In article i,
Kari B wrote:

I have got very simple noise source for receiver measurements, but
noise power verification makes some trouble...
Old analyzer noise floor is around -125dbm without input and noise power
using 10kHz BW is -95dbm using display 'peak' setting and long video
filter at 30MHz...

Same setup with display 'inf' setting shows -105dbm for the noise.

Sorry I don't have a manual for the analyzer to find what the 'inf'
means... but which value would be acceptable ?
I would use the first one ?

Kari-

I was hoping your question would elicit comments from others who are
experienced in noise measurement.

Back in the 60s I had a job as a test technician. One assignment was to
measure noise power in an AM aircraft receiver. I didn't know anything
about noise, but had a procedure to follow.

As I recall, you set up the receiver with audio output that would drive
an AC voltmeter calibrated in dB. You were to inject a modulated signal
at the frequency where the receiver was tuned, and increase the signal
level until the audio output increased by 3 dB above the noise level.
The theory was that that signal power would be equal to the noise power
in the channel.

I still don't know much about noise measurement. I can see how this
test might be valid for an SSB or CW receiver, but doubt it was any more
than a go/no go test for the AM receiver.

In your case, wouldn't RMS noise measurement be required rather than
peak measurement?

Fred
K4DII