Topband: Beverage antenna isolation measurements

[Charles Stackhouse via Topband]

     I have just finished installing 6 Beverage antennas in the woods north of my house. Details are below. My question to the group involves measuring the isolation of these antennas from my transmitting antennas. I am testing by transmitting on all bands with either 100 or 500 watts and simultaneously measuring RF levels on the Beverage feedline in the shack.  Eventually I want to set up an SO2R station.     To measure the power, I am using a homebrew RF power meter kitted by Kanga and based on the AD8307 (QST June 2001 - W7ZOI and W7PUA). This is a 50 ohm instrument and I feed the Beverages with 75 ohm RG-6.   This means a nominal VSWR of 1.5:1 (14 db return loss).  I placed an ICE Model 401 bandstop BCP filter in front of the meter. 
     The raw results look good with the worst isolation to date being 65 db, i.e. 160 microwatts while transmitting with 500 watts.  I have some more measurements to finish with the triband yagi aimed in a few different directions.
     How much inaccuracy is there in this method due to the 75/50 ohm mismatch at such low power levels (16 nanowatts to 160 milliwatts)?  Is there an easy mathematical correction (for a guy who took calculus 39 years ago) for the impedance mismatch or does it matter?

   
Beverage antenna details: (map of Beverages is on my QRZ.com page)
My Beverage antenna project is completed and the 6 antennas seem to work well.  They are between 470 and 860 feet long so as to stay on my property and radiate from a central hub. I used 17g galvanized fence wire 7 feet high held up with plastic fence insulators nailed to trees.   They point to 40, 80, 160, 240, 280 and 340 degrees.  I terminate them with Ohmite 470 ohm 2 watt resistors. The transformers are the usual 6.25:1 (5T,2T) on a BN73-202 core. Four foot ground rods are driven at the ends of each Beverage. There is 20-30 feet of RG-6 coax from the transformers to the homebrew switchbox. I use 24v relays (RK1-24V) and run almost 600 feet of CAT5 for a control line.  Unused antennas are not grounded. The RG6 feedline goes about 30 feet from the switch towards the shack where the braid is grounded.  I put a K9YC choke (8 turns of the RG-6 wound around a #31 Big Clamp-On) on the shack side of the braid ground and then the coax runs almost 600 feet back to the shack. I ground the braid again just outside the shack. In the shack is a homebrew switch box containing a 24v linear power supply and a 7 position rotary switch. (The 7th position is labelled for an external switch when I get around to building one similar to LA4HIA).
Thanks for reading this verbose question. I spared readers the details of cutting paths for these antennas totalling 4090 feet through woods infested with an understory of invasive European buckthorn, multiflora rose, and honeysuckle. Of course it was done in the hottest, most humid and most mosquito infested part of our late summer.
73, Charlie W2GN