It does not have to do with the ports being arranged backwards, nor
MIMO-A vs MIMO-B.
We were all like WTF when Cambium started shipping the ePMP 2.4 AP
antennas as -/+45 while the SMs are V/H. We all said, how the hell does
that work when the 45 antenna will be seeing the V/H signals at -3dB?
Does not compute! Well, as I said, Cambium explained how it works this
way. Way above my head and pay grade. Dan or Sri posted a PDF or
something that shows it.
On 12/4/2015 3:15 PM, Joshaven Mailing Lists wrote:
The Atheros chipset differentiating between the Mimo chains does not
have to do with 45º slant vs 90º… it means that chain A and B on one
radio can be H&V or V&H while chain A and B on the other can be H&V or
V&H interchangeably with no loss because and the chip will just cross
them over as needed. This means that hooking up the antenna cables
“backwards” won’t effect anything. It doesn’t mean that the
orientation of the antenna is irrelevant.
The energy received by an antenna that is out of phase with another
antenna is much less then if it was in phase that is a principal of
radio that no chipset will ever overcome. Now… maybe you can make a
magic antenna that is “multi phased” such that it can tune in a 45º
phase offset signal well. I suspect that the Cambium equipment when
properly matched will both have the same polorization.
I believe that if your having the same outcome on slant or not slant
it would be due to one chain being refracted. For example if your
horizontal chain was fine but the virtual chain was refracted off
something such that the wave was on or near a 45º slant then you would
have the same basic performance regardless of a standard or slant
orientation. This however is a path issue not a design characteristic
of the chipset or antenna.
FYI, linearly polarization is not a reference to the polarization
being on a vertical, horizontal or slant axis but a reference to the
way the wave propagates, circular polarization is an alternative to
linear polarization not “slant".
Sincerely,
Joshaven Potter
MTCNA, MTCRE, MTCWE, MTCTCE, UACA
Google Hangouts: yourt...@gmail.com <mailto:yourt...@gmail.com>
Cell & SMS: 1-517-607-9370
supp...@joshaven.com <mailto:supp...@joshaven.com>
On Dec 4, 2015, at 2:13 PM, Adam Moffett <dmmoff...@gmail.com
<mailto:dmmoff...@gmail.com>> wrote:
The DSP in the ePMP can do some kind of processing to correct for the
45 degree offset when you have slant on one end and V+H on the
other. I might not be stating it with the correct technical jargon,
but that's the gist of it. They sell a dual slant sector for the AP
with the intent to use it with V+H integrated SM.
This is a built in feature of the Atheros chipset, so presumably UBNT
radios can do the same thing.
That all said...I would try to match them up as just a matter of
principle.
On 12/4/2015 2:07 PM, Joshaven Mailing Lists wrote:
Your AP & SM should always have the same antenna orientation. I
promise you that you don’t want slant on the AP and not on the CPE.
If the signal is so obstructed that the orientation is screwed up
to the point that unmatched polarization is actually a benefit then
there is something seriously wrong…
Sincerely,
Joshaven Potter
MTCNA, MTCRE, MTCWE, MTCTCE, UACA
Google Hangouts: yourt...@gmail.com <mailto:yourt...@gmail.com>
Cell & SMS: 1-517-607-9370
supp...@joshaven.com <mailto:supp...@joshaven.com>
On Dec 4, 2015, at 1:59 PM, Paul McCall <pa...@pdmnet.net> wrote:
We are looking at smaller sector sizes for a 5 Ghz ePMP cluster (60
degree probably), and am considering my options, which might also
increase my gain quite a bit. Using a non-Dual Slant sector such
as AM-5AC21-60, would increase my options. There have been a
calling threads on Cambium’s sites about whether Dual Slant was a
big factor at the AP if the SMs aren’t dual-slant.
Cambium’s Daniel Sullivan made this comment … The thread was
originally about 2.4 Ghz options, so not sure if it applies exactly
to 5 Ghz.
Paul
