I have constructed an NEC model of an 80/40 dipole using bootstrap coax
traps. I have generated a helix of GW elements for the RG58C/U coax coil,
and use NT cards to model the coax interior.

Plain conductors are loaded with copper conductivity.

I am grappling with approprate conductivity for the outer surface of the
coax coil, thinking that it must be greater than copper because of
proximity effects and braid effects. I have initially tried loading those
segments with conductivity of one tenth of copper... but I don't expect
that will make the RF resistance tenfold.

With those numbers, the loss in the interior of the coax is more than
half the total loss.

Has anyone experience or thoughts on an appropriate way to load the
conductor representing the outer surface of the coax coil forming the
trap?

The model is below for anyone wanting to play with it. NOTE: the NT cards
are appropriate only at 7.05MHz. Some wrapped lines will need fixing.

Owen

CM 80/40m trapped dipole using coax traps
CM Alternative to NT cards:
CM TL 24 1 101 1 50 2.4 0 0 0 0
CM TL 14 1 103 1 50 2.4 0 0 0 0
CE
GH 1 80 0.005 0.05 0.0275 0.0275 0.0275 0.0275
0.00165
GW 2 1 0.0275 0 0 0 0 0 0.00165
GW 3 1 0.0275 0 0.05 0 0 0.05 0.00165
GW 4 1 0 0 0 0 0 0.05 0.00165
GM 10 0 0 90 0 8 0 10 1
GM 10 1 0 180 0 0 0 20 11
GW 101 20 -8.05 0 10 -14.5 0 10 0.001
GW 102 31 -8 0 10 8 0 10 0.001
GW 103 20 8.05 0 10 14.5 0 10 0.001
GE -1
LD 5 0 0 0 58000000
LD 5 0 1 80 6444444.4
LD 5 0 84 163 6444444.4
GN 2 0 0 0 13 0.005
EK
EX 0 102 16 1 0
NT 24 1 101 1 1.992e-3 -5.340e-2 -1.983e-3 5.701e-
2 1.992e-3 -5.340e-2
NT 14 1 103 1 1.992e-3 -5.340e-2 -1.983e-3 5.701e-
2 1.992e-3 -5.340e-2
FR 0 0 0 0 7.05 0
EN

Owen Duffy wrote:
I have constructed an NEC model of an 80/40 dipole using bootstrap coax
traps. I have generated a helix of GW elements for the RG58C/U coax coil,
and use NT cards to model the coax interior.

Plain conductors are loaded with copper conductivity.

I am grappling with approprate conductivity for the outer surface of the
coax coil, thinking that it must be greater than copper because of
proximity effects and braid effects. I have initially tried loading those
segments with conductivity of one tenth of copper... but I don't expect
that will make the RF resistance tenfold.

With those numbers, the loss in the interior of the coax is more than
half the total loss.

Has anyone experience or thoughts on an appropriate way to load the
conductor representing the outer surface of the coax coil forming the
trap?


Try changing the resistivity by a factor of 2 or 10 and see if the model
results change significantly. It might be a non-issue.

Jim Lux wrote in
:

....
Try changing the resistivity by a factor of 2 or 10 and see if the
model results change significantly. It might be a non-issue.


Thanks Jim, I did such an analysis and should have posted the results.

The majority of loss is in the interior of the transmission line as
posted. If I change the conductivity of the segments used by the coil,
total conductor loss halves, so it seems signficant.

I am thinking that with proximity effect likely to confine current
distribution to perhaps one fifth of the shield circumference, and
effects of braid on RF resistance, the effective RF resistance of the
outside of the shield might be five times or so the resistance of a
copper tube of same diameter.

I assume that NEC will not be modelling proximity effect, but I could be
wrong.

Do you have any suggestions?

Owen

Owen Duffy wrote in
:

proximity effects and braid effects. I have initially tried loading
those segments with conductivity of one tenth of copper... but I don't
expect that will make the RF resistance tenfold.


I wasn't very clear there... I loaded the segments with one ninth of
copper, and that was additional to the overall loading equal to copper, so
the effect should be resistance of ten times copper for the coil segments.

Owen