Long Range Wifi
 

Possible project needing designing.
Looking for the design of an antenna aiming device that can be controlled remotely (hardwired) from ground level, and need to be able to be controlled within a ~ 120 degrees left/right, and +- 15 degree elevation with 1 degree control steps or less in each direction. The weight of the dish is ~10 pounds. The entire assembly needs to be weather proof able to stay locked in position during heavy wind, rain, sleet, and snow. And as light as possible. It will be mounted atop a 21’ galvanized steel pipe 1-1/4” ID. The dish requires a ~2” mounting pipe to bolt to.
The cost needs to be as low as possible. Possibly two units required.
One dish mounted 80+ feet in the air atop a 60’ communications tower, the other 60 feet mounted in a tree.
The power source for the control needs to able to be carried either to base of tree/tower, or small enough to be carried up ladder with control panel.
This is for a long range wifi link, that needs to be fine tuned, but the actual dishes are too high above a safe standing/tie in point to tune other than lowering dish, setting elevation and re raising dish.:D

Re: Long Range Wifi
 

Take a look at the offerings from Ubiquity.

Re: Long Range Wifi
 

I'll say the obvious.

It sounds like you need a commercial pan/tilt unit (PTU), with an appropriate weather resistance rating.

There are reasons why vendors can stay in business mass producing those, instead of being put out of business by home-brew devices.

Re: Long Range Wifi
 

Unless it does really need to be wifi, I recommend looking at the xBee line of products. They offer a wide variety of ranges and bandwidths to carry a wireless serial connection between two endpoints.

Re: Long Range Wifi
 

They make some off the shelf devices that hams are using locally to try to set up a network. You listed some tight tolerances, remember that trees are designed to move and twist in the wind. In my area there is always a 5-10 knot breeze blowing

Re: Long Range Wifi
 

I actually am using 2 AC pbe-500 Ubiquity dishes. They are great.

Some tree movement is ok, the dish returns to the same position when the wind dies down. The link is ~ 1.1km

Re: Long Range Wifi
 

If you are trying to link a pair of dish antennae, why do you need to be able to swing one of them through such a wide range (~120 Deg Az & +/- 15 Deg Elev)?

Also, my gut tells me that putting a weather-proof PTU, and a (small) dish antenna, and cabling, atop a 21' pipe atop a 60' tower, isn't going to end well. My gut is often wrong, but that seems like a bunch of mass at then end of a floppy lever arm.

It might not crash and burn, but don't you think it has a good chance of swaying/oscillating enough to foul up your data link?

Re: Long Range Wifi
 

rethinking the specs, a pan of 45 degrees, and elevation of +-5 degrees would be more than enough for fine tuning. The elevation control is most important, because the entire mast pipe can be rotated. The 60' tower is guyed and stands atop 2+ cubic yards of concrete and with me at 60' has only slight sway. and the 21' foot pipe extends maybe 18' above that.

the dish mounted in the tree extends maybe 14' unsupported, and a tree was chosen as the cheapest option,$120, as opposed to installing another 60' tower at a cost of ~$1800 :yikes:

Re: Long Range Wifi
 

I'm worried most about the final 18' of pipe plus PTU and other payload, in the winds (different from ground level) at 80'.

Separate thought:
Can you align the router dish while at the top of the tower, then tweak the mounting fixture a known amount (use a shim?) (the tweak should be small) to get the dish close enough to the right elevation when it's later on the pipe, and has only moved a known distance higher than the spot whet you aligned it?.

Blake
PS: over several years, expect the tree to grow taller. The trees around my house have grown more into my satellite TV line-of-sight over the last 15 years. :mad:

Re: Long Range Wifi
 

Scott,
The tree with a dish is not a solution. Trees grow...
I would move to a yagi at the tree location. It has a bit wider aperture and doesn't need to stay pointed in exactly the right direction.
In reading your original specs, you have one dish mounted 80' feet up on a 60' tower? There is also some panel antennas that have pretty decent gain and a great front to back ratio. We are using those to link to the university next door (they rent space here). Approx path is a few hundred meters. They are powered over ethernet. I have to ask why you are mounting one at 80 feet. Do you have some obstacles?

Re: Long Range Wifi
 

Yes, my obstacles are trees. The theoretical throughput is ~400mbps. When tuned the best can at the moment, I usually get 90-160mbps, which I'm fine with, but I have trees entering the fresnel zone, and link drops to unusable in rain with the leaves on the trees.

Re: Long Range Wifi
 

Yagis at both ends will give you plenty of link margin if it is only trees. I fear you are really hitting up against the speed of light.

WiFi has some very tight send-and-respond timing, particularly at high throughput. 1.1 km is a significant fraction of the ultimate limit for 802.11 distance, specifically because of the time it takes for the signal to travel - at the speed of light.

Oh, and your dish-on-a-pole-on-a-tower will come down in the first or second ice storm - you get those in PA, right? Don't ask how I know.

Oh, and what Al said about trees. (But, perversely, although a tree WILL grow, your antenna won't get any higher off the ground as it does).

I encourage you to learn about WiFi link timing and do the math for 1.1 km. You will be surprised at how tenuous that link is, but not for signal strength. Try 802.16 instead, built exactly for what you want to do.

