[AMRadio] Pi-Net vs Link Couple

[Donald Chester]

>Actually, it isn't difficult if you follow the instructions in the BC610
>manual. First the tuning unit for 6 - 8 Mhz is used and also the Plate
>coil (with swinging link) also 6-8 Mhz. Then disconnect the output leads
>from the transmitter. Turn on the filaments, make sure the plate HV is
>off set excitation meter to PA grid, set exciter plate sw to on. Adjust
>the tuning unit controls to resonance at 7 Mhz, then tune the RF PA
>tuning control slowly to get a resonance indication. If neutralization
>is faulty, resonance is indicated by a sharp dip in the reading of the
>PA grid current reading. Adjust the neutralizing cap until there is
>little or no dip in the grid current meter at resonance. when this is
>properly neutralized the dip will NOT exceed 3 MA. That's all one time
>does it for all bands Except 160 meters.


That held it for 160m as well, when I helped Roger N4IBF(SK) get his BC-610E 
going.  We used one of the 1.5-2.0 mc/s coil sets.

I  have an old Mirage SWR/Wattmeter.  The power output calibration has 
always sucked, but it is a good SWR indicator and can be adjusted to be very 
sensitive, by switching to the lowest power range (20w), and turning the 
sensitivity (for SWR readings) all the way up.  I can get good readings with 
a fraction of a watt of rf.

I have a little 50Ω 50-watt (they call it 100 watt) dummy load that I 
terminate the Mirage into.  I connect the input side to the link output of 
the transmitter and it makes an excellent neutralisation indicater.  I set 
it way below maximum sensitivity, and tune grid and plate tanks until I get 
an indication.  Then adjust sensivitivity for full scale.  Then adj neut 
cap(s) for the deepest null, then repeat the process, including reresonating 
the tank circuits.  On all my rigs I can completely null out the rf 
feedthrough.

One word of caution, especially with single ended finals.  Just turning off 
the HV isn't enough.  You need to actually physically disconnect the B+ lead 
from the final.  Otherwise, I have found that I get a plate current 
indication and cannot completely null out the rf in the plate tank no matter 
what the setting of the neutralising caps.  Evidently if the plate has a dc 
path to ground, some electrons that escape through the grid structure of the 
tube will hit the plate and return to ground thru the tank circuit if the 
plate is not floating DC wise.  This causes rf to appear in the tank circuit 
even when the rig is perfectly neutralised.  The effect is less noticeable 
in pushpull finals, but I always disconnect mine anyway.

Another method is to adjust the grid tank for maximum grid current.  It is 
particularly useful for verifying that the final is neutralised following 
one of the above procedures.   With the final approximately neutralised, 
turn on the HV and tune the transmitter up normally.  Making sure the plate 
tank is perfectly dipped, and the grid tank exactly at resonance, tune the 
final tank capacitor back and forth above and below resonance.  If the rig 
is properly neutralised, the grid current will peak when the final tank is 
at resonance, and drop off symmetrically each side of the resonant dip.  If 
the neutralisation is off, the grid current will keep on increasing or 
decreasing as you tune through resonance, or it will drop off faster on one 
side than on the other.  You can even tell whether to increase or decrease 
the neut capacitance by which direction the plate current moves relative to 
the direction you are tuning the grid capacitor, but I always have to  look 
that up in a book and never trust my memory to do it that way.

Usually when I neutralise my final, I check and double check using all three 
of the above methods.

Even if the final does not actually self-oscillate, you should always 
neutralise the stage as completely as possible.  An improperly neutralised 
final can cause splatter, since it introduces some phase modulation onto the 
signal in addition to the amplitude modulation.

Don k4kyv