Stefan,

>So a high Q must be the goal, loaded or unloaded.

Regarding unloaded Q I think there is agreement

Regarding unloaded Q it depends HOW it is increased.
As Jim has pointed out increasing unloaded coil Q also increases somewhat
the overall system loaded Q
by reducing circuit total R, this is perhaps what Graham was concerned
about.
But the benefit overweighs this small disadvantage. 

As a matter of fact with a matching network consisting of two reactive
elements (L,C) or of one coil
and an autotransformer as we have here it is not possible to choose a
desired loaded Q (without the exception above).
You get the lowest possible loaded Q anyhow and it is determined by the
complex impedance of the antenna.
But if you match with three reactive elements like a Pi- or T-Network then
you can choose - within certain limits - 
the loaded Q by choosing the components values but it will be always higher
than with two elements.
High XL and XC components will result in a high loaded Q.
In this case time the high loaded Q is detrimental for efficiency,losses and
heat go up, so THIS way of increasing 
loaded Q should be avoided.

The source resistance of a transmitter or a vector network analyzer
decreases the overall loaded system Q  
which increases bandwidth.
If you would feed the system by a voltage source the 3dB BW-points would
coincide with a VSWR of 5,9.
With a 50 Ohm source you have the 3dB points at a VSWR of 2,6.

My apologizes if this all has been too long and boring. :-)

73
Clemens
DL4RAJ 






>-----Original Message-----
>From: [email protected] 
>[mailto:[email protected]] On Behalf Of Stefan Schäfer
>Sent: Saturday, February 16, 2013 10:11 PM
>To: [email protected]
>Subject: Re: LF: Re: how to increase the Q of my loading coil?
>
>
>Am 16.02.2013 11:10, schrieb Clemens Paul:
>>   [...]
>> The higher the *loaded circuit* Q the higher are the losses in all 
>> involved components including the coil due to increasing 
>circulating currents.
>> Given a certain circuit increasing the Q of its components 
>always will 
>> reduce overall losses of that circuit.
>>
>> 73
>> Clemens
>> DL4RAJ
>
>...however if the transmit power is kept constant, then the 
>radiated power is higher, due to the higher currents through 
>the radiation resistance. Thus the losses must be lower here. 
>It is not a constant voltage source, at least Dimitris' class 
>E PA isn't :-) So a high Q must be the goal, loaded or unloaded.
>
>73, Stefan/DK7FC
>
>
>
>-----
>E-Mail ist virenfrei.
>Von AVG überprüft - www.avg.de
>Version: 2013.0.2899 / Virendatenbank: 2639/6104 - 
>Ausgabedatum: 14.02.2013 
>