Using a TDR is a good start. I would terminate with a good, low VSWR,
load and look for any abnormalities, then reverse the ends and do it
again. You will then need to evaluate any glitches in the TDR plot. My
preference is to make a swept transmission magnitude test, but this
requires a two-port network analyzer. You can approximate this using a
transmitter and a power meter (preferably two) with a dummy load so you
can compare the power going in to the coax versus the power coming out.
If the loss is within the manufacturer's specification for the coax,
then you are good to go. The problem with the transmitter and power
meter approach is that some defects may be narrow-band (I call them
"suck-outs" on a swept gain plot), but the TDR plot of the terminated
cable should show these. A defect that would show up only at high power
(but within specification for the coax) would be most unusual.
73, Avery, WB4RTP
TowerTalk mailing list