MikroTik

I currently have an omni on a large tower and I have been thinking about switching out to three 120 deg sector antani's. My question is how should I set this up?

1) Three sectors into a splitter then to the single radio & routerboard like I have now. (400mw)

2) Three sectors into a three radio's onto the same routerboard?

3) Three sectors into three radio's using three routerboards?

What do you suggest is best? or most common way of doing this?

Thanks,
Michael

3 sectors, 3 radios, 2 RBs,

Would you set all three radio's to the same ssid, and set them in a bridge? And if your using two routerboards would you bridge radio-->Radio-->ethernet-->ethernet-->radio?

Kinda' late for answer.
You can put the three radios into a single rb ( there are at least 4 choices now with 3 minipci) and put them on a bridge. As the simplest solution. You can put a single ssid, if you want. If you put different ssid's, you can see them better from the cpe's situated on the "border" of two radios coverage.
Also, keep in account, putting three radios on bridge, creates a single broadcast among them, so you should count that when linking your client. If they are many, put different ssid's and put different networks for them.
Best, but most expensive would be to put different boards with single radio for each. Will also give you more separation of the traffic to clients.......... and more headaches....

You can do ths fine with rb333 and 3 r52h cards, using seperate channels.
Use nstreme with polling, and disable csma.

hi all,

I am also have almost same issue. We have three towers. Two of them are 45M and one is 24M but on top of a small hill. All towers carrying same devices with same configuration. One tower is as Base tower making point-to-point connection with two other towers. Currently my config is one RB532 with two R52 Mini PCI card, one card connected to 2.4GHz Omni antenna. Other R52 is connected to four 90 Degree Sector antenna of 16dbi (Model No. 5GD16) using a splitter. The four antennas are fixed on a square bracket by dividing 360 into 90. The performance is poor on 5 GHz. Its not reaching more than 1.5Km. How can I rectify thhis problem? I heard that 5GHz sector has more impact and long coverage than that of 5 GHz Omni????? Did my understanding is wrong? 5Ghz and 2.4 GHz AP are serving two different /24 IP. Now we urge to upgrade/reallocate the existing settings. We have one more RB532 and RB502 daughterboard(2 Mini PCI) for each tower. What is the ideal way to proceed.
One RB532+RB502 with 4 sector antenna+ One RB532 with 2.4 GHz omni,
Splitting the existing 5GHz /24 IPs into /26 with same/different SSID?

OR
One RB532+RB502 with 3 sector antenna+ One RB532 1 sector antenna and one 2.4 GHz omni,
Splitting the existing 5 GHz /24 IPs into /26 with same/different SSID?

OR

One RB532+RB502 with 2 sector antenna+ One RB532 2 sector antenna and one 2.4 GHz omni,
Splitting the existing GHz /24 IPs into /26 with same/different SSID?

I am really confused........ Which configuration is mangable? easily deployable? and reliable???? I hope somebody can help me..........

The reason I was asking about sectored antani's is I was wanting to know if it would have better performance than an omni. As in obtaining better signal strengths or distances.

Two examples:

One site is on the side of a small mountain that over looks the town. But on the back side of that antani your looking into the mountain and trees. I have about a 180 degree arc of the town. So my question I have is should I replace the omni with a sectored 180 so that I'm not wasting "signal energy" shooting into the hill or up into the air.

The second site is on a fm radio tower that is on the highest mountain in the area and at this moment I have about 45 clients connected to it. And the two longest links are 13 miles one north and one south. So I'm getting about 26 miles in a hilly mountain area. And my question with this setup is would sectored would be a better choice as well.

btw I'm using 802.11b

I have tried like this: 1 RB532+1 RB502 with one R52H and three R52 is connected with One 2.4 Omni and 3 sector antenna respectively and another RB532+R52 with 1 sector antenna. The result is frustrating one. No improvement. The same routerborad and its configuration with 5GHz omni (omni has only 10 dB while sector has 16 dB) is hitting more than 4Kms. The sector is hitting less than 1 km. I suspect that the Mikrotik 5g D16(Sector Antenna) is not propagating properly. How I can check this. According to the experiences the sector should perform better than the omni. Any advise or experiences to share ????? please

