I was a bit concerned to see Wes' data having a column of N/R for the copper
suppressor, when there clearly was a resistor built in there. Calculating
the missing values from L and Q gets a set of numbers consistent with the
parallel values.
I don't have time to go through thorough analysis, but it seems to me that
the Q of the suppressor network is only one small part of the overall
picture, and something of a distraction. I also think that it's unhelpful to
focus in on it in isolation as a parallel network. The suppressor works in
series from the anode to the tank circuit, so it's the series equivalent
that's most useful in visualising what is going on.
Comparing series impedance numbers for the W8JI/109 ohms and the NiCr60/100
ohms from Wes' table;
the copper version goes from .36 resistive + 7 inductive (112nH) at 10MHz to
98 resistive + 78 inductive (62nH) at 200MHz.
NiCr goes from .8 resistive + 6.7 inductive (106nH) at 10MHz to 79 resistive
+ 44 inductive (35nH) at 200MHz.
You cannot look at the suppressor in isolation - you have to view the whole
series effect between the anode and the tank circuit. The series inductance
of the suppressor is going to be in series with the connecting leads between
the anode and the tank circuit, which will probably be another 50-200nH,
depending on layout - this will reduce, maybe even swamp, the effect of
differences in inductance between the suppressors themselves.
In this example, for suppressors which are sort of similar at LF, NiCr
results in higher series resistance at frequencies up to 120MHz, and lower
above that.
I reckon it's not as easy as it might appear to say whether higher or lower
is better, but that's another story.
Steve
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|