I dug that article out of my stack of to-read stuff. And just like you say,
there is a very nice looking article in there with a very slick test setup.
I'm sure it gives repeatable results. Meaning if he takes the measurement
it will come out the same each time. great.
But I am in doubt as to the absolute accuracy of the indicated resonant
frequency. Reference figure 4, see the long pipe stubs on each end? In my
testing I found that changing the length of those stubs changed the
indicated resonant frequency. Which means that the fixture, as shown in
figure 4, is measuring the trap and the stubs on each end as a system - not
the trap in isolation.
It's possible I am not doing the test right. To hedge against that, I did
ask for suggestions on the N2PK reflector and my test method was OK.
I gave up my effort at that point having invested a couple of days in the
shop tinkering around with it. I'm sure the guy who wrote the article is a
far more savy RF engineer than I am and his method may be 100% perfect. But
I sure wonder what the indicated resonance point of the trap in fig 4 would
be if he whacked a couple of inches off one end of that trap tube...
73/jeff/ac0c
www.ac0c.com
alpha-charlie-zero-charlie
-----Original Message-----
From: Paul Christensen
Sent: Monday, April 13, 2015 11:31 AM
To: 'towertalk'
Subject: Re: [TowerTalk] trap antenna performance
Recently, an article on p.43 in the February, 2015 issue of QST discusses
trap performance and test fixture criteria. The article is titled: "Test
Setup for Measuring Traps." The author used a miniVNA device. It's worth
having a look.
Paul, W9AC
-----Original Message-----
From: TowerTalk [mailto:towertalk-bounces@contesting.com] On Behalf Of Jim
Lux
Sent: Monday, April 13, 2015 10:20 AM
To: towertalk
Subject: [TowerTalk] trap antenna performance
On 4/13/15 12:40 AM, Markku Oksanen wrote:
On "Low Performance Trap Antennas" : Is it really known what is the >
cause for the apparently very large difference between a correctly >
designed trap beam, say a TH6DXX or similar and one of the not so > good
performers? Trap loss can't be a big part because even > relatively low
power loss at a given trap, say 50 W, will burn the > thing very quickly.
Assume 3 elements, 6 traps that get exposed on a > given band and -3dB gain
compared to better beam, this would be some > 1500/2/6 watts, a lot of
watts per trap. Even it not this much just > loss, there would be ample
power to melt the insides of the trap
this is an interesting question...
I wonder if it's a "sensitivity to small changes in component values"
issue? Designing a trap antenna so that when its brand new it has the right
performance shouldn't be challenging: there's cut and try, if nothing else.
But traps are kind of difficult to model (at least in NEC) at a very
detailed level. So maybe if the L or the C of the trap changes
significantly, then the trap resonance changes, which then changes the
apparent series L or C in the element.
As Markku pointed out, the Q of the trap has to be fairly high, or they'd
melt from the losses.
So it could even be things like manufacturing variability. A trap change in
resonant frequency by 2 or 3 percent (a change in L and/or C of 4-6%.. is
that reasonable?) could change the apparent impedance of the trap at the use
frequency from inductive to capacitive.
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|