Technical data

Performance Verification
2400 Series Operation Manual, 33237 Revision A, May 2004 4-23
4.4.3.2 Frequency Modulation Test
4.4.3.2.1 Description
The following procedure tests Frequency Modulation Accuracy and Bandwidth at frequencies within
each band. The tests require the use of an FM discriminator
4.4.3.2.2 Equipment Required
• UUT
• Local Oscillator
• FM Discriminator
• Function Generator
• Measuring Receiver
• RF Cables
4.4.3.3 Delay Discriminator Description
The FM Discriminator includes an RF splitter; two coax lines of unequal length, and an RF mixer. The
shorter cable is connected from the output of one splitter port to the RF input of the mixer. The
longer cable is connected from the remaining splitter output to the LO input of the mixer. For this Fre-
quency Modulation performance test, the frequency range of the mixer LO and RF is 4 to 8 GHz. The
like-frequency two signals beat against each other at the mixer. The mixer produces a DC voltage at
the output of the mixer based on the phase differential of the signals. This procedure does not iden-
tify a specific mixer for the discriminator circuit. There are two factors to consider when choosing a
mixer for this test.
• The mixer RF and LO frequency range must be at least 4 to 8 GHz.
• Low conversion loss. Low conversion loss will result in higher output levels
The output of the mixer should be +15 mV or higher for a +1 MHz change in UUT frequency. Increas-
ing output power will improve the output range of the discriminator. It is recommended that a large
cable length ratio be used. Cable length ratios of 16 to 1 (48 inches to 3 inches) will produce a large
number of zero crossings within the 4 to 8 GHz frequency band and improve the display resolution of
the test. The rate of change for the output voltage per unit frequency (V/ F) must remain constant (lin-
ear) for the ±1 MHz deviation range for Narrow Mode and ±20 MHz deviation range for Wide Mode.
Rate changes within this deviation range will result in a non-symmetrical waveform and will produce
invalid test results. Zero crossings can be identified using the List mode of the 2400. A 4 to 8 GHz list
using a 1 or 2 MHz step size will provide excellent resolution for identifying an optimal frequency zero
volt crossing that is relatively linear above and below the zero volt crossing. WaveMaker software
included with the 2400 should be used to create the list. Identify several zero crossings across the 4 to
8 GHz frequency range. Approximately 1 to 2 points per 1 GHz step is recommended. I.E. 4.5, 5.5, 6.5
7.5 GHz are within 4 to 5, 5 to 6, 6 to 7, 7 to 8 GHz respectively.