03-13-2018 09:32 AM
I am in need of connecting two buildings that are just over 4000 ft (1.2 km) from each other. The line of sight is NOT perfectly clear. There do not appear to be any major obstructions, i.e no tall building etc, just mostly some trees and a few roof peaks. The connection is not likely to need heavy bandwidth. Mostly to tie into active directory login and general file sharing. No video or minimal video on occasion.
My thoughts are that I should use one of the units capaple of 900 mhz to cut through the clutter. I'd like to try avoid having to put up raised antenna mast if possible. Both buildings are already 2 1/2 story in height. The RocketM and NanoStationM both appear to be possibilities, but I have never used either up to date. Am I headed in the right direction? Can anyone suggest for or against a specific setup?
I've looked at a few other similar posts in the forum, but most seem to have extremely short distances- so I can't exactly find a comparable situation at the moment.
Thank you in advance!
03-22-2018 03:30 AM - edited 03-23-2018 03:16 AM
first ot all, 4000 ft is not exactly a "short distance", I just installed 100ft PtP link
my rule: no clear line of sight, no link
trees are the worst, even if they look like the least problem - they move from the wind and they contains water
finally - lower frequencies penetrate obstacles better, but lower frequencies also have bigger Fresnel zone, actually I don't know which one is more important
03-22-2018 07:15 AM
Thanks for the thoughts so far. Short distance is obviously relative. I understand your statements, but if I was talking about 100 ft I would just use a regular access point. When I see the specs on some of the units being rated for 15 to 20 km, my 4000 ft is just over 1 km- hence my assessment as short distance.
So far it looks like I may be able to attain a direct line of sight if I elevate one end approximately 15 feet. While I was hoping to avoid raising a pole mount on either end- I think a 15 to 20 ft rise to achieve direct LOS is better value than attempting to power through a few obstructions. I'm hoping to have some exact measurements tomorrow.
03-22-2018 05:39 PM
Yep, LOS always wins. Free-space path loss attenuation is a fraction of the attenuation of a tree or building. Use a smaller gain (or grid style) antenna at your elevated end to reduce wind load and make up the gain with a larger antenna at the other end.
03-23-2018 03:25 AM - edited 03-23-2018 03:38 AM
When I see the specs on some of the units being rated for 15 to 20 km
Even ubiquiti employees on this forum often write this is not real life achieved distance. Why? Because that distance in specs means you can create link on that distance (maybe), but it will be lowest possible speed on the world, but yes - link will be established. To achieve usable tens or hundrets megabites speed, distance need to be much lower.
I like the old version, because you can select different radio models and then see achievable speed.
03-25-2018 07:08 PM - edited 03-26-2018 04:01 PM
to say airlink is optimistic is a huge understatement
simply because that maps knows only terrain, not buildings, trees etc
03-26-2018 12:14 PM
I finally made it to the roof this weekend, and sadly I can say- even with a 20 foot rise over roofline- I cannot clear enough trees.
I do have a smokestack on a neighborhood school in clear view from both facilities, but no clear line to each building so far.
Unless the school may be willing to let me mount two antenna or dished on that stack, I may be stuck using a wired VPN, which will limit me drastically.
03-27-2018 02:30 AM
I definitely would not. Trees are moved by wind, they containt water, depending where you are they can be covered by snow (what is more water). At least ask the school so you know your options.
03-27-2018 06:27 AM
I've probably got at least 3 to 4 good sized trees in the way. Winter they are bare except for possible snow load. Summer they will be pretty full. I'm going to see if the local school can work with me or not. I have a local private tower nearby as well that may be an option, but I'm trying to avoid any major or recurring costs.
03-27-2018 01:46 PM
03-30-2018 01:49 PM
if you think outside the box then the smoke stack maybe your saviour.
If you used 5GHz narrow beam kit like the 400 or larger dishes then point each end at the smoke stack.
If it is a reasonable diameter you should be able to get enough reflected signal to make a working link.
If you have a bulding both ends can see that is in the same plane (so both aerials would point the same way to it) then even better. If it is metal clad even better still
It will all be down to angles and EiRP limits etc..
This works better with flat walls but we have customers bouncing signals off a wall to get around trees on a similar length path and it works fine.
UK 5 GHz WISP - ISPA UK, RIPE Member
Senior Design Eng , RF / Network Eng.