Radiating Efficiency
 

Frank,

To complete the graphs of Rin + jXin versus length for one radial, I
need data for lengths between 0.5 metres and 2 metres at intervals of
0.2 metres.

Could you oblige please?

The nice smooth graphs I have produced so far are free of
'measurement' errors which could be attributed to your hard, tedious
work.

The uncertainty due to NEC4 is as yet unknown. In the NEC4 model of a
single radial, what is the length of the short bits of radial used to
model it? Alternatively, into what number of short lengths is the
radial under test divided?
----
Reg, G4FGQ.

Radiating Efficiency
 

"Reg Edwards" wrote in message
...
Frank,

To complete the graphs of Rin + jXin versus length for one radial, I
need data for lengths between 0.5 metres and 2 metres at intervals of
0.2 metres.

Could you oblige please?

The nice smooth graphs I have produced so far are free of
'measurement' errors which could be attributed to your hard, tedious
work.

The uncertainty due to NEC4 is as yet unknown. In the NEC4 model of a
single radial, what is the length of the short bits of radial used to
model it? Alternatively, into what number of short lengths is the
radial under test divided?

No problem Reg, here are the data you requested:

0.5 m -- Radial Z = 167.6 - j 161.0
0.6 m -- Radial Z = 146.6 - j 136.3
0.8 m -- Radial Z = 115.9 - j 102.3
1.0 m -- Radial Z = 101.4 - j 79.1
1.2 m -- Radial Z = 89.4 - j 61.5
1.4 m -- Radial Z = 80.9 - j 57.2
1.6 m -- Radial Z = 74.6 - j 35.0
1.8 m -- Radial Z = 70.2 - j 24.1
2.0 m -- Radial Z = 67.2 - j 14.2

Concerning your previous question about the Smith Chart.
It is a powerful visual aid which gives a better understanding
of what is really happening to the impedances. Exactly what
I got out of graphing the data was really only a confirmation
of the expected spiral toward Zo. Also the "zero-crossing"
which occurs at quarter-wave multiples.
Normalization of the data (Division of each complex datum
by the complex Zo), and linear graphing, will also provide
the same information; including the gradual approach to the
normalized Zo of 1 ohm.

Where the Smith Chart really proves its worth is in the
design of matching networks involving: L, C, R, and
transmission lines. It is hard to imagine how anybody
could do such design without this aid. I am not sure
if anybody actually uses the paper versions these days,
but the software equivalent, combined with programs
such as (Now Agilent's) Eagleware, are pretty much
design lab standards.

In this model all segments are a constant 10 cm in length.
The number of segments will therefore be L(m)/0.1.

Frank

Radiating Efficiency
 

On Wed, 02 Aug 2006 14:03:48 GMT, "Frank's" wrote:


"Reg Edwards" wrote in message
...
Frank,
snip
Concerning your previous question about the Smith Chart.
It is a powerful visual aid which gives a better understanding
of what is really happening to the impedances. Exactly what
I got out of graphing the data was really only a confirmation
of the expected spiral toward Zo. Also the "zero-crossing"
which occurs at quarter-wave multiples.
Normalization of the data (Division of each complex datum
by the complex Zo), and linear graphing, will also provide
the same information; including the gradual approach to the
normalized Zo of 1 ohm.

Where the Smith Chart really proves its worth is in the
design of matching networks involving: L, C, R, and
transmission lines. It is hard to imagine how anybody
could do such design without this aid. I am not sure
if anybody actually uses the paper versions these days,
but the software equivalent, combined with programs
such as (Now Agilent's) Eagleware, are pretty much
design lab standards.

In this model all segments are a constant 10 cm in length.
The number of segments will therefore be L(m)/0.1.

Frank

May I insert a few words about the Smith Chart?

During the period between 1958-59 I designed the entire antenna system and
matching harness for the World's first weather satellite, TIROS 1, which was
launched April 1, 1960.

