Fw: Re: Fw: Re: Fw: Re: [Test-Equipment] Spectrum Analyzer BW measurement

Sat, 31 Jan 2004 21:20:38 -0600

Excellent !! My take on it............ 

Your input signal didn't change, however it shows me that a 10x RBW
change affected the "measured" BW as follows:

The -3dB comparisons should be right at 10x, the -50dB numbers should be
between 5x 
and 10x depending on the filter characteristics.

Based on your data, Dave. 
---------------------------
300Hz to 3Khz       11x at -3dB, 7.2x (11.6/1.6) at -50dB, 6.4x
(12.8/2.0) at -60dB
3Khz to 30khz        9.4x at -3dB, 5.8x at -50dB, 6.3x at -60dB
30Khz to 300Khz   13x at -3dB, 10.4x at -50dB, 10.3x at -60dB  

My input
---------
That compares to my previous measurements in an earlier note (putting it
in your format):
500Hz    500Hz @ -3dB     1.6Khz @ -25dB   2.4Khz @ -45dB     3.2Khz @
-55dB
5Khz      4.7Khz @ -3dB    12.6Khz @ -25dB  21.6Khz @ -45dB  30Khz @
-55dB  

My numbers (above) are going to be a little higher because the RBWs are
wider, the measured signals are not the same, and the filters are
different.

Based on my data.
------------------
500Hz to 5Khz      9.4x at -3dB, 7.9x at -25dB, 9x at -45dB, 9.4x at
-55dB 

What that confirms/allows me to make, is a rough approximation (that will
suffice) of the effect of a spec based on a 3Khz RBW and -55dB vs my
measurement equipment which allows 500Hz RBW or 5Khz RBW in this range.
However the 500Hz (few % of BW) RBW in "peak detect" mode will still
provide a far more accurate representation of the ACTUAL signal BW per
the FCC requirements. I think it also tells me that the filters in this
Marconi are pretty good (steep sides). Looks like the AMA 3Khz RBW spec
was put in place as "quick test" due to readily available test equipment
in the field, ease of use (fast scan), not requiring peak detect, and
adequately approximates the FCC spec for older R/C transmitters to
prevent interference. 

And yes, the noise floor should drop 10dB for a 10x reduction in RBW
...and it does.

Thanks, guys.

73 Kees K5BCQ

--------- Forwarded message ----------
From: Dave Emery <[email protected]>
To: [email protected]
Cc: [email protected]
Date: Sat, 31 Jan 2004 20:32:02 -0500
Subject: Re: Fw: Re: Fw: Re: [Test-Equipment] Spectrum Analyzer BW
measurement
Message-ID: <[email protected]>
References: <[email protected]>

On Sat, Jan 31, 2004 at 08:53:45AM -0600, [email protected] wrote:
> That, I understand. Thanks Dave. 
> 
> Just for curiosity what do you read with your SA when you look at a
> signal at the -55dB points using the "few %" RBW vs 10x that RBW. I
> realize actual filter passbands are going to be different, today's
> spectrum analyzers probably compensate, etc. but I'm interested if you
> see something around 5x-10x BW at -55dB.

        I measured my antique IFR A-7550 (it being the one that
was handiest and not needing work)...

        300 hz                 300 hz  @ 3db        1600 hz  @ 50 db  
2000 hz  @ 60 db
        3   khz                3.2 khz @ 3db        11.6 khz @ 50 db  
12.8 khz @ 60 db
        30  khz                30  khz @ 3db        68   khz @ 50 db   80
  khz @ 60 db
        300 khz                380 khz @ 3db   710  khz @ 50 db   820 
khz @ 60 db

        I think its 300 hz and maybe 3 khz values were possibly somewhat
contaminated by its inherent LO phase noise (incidental FM).   It
ain't all that wonderful compared to better instruments, and its noise
floor was about -78 dbC in this case.

        I tried a couple of different signals including a precision
GPS derived 10 mhz from a standard and they gave similar results.

-- 
   Dave Emery N1PRE,  [email protected]  DIE Consulting, Weston, Mass
02493