSP. May be Tim will bring light on this one more time, i can not find this here.

George KF2T

I agree with Sergey. Interestingly, I don't see a change in received noise level when changing panadapter width, even with AGC off. Reception of weak signals, using WSJT-X, seems unaffected. It is a little disconcerting to see the baseline moving up and down as I zoom - as a "comfort" it would be nice for the line to stay put.

Michael - N5TGL

Sergey, Tim mentioned it on July 8th while we were still on the email mailing list, which is why it's not here. I did manage to find it so here's the quote: Tim said: "As you zoom out, the bin size of the FFT producing the display must change at certain points. When this happens, the bin size will change by a factor of two. If you zoom out, the bin will now contain twice as much noise because it is twice as wide. This changes the noise floor of the bin and you see the noise rise. The same thing happens in reverse when you zoom in. The receiver also has a certain amount of noise that is proportional to the size of the filter, but is not related to what you see. This makes it possible to hear very small signals even though you may be zoomed out in the panadapter."

Al K0VM

Sergey, First .. the panadapter width does not change the noise floor in a receiver slice.. You can think of the spectrum display as a receiver, independent of the rx slice. It only has a display output. Second - the spectrum display is made up of a fixed number or bins ( Filters ). As you narrow up the span of the display, the width of each bin also narrows.. A narrower bin will contain less noise. A receiver slice has it own sense of bin size and its s-meter level will change with RX filter band width. Hope this makes sense.. AL, K0VM

Sergey R5AU

Well Michael, thank you for note from Tim, means bandwidth of the FFT recognized as separate receiver with equal to the FFT effective bandwidth (explanation a bit complicated)

Sergey R5AU

Thank you Al, you was fast with reply !

Sergey R5AU

All clear, closed