The AM Forum

By searching this forum, and other places, it's my understanding that, in a class C stage, having maximum power output not co-incident with plate current dip, is an indication of improper neutralization.

Nowhere have I found a way to pre-determine whether I need more or less neutralization capacitance to correct this. Before I start screwing with the neutralization capacitor, can it be determined by which side of the dip, with respect to the pi-network "tune" capacitor, that maximum power output is obtained? Is there a rule of thumb for this? ???

Or, if I'm lazy and do nothing to improve the neutralization, should I just tune for the dip, or the maximum power output? :-\

The subject transmitter is a Heathkit DX-100. This one does has the 'stiff wire' neutralization capacitor poking through the firewall. As I understand it, not all DX-100's have this wire. There's no mention of it, or any neutralization procedure, in the pdf manual that I have. What's a poor boy to do? :'(


Don

Does the transmitter meter grid current?

If so, see my post on dynamic neutralization.

Hi Don,

Another neutralization technique is to adjust the cap for minimum feedthrough RF called "cold neutralization."  

Turn on the filaments, peak the tank,  no HV, key the rig and hook a scope probe to the loading cap output. Make sure the screen current doesn't soar - try it with the fils off first.  Then inject some signal generator RF into the grid input.

Get a full trace on the scope.  You should be able to find the point where the neut cap balances out  the feedthrough grid-to-plate capacitance and tunes (nulls the RF) to minimum on the scope.

This is a good starting point and will tell you if the neut is really working right. If you later want to hot neut it,  use the plate dip technique.  It should already be very close if not perfect.

Generally, in the end, the grid current should peak, the plate current dips while the power output peaks.

I don't know which direction to tune your neut cap, but it will not matter if you null out the feedthrough signal accurately.


T

In the the Johnson Valiant manual manual page 15. It gives a good explanation of the neutralization procedure they suggest. I have followed it and it works well.

John N8QPC

The Dynamic Neutralization thread by WA4WAX is here:
http://amfone.net/Amforum/index.php?topic=41892.0
In it, he refers to an article that I reprint now (emphasis is mine):

---

RADIO-ELECTRONICS
March 1951

DYNAMIC NEUTRALIZATION OF CLASS C AMPLIFIERS
By MARVIN H. KRONENBERG, W2IJU

The need for neutralization is one of the disadvantages of using triodes and poorly shielded tetrodes as r.f. amplifiers in transmitters. The usual procedure for neutralizing an amplifier is to remove its plate supply voltage and apply normal r.f. grid drive. Set the neutralizing capacitor or capacitors so grid current is constant and there is no indication of r.f. in the plate tank when the plate tuning capacitor is tuned through resonance. Although this method takes considerable time, it is satisfactory when performed correctly on an amplifier which is operating properly. A considerably faster and more accurate neutralization method is used by many commercial operators. This operation, which may best be described as dynamic neutralization, is carried out while the amplifier is operating with normal load and plate voltage. As most amateurs and commercial operators have noticed, the maximum grid current and minimum plate current (resonance) occur at the same setting of the plate tank capacitor when the amplifier is fully neutralized. All too often, this does not happen when the rig is neutralized using the method just described. Instead, the grid current may peak when the plate circuit is tuned slightly above or below resonance. It is possible to tell if there is too much or too little neutralizing capacitance by noting whether the peak occurs on the low or high capacitance side of plate circuit resonance. The procedure for dynamic neutralization is based on these observations. Follow these rules when neutralizing any r.f. amplifier by the dynamic method :1. Set the neutralizing capacitors to a reasonable value. If the amplifier has been neutralized previously, the approximate settings will be known. Otherwise it may be necessary to determine the approximate settings by the usual neutralizing method. 2. Apply normal excitation, then turn on the plate supply and adjust the plate load to normal. Carefully tune the plate circuit on both sides of resonance and note the grid current. (a) If the grid-current maximum coincides with plate current minimum leave it alone. The stage is fully neutralized. (b) If the grid current peaks on the low capacitance side of plate circuit resonance, more neutralizing capacitance is needed. (c) Reduce the neutralizing capacitance when the grid current peaks on the high capacitance side of resonance. Only a few trial settings are required to reach perfect neutralization as indicated by maximum grid current and minimum plate current occurring at the same setting of the plate tuning capacitor. Any amplifier neutralized by this method will pass any test for neutralization if it is free from parasitics and other defects. It is simple to test the neutralization of any transmitter. Simply de-tune the amplifier. The grid current meter will tell the story. A final word of caution: remove the plate voltage from the stage before touching the neutralizing capacitors if they don't have long insulated shafts which can be reached while standing well away from the rig. (If your rig is one of those compact jobs in a relay rack or cabinet, better pull the big switch and discharge the filter capacitors before making any adjustments. Those innocent looking coils and plate caps can do a lot of damage. We had to learn the hard way. Editor)

---

Great stuff! Just what I was looking for. Apparently, the rig itself has all the test instrumentation needed to do the dynamic neutralization procedure. It's already 'sufficiently' neutralized, but if it were a new build, then the 'cold neutralization' procedure that Tom suggests would probably be a safer starting point.

Thanks to all!
Don

I just ran the dynamic neutralization procedure. This is what I found:

The rig was pretty close with respect to grid current peaking either side of the plate current dip. Still, I balanced it a little better. But, the maximum power output still did not coincide with the dip. ???
Upon loading the amplifier a bit more heavily, I found that they did coincide quite nicely. :)

My conclusion is that the dynamic neutralization probably provides the best stability of the amplifier, but there may still be a sweet-spot for loading where the maximum power output is at the plate current dip. At least, that's what I'm seeing now. I had been running the rig at 154 watts input (770V@200mA) but it looks like it's happier at 173 watts input (770@225mA). The dip feels a bit 'softer' at the higher loading so I think the tank Q must be a little lower this way.

Problem solved — I guess.


Don

By softer do you mean the dip seems broader, and not as sharp or deep? Thats a result of increasing the loading, the plate dip at resonance is not as deep or as narrow as it is at lighter loading. If you had no loading at all on the final, the dip will be very sharp and deep, almost to zero plate current as you tune through resonance. I have used this before on a link coupled rig, unhook the antenna or remove the coupling coil and key it up and tune it to be sure that the coil/cap combo resonated.

Yes, softer/broader...  exactly.

When I produced 13.56 MHz pulsed 10 kW generators using a 4PR400, I always neutralized without HV and without filament. Just tune the tank at max output and neutralize at minimum output while applying a drive. Worked perfectly.
And the neutralizing was critical with an amp that amplified 5 Watt pulses into 10 kW

Hi,

I seem to recall the following relative to the dip in PA Plate current:

 if the maximum power output occurs as the PA Tune cap value is reduced, you need to increase the neutralizing cap value
 if the maximum power output occurs as the PA Tune cap value is increased, you need to decrease the neutralizing cap value

I generally use the "remove the plate and screen voltages and adjust the driver and PA tune caps for maximum output (usually monitor with a scope on the output, then adjust the neutralizing cap for minimum feed through" method. This is often close enough, but sometimes I use the above steps to "tweak" the neutralization a bit.

Bob,  AB1MN