[Elecraft] Your "slide 8" - phase noise

[Jim Brown]

On Tue,9/1/2015 2:52 PM, Rick Stealey wrote:
> Jim,
>
> I'm duplicating your slide 8 test with my K3 going into a flex 6300.
> 25 watts
> At 80 db down I'm seeing a little over 400 cycles, and a pedestal.
> This is K3 with the original synthesizer.

Studying this measurement, I have several thoughts. First, the 
resolution bandwidth appears to be about 30-40 Hz at -70 dBc, compared 
to 60 Hz for the K3 with P3SVGA. That's very good.  Also, those peaks 
that show up in the range of 260-310 Hz either side of the carrier are 
visible in the waterfall and are moving in the waterfall, so I suspect 
they are some sort of modulation. Note that they are not visible earlier 
in the waterfall.

> Do you know, when you say you see X dbc what effective bandwidth that 
> is?  i.e. how to translate that into signal power in a 1 hz bandwidth 
> like ARRL does?

No, I don't know, because we don't know the effective bandwidth of our 
measurement systems.

> Because with the FFT spectrum analyzers there is no fixed bandwidth 
> filter sweeping across the spectrum.

Right. We've been working with both swept measurement systems like TDS 
(Time Delay Spectrometry) and dual channel FFT since 1980. I had a TDS 
system in 1982. Cost me $12K.

> ex: at 1 khz offset (14051= 5 divisions) phase noise is ~-95 dbc.  but 
> what would that be in a 1 hz bw?

No good answer, for the reason noted above.  But this measurement shows 
that this test setup is very good for measurement occupied bandwidth as 
long as you stay out of non-linearity (that is, either too close to the 
top or bottom of the A/D).

By all means save this measurement and compare it with what you see with 
the new synth board.

73, Jim K9YC
>
> Rick  K2XT
>
>
>
>