I have tried like this: 1 RB532+1 RB502 with one R52H and three R52 is connected with One 2.4 Omni and 3 sector antenna respectively and another RB532+R52 with 1 sector antenna. The result is frustrating one. No improvement. The same routerborad and its configuration with 5GHz omni (omni has only 10 dB while sector has 16 dB) is hitting more than 4Kms. The sector is hitting less than 1 km. I suspect that the Mikrotik 5g D16(Sector Antenna) is not propagating properly. How I can check this. According to the experiences the sector should perform better than the omni. Any advise or experiences to share ????? please

What kind of Omni antenna's are you using? and what kind of sector antenna's are you using. It is important apon the type antenna's you use. Hyperlink is a good choice for sector antenn's in my opinoin. Also not to be harsh apon what your doing but this is good news to me due to the fact i will be setting up AP's with Omni antenna. that is why im slightly curious to see what omni you used to get 4km's? thanks -Jordan

it@mikrotik,

4 cards in a rb532 ?
Sounds like alot.

Try running each sector from a seperate rb532s and enclosures.
Be sure to use substantial channel spacing.
Disable the radio with the omni, while testing, then test again with the omni radio enabled.

I beleive the answers may pop out at you.



jordantrx,

I have a links >10 miles off 2.4GHz with 100%/100% signal quality.
It is irrelevant what omni is being used, to a degree.
My links mentioned are from a 9db PacWireless omni.
Terrain and noise, figure into the equation in a huge way.
Do not forget about altitude.....

jordantrx,

I have a links >10 miles off 2.4GHz with 100%/100% signal quality.
It is irrelevant what omni is being used, to a degree.
My links mentioned are from a 9db PacWireless omni.
Terrain and noise, figure into the equation in a huge way.
Do not forget about altitude.....

Correct, the reason i wanted to know what type of sector antenna's he had is because it does matter much apon the antenna manufacture credibility to quality, It could be somebody making home made job antenna's and they have more Uptilit than downtilt to them. Buying good antenna's and most likely expensive is the key to getting good distance/less interference/and better signal quality.... I beleive what antenna he is using is relevant to determine the quality of the product, But when i asked the question it was more out of curiosity of to what distance can be acheived with certian DBI omni antenna's.. Thats all- Jordan

P.S Those are very good distances to be getting with an omni antenna i admit, You must have had them quite level with eachother with LOS?

omni is at ~115ft
client is at ~30ft

the client is on a grade which puts them at 50ft relative to the omni.

omni is at ~115ft
client is at ~30ft

the client is on a grade which puts them at 50ft relative to the omni.

Not to get off topic but what kind of antenna and radio did the client have? thats really good distance!!!

I have tried like this: 1 RB532+1 RB502 with one R52H and three R52 is connected with One 2.4 Omni and 3 sector antenna respectively and another RB532+R52 with 1 sector antenna. The result is frustrating one. No improvement. The same routerborad and its configuration with 5GHz omni (omni has only 10 dB while sector has 16 dB) is hitting more than 4Kms. The sector is hitting less than 1 km. I suspect that the Mikrotik 5g D16(Sector Antenna) is not propagating properly. How I can check this. According to the experiences the sector should perform better than the omni. Any advise or experiences to share ????? please

What kind of Omni antenna's are you using? and what kind of sector antenna's are you using. It is important apon the type antenna's you use. Hyperlink is a good choice for sector antenn's in my opinoin. Also not to be harsh apon what your doing but this is good news to me due to the fact i will be setting up AP's with Omni antenna. that is why im slightly curious to see what omni you used to get 4km's? thanks -Jordan

Hi jordantrx,
I have explained all details in my preivous posts. I will explain again. We have 3 towers. two of them are 45 m and other one is 25m but on a small hill. Each tower carrying one 2.4 GHz omni and 4 Sector antennas of 90 Degree each(5G D16, provided by mikrotik, 90Degree horizontal beamwidth, 5 Degree vertical beam width,16 dBi gain,5GHz). The 4 sector antennas are connected in a square bracket to get 90 Degree angle. The very first config was like this: One RB532 with 2 R52 wlan card, one is connected to 2.4 GHz omni and another one is connected to 4 way splitter connecting four sector antennas. The signal is not going more than 750 Meters..
Now the config is this: One RB 532+RB502(daughter board), 4 R52 Wlan card, Among these 4 Wlan card, one is connected to 2.4 GHz and three are connected to sector antennas individually. One more RB 532 with R52 wlan card is connected to fourth sector antenna. Still the result is same.
Tried with 5GHz, 10 dBi Omni, the result is amazing..... its shooting a range of 4+ KMs......
All equipments, components, accessories are from Mikrotik(2.4 Ghz Antenna, 5GHz Antenna, RBs,Daughter boards, connecters, factory crimped cables, Wlan cards etc ).. According to the experiences the sector should work well other than omni????? Any body can help...If you need debug these again I am very happy to provide...........

