I also find that external effects affect the depth of null you get - in my field strength measuring experiments, I found some locations where there was almost no null - This seemed to occur near overhead wires and power lines, which were presumably acting as parasitic antennas, inducing signals in the loop.