David Gilbert wrote:
>In response to N3OX's posting on cage dipoles, I searched around to find
>more information on them and happened to find the website of a
>manufacturer who also claimed better performance than a standard
>dipole. I have no idea whether or not this is the same one who had
>corresponded directly with N3OX, but I decided to write this
>manufacturer directly to see how they derived their claimed results in
>order to better understand the design in general, and in particular why
>modeling programs like EZNEC don't properly deal with a cage antenna.
>
>I received a very quick response from one of the owners of the company,
>and since this person states he is aware of the thread here on
>TowerTalk, as well as my own personal skepticism, I am taking the
>liberty of forwarding his comments verbatim (cut and paste ... no
>editing whatsoever on my part) so that his side of the discussion may
>get fair representation.
>
>I won't divulge the name of this manufacturer so don't bother to ask. I
>also think it best if I personally make no further comments in this
>forum, either supportive or critical, on the topics of field testing or
>time diversity in modeling. I will say, though, that I hear as well as
>my neighbors.
>
>73,
>Dave AB7E
>
>
><quote>
>
>Sure the antenna was tested on an antenna range vs a standard resonant dipole.
>An independent engineering company srt up the trst doing firld intensity
>measurments in all planes at all practical distance using hewlett packard
>spectrum analyzers that are capablr of .1 db resoution.
>The antenna was tested at distances of 1,5,and 10 miles. The antenna was
>rotated to check all lobes. Reference data was compared at 15 degree
>intervals. Antenna height was 1/4 wave and tested again at 1/2 wave. results
>were surprisingly good. In all cases the cage exceeded the standard by 5 db or
>better...We were surprised ourselves. However the company used assured us of
>the correctness of the measurements. They commonly do proof of performance
>testing for broadcast and cable. Their instrumentation is current calibration
>and correctly used by experienced engineers.
>Standard modeling programs are based on assumption of single piece radiators
>and have no provision for time diversity in them. The cage uses this
>phenomina. When one uses most modeling programs, the results err due to this
>fact. The field data reflects the differences.
>I have read your tower talk posts and hope you are amused. The fact is still
>that the damn thing actually works as advertised..LOL!
>The cost of this testing was not cheap. We chose to do it because we are
>sincere in our desire to provide a first rate product. The guarantee we give
>on this product is 100% sarifaction or your money is refunded less our
>shipping..So you risk less than 25 bucks to try one...To this date, 486 units
>in the field with no returns and no complaints ..NOT EVEN ONE....
> That to me is the greatest endorsement..Satified clients.
>Most of the detractors we have spoken with have never used a cage of any
>design and certainly not ours. We also have found they are too cheap to take
>our challenge and try one even if we agree to pay shipping back. They
>basically have their opinion..And that is "if I can't afford it it must not be
>any good". With our warranty there is no risk to try it! Now would we do that
>if we had no confidence in our products. I put my money where my mouth is
>daily. HOW ABOUT YOU! Or are you a person who buys a $3000.00 radio to use as
>$20 dipole for an antenna and then bitch because the rig isn't hearing as good
>as your neighbors.
>
></quote>
>
>
Thanks for posting this, David. Having read the email you received from
the cage antenna manufacturer, I am still extremely skeptical that this
antenna if operated at a frequency where its length is ~ 1/2 wavelength
would outperform a single wire 1/2 wavelength reference dipole placed in
the same location by 5dB. This goes against every bit of antenna theory
that I have ever studied. Also, I can see no reason whatsoever that NEC
would not properly model this antenna. The time delay between wires in
the cage would at most be 1 or 2 nanoseconds. Twelve inch wire spacing
for instance would only produce 2.5 degrees of electrical phase
difference at 7 MHz. This is too small to be of any consequence. And
even if it wasn't inconsequential, NEC would still model it properly.
By the way, I am not saying that the test results referenced in the
email are fabricated. It is quite possible that the independent testing
lab did measure 5dB gain over a reference dipole. If that is the case,
then the methodology used by the test lab needs to be carefully
reviewed. Whenever one acheives an experimental result that is not
consistent with well established theory, one should be skeptical of the
test methodology. If this manufacturer really has figured out a way, to
acheive 5dBd gain over a single-wire reference dipole, then they should
patent the method immediately as this would truly be a breakthrough
(that is what I would do). Once the methodology is patented, I would
then submit the design for independent peer review by experts in
industry and academia. A Nobel prize might even be in the mix if the
antenna performed as advertised (I am totally serious about this
statement since that kind of performance would be overturning bedrock
antenna theory that has been in place for at least 70 to 80 years).
To reiterate, I am skeptical of the claim that this antenna produces a
5dB improvement over a reference half-wavelength dipole. If I am
correct, it doesn't mean that this is a bad product. The cage likely
produces an noticeably better VSWR bandwidth when compared with a single
wire dipole. If it is well constructed and the improved VSWR bandwidth
is desired by a prospective customer, then the antenna may be worth
every penny of what the manufacturer is asking even if it turns out that
it doesn't exhibit 5dB gain over a single wire reference dipole. If I
had a requirement for a cage dipole and I didn't have time to build it
myself, I would certainly consider buying one from this manufacturer.
73, Mike W4EF................
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|