Re: Long Range Wifi
 

https://dl.ubnt.com/datasheets/Power...Beam5ac_DS.pdf

The dishes are the PBE-5AC-500. 5.8GHZ Carrier using 80MHz channel.
Light travels 300,000km/s, so a 1.1km distance would take 3.67uS one way
Tower is a rohn 25G series.
21'pipe is 1-1/4" ID with 1.66"OD

Re: Long Range Wifi
 

If you do have any doubts about your expected throughput, and if you already have the equipment, take it to the two sides of an easily accessible 1.1km wide unobstructed "valley", and then see how much throughput you can get. That will put an upper bound on what you can try to achieve shooting through the trees.

In the same vein, attach some weight lifting weights to the pole to represent your equipment, plus ice & snow, then mount the pole on something sturdy (at ground level), then pull hard on its top with a rope to mimic a strong wind. See what happens (be out of the way if anything breaks loose).

For the weight and wind loadng test, if you can attach the pole to something at ground level using the same brackets you want to use to lock it in place up in the air, even better. See how they hold up when you abuse them.

Re: Long Range Wifi
 

Not wanting to derail Scott's question, but much along the same lines...

Any suggestions or ideas on a ~750' run thru some trees and the tops of 2-3 houses?

I could try to elevate but that would only get me away from maybe 2 of the houses, trees would still be in the way.

Re: Long Range Wifi
 

OK, one thing is that it boasts "automatic distance selection (ACK Timing)" so my comments about lightspeed are somewhat less relevant.

The spec for the 500 also calls for a 60 lb wind loading force at 60 MPH. Your 6061 aluminum pipe (nominal 1.25" Sch 80) yields at 24000 lbs, you have a Section Modulus of 0.291 cuin. Bending stress at the bottom of the 18' pole is (60 lb x 216 in) / .291 = 44,536 psi which is nearly double the pipe's strength - it'll bend long before 60 MPH is seen. Source (pdf)

And I still thing you're underestimating the effects of tree sway. The 3 dB beamwidth is shown as about 5 degrees, which is pretty tight (but typical for a dish, or a Yagi with comparable gain), wobble will kill your performance.

I don't disagree you need to get above the trees.

Re: Long Range Wifi
 

Don - The OP says the pipe is galvanized steel - Blake

Re: Long Range Wifi
 

Ah, missed that. Steel should be OK, at about 54k. Thanks Blake.

750 feet through the trees: The lower the frequency, the less the trees matter. So 2.4 GHz is better than 5 GHz, 900 MHz even better. Not sure there's a good commercial solution below that. Even 440 MHz is affected, while 150 hardly at all.
No chance you could run some Ethernet, is there? drainage culverts sometimes offer a solution to cross streets...

Re: Long Range Wifi
 

Yep, just don't ask me to be one of the poor saps who has to lug a steel pipe, plus the payload, up that 60' tower; and then hold the pipe & payload upright and aloft to attach it to the tower.

That's not a desk job. I'll create the drawings, if someone else will do the climbing ;) .

Re: Long Range Wifi
 

I raised the pipe using ropes, and a jin pole, same with the dish and the top tower section comes to a point with a mast bushing in its center. So the pipe was raised, than lowered through the bushing, dish was attached and the whole assembly was raised back up. I was securely attached to the tower with my work at heights harness.

As for the 750' culverts are the way to cross streets, if used I would run the cable through 1/2" black water line to help protect it. as for the 100m ethernet limitation, a poe extender solves that

http://www.bhphotovideo.com/c/produc...x_Outdoor.html

Re: Long Range Wifi
 

Scott,
Sorry but work got in the way of responding. I have to weigh in on the pipe moving around. We have ours mounted on 4" galvanized pipe and it is only 10 feet high (roof mounted) to keep the patterns aligned. Mounting another dish on a tree flexes even more that a small pipe. With both the tower and the tree moving around and twisting, it is very likely that the dishes will pull out of alignment on a regular basis that will be nearly impossible to compensate with a moving dish.
A system to aim antennas real time, requires some method to determine the direction of the move and magnitude. I have seen this done with a system that "wobbles" the device so that a feedback mechanism can detect both direction and magnitude. You would not want to due that with a dish for mechanical stress. You could move the feed horn which is a much smaller mass but it may not give you the range needed.
As to trees in the fresnel zone, if they were simply into the zone and not obstructing it, the path loss would seem to overcome the phase rotation introduced by the leaves. However, I think the trees are likely obstructing the path and you have significant path loss that this dish pair cannot overcome. Just running a quick path loss calc it looks like you could easily have 80 dB loss without the trees with this dish pair. I would bet that the tress introduce a variable or between 20 and 40 dB additional loss. That eats all of your available margin. The only way to overcome path loss is to increase transmit power, receiver gain or antenna gain(size). With the antennas you are using, none of those are possible. Even going to the 620 antenna only gets you another 2dBi. I think you have to pick up another 10-20dB of gain.
For my money, a larger antenna mounted on the tower and not a pipe is preferred. I can speak to leaves being a significant attenuator at 450 MHz. Many digital broadcast stations are finding that antenna placement for our viewers need to be away from trees. Many folks who had no problem with reception of our high VHF station in the past, have issues on our Ch47 (660MHz) UHF digital station. When it rains, data loss is significant to the point of no reception at all.