Four transmitters operating simultaneously on separate frequencies fed the
antenna that comprised four crossed monopoles radiating right circular
polarization from two transmitters and left circular polarization on the other
two frequencies.

The point concering the Smith Chart is that the only tools I had for the
development of both the antenna system and the matching harness was the
Hewlett-Packard HP-805 slotted line for impedance measurements, the slide rule
for calculations, and the paper Smith Chart to tell me where I was in the
desert. The resolution available with the Chart was sufficient to make the
system work successfully. Philip Smith was my hero.

Incidentally, the matching harness was fabricated entirely of printed-circuit
stripline, and that was 47 years ago.

Walt, W2DU

Walt, W2DU, and Tiros
 

On Wed, 02 Aug 2006 11:18:07 -0400, Walter Maxwell
wrote:


During the period between 1958-59 I designed the entire antenna system and
matching harness for the World's first weather satellite, TIROS 1, which was
launched April 1, 1960.


Walt,
Could you please tell me a little more about that satellite? Reason I
ask is there is a place I am sure you can enlighten us on, called Camp
Evans.
I know there is still a big dish antenna there, although the wooden
structure surrounding it is failing. Supposedly it was used for
communicating with Tiros. Is that right?

A NJ radio club had a hamfest there in April, and I was able to walk
under the antenna and they have some displays there. Supposedly Camp
Evans was also the place where the first signals were bounced off the
moon, somewhere around 1946.
Camp Evans looks like at least parts of it are being renovated.
Brookdale Comm. College has some classrooms there, very modern.
Oh yes, and on the road leading to Camp Evans (Marconi Road !!) there
is a display of the top portion of one of Marconi's towers, and a
plaque noting that there were many such towers in the area in the
early 1900s.
Quite a bit of radio history, in these parts of New Jersey, and I am
glad to know one of the pioneers frequents this newsgroup.
Not far away, in Holmdel, is where AT&T communicated with Telstar, I
believe, about the same time period. The Holmdel facility, home to
6000 Bell Labs employees as recently as1999 has been sold and is going
to be demolished.

Rick K2XT

Radiating Efficiency
 

May I insert a few words about the Smith Chart?

During the period between 1958-59 I designed the entire antenna system and
matching harness for the World's first weather satellite, TIROS 1, which
was
launched April 1, 1960.

Four transmitters operating simultaneously on separate frequencies fed the
antenna that comprised four crossed monopoles radiating right circular
polarization from two transmitters and left circular polarization on the
other
two frequencies.

The point concering the Smith Chart is that the only tools I had for the
development of both the antenna system and the matching harness was the
Hewlett-Packard HP-805 slotted line for impedance measurements, the slide
rule
for calculations, and the paper Smith Chart to tell me where I was in the
desert. The resolution available with the Chart was sufficient to make the
system work successfully. Philip Smith was my hero.

Incidentally, the matching harness was fabricated entirely of
printed-circuit
stripline, and that was 47 years ago.

Very interesting Walt. Judging by the type of work you were involved in
you must be familiar with the classic text by: Matthaei, Young, and Jones.
I am not familiar with the HP-805, but probably because most of my
earlier lab work was concerned only with HF. I realize that slide rule,
pencil, and Smith Chart was the only way in those days. When I started
work any intense analysis was done in FORTRAN on main frame
computers. One of the hot topics I remember was mini-computers,
but never got to see one. I don't think they lasted long, since soon
after programmable desktops started to become available.

One thing I remember was a guy designing HF antenna tuners, and he
used the Smith Chart. What is interesting is that the charts were
very large, about 2 ft X 3 ft, and came in pads. Much greater
accuracy than the standard chart. I think they may have been published
by a company called K-electronics. I used to needle guys in the lab
by telling them that Smith was a ham.

I now have a nice high res pdf of the Smith Chart, but
usually use an electronic version from
the "Berne Institute of Engineering".

