CATV Bench Sweep Setup
When a CATV bench sweep is used to align all RF modules, any errors in the bench setup will translate to the system as an RF signature which can become additive with the cascade length.  It is therefore very desirable to eliminate or minimize any sources of impedance mismatch on the test bench itself.  The sources of the mismatch are power inserters, splitters, couplers and a low-grade of coaxial drop cable instead of a solid dielectric headend lash-up type cable.  This procedure will give the hints and kinks that are most commonly found in a test bench setup. The modern bench sweep has the ability to minimize the errors mentioned above by the use of digital signal processing and internal calibration routines.  This will reduce the errors somewhat but is still open to some degree of error that can be removed by following good setup practices of the mechanical portion of the bench that is described below.   These changes will improve the performance of any test bench with either old or new test equipment.
Figure 1 shows the typical test bench which quite often has a power inserter and regular drop cable as the key offending components.   The power inserter is necessary to power a line extender under test but is not required when testing a trunk amplifier.  Why, because the AC power source can be connected to any feeder port on the trunk housing.  Fuse only that feeder line and remove any trunk power plug, fuses, etc.  With the arrangement shown in figure 2, the power inserter can be removed from the trunk test procedure and with its removal a device that has only a 16 dB return loss is eliminated from the circuit.
Next, the cable should be the solid dielectric cable used in headend wiring.  It will hold up better in the day-to-day handling that drop cable will not survive.  The return loss on the solid dielectric cable is also much better than foam dielectric cables.  The next thing that can be done is to put a 3 dB, type "F" inline pad on the input to the housing at the trunk input and trunk output ports.  The idea is to force the housing to see impedance match conditions that are 26 dB return loss or better since this is what a reel of good trunk cable will have as its return loss after it is installed on the strand.  A good trunk cable could easily have 30 dB return loss while it is still on the reel.  Remember that we are providing the trunk amplifier housing with a test environment that is similar to what it sees in the field -- it does not feed drop cable.  The 3 dB pads will be more effective on drop cable if you do not have a good solid dielectric headend cable.   Either way, all you need to do is compare the peak-to-valley RF sweep response with and without the pads to see if you are benefiting from their use.        (Click here for page 2.)