[Elecraft] P3 and continuous 5-15 db 'spike'

[N8LP]

The LO leakage of LP-PANs after serial number 200 is about -120dBm. Earlier
serial numbers had a bit more leakage, but this is easily fixed by changing
the value of C40 to 0.001uF. -120dBm is overkill for LP-PAN's intended
purpose, but connecting LP-PAN directly to a receiver (P3) tuned to the LO
frequency changes things a bit.

I have run some tests to simulate this situation. First, I connected an
LP-PAN directly to a receiver tuned to the LP-PAN LO frequency (8.209MHz for
most units, 8.215MHz for the first couple runs). I verified that the leakage
was in the -120dBm range by comparing the received signal to my calibrated
HP signal generator. Then, I connected the second receiver, LP-PAN and K3
together using a 3dB hybrid splitter . This is reportedly what the P3 uses
to split the K3 IF signal.

I could detect nothing on the second receiver at the LP-PAN LO frequency.
This makes sense since the hybrid splitter has an isolation spec of -25dB
between output ports, which should provide an LP-PAN signal of about -145dBm
at the receiver input. 

If you make the change to C40 and if the hybrid in the P3 is working as mine
does, you should see nothing from the LP-PAN. I repeated the test using a
couple different LP-PANs, and I also tried a couple different models of
Mini-Circuits splitters. The results were consistent.

BTW, the LP-PAN preamp kit adds another 15-20dB of LO isolation, and lowers
the LP-PAN NF to 7dB to help offset the additional loss of the splitter in
the P3.

73,
Larry N8LP 



VE3NFK wrote:
> 
> Yes, Alan
> 
> an LP-Pan  
> 
> and yes that is where the spike comes from...
> 
> I checked Brett's suggestion that the in/out might be backwards.. but it
> is fine.
> 
> So - now I need to find out how to stop the leakage.
> 
> 73
> 
> John VE3NFK
> 

-- 
View this message in context: http://elecraft.365791.n2.nabble.com/P3-and-continuous-5-15-db-spike-tp5595056p5596501.html
Sent from the [K3] mailing list archive at Nabble.com.