jordantrx,

I have a links >10 miles off 2.4GHz with 100%/100% signal quality.
It is irrelevant what omni is being used, to a degree.
My links mentioned are from a 9db PacWireless omni.
Terrain and noise, figure into the equation in a huge way.
Do not forget about altitude.....

Correct, the reason i wanted to know what type of sector antenna's he had is because it does matter much apon the antenna manufacture credibility to quality, It could be somebody making home made job antenna's and they have more Uptilit than downtilt to them. Buying good antenna's and most likely expensive is the key to getting good distance/less interference/and better signal quality.... I beleive what antenna he is using is relevant to determine the quality of the product, But when i asked the question it was more out of curiosity of to what distance can be acheived with certian DBI omni antenna's.. Thats all- Jordan

P.S Those are very good distances to be getting with an omni antenna i admit, You must have had them quite level with eachother with LOS?

All antennas and accessories are from Mikrotik. The 2.4 Ghz is 7 dbi Omni and 5Ghz is 16 dbi Sector(90 degree horizontal beam width and 5 degree vertical beam width)


Tgrand,
Listen 4 R52 wlan in not connected directly to RB532. Two of them connected directly and rest of two are connected using a daughterboard (R502) In a big outdoor enclosure.
Each sector antenna are running on different channels (5220,5240,5280,5260) with different SSID. Am I doing some thing wrong??? any suggestions or advises????

I have tried like this: 1 RB532+1 RB502 with one R52H and three R52 is connected with One 2.4 Omni and 3 sector antenna respectively and another RB532+R52 with 1 sector antenna. The result is frustrating one. No improvement. The same routerborad and its configuration with 5GHz omni (omni has only 10 dB while sector has 16 dB) is hitting more than 4Kms. The sector is hitting less than 1 km. I suspect that the Mikrotik 5g D16(Sector Antenna) is not propagating properly. How I can check this. According to the experiences the sector should perform better than the omni. Any advise or experiences to share ????? please

What kind of Omni antenna's are you using? and what kind of sector antenna's are you using. It is important apon the type antenna's you use. Hyperlink is a good choice for sector antenn's in my opinoin. Also not to be harsh apon what your doing but this is good news to me due to the fact i will be setting up AP's with Omni antenna. that is why im slightly curious to see what omni you used to get 4km's? thanks -Jordan

Hi jordantrx,
I have explained all details in my preivous posts. I will explain again. We have 3 towers. two of them are 45 m and other one is 25m but on a small hill. Each tower carrying one 2.4 GHz omni and 4 Sector antennas of 90 Degree each(5G D16, provided by mikrotik, 90Degree horizontal beamwidth, 5 Degree vertical beam width,16 dBi gain,5GHz). The 4 sector antennas are connected in a square bracket to get 90 Degree angle. The very first config was like this: One RB532 with 2 R52 wlan card, one is connected to 2.4 GHz omni and another one is connected to 4 way splitter connecting four sector antennas. The signal is not going more than 750 Meters..
Now the config is this: One RB 532+RB502(daughter board), 4 R52 Wlan card, Among these 4 Wlan card, one is connected to 2.4 GHz and three are connected to sector antennas individually. One more RB 532 with R52 wlan card is connected to fourth sector antenna. Still the result is same.
Tried with 5GHz, 10 dBi Omni, the result is amazing..... its shooting a range of 4+ KMs......
All equipments, components, accessories are from Mikrotik(2.4 Ghz Antenna, 5GHz Antenna, RBs,Daughter boards, connecters, factory crimped cables, Wlan cards etc ).. According to the experiences the sector should work well other than omni????? Any body can help...If you need debug these again I am very happy to provide...........

