[Amps] T networks and harmonics

[Gary Schafer]

Hi Bill,

I guess this was in response to your response to someone that said that "a T
network tuner after the amp would not help reduce harmonics as it was
basically a high pass filter". You said that "it would help some". 

So I assumed you were,in this case, talking about T network type antenna
tuners. Capacitors in series coil to ground at the center.

I did some more measurements this time using an HP 3586C selective level
meter which has a tracking generator and level readout in DB.

At 10 mhz the tuner was loaded into the input of the 3586C with the input
impedance set to 10k ohms with 50pf in parallel. There was about 5 feet of
rg58 cable between the tuner and the 3586C input. Not sure what the load
actually looks like at the tuner. This is what I got:

10 mhz = 0 db.
20 mhz = -9 db
30 mhz = -8 db

There was a notch around 15 mhz of about -15 db.

At 9 mhz the level was -12 db.
At 8 mhz the level was -20 db.
At 7 mhz the level was -25 db.
These frequencies are of course below the 10 mhz where the tuner was tuned
to. Shows how much the low end drops off compared to the high side.

With a 560 ohm resistor at the input of the 3586C and the tuner retuned at
10 mhz for this load this is what I got:

10 mhz = 0 db
20 mhz = -4 db
30 mhz = -5 db

With a 50 ohm load at the input of the 3586C the response was almost flat.
Less than a db difference between 10 mhz and 30 mhz with tuner tuned to 10
mhz.

I don't know how much the input capacitance of the 3586C affects the load
seen by the tuner or how much the coax effects things. 

I wish I had a way of isolating the input loading on the measuring devices
for higher impedance loads. Same problem with the spectrum analyzer except
its' input is a fixed 50 ohms.

73
Gary  K4FMX


> -----Original Message-----
> From: amps-bounces at contesting.com [mailto:amps-bounces at contesting.com] On
> Behalf Of Bill L. Fuqua
> Sent: Monday, February 06, 2006 7:25 AM
> To: garyschafer at comcast.net
> Cc: amps at contesting.com
> Subject: Re: [Amps] T networks and harmonics
> 
>      If the impedances are quite different there will be attenuation. You
> have the same input and output impedances.  I was referring to a T network
> as an output impedance network which has to match very different
> impedances.
>      I thought we were talking about pi-l, Pi networks as   output
> coupling networks.
>      In fact any time you try to design a matching network  with out
> transformers and use a pi or t or what ever you  can't avoid resonance and
> thus attenuation of both the high  and low frequencies.
> 73
> Bill wa4lav
> 
> 
> 
> -----Original Message-----
> From: "Gary Schafer" <garyschafer at comcast.net>
> To: <amps at contesting.com>
> Date: Sun, 5 Feb 2006 22:48:48 -0500
> Subject: [Amps] T networks and harmonics
> 
> There was some discussion a few days ago about pi/L networks and
> harmonics.
> Someone suggested that a T network tuner would also help reduce harmonic
> energy out of an amp as "it did not only act as a high pass filter, but
> being a resonant circuit it would provide some harmonic attenuation".
> 
> I had always thought of a T network with series capacitors looking like a
> high pass filter. I drug out the spectrum analyzer and a dentron KW T
> network antenna tuner. Hooked it up to the dummy load and the transmitter.
> Tuned things up and checked 2nd and 3rd harmonic levels on the spectrum
> analyzer. Then disconnected the tuner and again checked the harmonic
> levels.
> 
> 
> They were exactly the same with and without the tuner in line!
> 
> I could see other spurious trash from the transmitter that was below the
> transmit frequency being attenuated some but nothing above the transmit
> frequency was attenuated.
> 
> I then hooked up a comb generator with 1 MHZ harmonics to the input of the
> tuner, still tuned as before, and observed the various harmonics of the
> comb
> generator.
> 
> No attenuation above the frequency that the tuner was tuned to was
> observed!
> It passed harmonics all the way up to around the 2 meter frequency range
> just as well as it passed 14 MHZ where the tuner was tuned. You could see
> the comb generator was attenuated below 14 MHZ, just as you would expect
> from a high pass filter!
> 
> Removed the tuner and observed the comb generator directly to verify
> levels
> out of the comb generator and they were the same as with the tuner at 14
> MHZ
> and above.
> 
> So from what I saw, a T network antenna tuner does nothing at all to
> attenuate harmonics.
> 
> Perhaps if it was feeding a reactive load rather than a fairly flat load
> it
> would make some difference?
> 
> 73
> Gary K4FMX
> 
> 
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
> 
> 
> 
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps