Minimum value for RF chokes

[PA4WM]

Hi,

Most tube transmitter designs call for a pie wound 2.5mh choke in either the anode, kathode or in the negative bias lead.
I just searched my junkbox, and found a few of those pie wound chokes.
Meassured them with an inductance meter, and found out I don't have a 2.5mH at all.
The values range between 0.4mH and 1.9mH

Considering a transmitter for 160m, 80m and 40m only, is there a minimum (or even a maximum) choke value to be used?

Regards, Martin

[PA4WM]

hi John,

I assume you mean the "big" anode choke in the final?
I'm more refering to the smaller chokes used in the osc. buffers and driver, to keep the RF off the DC power.

Martin, PA4WM

[WA1GFZ]

10X the plate Load resistance at the lowest operating frequency is plenty.
Check it for series resonance

[PA4WM]

10X the plate Load resistance at the lowest operating frequency is plenty.
Check it for series resonance

Not sure if I get this right...

For instance:
The transmitter buffer/amplifier will be a 6146. At 500volts I recon it's plate wil draw approx. 75mA.
the buffer/amplifier plate resistance will be approx. 6500 ohm
So 10 x is 65000 ohm

X=2.pi.f.l
65000 = 2 x 3,14 x 3500000 x l
l= 65000/21980000
L=0,00296
L=2,9 mH

The same calculation comes up with 5,7mH when used on 160m (1.8 mHz)

If I use the calculation on the negative grid impedance with 150 volt and 10mA, I'll have to use a choke of 65mH at 3,5mHz......

I'm sure I'm doing something wrong here, since most of the schematics call for a "standard" 2.5mH pie wound choke in the various leads.....

[WA1GFZ]

Z = V/2XI for class C so the value is 1/2

[KM1H]

A 1.5mH is more than sufficient even at 160M as the bypass cap will remove any traces of RF heading to the PS.

You can actually use a lot less L and all that means is the choke will become part of the pi-network and require a little retuning. Commercial amps use 225uH down to 160 and the choke shows little heating even at 2000W+ out. Since electrolytics react violenty when goosed by RF the bypass cap reactance at the lowest frequency is critical. A .01 or a pair of .0047's in parallel do the job at 160.

I rarely use a RFC in the bias line as strange LF oscillations have been known to occur. A 2.7 to-3.3K carbon resistor and well bypassed works well, otherwise use a similar resistor in parallel with the RFC to swamp any resonances.

Carl

[PA4WM]

Thank you guys.

It is nice to do the math, but also good to know that 1.5mH is more then sufficient in combinatiuon with the by-pass cap....

Martin