[TowerTalk] Using Stubs to Reject Harmonics

[Jim Lux]

-----Original Message-----
>From: "Joe Subich, W4TV" <lists at subich.com>
>Sent: Feb 2, 2010 4:48 PM
>To: "'Tower and HF antenna construction topics.'" <towertalk at contesting.com>, TexasRF at aol.com
>Subject: Re: [TowerTalk] Using Stubs to Reject Harmonics
>
>
>> 1/4 wavelength at the second harmonic is only 1/8 wavelength at 
>> the  operating frequency. If you make the connection 1/4 wavelength 
>> away at the operating frequency, the described transformation does 
>> not happen at the second harmonic.
>
>Which is why W2VJN's stub cookbook shows some interesting behavior 
>with 1/8 wave connecting lines <G>.  Without getting the book off 
>the shelf, I believe he was using 1/8 wave between the amplifier 
>and first stub and 1/8 wave between first and second stubs in some 
>cases to optimize 2nd harmonic rejection.  
>
>1/16 wave would be another interesting connecting length if the 
>4th harmonic were the bigger problem ... the combinations are 
>endless <G>. 
>
>73, 
>


Which is why a computer to model it is nice<grin>... 
Or a set of decent design equations.

I'll bet there's a nomogram or set of templates for a Smith Chart that does this kind of thing, at least for a fairly simple 1, 2, or 3 stub setup.

It's not much different than using a multistub tuner at microwave frequencies.  After you do it a few times, you get a feel for what the various stubs are doing.  If you have a network analyzer, it's even more fun, because you can see the S11, S22, and S21 all at the same time.

I will say, though, that for multiple stubs, doing without explicit calculation is somewhat of an art. There are people who can tune up a multi section filter quickly, and those who can't.

Jim