On Fri, 9 May 1997 08:34:50 +0200, you wrote: >When performing the emission test RE102 according to MIL-STD-461/462D we >observe that the test system noise floor is at some frequencies just >below the most stringent limit (navy mobile and army). This is when we >use the bandwidths prescribed by the standard. We can lower the noise >level when we reduce the bandwidth. <snip>
At the risk of stating the obvious ... Have you tried minimizing attenuator settings in the test setup? (both internal and external)? The reference level at the top of the screen is determined by the attenuator settings and by the amplifier gain. High values of attenuation require high values of amplification to create the same reference level. And the higher the amplification, the higher the noise floor. The downside to minimizing the attenuation is that the analyzer is more susceptible to overload and input preamplifier damage. As an example, I set my analyzer for a reference level of 100dBuV, and a frequency of 100MHz. There was no connection to the input (internally terminated). I varied the attenuator setting, keeping the same reference level, and recorded the noise floor: Attenuator Noise floor (dB) (dBuV) ---------- ----------- 0 27 10 35 20 45 30 55 40 65 50 75 Pat Lawler [email protected]
