All wifi specialist journalist and blog writer tell that ceiling mounting is way more optimal than wall mounting. It is clear that they are right when we check the antenna pattern plots of most access points.
But i just checked the AP-300 Aruba's AP, and it seems that they emit more power in front of them than on their sides :
Antenna pattern plots : https://i.imgur.com/yI2R6I2.png (the Azimuth is pure omni)
According to this plot, it is not an awful thing to mount them on a wall. Oddly, the 300's datasheet tell us that they are optimized to be mounted on the ceiling https://www.arubanetworks.com/assets/ds/DS_AP300Series.pdf .
What do you guys think about this ?
A careful read of the specifications indicates that there are two models of the AP. The model 305 antenna complement has built-in "down-tilt". This should make this version more effective on ceiling mounts, and conversely less so in a wall mounted position.
The down-tilt is specified as approximately 30 degrees - a non-inconsequential amount. Especially useful with taller ceilings.
The pattern diagrams you link to are not directly linked to other documents, so it is difficult to get much information from them. However they (perhaps) imply that there are three antenna elements possible (A1, A2, A3). Obviously A3 was used for 5 GHz in the one view.. It might prove useful to read thru the GUI or CLI descriptions. Without further explanation, it is hard to know exactly which AP models the diagram(s) refer to.
Obviously, either model should work (somewhat) in either orientation. However, I would stick with the recommendations from Aruba, based on your planned mounting position.
I have worked with both ceiling and wall mounted AP's in a large test lab environment. For ease of access in that environment , and uniform coverage, I prefer centrally mounted ceiling models. At home, where I do not have drop-ceilings, I prefer wall or tall bookcase mounted ones.
Thanks for your reply.
The antenna pattern plot is directly extracted from the datasheet linked in my post. Check the page 6 of the datasheet https://www.arubanetworks.com/assets/ds/DS_AP300Series.pdf , it is the antenna pattern of the 300 series. I guess it is the 305's one, since the 304 is the model with external antenna connectors.
I'm not sure to understand what they mean by downtilt -30°. If it is an electrical downtilt, the antenna main lobe should point in complelty another direction. Right now, when we check the antenna pattern, we can see that the AP emits way more directly in front of its. It is like a low gain patch antenna ! It shouldn't have this pattern if it was optimized for omni-ceiling
A lot more data on these pages.
They have "already" turned the AP 90 degrees for the ceiling mount version's diagram, so that the "hole" in the pattern at the top of the page, is towards the ceiling.
Obviously, the ceiling mount version would have less bleed-through towards the floor above, and a greater amount of power directed downwards. Depending on the installation, this might be exactly what you want.
For example, If there were no floor above, or a different companies space above, it could be perfect.
Rarely have I found two, or more, companies to agree on channel selection or AP management methods. Anything that promotes diverse Wi-Fi (RF) overlap, the better. My company may have long term occupancy planned for the building but you never know when, or if, some small company will come in and ruin all your channel planning.
I had that happen several times in the last building I was in, and it always caused a minimum of several weeks of unpredictable performance each time it happened - especially in my product evaluation lab.
This strong emission towards the floor could also leads to STA connected from the floor below, which is not what we want. Even if the AP below will not be interfered by the AP above (due to antenna rx gain reciprocity to tx), clients could suffer interferences.
I made a simulation with Ekahau with this exact model (AP-305), and what i found was that the AP behaves like a patch antenna when mounted on wall. Which is really strange since traditionnal omni looks like a crushed donut when mounted on wall.
Ceiling : https://i.imgur.com/lV4xvt7.png (not a very good omni)
Wall : https://i.imgur.com/oS6uhct.png (looks like a patch)
Traditionnal omni ceiling : https://i.imgur.com/SEcxUwh.png
Traditionnal omni mounted on wall : https://i.imgur.com/ppDbpBE.png
The datasheet seems wrong when they say that it is optimised for ceiling (for the 305), according to the antenna pattern and ekahau simulation the AP will be highly visible from the floor above and not so much on the side.
From the looks of these plots, I'd guess that Ekahau has the antennas modeled incorrectly. That kind of mistake does happen.
Also these are NOT high powered AP's. They are more like a client than an AP, in terms of power. If the power from the floor above is too high, then the power should be turned down. I doubt that you will have any problems caused by too much power directed into the floor below with either of these models.
Stacking AP's vertically is not usually an optimal practice either. You should also be planning on a follow-up post installation (verification) test.
Have you spoken with an Aruba sales engineer on the subject ? They should be able to straighten everything out.