Han Higasa wrote:
>Hello OMs.
>
>I am using a home brew made six-meter linear amplifier with
>a 3CX1200A7, pi input and pi-L output network.
>
>Now I have a difficulty to find a design/calculation method for
>the very popular input matching network:
>
>input - VC1 - L1 - VC2 - L2 - filament (cathode) + Cin//R
>
>Where VC1 is in series to the circuit, and VC2 is connected
>in parallel to the junction of L1-L2 and the ground.
>
>
>This type of input matching networks have been used on
>various VHF amplifiers including the publication of the ARRL
>handbook, GS35B and 8877 amplifiers seen on the web.
>
>With a spectrum analyzer, tracking generator and an old Anzac
>return-loss bridge I found that VC1 adjusts the tuning frequency
>and VC2 adjusts SWR by cancelling any reactance on the frequency.
>
>Would you please tell me any basic theory or references for the
>circuit?
You can think of it as two back-to-back L networks. Analysing it in more
detail...
The net reactance of VC1-L1 is always inductive. (VC1 is simply a
convenient way to make this reactance adjustable; other constructors use
a variable slug-tuned coil instead.) To simplify the network, delete VC1
and think of L1 as having a smaller value.
Now divide VC2 into two parallel-connected parts (VC2a and VC2b). We now
have two impedance transforming L-networks which are connected
back-to-back at point X:
50 ohms - L1 (series) - VC2a (parallel) -X- VC2b (parallel) - L2
(series) - Zin of tube.
For design purposes you need to assume some value for the "intermediate
impedance" at X. To obtain valid solutions for both L-networks, Z(X)
must be greater than 50 ohms and also greater than Zin of the tube.
Zin also has the grid-cathode capacitance in parallel. This adds more
complications, but I hope you can see the general idea.
The most common practical mistake is to use too much inductance at L1
and L2. The network will still give an impedance match, but the
intermediate Z(X) may be quite high. This gives the whole network a high
loaded Q and narrow bandwidth, and makes it too sensitive to changes in
the input impedance of the tube.
--
73 from Ian GM3SEK
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|