Vic, Shane, and others,
Changing the circuit to be 2 inductors in series with a capacitor to
ground at the connection point between the inductors changes things but
does not fix the possible low overall circuit impedance from the tube to
ground at some frequency in the range of operation!
A few hours with LTSpice (free) shows this very clearly. Carl's post
about using either only the large inductor or both inductors from 160m
up through say 20m and then to ONLY use the small inductor for higher
frequencies works and is easy to do. Also a bypass cap can be switched
in for the higher frequencies which also "eliminates" the large
inductor. In every case I would measure the impedance of the choke in
the cabinet before applying RF power. Again anyone building equipment
without modest test equipment is "flying blind". Some of the
inexpensive RF impedance meters like the AIM4170 and others will teach
you more than you can imagine and ACTUALLY let you see what is going on.
A comment that was made about the large inductor being only +j2500 at
1.8MHz is not really relevant. The effect of this inductance on the
output network is to cause the tuning capacitor to be increased
slightly. The tube has no idea this inductance is present. The bigger
problem is the RF resistance of the plate choke which dissipates power.
This is why attempts to use ferrite cores need to be analyzed carefully
since the ferrite material lowers the Q. Ferrite works great for RF
suppression where the current is low but for high power plate chokes and
baluns you want the Q to remain high so ferrite material such as #61 or
#64 could possibly be used but not #75, #43, or #31 for this application.
Larry, W0QE
On 6/10/2011 11:22 AM, Vic K2VCO wrote:
> Correct, but if there is a bypass capacitor to ground in between the chokes,
> then the
> additional impedance provided by the big choke -- even if lower than it could
> be due to a
> nearby resonance -- is still useful. The voltage applied to it will be much
> lower than if
> it were the only choke in the circuit and it won't burn out. So at worst the
> second choke
> would not be of much help on certain frequencies.
>
> On 6/10/2011 8:09 AM, Larry Benko wrote:
>> Vic,
>>
>> Look at my post of a few hours ago. Two non-resonant chokes in series
>> can together be resonant. This is basic circuit theory and there is no
>> way around the issue.
>>
>> Larry, W0QE
>>
>> On 6/10/2011 8:43 AM, Vic K2VCO wrote:
>>> Because if I do this, the bigger choke will be out of the circuit
>>> completely on the high
>>> bands. My method will leave it there, after the bypass, to provide
>>> additional filtering on
>>> the HV lead to the power supply.
>>>
>>> On 6/9/2011 11:12 PM, Angel Vilaseca wrote:
>>>> Vic,
>>>>
>>>> Why not use the vacuum relay to short the big choke? That is what is done
>>>> on the tank
>>>> circuit when bandswitchig.
>>>>
>>>> 73 de HB9SLV
>>>>
>>>> Vic K2VCO a écrit :
>>>>> I'm planning another amplifier (2 x 4-400A, 1.8-28 mHz) and I was
>>>>> thinking that I would
>>>>> deal with the "RF choke problem" by having two chokes in series. The one
>>>>> nearest the
>>>>> plates will have enough inductance for 20-15-10 meters with no series
>>>>> resonances below
>>>>> 30 mHz. The second will be 1 or 2 mh to provide enough inductance for the
>>>>> 1.8-7 mHz
>>>>> bands. I will use a vacuum relay to connect a bypass capacitor from the
>>>>> junction of the
>>>>> two chokes to ground on the high bands. Of course there will be further
>>>>> bypassing at the
>>>>> cold end of the big choke.
>>>>>
>>>>> That way, both chokes will always be in the circuit to help keep RF out
>>>>> of the power
>>>>> supply but the resonances and distributed capacity of the large one won't
>>>>> upset the
>>>>> higher bands.
>>>>>
>>>>> Is this a good idea or am I missing something?
>> _______________________________________________
>> Amps mailing list
>> Amps@contesting.com
>> http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|