dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
3791
share rss forum feed

whoever

join:2006-03-13
Avon, CO

Shield kits for Ubiquiti?

OK, need opinions.

Who makes the best, most reliable shield kit for the Ubiquiti sector antennas. There are a few manufacturers now. Need feedback before I order 8 kits.

These are on separate towers. GPS not an issued just yet.

Thanks.

gunther_01
Premium
join:2004-03-29
Saybrook, IL

1 recommendation

Buy AMERICAN. That's the answer. It's darn near the same thing in all cases. So spend a dollar or two extra and support someone in this country.!!!
--
»www.wirelessdatanet.net

OHSrob

join:2011-06-08
reply to whoever
Be careful with your wind loading.

Newbie

join:2011-04-18

1 recommendation

reply to whoever
I buy exclusively from RFarmor, they work wonders.

mj3431

join:2003-04-21
STL, MO
reply to whoever
+1 for RF Armor.


John Galt
Forward, March
Premium
join:2004-09-30
Happy Camp
kudos:8
And +1 for wind loading...


refused

join:2005-10-10

1 recommendation

reply to whoever
while everyone seems to be happy with their rfarmor shields, i'd almost go with something else just in spite of his constant advertising on the ubnt forums. dude is seriously annoying.

wirelessdog

join:2008-07-15
Queen Anne, MD
kudos:1

2 recommendations

reply to whoever
I wish ubnt would get their shit together so you don't need these shields.


WHT

join:2010-03-26
Rosston, TX
kudos:5

2 recommendations

reply to refused
Your metric for "advertising" must be pretty low. He usually mentions it when it is a solution to a users problem.

OHSrob

join:2011-06-08

4 edits

1 recommendation

reply to whoever
FYI wirelessdog, rf armor is just for people who cannot plan their deployments properly.

Ubiquiti's 2.4ghz and 5.8ghz gear is rock solid they may not have figured out GPS sync yet but im sure that someone will and open it (Probably a college or university) and ubiquiti will integrate it in their products if they are not stupid.

I have tested it and it is not worth it for us as it seems to add too much multipath reflection. It makes maintaining the 64QAM rates very difficult to maintain as you need a higher SNR. 16QAM modulations don't seem impacted tho (tested with both the rocket dish version and the nanobridge version).

If the RF armor had a RF absorbing foam inside it like the andrew point to point parabolic's. It would eliminate the multipath reflection and would be good enough for me to consider wide scale deployments.

edit: Ubiquit does not like signal's stronger than -45dbm, I don't know the technical reason behind why it handles signal overload so poorly but if its stronger than -45dbm it will have a line in airview right across -70 to -75dbm when it transmit's that significantly impacts the receive sensitivity. (Regardless of what channel the other equipment is on).

edit: I don't have the diagrams on what magic is occurring inside the atheros chip SOC but it might be that the down converter is leaking too much signal into the IF when you have too strong of a signal.


WHT

join:2010-03-26
Rosston, TX
kudos:5
UBNT gear doesn't handle signal overload for same reason as other manufactures' gear. Nothing surprising.

gunther_01
Premium
join:2004-03-29
Saybrook, IL

1 recommendation

reply to whoever
So the shields that someone took upon themselves to build for themselves and then the community are because no one in the industry knows how to build and deploy their network? Pfft and LOL

The same shields that he built, are being built by others now, AND UBNT integrated in to their products. Yeah, I think no body knows what they are doing. It started with UBNT in their design, not everyone elses.

You must not have been using this gear very long or something. Most of us know why the shields are important, and work rock solid afterwards.
--
»www.wirelessdatanet.net

OHSrob

join:2011-06-08

1 recommendation

said by gunther_01:

So the shields that someone took upon themselves to build for themselves and then the community are because no one in the industry knows how to build and deploy their network?

I would say from what I have seen at least a good 80%+ of wisp's (probably higher) sure don't have a clue in the slightest how to deploy or design a network properly or even how the most basics behind how the protocol's they are using works.

They think because they put a few 802.11 ap's up in a building before that they are RF masters.

