if it were me, I'd use the Omni. 1 mile just isn't far enough to justify the expense of sectors which could cost 3 times as much. If the requirement was 5 or 10 miles I would agree completely with my esteemed colleague.
With four 90 degree sectors you have kind of a four leaf clover radiating pattern. With three 120 degree sectors you have more of a three leaf clover radiating pattern. You need to take into account where your nulls will fall and try to have them where your users are not located.
Also with the three or four sector antennas it is much more important to have RF shielding (sometimes called RF armor) on each antenna to keep the RF from one antenna from getting into and de-sensing the other two or three antennas at the same site.
So, you might ask, "With the additional cost of three (or four) antennas and radios (and cat5e STP, and POEs...) why go with sectors rather than one omni?"
The answer is to isolate local traffic. If most of the users on the North side of your site talk primarily with other users on the North side of your site, and similarly for users on the West, South, and East sides of your site, then the traffic will go in one antenna and come out the same antenna with no impact to users on the other two or three served areas.
And, for the traffic that comes in on one antenna and goes out on another antenna, you can use frequency diversity so that traffic can be received and transmitted simultaneously.
One more thing to consider... When mounting sector antennas, they are mounted facing away from the tower. When mounting an omni antenna, your radiating pattern is going to be affected by the tower. Depending on the spacing from the tower leg the radiating pattern can be changed dramatically from the perfect circle we think of an omni antenna having.
One final point is down tilt. This document shows the beam width and down tilt of the Ubiquiti omni antennas. https://dl.ubnt.com/datasheets/airmaxomni/amo_ds_web.pdf
If your node is significantly higher than than your users you may be shooting over the heads of your users when using an omni. With sector antennas it is much easier to adjust the amount of downtilt/uptilt to cover the desired target area.
Earlier in this tread it was mentioned that sector antennas need shielding to avoid de-sensing. What if you have three point-to-point antennas such as the PBE-M5-300? And if they are all on 5.9GHz and configured for 10 MHz bandwidth, how many channels do you need as "guard bands"?
We have just used our first mesh system operationally, with a single 5 GHz backbone link between mountains, and a 2.4 GHz link down to the valley. It worked great, but we didn't need to compete with anyone else for spectrum. Now we'd like to add a couple permanent sites, but those will be in areas where commercial wireless internet companies have nodes near by. So... time for my dumb question... are they operating between 5.665 and @5.875 GHz, or are they going to be somewhere else? Are those frequencies just for public WiFi? If the latter, I won't worry much, as these sites are miles away from the general public. I just don't want to interfere with... or receive interference from... the wireless internet providers.
If you stay in the Part97 only portion of the bands, you should be clear of any WISPs. If they are in part97, you can call them and ask them to move, or, report to the FCC. If you use directional antennas like the PowerBeam, that will help as well since the beam is narrow.
But that relates to my earlier question... how much space do I need to maintain between co-located channels? If with 10 MHz bw, I can operate every other channel, staying in Part 97 should be easy for our purposes. If I need to have three channels in between, I run out of space fairly quickly. Channel 178 might not be available, for instance, if a WISP is operating on ch 177... or maybe even ch 176 depending on how much guard band I need. In one case, I may be on the same tower within a few feet of their nodes.
if it were me, I'd use the Omni. 1 mile just isn't far enough to justify the expense of sectors which could cost 3 times as much. If the requirement was 5 or 10 miles I would agree completely with my esteemed colleague.
Andre
How far out do you think an omni would be good for?
I have omni's that are linking 25 miles. one is at 450', one at 200', one at 200'
tx
Ron K7OPA
channel -2
5Mhz
TxMbs ranges between 3Mbs and 5Mbs
If you open the bandwidth to 10 MHz what does that do to the mileage?
73,
N0KMO
Q: When does 120 + 120 + 120 not = 360 (or when does (90 + 90 + 90 + 90 not = 360)?
A: When you are looking at coverage area and localizing traffic.
Look at the radiating pattern for the sectors here: https://dl.ubnt.com/datasheets/airmaxsector/airMAX_Sector_Antennas_DS.pdf
With four 90 degree sectors you have kind of a four leaf clover radiating pattern. With three 120 degree sectors you have more of a three leaf clover radiating pattern. You need to take into account where your nulls will fall and try to have them where your users are not located.
Also with the three or four sector antennas it is much more important to have RF shielding (sometimes called RF armor) on each antenna to keep the RF from one antenna from getting into and de-sensing the other two or three antennas at the same site.
So, you might ask, "With the additional cost of three (or four) antennas and radios (and cat5e STP, and POEs...) why go with sectors rather than one omni?"
The answer is to isolate local traffic. If most of the users on the North side of your site talk primarily with other users on the North side of your site, and similarly for users on the West, South, and East sides of your site, then the traffic will go in one antenna and come out the same antenna with no impact to users on the other two or three served areas.
And, for the traffic that comes in on one antenna and goes out on another antenna, you can use frequency diversity so that traffic can be received and transmitted simultaneously.
One more thing to consider... When mounting sector antennas, they are mounted facing away from the tower. When mounting an omni antenna, your radiating pattern is going to be affected by the tower. Depending on the spacing from the tower leg the radiating pattern can be changed dramatically from the perfect circle we think of an omni antenna having.
One final point is down tilt. This document shows the beam width and down tilt of the Ubiquiti omni antennas.
https://dl.ubnt.com/datasheets/airmaxomni/amo_ds_web.pdf
If your node is significantly higher than than your users you may be shooting over the heads of your users when using an omni. With sector antennas it is much easier to adjust the amount of downtilt/uptilt to cover the desired target area.
We have just used our first mesh system operationally, with a single 5 GHz backbone link between mountains, and a 2.4 GHz link down to the valley. It worked great, but we didn't need to compete with anyone else for spectrum. Now we'd like to add a couple permanent sites, but those will be in areas where commercial wireless internet companies have nodes near by. So... time for my dumb question... are they operating between 5.665 and @5.875 GHz, or are they going to be somewhere else? Are those frequencies just for public WiFi? If the latter, I won't worry much, as these sites are miles away from the general public. I just don't want to interfere with... or receive interference from... the wireless internet providers.
Thanks
Dave
Andre