Have you tried using another type of sector antenna's such has hyperlink, i know those antennas will shoot up to 10 miles.. It is ether your wireless cards are damaged slightly and not putting enough power out, or your cable pigtails and MMCX connectors are weak and the connect points are broken inside. Or you have a crapload of interference, But if you have interference on your sectors you would have it on your omni so it cant be that is the problem..... I think your best soulotion is to make sure your wireless cards are good (if you used the omni and the same wireless cards you used for the sectors i imagine they are good aswell). Make sure your cables are in perfect condition and tight... if all else fails. The only thing possible it could be is the sectors are not good, I have good experience with Hyperlink tecnologies sector antenna's very well built....... Well good luck ! -Jordan

Ummmm..... is everyone not seeing the obvious here? He's using a splitter....

Lose the splitter.

jwcn:
the splitter was 1 of 3 proposed scenarios.
please, re-read the first post.


it@mikrotik:
I would not recommend anyone use 4 radio cards within the same enclosure, let alone on the same router (daughterboard or not).

it@mikrotik is currently using the splitter...

Hi Jordantrx,

We have tried with new R52 wlan cards. and have same problem. Check the cables and connecters by connecting one omni. works fine. when we connected the sectors using the splitter, the signal strength on clients side within 500Mtrs from the tower was -87 after removing the splitter, connecting to individual wlan card, the strength is around the same figure(-85 to -87). What it could be????

jwcn:
the splitter was 1 of 3 proposed scenarios.
please, re-read the first post.


it@mikrotik:
I would not recommend anyone use 4 radio cards within the same enclosure, let alone on the same router (daughterboard or not).

tgrand,
Currently we have RB532 and RB502(daughter board) only. According to you, can we try 2 wlan using one RB532 and another 2 wlan using second RB532 instead of RB502(daughter board)?

it@mikrotik is currently using the splitter...

We are not using splitter now. In my previous scenario it was. Now RB532, RB502(daughter board) and R52 wlan card.

Hi Jordantrx,

We have tried with new R52 wlan cards. and have same problem. Check the cables and connecters by connecting one omni. works fine. when we connected the sectors using the splitter, the signal strength on clients side within 500Mtrs from the tower was -87 after removing the splitter, connecting to individual wlan card, the strength is around the same figure(-85 to -87). What it could be????

i Would say Use different sector antenna, you can see that your omni antenna and cables are good because when you hook your wireless card and cables up to the Omni antenna performence is great, but once you hook it up to the sector it sucks butt... To me atleast that seems like bad sector antenna..... Or SOME how SOME way that r52 dosent like those sectors and you could try using an Sr2 or an Xr version card...... I highly dought this but you NEVER know with wireless... best bet is to buy a different brand (Goood) brand of say 1 sector and give it a shot it it is just as bad send the sector back it isnt your sectors..... Good luck....

jwcn:
the splitter was 1 of 3 proposed scenarios.
please, re-read the first post.


it@mikrotik:
I would not recommend anyone use 4 radio cards within the same enclosure, let alone on the same router (daughterboard or not).

tgrand,
Currently we have RB532 and RB502(daughter board) only. According to you, can we try 2 wlan using one RB532 and another 2 wlan using second RB532 instead of RB502(daughter board)?

Yes avoid the daughter card for this con fig.
Use the R52 or R52H, avoid the xr series of cards on the sectors.
Even a poor grade of sector should net you better results than what you are getting, so I do not beleive it to be the sectors.
A splitter will only attenuate the signal by 4.5db on three way, which does not match your poor results either.

A Good test would be to try the RB532 with no daughter board and only 1 radio connected to one sector, and check you signal levels.
Next try adding a second card connected to a second sector, and check signal levels again, off both sectors, and log your results.
Next add the daughter board and another radio connected to third sector, and repeat measurments... etc, etc.

Make conclusions based upon your results, and report findings here.

If you have multiple sectors on one site, they will surely interfere with each other.
One of the ways of avoiding this, is by studiing the tipical side lobes of the antenna, to try at least to tilt them down in such a manner that they will get side lobes on the back pointing upwards. Also, try putting them one on top of each other, separated of as much space as you can afford in your specific scenario.
I have sites where i'm better with an omni but with three 120 degrees sector. Weird, no ?
My guess is that you are getting noise on each card from the others.......
Do some tests with only one sector at a time enabled, see what happens. If it's the same as with all of them, you might try checking again the hardware you are using.

Hi Jordantrx,

We have tried with new R52 wlan cards. and have same problem. Check the cables and connecters by connecting one omni. works fine. when we connected the sectors using the splitter, the signal strength on clients side within 500Mtrs from the tower was -87 after removing the splitter, connecting to individual wlan card, the strength is around the same figure(-85 to -87). What it could be????

