As I have designed several amps in this manner, I am a firm believer in the
PI filter only
output stage. It is recommended to use a progressive thread variable
inductor so as on the 6 and 10 m bands the inductance per turn diminishes.
If you have the mechanical skill or you can buy one, I will strongly
recommend a variable inductor coil with at least 4 rollers that make contact
with the coil(for mechanical reasons).
Also, fully shield this stage form the rest of the amp, and you are set to
see your tube IMD specifications fulfilled, which, in most cases and for
amateur use, is more than FCC requires :).
Which brings me to a different issue. Just try this experiment. Get your amp
exited by your rig
and measure it's harmonical output (in CW mode). Then replace the rig with a
good quality
signal generator-preamp (one that has less than -120dB spurious on the
second harmonic- and you'll be surprised how good your amp output looks.
As a reminder, harmonics are not only produced by a Amplitude distorted
signal (and other minor causes) but also by its Phase jitter (poor Local
Oscillator designs).
Alex
> -----Original Message-----
> From: amps-admin@contesting.com [mailto:amps-admin@contesting.com]On
> Behalf Of Richard
> Sent: Friday, June 28, 2002 9:17 AM
> To: Tom Rauch; AMPS; Norman Hockler
> Subject: Re: [Amps] Pi tanks vs Pi L + LOW PASS
>
>
>
>
> >> Wondering about the harmonic attenuation of a Pi L tank vs a standard
> >> Pi tank followed by a good quality low pass filter.
> >
> >Having wrestled with this stuff to meet FCC TA guidelines, I've
> >learned to watch the layout carefully.
> >
> >On paper the Pi-L looks a bit better, but in real life the actual
> >layout is generally significantly more important than the change of
> >components or network used.
>
> * In my experiences, this is not rocket-science. The idea that only
> recognized amplifier experts can understand the esoteric secrets of
> proper layout smells kinda like bologna.
> - After I added 160m coverage to a SB-220, I was curious about
> harmonic radiation. I measured second-harmonic radiation in the
> near-field at c. 58db down. The L-network antenna tuner that was feeding
> the 160m halfwave end-fed antenna was probably responsible for some of
> the attn.
>
> >Subtle things like how you route the RF
> >leads to and through the capacitors and how things are grounded means
> >more than network Q, especially for higher order harmonics.
> >>For bands close to the operating band, like the second harmonic, by
> >far the best results come from using a series-resonant trap on the
> >output. A simple single-section trap can add 20 dB or more of
> >harmonic suppression, and only takes one large component (a doorknob
> >cap) and a small series inductor.
>
> * For better vhf harmonic suppression, connect a 50pF doorknob ceramic
> cap from the output connector to chassis gnd. The Xc will also help to
> cancel the intrinsic L in the T/R-relay when operating in bypass-mode.
> The inductance of the lead connecting the cap to the output terminal
> forms the L part of the vhf filter.
> >
> >> I am interested in building an amp using a tune and load cap with a
> >> good roller inductor in between to eliminate the band switching mess
> >> and adding a good low pass filter.
> >
> >Rollers sometimes are great to use and they look neat, but they can
> >occasionally cause problems.
> >
> * When a roller coil-form is made from a material which does not have a
> low D-factor - such as Delrin, PVC or Nylon, especially when feeding
> high-Z loads, the coil form can catch fire. Better materials are:
> alumina [ceramic], ABS, G10 [glass-epoxy], Micalex, and Rexolite.
>
> >...
>
> >As for the roller.... while it is mechanically and electrically more
> >simple wiring in a roller vs using a switch, rollers can create
> >problems.
> >
> >For one thing, rollers almost never short unused turns at multiple
> >points. They generally have one shorting roller that rolls along the
> >coil. That means they commonly have "strong" spurious resonances far
> >above the operating frequency, and at those resonances can cause the
> >system to have very poor harmonic attenuation on some harmonics.
>
> * 160m and 80m tank inductors also have such vhf resonances.
> >
> >Another issue is current rating. In order to handle enough current at
> >the ten or fifteen meter end, the component will have to overkill
> >the lower bands. Be sure to size it for ten meters, and hope
> operating ten or
> >fifteen with hardly any turns cranked in it will
> >have sufficient reactance on harmonics to help suppress them!
>
> * One solution is to use a separate, 10m tank-L made from larger surface
> area conductor material -- with the roller-L set to zero turns on 10m.
>
> >Close to the operating frequency, everything works the same for
> >spurious signals. The further off from the operating frequency we
> >look at the differences, the more critical component selection and
> >layout becomes.
> >
> >73, Tom W8JI
>
>
> - R. L. Measures, a.k.a. Rich..., 805.386.3734,AG6K,
> www.vcnet.com/measures.
> end
>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
|