[TowerTalk] (no subject)

[K7GCO@aol.com]

In a message dated 9/19/00 7:36:02 AM Pacific Daylight Time, 
w8ji@contesting.com writes:
<< 
 > I am most interested in seeing your trap measurement results. Apparently I
 > missed an earlier E-mail telling of this venture.  I am particularly
 > interested in some comparative data between: W2AU, RG58 coax traps, RG59
 > coax traps, etc.
 
 Send me a coax trap quickly, and I'll measure it.
  
 > What parameters will you be looking at on each trap?
 
 Only ESR on frequencies where the trap is passing signal and 
 EPR on frequencies where the trap is "trapping".
 
 The concern was loss.
  
 > I find it particularly interesting that you have built a LARGE shielded
 > box in which to do your measurements. I would tend to think that in order
 > to minimize the capacitive effects, etc., on the trap the "box" would
 > become so large as to become unwieldily!  
 
 I am not worried about five or ten percent shift in resonance, but I 
 am worried about the coupling to everything on my bench including 
 the test cables and lights. As I said, if I even get close to the trap I 
 can see the EPR of the trap drop drastically. That's because the 
 traps mostly have one foot or more of conductor exposed to the 
 environment around the trap. At the parallel resonant frequency, 
 and dissipative load coupled to the trap through stay capacitance 
 will de-Q the trap drastically.
 
 I have to go through a similar exercise to measure physically large 
 inductors.
 
 >Being an EMC engineer, I would
 > tend to think that an RF anechoic chamber would be just the ticket for
 > doing these measurements.
 
 An RF anechoic chamber would do nothing for this problem, 
 because the problem isn't the radiated field. The last thing you'd 
 want to do when measuring Q is add energy absorbing structures 
 around the system under test that are not there in the normal 
 environment.
 
 An outdoor test over a "lossless" ground plane would work for 
 measuring trap resonant frequency in the actual system, and the 
 exact value of reactance added, but I can't thing of any way to do 
 that accurately with test equipment near the trap.
 
 What I would do, if I had the time, is put the trap in a vertical 
 mounted over a ground screen and determine the resonant 
 frequency and inductance (at pass frequencies) of the trap by 
 measuring it's effects on various height radiators.
 
 The problem is the physical size of the trap and the extra tubing 
 sticking out each end. That doesn't make for something that can 
 be measured using normal "hook it up" techniques. I can get a 
 reasonable idea of the parallel resonant frequency in my test setup, 
 but not enough to be useful in a model because of how critical that 
 effect is on resonant frequency of the elements.
 
 If the trap was physically small, I wouldn't have had to go through 
 this mess. I could have laid it on my wooden bench. 
  
 73, Tom W8JI
 w8ji@contesting.com
  >>
 
  This chamber to measure very Hi-Q traps has some merit.  After measuring 
traps for years I came up with a design observation and simplification.  How 
totally valid remains to be seen.  After tuning many trap beams and dipoles I 
found that the assumed and desired Hi-Q trap to stop RF at a particular point 
for resonance wasn't as necessary as logically assumed.  I found that Lo-Q 
traps will also establish a high enough impedance to establish resonance--up 
to a point.  Even then that could be compensated for--up to a point.  I've 
also found that some Hi-Q traps can even heat with all that RF current 
spinning around inside creating a Hi-Z.  Anyhow sufficient decoupling is the 
desired goal and the finer points don't always get addressed.

In regard to the blue plastic traps Hy Gain used, I'd have to say that they 
were on the Lo-Q side with all that dielectric filling every space and their 
other characteristics.  They were easy to GD.  I'd lay them on the arms of a 
wooden chair and GD.  Non metal surrounding objects didn't seem to be much of 
a problem for them or the ones with the metal 1 3/4" sleeves.  I used to mark 
the resonant frequency on the metal sleeve ones with a vibrator pen and on 
the plastic ones with a staining ink pen.  It never changed it seems 
regardless how I GD'ed them.  I admit I never had any real Hi-Q traps as 
there didn't appear to be any useful reason to use them with all their 
problems.  I may try some just to see what is gained.  There is evidence they 
can sharpen the SWR curve.

Further more traps in tribanders and trapped verticals aren't tuned to some 
frequency in the band.  They are tuned as low as 1.7 MHz lower.  The reason 
was they were wound with aluminum wire and would heat if tuned in the band.  
To compensate for the low resonant frequency of the trap all one has to do is 
to shorten the tubing.  You still end up with resonance.  Overall a multiband 
vertical will be much shorter from trap low resonant frequencies along with 
the trap inductance and shielding.  It sold at a low price, had reduced 
bandwidth but didn't really work too well--except on saltwater.  I used to 
resonant the traps on the high side of the band.

I then rewound the traps with copper wire and this allowed me to resonate 
them closer to the band before heating.  I made double connections also for 
the wire tips as they were often bad.
 
Then for one of the "All Time Blunders" in trap manufacture.  Those supported 
with aluminum tubing had 2 shorted turns in the trap field. The ends of 
tubing are "shorted turns" when too close.  I slit the 3/4" tubing back 3/4" 
and the trap resonant frequency went down 220 KHz on 20M.  At first I was not 
aware of all these Q lowering factors nor was Hy-Gain I found out.  They 
later added copper wire.  I have even Lower-Q examples of no Xc traps just of 
certain amounts of inductance (5-20 times more) that work just fine creating 
just enough impedance to create resonance and good SWR curves.  They even 
leak and adjusting the end tips on the lower band affect the higher band some 
but with some creative and open minded adjustments the desired resonance's 
are easily obtained and a stable Lo-Q circuit and resonance exists.  I've 
even duplicated them in Eznec.  A w4 in advertised them in QST for years.  
The KISS concept shines very brightly.  I normally make every effort to 
squeeze every element of efficiency out of a circuit but in this case could 
not see any areas of loss without warm components.  Hi-Q and all it's 
problems just wasn't the direction that seemed productive or pursued by the 
manufactures.  I will actually try some Hi-Q coils but at this time see no 
need for them.  It seems like another example of where "conventional wisdom" 
has led many (even I) down the wrong path.  When I see conflicts of 
"conventional wisdom" I jump on them.  

I have increased the Q of tuned grid circuits in input RF stages of receivers 
like the Drake TR4.  I picked up 10dB on 10M and 6 dB on 15M.  Receiver input 
circuity has ALWAYS been wound with too small diameter of wire.  Hi-Q trap 
circuits has it's applications but in antenna traps I'd like to see some data 
to really support it.  Lo-Q traps work just fine.  Data on Hi-Q traps seems 
to be just "academic."  No one seems to be using them in antennas worth 
observable benefits.    K7GCO  


--
FAQ on WWW:               http://www.contesting.com/FAQ/towertalk
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com