Check the client's config. Might be something there. I have -85 1500m from a r52+12db omni, on about the same level ( altitude ), with a rb133+r52+rubber antenna (the ones for interior use)

Just to make sure.... your sectors are the same polarization as your omni, right? I skimmed through the posts again and didn't see if it was specified.

Joe

Just to make sure.... your sectors are the same polarization as your omni, right? I skimmed through the posts again and didn't see if it was specified.

Joe

Yes all the antennas are on right polarisation

I got some suggestions, that i havent tried. Will do and post the reply soon, thanks.

We have 8 sites all using sector approach.

Each tower uses 3 RB532A with a SR5 and a SR9. When sectoring you MUST get good vertical seperation on the tower with antennas in the same frequency range. I separate my antenna's at least 4-5 feet to avoid self interferance. ( If you use a combination of vertical and horizontal antennas you can lessen the physical spacing if required). With the SR9 we get a maximum of 7Kms NLOS. Sr5... upto 12-15km LOS in a PTMP environment. We use 5Mhz channel spacing on the SR9 but beware... the SR9 will interfere significantly each other in a sectored approach unless you get really good filters. The Ubiquity filters are good and give you the ability to switch channels if needed, but if you are in a noisy area you can buy filters like DCI which you can get custom tuned for a channel @ 5.5MHZ channel spacing. (Drawback is, you can't switch frequencies)

Using a splitter is a very bad idea!

The advantage to sectoring:

1. Less clients per RB therefore better performance
2. More avalable radio time slots / better performance
3. (3 sectors).. if you have a single sector failure only 1 third of your clients on that site are down (You can move nearby clients to the other sectors therefore even fewer clients are down) ... with an omni.. everyone is down.
4. Main advantage... MORE CUSTOMERS PER SITE = MORE $$ / month

Hope this helps!

We have 8 sites all using sector approach.

Each tower uses 3 RB532A with a SR5 and a SR9. When sectoring you MUST get good vertical seperation on the tower with antennas in the same frequency range. I separate my antenna's at least 4-5 feet to avoid self interferance. ( If you use a combination of vertical and horizontal antennas you can lessen the physical spacing if required). With the SR9 we get a maximum of 7Kms NLOS. Sr5... upto 12-15km LOS in a PTMP environment. We use 5Mhz channel spacing on the SR9 but beware... the SR9 will interfere significantly each other in a sectored approach unless you get really good filters. The Ubiquity filters are good and give you the ability to switch channels if needed, but if you are in a noisy area you can buy filters like DCI which you can get custom tuned for a channel @ 5.5MHZ channel spacing. (Drawback is, you can't switch frequencies)

Using a splitter is a very bad idea!

The advantage to sectoring:

1. Less clients per RB therefore better performance
2. More avalable radio time slots / better performance
3. (3 sectors).. if you have a single sector failure only 1 third of your clients on that site are down (You can move nearby clients to the other sectors therefore even fewer clients are down) ... with an omni.. everyone is down.
4. Main advantage... MORE CUSTOMERS PER SITE = MORE $$ / month

Hope this helps!

Yes, a really interesting post. But please make it clear that make and specs of the sector antenna u r using. I haven't worked with SR series of cards. Can you please compare SR cards with R52 cards. The way and idea you implemented is a good one. I am sorry to say that I dont have experience using the filters. Now I want to make sure that the sector antenna with us is propagating the signal properly. One of the antenna is in ground. How can I check it effectively? Thanks in advance:)

On my main tower I have a single 15dbi omni 2.4ghz and I have clients in a 13 miles in each direction (26 mile radius). Using nmp- cards 400mw and a single rb532. I have more issues out of my 5.8ghz links which typically are less than 3 miles with gird antani's on both sides than I do 2.4ghz.

But on a different tower with the same anti I do 7 mile links but the back 180 degrees of the antani faces into a hill side of trees. I had wondered if I should use a sectored antani to basicly push the power foward and have less shooting into the trees which does no good.

-Michael

mperdue,

You may be getting reflections from the hillside.
Try using a smaller panel for the panel that aims towards the hills, and larger antenas on the other side of the link.

I don't think anyone pointed out or asked the original poster how far apart the sectors were? Could be self interference. What the tilt angle was, he said he was getting a short distance, and If I recall from reading the post he said a 5 deg vertical range, I think that at the 1.5km's and the height of the tower would explain the lower signal, perhaps try going out further and testing.

