Hello,
An attempt was made to hook-up single 10 meter a pi-network input to
a GS-35B. Unfortunately, it didn't work although seemed to check out
okay on the bench and when it was installed in the amp on the bench.
Details:
The input network was designed using the spread sheet pi calculator
by Phil, VE3OZZ, with the following entered parameters; Freq. =28.3
MHz, Q=2, Rin=50 Ohms, Rout=30 Ohms. The 30 Ohms was derived from
review of the archives on GS-35B input impedance and from the
equation Eg/(Ip+Ig) where likely operating parameters would be Eg=30
V, Ip=0.8 A, Ig=0.2 A.
The spread sheet produced the values of C1=142 pF, C2=139 pF, L=0.22 uH.
L was wound on a T68-2 with six turns of number 18 gauge wire. The
six turns was calculated using Amidon information.
The circuit was assembled on FR4 perf. board with a 30 Ohm
composition resistor across C2.
The pi-network was tested with a MFJ259 connected across pi network
input, C1, by ~3 feet of RG58.
The circuit didn't work as the resonance was ~26 MHz.
C1 and C2 were adjusted for a minimum SWR (~1.1 : 1) at 50 Ohms on
the MFJ259 at 28.3 MHz. The values for C1 and C2 are, respectively,
130 pF and 80 pF.
The pi-network C2, less 20 pF to compensate for the tube's input
capacitance, was connected to one of the GS-35B's filament
connections. L was attached to the same point on the filament and C1
and the 0.01 uF mica blocking capacitor to the other end of L. A
similar 0.01 mica capacitor is connected between the filament
terminals. A 30 Ohm resistor was connected across C2 and the MFJ259
connected to the free end of the 0.01 uF blocking capacitor. The
circuit checked out okay. The filament choke was connected to the
filament connections when the measurement was taken. The 30 Ohm
resistor and MFJ259 were removed and the RG58 cable from the input
relay was connected to the 0.01 blocking capacitor.
The xcvr connected and amp. turned on. When the amplifier was keyed,
the xcvr registered a SWR greater than 3 to 1 at 28.3 MHz. No
resonance was seen by a dip in the SWR reading in the xcvr over the
10 meter band.
Okay, where did I go wrong in my thinking and/or implementation?
73,
Mike, K4GMH
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|