Alberto, Thanks for getting me on track here.
I looked at the outputs with a dual-trace scope and the service monitor feeding it 10 khz off at 7050 khz. I see what looks close to 90 degree phase difference in the 10 khz audio out...but large amounts of RF on top of it. A .01 across each output cleared that up pretty well (soft66 radio issue!). Now back to the computer. With the sig-gen set for 7050, I still see the mirroring in SDRadio. Changed to Rocky, and clicked on the "+1 sample offset", and eureka, the image was immediately nulled by Rocky. Changed the sig-gen to give it a good signal every 5 khz across the band, and Rocky nulled at each freq. I now see the un-mirrored spectrum that I expected. I guess that SDRadio doesn't have enough "skew" to make up for the full 1 sample error. That also makes me think that every sound card I own has a sample delay problem. Guess the next step is to buy a Delta. Thanks for your help. 73's Steve
