OK

This file  http://www.g4jnt.com/DownLoad/DCF77_TuningTest.WAV    was recorded at 31.25Hz sampling rate and is from the 77.5kHz time signal (MSF is off the air at the moment, just when I wanted a stable test signal typical !!!!!!!)

I started with the tuning dial showing 5Hz high (effective LO = 77505Hz) , so the IF / baseband coming out was 5Hz low, at 995Hz audio from the converter  and at -5Hz in the data stream.   This was then stepped down  1Hz at a time, at roughly one minute intervals so the resulting output frequency  moved UP over the 11 minute interval, stopping with the tone/baseband data 5Hz HIGH in frequency with the tuning showing 77495Hz

Reading the file into SpecLab shows a plot going from bottom left to top right, with the left hand, lower frequency terms having  earlier time stamps.  This plot is the same direction as the one I had on screen during the recording.

So it looks as if my I/Q saved to Speclab compatible .WAV format is the correct way round.

Andy  G4JNT