Page 112 - Antennas for Base Stations in Wireless Communications

Base Station Antennas for Mobile Radio Systems 85

this kind are sometimes used to carry out routine testing of production
antennas; they provide rapid confirmation that connections have been
correctly made and that the array has no serious problems.
If the feed network appears to provide the correct output currents
when measured on the bench with matched loads, yet the complete
array shows variability of element currents with frequency, the cause is
almost certainly mismatch at the input of the elements. If the elements
(or element groups comprising each tier) are known to match correctly
when measured separately, then attention must be given to the mutual
impedance between them. The mutual isolation between tiers is easy
to measure, and the driven impedance of a tier in the presence of its
immediate neighbors can be measured using multiple directional cou-
plers or a multiport VNA. The coupling to neighboring elements at 2l
spacing or more can generally be neglected.

2.3.12.3 Passive Intermodulation Products PIM is usually measured with
the AUT placed in an anechoic box with a typical clearance between the
antenna and the absorber of around 1 m. The supporting frame or slide
on which the antenna is placed during measurement should be made of
wood, GRP, or other insulating material. While a workable measurement
system can be made using two signal sources, two power amplifiers, a
triplex filter, and a spectrum analyzer, a complete integrated system is
available for this task, allowing fast and flexible measurement down to
a level of around −160 dB for two input signals at +43 dBm. Most BS
antenna manufacturers measure PIM on a routine basis on 100% of
production. As noted, PIM is likely to be generated at improperly com-
pressed conductor joints, and the general practice is to measure PIM
while the antenna is vibrated, sometimes by a motor-driven device, but
often using a rubber-faced hammer or a walking stick!
The field intensity close to the antenna created during a PIM mea-
37
surement is likely to exceed safe limits, and access to the antenna by
personnel should be restricted while power is applied by a physical bar
or walk-on pressure pad.
The diagnosis of PIM problems in production is not easy. Consistent
failure suggests inadequate design and specification, while occasional
failing of batches of product may point to a defective batch of material.
A sudden increase in failures may indicate a change in a material, a
process, or a member of the production team. Isolated failures may relate
to a production error (wrong part, loose screw, bad solder joint, or con-
tamination). As with VSWR performance, the use of statistical process
control (SPC) is a powerful method for spotting trends and understand-
ing the most likely reasons for failure. Practical measured results will
always display statistical variability, but by understanding the scatter
of results and the mean values obtained on each product over a period