Amps
[Top] [All Lists]

Re: [Amps] Pi-L Network

To: "Peter Voelpel" <df3kv@t-online.de>, <amps@contesting.com>
Subject: Re: [Amps] Pi-L Network
From: "Tom W8JI" <w8ji@w8ji.com>
Date: Wed, 28 Mar 2007 21:27:40 -0400
List-post: <mailto:amps@contesting.com>
In the minimum I got 6db improvement, it depends also on the 
output
impedance and layout of the components. I gained 10db more 
at high output impedance.
Most are 1000-1500 ohms, 160m is 3 kohm.>>

Are you saying the mid-impedance of the network, or the 
impedance across the loading capacitor, is 1000-1500 ohms? 
Or 3000 ohms at that point on 160?

If so, I agree with you. You could get 6dB or more 
additional harmonic suppression.

The problem is the loading capacitor has 3000 volts peak 
voltage at 1500 watts using a 3000 ohm mid impedance, and 
2100 volts peak using 1500 ohms at 1500 watts

So while you might gain 6dB or more harmonic suppression 
into a 50 ohm resistor load with a good layout, you now have 
to deal with 3000 volts where most amplifiers only have 700 
volts peak. This is OK in a monobander but not easy to 
handle with components in a production bandswitched 
amplifier.

On the other hand I can get more than 20dB  addition 
suppression with a regular pi if I add a simple series tuned 
trap to ground, and the added cost might very well be close 
to nothing. Why not just use a shorted 1/4 wl stub?

I'd hate to have people think that a conventional pi-L with 
a mid impedance of 200 ohms or so would do anything remotely 
close to what a system with a mid impedance nearly the same 
as the anode impedance would do. Wen should probably all be 
careful to not generalize and just say "a pi-L does this" 
because it really depends on the layout, the load impedance, 
and the mid impedance of the pi-L.

In the case where we can use a normal size bandswitch, cover 
160-10, and a reasonable voltage loading cap we are lucky to 
gain 3dB for all that work....and very likely could make 
things worse if we are not careful with the layout. And that 
is into an ideal load...not an antenna where things can go 
very bad and the pi-L actually increase the harmonic 
levels..

It's pretty tough and very expensive to obtain the perceived 
harmonic advantages of a pi-L in the real world of 
multi-band amps. It's much cheaper to do it other ways.

73 Tom








Before I used simple Pi, but for SO2R M/M operation I 
converted them all.
Vacuum output capacitors were used already except on the 
160m amp, no padder
anymore.
Usually the Pi-coil is above the chassis, the L is below it 
together with
the load cap.
In the 160m amp both coils are above chassis but 90 degrees 
to each other
and there the difference is very remarable.
The middle impedance on the lower z amps is 600-800 ohms and 
I am using 1200
ohms on the 160m amp.
Antennas are monoband as well.
M/M is possible while antennas are 3m apart on the same 
mast.
I tried Pi-L on a new built multiband amp but gave up and 
went to Pi again,
to many problems with arcs in the bandswitch.

73
Peter

-----Original Message-----
From: Tom W8JI [mailto:w8ji@w8ji.com]
Sent: Donnerstag, 29. März 2007 01:04
To: Peter Voelpel; amps@contesting.com
Subject: Re: [Amps] Pi-L Network


> Would be very interesting.
> And the Pi-L  IS better then the Pi on harmonic 
> suppression.
> I use it in all my monoband amps.


So how many dB better is it into your actual antennas?

What middle impedance do you use?

Do you partition or shield the L coil from the main tank 
coil and tube
leads??

73 Tom







_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>