Then they hang 10 square feet worth of gear on a tower rated for 4 square feet of wind loading and run the network unencrypted with all the gear crammed in the smallest space at the top of the tower.

edit: I have not tested the RF armor sector sheild's as I do not wish to put big sails on my tower they may not add multipath reflection. But they are of no use to me anyways since I plan my deployment's.

gunther_01
Premium
join:2004-03-29
Saybrook, IL
reply to whoever
I would partially agree with you. But keep in mind a lot of us have been doing this long before UBNT came out with the M series. Of course there are a lot more WISP's, (and I say that term very lightly, since most are not) doing it afterwards, that don't know a thing also.
--
»www.wirelessdatanet.net

Newbie

join:2011-04-18

1 edit

1 recommendation

reply to OHSrob
"FYI wirelessdog, rf armor is just for people who cannot plan their deployments properly."

This is hands down one of the most ignorant blanket statements I have ever read on these forums..... And that says something!


WHT

join:2010-03-26
Rosston, TX
kudos:5

1 recommendation

reply to OHSrob
said by OHSrob:

But they are of no use to me anyways since I plan my deployment's.

You are fortunate that your competition will work with you on mitigating the interference they are causing you.

wirelessdog

join:2008-07-15
Queen Anne, MD
kudos:1
reply to whoever
Sometimes the truth hurts.

Somebody explain to me how this cluster works with all the radios in the same band and each non-overlapping frequency is used twice.

NO Shields required LOL


treichhart

join:2006-12-12
reply to whoever

gunther_01
Premium
join:2004-03-29
Saybrook, IL
reply to wirelessdog
The little pointy thingy on the top, that's how Oh, and a proper RF set. But hey, it costs more, and we can't have that in this day and age. Well, except the companies that plan to stay around longer than the competitor who doesn't have the pointy thing on top LOL

(GPS receiver/polling, just in case you didn't get the sarcasm)
--
»www.wirelessdatanet.net

Chele

join:2003-07-23
kudos:1
reply to whoever
we actually roll our own! We bought a sheetmetal roller and buy the metal locally. We use enough of them that it makes it cost effective to make our own stuff, plus we have a strong background in manufacturing.


WHT

join:2010-03-26
Rosston, TX
kudos:5
reply to wirelessdog
said by wirelessdog:

Somebody explain to me how this cluster works with all the radios in the same band and each non-overlapping frequency is used twice.

Easy....

1) All the transmitters are synchronized, so there isn't any receiver interference.

2) Adjacent radios not on same channel, so one radio won't be hearing any CPE on that adjacent channel.

3) Beamwidth of every third AP is such that they AP will only hear it's own set of CPEs.

Now what happens when there is any interference just outside of any one radio's beamwidth that is sufficiently strong enough to cause a problem? This is what the external shields do, they tighten the beamwidth.

Added...
said by wirelessdog:

NO Shields required LOL

BTW, you don't need shields to make that work. Unless you have uncontrollable third-party interference where shields have been shown to help.

OHSrob

join:2011-06-08

3 edits
reply to WHT
said by WHT:

said by OHSrob:

But they are of no use to me anyways since I plan my deployment's.

You are fortunate that your competition will work with you on mitigating the interference they are causing you.

They do not co-ordinate with me, I deal with over 10 different wisp's in my coverage area all using 20mhz channel's.

One waveon wisp, three canopy providers, two using mikrotik, three ubiquiti (two airmax) and one tranzeo wisp.

One of my 2ghz sectors picks up over 40 access points from various wisp's. (I have line of sight for about 50km from the sector)

I simply find the most clear part of the radio spectrum and size my channel and amount of equipment I put up accordingly. I also keep the amount of equipment I put up to just the minimum that I need.

I avoided a deployment of 900mhz for over two year's because I was not able to find a low noise suitable spot until only recently where I found a spot that I could not hear canopy on horizontal. I noise test from the ground before I even consider building a tower or hanging equipment.

I take advantage of large groups of trees and geographical isolation to limit self interference. I also only deploy 2.4ghz in very specific situations and at most only two 2.4ghz sectors per tower. (The majority being only one sector deployed)

None of my equipment has a transmit power higher then 24dbm and this is only on my macrocell's.

My Microcell's are all below 10dbm same with all my short backhaul link's.

I maintain the minimum transmit power required to maintain at a 3/4 64qam rate to my client's. I also ensure that no client needs more then at most one or two re-transmit's.

Also to prevent signal overload the best thing I have found is to try to offset sectors both horizontally and vertically with sufficient space to prevent them from hearing each other too loud.

This way I don't suffer from the increased ping and jitter people post about on the Ubiquiti forum that use GPS sync complain about.

