Topband: Stub measurement error

[Tom Rauch]

> Regarding tee connectors, if you look at them carefully 
> you will see they
> are not symmetrical.  The two straight arms are the same 
> length, but the
> center arm is shorter.  This is true for F-M-F UHF and BNC 
> tees.  This is
> why you cannot just average the two readings mentioned 
> before.

Then the simple solution is to use the center arm for the 
stub, if indeed you have a system that critical. If you use 
a short barrel you'd have about the same test distance from 
the analyzer to the stub as most cheap analyzers are 
calibrated at.

It's also important for everyone to understand that the 
things we measure in a 50 ohm broadband test fixture don't 
repeat in the practical world of our systems. This is 
particularly true for filters.

It's also important to understand even-order harmonics in 
the output of an analyzer can really mess up adjustment of 
stubs.

73 Tom