73,

Frank

Radiating Efficiency
 

On Wed, 02 Aug 2006 16:50:39 GMT, "Frank's"
wrote:

"Berne Institute of Engineering"

Hi Frank,

Reverse engineering this clue, I've found the author is Prof. F.
Dellsperger, and that his Smith Chart software can be found at:
http://www.fritz.dellsperger.net/downloads.htm

73's
Richard Clark, KB7QHC

Radiating Efficiency
 

"Richard Clark" wrote in message
...
On Wed, 02 Aug 2006 16:50:39 GMT, "Frank's"
wrote:

"Berne Institute of Engineering"

Hi Frank,

Reverse engineering this clue, I've found the author is Prof. F.
Dellsperger, and that his Smith Chart software can be found at:
http://www.fritz.dellsperger.net/downloads.htm

73's
Richard Clark, KB7QHC

Thanks Richard, I had recently tried to locate a source for this
program, but could no longer find it. I see he has updated the program
so will try the upgrade. I see he also has a lot of other stuff that looks
interesting.

73,

Frank, VE6CB

Radiating Efficiency
 

No problem Reg, here are the data you requested:

0.5 m -- Radial Z = 167.6 - j 161.0
0.6 m -- Radial Z = 146.6 - j 136.3
0.8 m -- Radial Z = 115.9 - j 102.3
1.0 m -- Radial Z = 101.4 - j 79.1
1.2 m -- Radial Z = 89.4 - j 61.5
1.4 m -- Radial Z = 80.9 - j 57.2
1.6 m -- Radial Z = 74.6 - j 35.0
1.8 m -- Radial Z = 70.2 - j 24.1
2.0 m -- Radial Z = 67.2 - j 14.2

=====================================
Frank,

Thank you for the above.

Can you do 36 radials from 0.6 to 8.0 metres in increments of 0.2
metres?

Must be the same input data as before :- F = 8.07 MHz, Ground =
150,16 Radial diameter = 1.64mm, Depth = 25mm.
----
Reg.

Radiating Efficiency
 

Reg,

I have taken Frank's numbers from NEC-4 (that he sent me) and compared the
results with Radials_3. Here is what I found:

NEC-4 Parameters (in parens are the values I get with Radials_3, note the
differences are LARGE)

Height 20 m;
Radials 15 m;
24 radials at 15 degree intervals;
Radials 25 mm below ground;
All wires # 14 AWG copper; (I USED 2mm wire)
Gnd. Er = 25, Resistivity 25 ohm-m;
Resonant Frequency 3.64 MHz, and; (Radial_3 shows Resonant Freq of 3.750)
Zin = 41.32 + j 0.156466 ohms at 3.64 MHz. (34.17 ohms and j -24.6)

If I change the Freq in Radial_3 to 3.750, I get Zin = 37.41 and j -0.0.

That is 110 KHz difference for the same antenna and ground characteristics.
Not very good agreement.
=================================================
Now, on to the real question I have been asking for a week: Does the current
along the radials fall off as rapidly as Radial_3 predicts? I used the
numbers from the above antenna, (supplied by Frank from NEC-4) and compared
them to what Radial_3 says for the same length. I do not have numbers
showing a 5 meter radial for NEC-4 and a 5 meter radial for Radial_3. What I
am presenting is the current along the full radial in NEC-4, vs. the current
in a radial of specified length in Radial_3. Length is in meters.

Since we are using ratios, I only used the integer and two decimal places
for the current in my calculations. All values for current are actually what
is presented * 10^-2

Length Current NEC-4 (dB) Radial_3 (dB)

0 9.15 N/A N/A
1 8.37 0.77 6.2
2 6.98 2.35 12.4
3 5.63 4.22 18.6
4 4.55 6.07 24.7
5 3.57 8.17 30.9
6 3.00 9.69 37.1
7 2.54 11.13 43.3
8 2.16 12.54 49.5
9 1.84 13.93 55.7
10 1.60 15.15 61.8
11 1.46 15.94 68.0
12 1.41 16.24 74.2
13 1.36 16.56 80.4
14 1.04 18.89 86.6
14.75 0.26 30.93 91.2

Notice at a lenght of 4 meters, Radial_3 is showing 24.7 dB of attenuation
of the current, while NEC-4 is only showing 6 dB...not even close.