My customer's can expect their ping to not vary more than 5ms or so 99% of the time. I do my best to ensure that all packet's are delivered on the first transmit to ensure the most free radio time. My network is also packetloss free even to my weakest customer that is something none of my competitors in my area can say about their networks.

said by WHT:

said by OHSrob:

But they are of no use to me anyways since I plan my deployment's.

You are fortunate that your competition will work with you on mitigating the interference they are causing you.

They do not co-ordinate with me, I deal with over 10 different wisp's in my coverage area all using 20mhz channel's.

One waveon wisp, three canopy providers, two using mikrotik, three ubiquiti (two airmax) and one tranzeo wisp.

One of my 2ghz sectors picks up over 40 access points from various wisp's. (I have line of sight for about 50km from the sector)

I simply find the most clear part of the radio spectrum and size my channel and amount of equipment I put up accordingly. I also keep the amount of equipment I put up to just the minimum that I need.

I avoided a deployment of 900mhz for over two year's because I was not able to find a low noise suitable spot until only recently where I found a spot that I could not hear canopy on horizontal. I noise test from the ground before I even consider building a tower or hanging equipment.

I take advantage of large groups of trees and geographical isolation to limit self interference. I also only deploy 2.4ghz in very specific situations and at most only two 2.4ghz sectors per tower. (The majority being only one sector deployed)

None of my equipment has a transmit power higher then 24dbm and this is only on my macrocell's.

My Microcell's are all below 10dbm same with all my short backhaul link's.

I maintain the minimum transmit power required to maintain at a 3/4 64qam rate to my client's. I also ensure that no client needs more then at most one or two re-transmit's.

Also to prevent signal overload the best thing I have found is to try to offset sectors both horizontally and vertically with sufficient space to prevent them from hearing each other too loud.

This way I don't suffer from the increased ping and jitter people post about on the Ubiquiti forum that use GPS sync complain about.

My customer's can expect their ping to not vary more than 5ms or so 99% of the time. I do my best to ensure that all packet's are delivered on the first transmit to ensure the most free radio time. My network is also packetloss free even to my weakest customer that is something none of my competitors in my area can say about their networks.

said by wirelessdog:

Sometimes the truth hurts.

Somebody explain to me how this cluster works with all the radios in the same band and each non-overlapping frequency is used twice.

NO Shields required LOL

When you don't have equipment on the same tower transmitting at the same time another device is receiving you cannot suffer from RF signal overload.

Proper GPS sync is nice, Canopy got it right first try.

said by Newbie:

"FYI wirelessdog, rf armor is just for people who cannot plan their deployments properly."

This is hands down one of the most ignorant blanket statements I have ever read on these forums..... And that says something!

Sorry but your right there is one instance where I would consider using them.

If I was co-locating with another wisp on the same tower or within 800 meters of my tower using the same band as me.
Expand your moderator at work


WHT

join:2010-03-26
Rosston, TX
kudos:5

Re: Shield kits for Ubiquiti?

Shield kits work to point some people don't bother with GPS.
Expand your moderator at work


WVNet

@lsnetworks.net

Re: Shield kits for Ubiquiti?

I find the mechanical benefits of the shields almost as useful as the RF shielding. With mountaintop towers at 7000 or 8000ft they offer protection from ice damage, contain the heat of the radio to protect from low temps, and offer a grounded metal shield around the radio, reducing the chance of lightning damage.

We use RFarmor, and he's been building them long enough to start making some good mechanical improvements to his designs.

raytaylor

join:2009-07-28
kudos:1
reply to wirelessdog
said by wirelessdog:

Sometimes the truth hurts.

Somebody explain to me how this cluster works with all the radios in the same band and each non-overlapping frequency is used twice.

NO Shields required LOL

Thats because the little triangle pointy thing on top is the GPS reciever. Canopy use gps timings.

I was just given an AP cluster like this with 7 AP's (6 for 360 coverage), two gps timing controllers (with the triangular antennas) and 50 client radios+reflector dishes

The instruction manuals say to have them on non-overlapping frequencies with the gps timing enabled.

They dont need shield kits because the timing is set up so that all AP's tramsmit at the same time - therefore one ap on the opposite side of the cluster wont be transmitting while another is recieving and therefore the self interference is removed.


WHT

join:2010-03-26
Rosston, TX
kudos:5
You still non-overlapping channels to prevent CPE interference at the AP.

Say if you had three sectors pointed N, SE and SW. A client off to the NE or E and associated with the N AP, would be seen by both the N and SE AP. The SE AP would see it as noise.