Band-Gain-Noise

  • 1
  • Question
  • Updated 5 years ago
  • Answered
Could some body comment following results of the my F6700 (# 2213-3088-6700-5462) with noise level on Panadapter:

Can we expect noise level on -150 or below ? with DVGA =+30db ?
Can we see on panadapter signals below sensivity level ?
Photo of Sergey, R5AU

Sergey, R5AU

  • 801 Posts
  • 96 Reply Likes

Posted 5 years ago

  • 1
Photo of George Molnar, KF2T

George Molnar, KF2T, Elmer

  • 1529 Posts
  • 541 Reply Likes
Not sure what we're looking at, Sergey. What were the test conditions?
Photo of Sergey, R5AU

Sergey, R5AU

  • 801 Posts
  • 96 Reply Likes
George, i switching only Band and Gain, where Ant=TRCVR, AGC-t = 60 on default, USB and BW= 2,9kHz make no sense
The issue is results are very different from band to band and with DVGA =0db, noise level is around -120db, means add 30db from DVGA should be -150db, but not achieved? BTW in the same time noise curve width is very wide where around 4-8 db. i need some comments on that.
Photo of Al K0VM

Al K0VM, Elmer

  • 566 Posts
  • 93 Reply Likes
What is DVGA ??

AL, K0VM
Photo of Graham / KE9H

Graham / KE9H

  • 4 Posts
  • 2 Reply Likes
Sergy:

Yes, you can see narrowband signals (such as CW) on the panadaptor that are below the sensitivity of the receiver. The reason is a result of the bandwidth of the receiver you are using.

500 Hz bandwidth is normally used for sensitivity specifications.
It is also a common bandwidth for copying CW signals.

For a receiver with a given noise figure, if you cut the bandwidth of the receiver in half, the receiver sees one half the noise, and the sensitivity improves by 3 dB.
Of course, the amount of information (data bandwidth) is also cut in half, so no "Laws of Physics" are broken.

The panadaptor display is made up of several thousand very narrow receivers sitting side by side. If you are zoomed in all the way, (14 kHz across the panadaptor), then the bandwidth of each panadaptor bin (receiver) is 5.8 Hz wide.
This panadaptor will be able to see CW signals 19 dB [10*log(BW1/BW2)] below the sensitivity of the slice receiver if set to 500 Hz bandwidth.

With the panadaptor set to 14 MHz, the bandwidth of the panadaptor bins (receivers) is 5.8 kHz, so when very wide it is less sensitive than a slice receiver in SSB mode.

But, if you are trying to find weak CW signals at the noise floor, if you set the panadaptor for say. less than 50 KHz across the display, it can be a read help. You will be able to see signals that are too weak to copy.

Now, all the above being said, normally, the thing that limits the usable sensitivity of the radio, is not the internal noise, but the atmospheric noise coming in the antenna. If you can see the noise floor of the antenna/atmosphere, then do not add more preamplifier gain. It will not improve anything and can cause problems. So normally leave the gain on "0" up to 20 or 15 meters. You should not need more than 10 dB on 10 Meters.

73's
--- Graham / KE9H
Photo of Steve - N5AC

Steve - N5AC, VP Engineering

  • 1007 Posts
  • 969 Reply Likes
As others have commented, I'm not sure exactly how you are collecting data, but perhaps this will help -- Both the panadapter and the meter in a slice receiver show the noise floor level when a signal is not present in a bin or passband, respectively. In the case of the panadapter, the bin size changes with zoom level and so the noise floor will consequently change. For example, if your pandapter is zoomed in to say 15kHz wide, your bin size is 5.86Hz per bin. If you zoom out to 30kHz, you are now looking at a bin size of 11.72Hz. Since there is twice the bandwidth, there will be twice the noise in the bin and the noise floor will rise by 3dB.

On the other hand, the receiver noise bandwidth is set by the filter in the slice receiver. So if you set it for 500Hz, your noise level is much higher than is in the panadapter bin which is being measured at only 5.86Hz (almost 20dB difference).

Another way to see this is to set your bandwidth in the slice receiver to 500Hz and then hover over the S-meter. Where I am, I get a noise floor reading of -121dBm on 20m. I then expand the filter to 2.5kHz and I read -114dBm. The calculated difference is 10*log(2500 / 500) = 6.99dB and of course -121dBm + 6.99dB = -114dBm. Then if I go down to 50Hz bandwidth in my receiver filter, I read -131dBm. The calculated value is 10*log(500/50) = 10dB better or -121 -10 = -131dBm. This is how noise measurement works and why you have to specify the bandwidth where you are measuring the noise.

Finally, if you are connected to an antenna, your noise floor will change (of course) based on the antenna type, the noise level and type at your location, etc.

Steve
Photo of KB5XE

KB5XE

  • 12 Posts
  • 0 Reply Likes
Steve, I am seeing noise on transmit using v.0.16.4. I have a pic., but cannot seem to be able to load it here. Did not have the noise on v.0.15.3.
Photo of Sergey, R5AU

Sergey, R5AU

  • 801 Posts
  • 96 Reply Likes
Al, DVGA is ADL5201 Digital Variable Gain Amplifiers (DVGA) for each
Spectral Capture Unit.
Well Graham, Steve, you are both right, thank you, my mistake was i am not control a panadapter zoom( ? :-) ), due i compare calculation with reads, what put me in to the deadlock. I have check at morning today, on all bands noise path on the panadapter with DVGA +30db and zoom 15kHz below -150 (151-153 ). Thanks for the greate radio and nice community. I assume with stepping forward it will be better and better. What i see DSP improvement and AVG for panadapter are really wanted features.
Photo of Steve - N5AC

Steve - N5AC, VP Engineering

  • 1007 Posts
  • 969 Reply Likes
You indirectly control the bin size of the panadapter. If you click the + and - buttons in the lower left of the panadapter, you will move in or out by a factor of two which will ALWAYS change the bin size. If you drag the spectrum graticule you will hit boundaries where the bin size will be adjusted and the noise floor will jump. Math we do in the background determines when to change the bin size for you.

Steve
Photo of Sergey, R5AU

Sergey, R5AU

  • 801 Posts
  • 96 Reply Likes
I get it Steve thank you ! It is really visualizing theory on practice.
Photo of k0eoo

k0eoo

  • 577 Posts
  • 78 Reply Likes
Thanks Graham and Steve for the detailed description of how the noise floor bin size for the panadapter and the relationship to the bandwidth....

Dennis, k0eoo