[Elecraft] K3 output buffer modification

[James Sarte]

Hi Don,

I don't have anything that sensitive, so that's out.  Perhaps I can borrow a
spectrum analyzer from someone near me... or just wait for the P3 to come
out.  That's the only reason I performed the mod in the first place.

BTW, thanks for the QST article info... will give that a read tonight.  

Cheers!
James K2QI

-----Original Message-----
From: Don Wilhelm [mailto:w3fpr at embarqmail.com] 
Sent: Wednesday, January 06, 2010 6:37 PM
To: James Sarte
Cc: 'Elecraft'
Subject: Re: [Elecraft] K3 output buffer modification

James,

What do you propose to measure a 50 uV signal with?  If you have 
equipment that will measure levels that low, then well and good, but if 
you are expecting to see it on an oscilloscope or RF Probe, you may as 
well forget it.  Recall that 50 uV is an S-9 level, and is -73 dBm which 
is a very low level to measure.

There are power measurement devices that will detect levels that low, 
but most of the commercial ones are priced out of the range of 
amateurs.  One that can be homebrewed and will measure levels down to 
-80 dBm is the W7ZOI  design using the AD8307 Logarithmic Amplifier as 
the front end (see Experimental Methods in RF Design, Measurements 
chapter or the QST article from June 2001 p38ff).  To calibrate it, one 
must have a known level source of RF at two different levels.  Those 
with a calibrated HP8640 signal generator (or similar) will be able to 
calibrate that meter with no problem.
For anyone considering building such an instrument, understand that it 
must be well shielded.  Stray coupling into the meter will produce 
erroneous results.  Mine is in a shielded enclosure with an internal 
battery, and even that is not enough, the front end circuits are 
shielded from the rest of the assembly.  Yes, it can even measure the 
total signal level coming in from an antenna, so it is a good performer 
if properly calibrated and is a very sensitive detector.  With the 40 dB 
Power Tap, it can accurately measure power levels up to +50 dBm (100 
watts), and will work up to 500 MHz with careful construction.  If you 
want accurate power measurements, consider building one, but you need 
access to a calibrated power source to calibrate it - calibrate at 10 
MHz, 50 MHz, 144 MHz and 450 MHz so you have a good idea of the 
variation with frequency - there will be some variation.

73,
Don W3FPR

James Sarte wrote:
> Hmm... just out of curiosity, couldn't I measure the IF output using the
XG2
> signal generator set to 50uV?  How strong does the signal have to be?
>
> I'll go ahead and check the junction voltage at R9 as you suggested.
>
> Mni tnx!
> James K2QI
>   
>