Notice throughout the lengths from 3 to 11 meters, the difference between
Radial_3 nd NEC-4 is a pretty consistent 4X, i.e, Radial_3 is showing four
times the attenuation at a given length that NEC-4 is showing. Patterns like
this in data are rarely the result of chance. Maybe this will lead to a
resolution of the problem.

If Reg's threshold of 25 dB attenuation is reasonable (at which point adding
additional length is of very little value), then NEC-4 says for this
particular set of variables that we don't attain anything near 25 dB current
attenuation along the length until we get to greater than 14 meters (and the
wire is only 15 meters long!)

How could the current in a 4 meter radial drop by 25 dB within its length?

So, I'm left with two questions:

1. Extra wire....what extra wire? It looks like one needs the entire 15
meter radial wire, not some 4 meter stubby....at least according to NEC-4.

2. Why such a discrepancy in resonant freq between NEC-4 and Radial_3?

Unless someone else can point out an error in the analysis, it appears that
the transmission line model that Reg is using to predict rate of current
drop along the radial vs. length is either the wrong model, or the wrong
application thereof. NEC-4 just doesn't seem to support the rapid current
fall-off that Radial_3 predicts.

BL&E says it ain't so.
Tom, W8JI, says his measurements say it ain't so.
NEC-4 appears to say "it ain't so".

If I were a betting person, I'd say it ain't so.

Back to the drawing board?

In the mean time, anyone putting in a radial field woud appear to be better
served by the references in this thread to "optimizing radial systems", I
think by K3LC...it's in several antenna books.

73,

p.s. How all this fits into efficiency someone else can analyze, but it
can't be good.

....hasan, N0AN

Radiating Efficiency
 

On Wed, 02 Aug 2006 16:50:39 GMT, "Frank's" wrote:

May I insert a few words about the Smith Chart?

During the period between 1958-59 I designed the entire antenna system and
matching harness for the World's first weather satellite, TIROS 1, which
was
launched April 1, 1960.

Four transmitters operating simultaneously on separate frequencies fed the
antenna that comprised four crossed monopoles radiating right circular
polarization from two transmitters and left circular polarization on the
other
two frequencies.

The point concering the Smith Chart is that the only tools I had for the
development of both the antenna system and the matching harness was the
Hewlett-Packard HP-805 slotted line for impedance measurements, the slide
rule
for calculations, and the paper Smith Chart to tell me where I was in the
desert. The resolution available with the Chart was sufficient to make the
system work successfully. Philip Smith was my hero.

Incidentally, the matching harness was fabricated entirely of
printed-circuit
stripline, and that was 47 years ago.

Very interesting Walt. Judging by the type of work you were involved in
you must be familiar with the classic text by: Matthaei, Young, and Jones.
I am not familiar with the HP-805, but probably because most of my
earlier lab work was concerned only with HF. I realize that slide rule,
pencil, and Smith Chart was the only way in those days. When I started
work any intense analysis was done in FORTRAN on main frame
computers. One of the hot topics I remember was mini-computers,
but never got to see one. I don't think they lasted long, since soon
after programmable desktops started to become available.

One thing I remember was a guy designing HF antenna tuners, and he
used the Smith Chart. What is interesting is that the charts were
very large, about 2 ft X 3 ft, and came in pads. Much greater
accuracy than the standard chart. I think they may have been published
by a company called K-electronics. I used to needle guys in the lab
by telling them that Smith was a ham.

I now have a nice high res pdf of the Smith Chart, but
usually use an electronic version from
the "Berne Institute of Engineering".

73,

Frank

Thanks for the nice response, Frank,

Walt