For another data point, I made a number of coax traps using RG-58, and
measured the L, C, and Q. I don't have the numbers handy right now, but
using the measured values I looked at overall loss when used in antennas
for various combinations of bands. In most 3-band configurations, there
was one band where the loss was a couple of dB, and not necessarily the
band where the trap was resonant. Some 2-band configurations were ok,
some not.

On Field Day, I sometimes use a supplementary 3-band dipole arrangement
pointed N-S for California -- my main antennas point to the east. It
consists of a 40-15 trap dipole, with a 20 meter dipole drooping under
it as a separate inverted vee but connected to the same feedline. With
RG-174 traps, the loss is tolerable with the 40-15 combination, but not
with 40-20-15.

Trap loss when used in an antenna is a function of many factors, and
there's really no practical way to determine what it'll be except for
modeling. Of course you have to know the equivalent L, C, and R or Q.
You have to check it on all bands, since it might easily be ok on the
band where it's resonant but lossy on others. Way back in 1998 I did a
brief study of the effect of L/C ratio on trap loss, assuming a constant
trap Q. The results, along with the EZNEC models I used, are at
http://eznec.com/misc/ as traps.zip. I haven't looked at them for years,
and see that I used Laplace type loads, since this was done before EZNEC
had RLC type loads available. Modern EZNEC has a special "trap" type RLC
load with the R in series with L, and C in parallel with the
combination, and with the ability to make R vary realistically with
frequency. Nonetheless, EZNEC users might find the models useful, and
others might be interested in the results given in the accompanying text
file. Note that the Q I assumed for the study, 400, is probably better
than you'll see with a coax trap.

No matter what kind of trap you make, it's essential to keep water from
getting between the turns of the inductor or coax. Letting that happen
is guaranteed to trash the Q.

Roy Lewallen, W7EL

Owen Duffy wrote:
On Sat, 11 Nov 2006 16:19:48 -0800, Danny Richardson
wrote:

On Sat, 11 Nov 2006 22:07:14 GMT, Owen Duffy wrote:

Programs tend to underestimate the losses of these traps.
Owen,

What is your estimation of the loses? Will it be more than a dB?

Danny,

I recall modelling dipoles incorporating such traps with NEC some time
ago. My recollection is that the loss is of the order you suggest,
which might seem insignificant in the context of a path budget, if it
wasn't for the fact that it might about to a significant amount of
power to be dissipated in the trap, depending on the power level /
mode.

Operation at resonance exacerbates the situation.

My sentiment is not so much that the traps are a bad idea, but the
design tools that are around seem to take shortcuts and are
inconsistent. The design tools would make one think that the trap
designs are better than they really are.

I tried Reggie's COAXTRAP and compared the predicted inductance with
his SOLNOID3, and they differ by a factor of 4, and SOLNOID3 produces
the more believable result. I don't know why that would be so.

Owen
--