I need some help and can not find the answer anywhere.

When neutralizing a triode, how much leakage between grid circuit and
tank circuit is allowed?

I am looking at the output of the final loading cap with a scope and get
a few volts of RF when the NC is nulled. This is with 20 ma grid drive
or about 10W input. The output just tickles the wattmeter on the 30
watt range and rises to about 10w when the NC in moved from the null
position.

Second question: should I see any leakage when the filament is off?

Thanks for your help.

Jack



--
PHOTO OF THE WEEK: http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silver http://schmidling.com

Jack Schmidling wrote:

I need some help and can not find the answer anywhere.

When neutralizing a triode, how much leakage between grid circuit and
tank circuit is allowed?

I am looking at the output of the final loading cap with a scope and get
a few volts of RF when the NC is nulled. This is with 20 ma grid drive
or about 10W input. The output just tickles the wattmeter on the 30
watt range and rises to about 10w when the NC in moved from the null
position.

Second question: should I see any leakage when the filament is off?

Thanks for your help.

Jack




What type of circuit: grounded grid or ground cathode?

Grounded grid with input circuit below chassis should require no neutralization.
The chassis provides isolation.

Leakage with filament 'OFF' is common in a grounded cathode circuit' The
inter-electrode tube capacitances are still there. So, circuit stray capacitance
and tube capacitances still need compensation.

Leakage with the tube removed indicates inductive coupling between the input
circuit [grid] and the plate tank circuit. Either a lack of input stage to
output shielding or the two tank circuit coils are not at 90 degrees relative to
each other [90 degree orientation yields minimum mutual inductance].

Dave wrote:


What type of circuit: grounded grid or ground cathode?

It's class C with the filament grounded through a meter.

Leakage with the tube removed indicates inductive coupling between the
input circuit [grid] and the plate tank circuit.

I just tried it without the tube and it seems to have nothing to do with
the tube. When both input and output are resonant, 5 watts in gives me
5 watts out.

The input is a toroid under the chassis so it is not likely inductive
coupling. What I just found was that it goes away when I disconnect the
NC at the plate end.

What on earth does this mean?

js

--
PHOTO OF THE WEEK: http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silver http://schmidling.com

Jack Schmidling wrote:
Dave wrote:


What type of circuit: grounded grid or ground cathode?


It's class C with the filament grounded through a meter.

Leakage with the tube removed indicates inductive coupling between the
input circuit [grid] and the plate tank circuit.


I just tried it without the tube and it seems to have nothing to do with
the tube. When both input and output are resonant, 5 watts in gives me
5 watts out.

The input is a toroid under the chassis so it is not likely inductive
coupling. What I just found was that it goes away when I disconnect the
NC at the plate end.

What on earth does this mean?

I suspect the NC is NOT providing feedback of proper amplitude AND phase. A
connection from the plate to the grid is inadequate. The feedback MUST connect
into the grid tuned circuit.

Another reply speaks to this issue.

Jack Schmidling wrote:
I need some help and can not find the answer anywhere.

When neutralizing a triode, how much leakage between grid circuit and
tank circuit is allowed?

I am looking at the output of the final loading cap with a scope and
get a few volts of RF when the NC is nulled. This is with 20 ma grid
drive or about 10W input. The output just tickles the wattmeter on
the 30 watt range and rises to about 10w when the NC in moved from the
null position.

Second question: should I see any leakage when the filament is off?

Thanks for your help.

Jack
The neutralizing capacitor compensates for the grid - plate capacitance
and possibly other
feed back paths. The maximum balance required depends on the gain of the
amplifier stage.

The balance has to be sufficient to keep the amplifier from oscillating
and a little bit more.
Assuming a gain of 10 dB, then the output with drive and the filament
off should
be more than 10dB below the drive level.

Usually a grounded grid stage doesn't require neutralizing because the
grid is grounded interrupting
the input to output capacitance

Cheers
Bill K7NOM

Hello Jack,

In order to neutralize a triode you must have a balanced grid circuit.
You build the input tank circuit just like you would if you had 2
tubes in push pull but rather than connecting one side of the grid
circuit to a second tube you connect it to your neutralizing
capacitor. The other side of the neutralizing capacitor goes to the
plate of the tube.

A bridge or single ended type grid tank circuit will work with a
tetrode but the triode needs the balanced circuit in order to get
enough feed back.

You could also use a balanced plate circuit like you would use with a
push pull circuit and a single ended or bridge grid circuit but a
balanced grid circuit is usually much easier to do than a balance
plate circuit.

73
Gary K4FMX


On Sat, 17 Feb 2007 13:46:35 -0600, Jack Schmidling
wrote:

I need some help and can not find the answer anywhere.

When neutralizing a triode, how much leakage between grid circuit and
tank circuit is allowed?

I am looking at the output of the final loading cap with a scope and get
a few volts of RF when the NC is nulled. This is with 20 ma grid drive
or about 10W input. The output just tickles the wattmeter on the 30
watt range and rises to about 10w when the NC in moved from the null
position.

Second question: should I see any leakage when the filament is off?

Thanks for your help.

Jack

Gary Schafer wrote:

A bridge or single ended type grid tank circuit will work with a
tetrode but the triode needs the balanced circuit in order to get
enough feed back.

I have a balanced input. It happens to be a toroid but I started with a
coil and two gang cap.

I got off on a tangent when I removed the tube so I now understand what
happened there.

My question really was looking for how much is enough? All the old books
say when the final does not effect the grid drive, it is neutralized but
everyone has a scope these days and it is so easy to hook it to the
output and measure the exact feed through.

I was looking for a number and I believe someone said 10 DB down from
input if the amp has 10 DB gain, so I guess I got the answer.

I am not quite sure how to translate this to volts at the output. Or
can I put in ten watts and only see the needle twitch on a 30 Watt FS
wattmeter at the output and consider it done?

js


--
PHOTO OF THE WEEK: http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silver http://schmidling.com

"I am not quite sure how to translate this to volts at the output. Or
can I put in ten watts and only see the needle twitch on a 30 Watt FS
wattmeter at the output and consider it done?"

Neutralizing a triode is not rocket science. If you are using a final with
a lot of plate dissipation, they do not get hurt when the neutralization is
not right. In the old days, many hams simply fired it up and tuned the NC
until bad things stopped happening. they did it with high voltage on. Some
of them got shocked.

A 30 watt FS output meter that barely twitches when driven by 10 watts and
the NC tuned for minimum may be enough. All you need is good enough. No
extra points for perfect neutralization.

To test, put a load on the amp and remove exciter. Turn on and tune input
and output tuning and see if it takes off. If it does, back to neutralizing
step 2. If it works as supposed to, then connect the antenna and start
calling CQ DX.

73, Colin K7FM