You are here

Ubiquiti antennas on 5.9ghz? Do they work?

13 posts / 0 new
Last post
k2viz
Ubiquiti antennas on 5.9ghz? Do they work?

The majority of my nodes are channels 133-150 (5Mhz wide), we have a new site available which already has a 2.4 and 5.8 WISP there, and they are very protective of their 5.8ghz spectrum(long story, they have priority with the landlord).

How well do the Airmax 5.8 sectors work in 5.9ghz or are there any 5.9 dish antennas out there which don't cost a fortune?

My experience with the Powerbeam 620 showed the performance at the top-edge of 5.8 to be poor and 5.9 to be near non-existent.

AE6XE
AE6XE's picture
There are a lot (probably 50+

There are a lot (probably 50+) of 120deg UBNT sectors in SoCal at the top edge of 5.8 and in 5.9.   I can't recall any issues of poor performance due to antenna not matching, hasn't been noticed or discussed.  I'm operating several of these sectors and 1 pbe-m5-620 (on ch 172) on a major tower site in SoCal saturated with WISP operators.   part 15 channels are generally not available or don't work at all given all the WISP operators. 

Joe AE6XE

k2viz
Good information - thanks

Thanks for the feedback -- upon some further experimenting, it appears the alignment on the Powerbeam 620's has a slight "shift" moving from Channel 133 to 180.   I tried this on a 25 mile link and (moving from 133 to 180) and the link failed.   At a closer site, it appears the 620's might have a slight shift in alignment at extreme ends of the spectrum.

I was researching this because one club was operating under the assumption that anything below channel 178 was "part 15", we disagreed so I began experimenting.

AE6XE
AE6XE's picture
In my local area we have

In my local area we have active ATV users on ch 180 blasting out a lot of power.   Any possibility this is also occurring in your area to explain why 180 isn't working?

Does this mean that when you shifted the dish alignment, it did work on ch 180, when before it did not?  ...or it has never worked with any alignment on ch 180. 

While there is future concern of additional overlap,  I understand part 15 and part 97 share below 5850 or ch 170.   

Joe AE6XE

K6AH
K6AH's picture
Side Lobes

You were likely on a side lobe initially.  These can shift around with frequency, but the primary lobe shouldn't.

Andre, K6AH
 

AC2OG
AC2OG's picture
2X2 Mimo sector antenna 19dBi 120°

Hello Joe,
How does it work a PtMP link with a 2X2 MIMO sector antenna and the neighbor nodes with SISO vertical polarized antennas?
Now I have a PtMP node made with a Bullet M5HP + a 60 ° 17dBi  sector and all the neighbors also with Bullet M5HP with a 27 dBi directive antenna and would like to replace the Bullet M5HP + the sector antenna with a Rocket M5 + a 2X2 MIMO sector 19dBi 120°. Is it worth to do? 
Thanks and 73 de Leo.
 

AE6XE
AE6XE's picture
Note, these radios are very

Note, these radios are very smart and continually test all the possible combinations of modulation, error correction coding, antenna polarities available, etc.   A multiple-point transmitter will test the options to each neighbor and based on the real time conditions (~10% of packets will test and measure a different option) will select the optimal combination that maximizes the data thoughput.  Every neighbor may have a different combination in use when transmitting to and change with conditions in real time.  Every other packet transmitted from a multi point transmitter could be a different combination.  See the 802.11n MCS rate table to see the combinations. 

Joe AE6XE

KE2N
KE2N's picture
sector

as Joe points out - a 19 dBi MIMO is really two 16 dBi antennas,  And your existing antenna is 1 dB better than that.
But if you need the 120 degree field of view, it may be worthwhile.
On the other hand, if all your "customer" stations are within a 60 degree field then you are going backwards - the only advantage being that the Rocket has more memory than a Bullet.

AE6XE
AE6XE's picture
Scratching my head to make

Scratching my head to make sure I'm thinking about this right.  If we modeled this in radio mobile to compare a SISO and MIMO client receiving a signal from a MIMO Sector Multi-Point:

1) -3dB because only able to receive ~half of the MIMO's transmitted power -- the 90 deg off polarity energy is mostly lost
2) -3dB because the single aligned polarity antenna gain on the sector is only 16dBi

This would be the modeling for 802.11n MCS0 to MCS7 modes where the same data is transmitted out both polarities (which are the only modes or link rates the SISO can do). 

This means the SISO is -6dB disadvantaged as compared to a MIMO when receiving the signal from the tower site.

Joe AE6XE 

AE6XE
AE6XE's picture
When the Sector MIMO device

When the Sector MIMO device transmits and splits the power on 2 polarities, the SISO single polarity device will not be able to receive ~half of the power.  It will be -3dB disadvantage.  The SISO device can only do half the max link rate, no capability to double up the data and send or receive over 2 polarities.  See this write up in more details:

https://arednmesh.readthedocs.io/en/latest/arednHow-toGuides/siso-mimo.html

Joe AE6XE

AC2OG
AC2OG's picture
Sector MIMO PtMP & SISO PtP

Understood. I thought the + 2 dB Pout of the Rocket (vs the Bullet) plus the +1 dB Gain of the sector compensated the -3dB of splitting power on the MIMO.
So covering a solid angle of 120° against the actual 60° at no cost could be worthy. The problem is that the SISO devices are going out of production and replacing all the SISO with MIMO is very expensive.
Thanks for your support

kj6dzb
kj6dzb's picture
Here in the bay area ive been

Here in the bay area ive been testing the return loss of the rocket antennas and found that they dont all preform in the ham band.

https://docs.google.com/spreadsheets/d/1kO3xYspySozyCfcIIhlbO8ASPL2D-ucJ...

KE2N
KE2N's picture
thanks

Thanks for that!   

I tested some 2.4 antennas and found that SWR tended to shoot up at the band edges more quickly for the higher gain antennas than the lower gain version of the same antenna. Not surprising suppose.

I would suggest that as long as you get more than 10 dB return loss, it will work just fine. 

Could you explain the S21 measurement? is that a measure of isolation between H and V (ch0 vs ch1)?

It would be interesting to see how gain varies with frequency.  That is a bit harder to do propoerly ...

Ken

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer