So what you are essentially saying is the K3 has a design fault because in a perfectly normal factory configuration it is leaving a port unterminated. Maybe it is the electronic equivalent of a Persian rug.
I'm not rigging anything, it's how it is right out of the box. I think the real objection here is simply that a low number is reported which is contrary to the perceived wish for the K3 to be somehow perfect. Regards, Mike VP8NO On 29/04/2012 21:47, Guy Olinger K2AV wrote: > In your context the ANT1 circuit was normally terminated, one port in the > transmitter, the other in an antenna. That's normal, a load on both ends. > ANT2 was NOT normally terminated unless you want to treat the reactive 10 > k+ Z termination of the tiny capacitance across the relay as "normal". > > If antenna two was actually being used for something while TX on ANT2, > there would be something on J43 when not in transmit. If you have a sub RX > on the non-TX line, then the crossover means something, as in do I need to > close the COR relay. But that now is the same as your ANT1 termination, TX > on one end, ANTenna on the other, except RX on one end and antenna on the > other. > > I could say that your method was the one rigged (as in not normal), and was > devised to come up with the worst possible reading. I actually USE ANT 2 > during TX on ANT 1. That's when you want to know the separation. I don't > get readings as bad as yours. > > 73, Guy. > > > On Sun, Apr 29, 2012 at 5:51 PM, Mike Harris<[email protected]> wrote: > >> I thought it was clear that all I was looking at was a transmit >> parameter, that is, what came out of ANT2 when ANT1 was was carrying a >> signal. This gives an idea of the coupling across the C/O relay K18 >> which selects ANT1 or 2. In this context both ports were terminated. >> >> I fail to see why I should also have to terminate J43 on the KAT3, it >> would in effect doubly terminate the port with one termination before >> the wattmeter and one after. Not a satisfactory situation for the >> measurement. Naturally, if the Sub RX is connected to J43 that job >> would be being done by the input of the RX and readings might be different. >> >> I have my Sub RX either listening to the same antenna as the Main RX or >> to the Aux antenna port. Nothing connected to J43. >> >> The idea was to see what happens with one of a set of standard set-ups, >> not one where the configuration has been rigged to try and maximise a >> reading. If there is a real need to absolutely minimise what came out >> of, or into, the unused antenna port then by all means terminate J43. >> It's yours to do with what you like. >> >> Regards, >> >> Mike VP8NO >> >> On 29/04/2012 17:06, Guy Olinger K2AV wrote: >>> You are getting some really worse numbers than I did. >>> >>> You may have ant2 only terminated on one end of its cable connections, >>> which would be really worse on 6 meters than 160. There is one of those >>> bitty jacks that needs a cable attached to it and then terminated. It's >>> the "other" antenna connection that allows you to listen on whichever of >>> ANT1/2 the TX is NOT using. >>> >>> Out on Core Banks, NC, we were using doublets at right angles on the same >>> pole to ANT1 and ANT2, receiving with diversity. There was 30 dB >>> separation with the two antennas hanging on the same pole. How much >> better >>> does it have to be inside the K3? >>> >>> You need to terminate the internal K3 connection of ANT2 when you make >> the >>> measurement. Also the shields of both cable connections at both ends >> need >>> to be grounded as well as terminated. >>> >>> 73, Guy. >>> >>> On Sun, Apr 29, 2012 at 9:28 AM, Mike Harris<[email protected]> >> wrote: >>> >>>> Interesting. I looked at it from the other direction. With the K3 ANT1 >>>> connected to a dummy load and ANT2 connected also to a dummy load via my >>>> WM-2 QRP wattmeter. >>>> >>>> With 100W out at ANT1, out of ANT2 I measured 30mW at 50MHz down to 1mW >>>> at 7.1MHz. A range of -30dB to -50dB. 50MHz is not in the WM-2 spec, >>>> however, 25mW was measured at 28.2MHz which gives -36dB. >>>> >>>> There are obviously potential measurement errors there so no one needs >>>> to take it as absolute. Perhaps someone with access to precision >>>> equipment could also do the test. >>>> >>>> Regards, >>>> >>>> Mike VP8NO >>>> >>>> >>>> >>>> On 28/04/2012 12:27, Mark - G4AXX wrote: >>>>> The isolation between ANT 1 and ANT 2 varies from -76dB at 1.8MHz to >>>> -46dB at >>>>> 50MHz. >>>>> This is measured with a signal generator, using the K3 S-meter, with >> ANT >>>> 2, >>>>> AUX RF and RX ANT IN terminated in 50 Ohms. >>>>> >>>>> http://elecraft.365791.n2.nabble.com/file/n7509068/ANT1-2_leakage.jpg >>>>> >>>>> The isolation between AUX RF and ANT 1/2 is>95dB (1-30MHz), rising to >>>> -70dB >>>>> (52MHz). >>>>> This was measured with an N2PK VNA. >>>>> >>>>> I have tried to dress the unscreened wires from ANT 1& 2 to the ATU >> in a >>>>> way to minimise coupling, but the coupling is dominated by the >>>> capacitance >>>>> of relays K18& K19. >>>>> >>>>> Using ANT1/2 for Diversity reception means the receivers are never >> truly >>>>> independent. There is likely to be more coupling inside the K3 than can >>>>> usually be achieved by physical separation of the antennas. >>>>> >>>>> Operating Diversity reception with the SUB RX fed from the AUX RF BNC >>>> input >>>>> is the way to go. >>>>> >>>>> 73 Mark G4AXX >>>> ______________________________________________________________ >>>> Elecraft mailing list >>>> Home: http://mailman.qth.net/mailman/listinfo/elecraft >>>> Help: http://mailman.qth.net/mmfaq.htm >>>> Post: mailto:[email protected] >>>> >>>> This list hosted by: http://www.qsl.net >>>> Please help support this email list: http://www.qsl.net/donate.html >>>> >>> >> ______________________________________________________________ >> Elecraft mailing list >> Home: http://mailman.qth.net/mailman/listinfo/elecraft >> Help: http://mailman.qth.net/mmfaq.htm >> Post: mailto:[email protected] >> >> This list hosted by: http://www.qsl.net >> Please help support this email list: http://www.qsl.net/donate.html >> > ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