I have also had a wierd issue with RB52H cards, running in 5.8, all of the sudden, my link that had great speeds and CCQ dropped to nearly 1/2 signal, and the ccq went from 90's to 50 -60. I don't use that one card anymore, but it still works for testing purposes such as when I tested it in a 532A running bandwidth test transmitting, for 36 hours on a 18AH gel battery. I decided to stop it since I didn't want to do a deep discharge on the battery, if it wasn't already done. Was testing to see how long it would go on 8 of those batteries for a possible solar site I might be setting up. Hopefully I get authorization to do it, could connect all of my buddies together!

Yes Off topic... sowwy..

Ok - so we are VERY lucky to have this "rock" (read mountain) which is 2000m (±6500ft) ASL (above sea level) and our average ground level for users is ± 500m (± 1600ft) .. so our "look-distance" is huge... and it's a range of mountains that runs along our "area" for 70km (± 40 miles)...

Getting back to the issue...

Our standard base stations are:-

Everything in 5.8GHz

3 x 17dBi 90 deg Sectors - at least 2 metres apart
2 x 27dBi Grids (for backhaul)
3 x RB333 inter-connected on the ethernet (inside the enclosure)
5 x XR5
1 x industrial 10 Amp Battery charger
1 x 109AH Deep Cycle Battery cabled to the enclosure and fused

Our Backhaul - on average is about 40 - 70Km apart - the shortest is 24Km

On average we have clients (RB133c - RB52H 27dBi Grid) between 17 and 50Km.

Average clients per base station ± 60 with the biggest load being 47 clients on one Sector.

It all works in MT bliss.. ie: trouble free and is envied by our opposition..!!!!

I thought I'd just drop this in because we KNOW it works and really is hassle free.

We apply really high standards on fixings and LMR's (only 400's) as well as water-proofing.

Hope this helps.

T

Hi Mperdue

Check the picture i provide here :
This is how we do it on 25 towers. If we have 4 sectors 90 Degree we keep them apart for 1 meter, we have this setup in the top of each tower that we own.

And we use always 1 routerbord with 1 R52 card in it for each antenna.
so we dont have anny problems with inteference at all.

//Rickard

I'd love to see the bottom of that How high is this tower? that looks like 1 inch or 1.5 inch galvanized tubing from a chainlink fence hey it works, but how is that in the wind, does it vibrate/shake?

I'd love to see the bottom of that How high is this tower? that looks like 1 inch or 1.5 inch galvanized tubing from a chainlink fence hey it works, but how is that in the wind, does it vibrate/shake?

You should see my tower heh heh... Looked like A C without any guyid wires last windy strom....

What is the make/model of the antennas in the picture?

The antennas is Steeladoradus 15 db 2,4 ghz 90 Degree...

And that picture is on a silo , but we use the same setup on towers that are 40 meter but the big tube is 2 inch and the smal is 1,5 inch ... and yes i think it moves in the wind, but hey it have ben working now for 3,5 years! without failures....
and it serv 40 customers..

//Rickard

The antennas is Steeladoradus 15 db 2,4 ghz 90 Degree...

And that picture is on a silo , but we use the same setup on towers that are 40 meter but the big tube is 2 inch and the smal is 1,5 inch ... and yes i think it moves in the wind, but hey it have ben working now for 3,5 years! without failures....
and it serv 40 customers..

//Rickard

My kind of setup lol.




Oo way off topic question for yall,, How come one sector can handel say 40 customers, and an omni can only handle 25 to 30? interference??

Sorry had to add this, it is 43 foot tall the tower itself and the silo is 50 feet. Totalling 93 feet tall. I dont know what i was thinking. I built it myself. It just needs guide wires to be finished, it folds down on a hing and folds back up, It cant hold sectors, to heavy Just an omni, maybe 2 one 2.4 and one 900mhz. Anyways tah dah

sheeeeeeeesh!!!!!!!!!

anybody seen this kind of sector antennas? I am may be ignorant. I joined my current organisation after installing these antennas on tower. So I couldnt see its physical appearnce. Now I bringdown one of it and took snaps. This is the sector antennas we have. How I can check whether this antenna is propagating 16dBi of signal effectively? Please provide some information......anybody...........

Looks like a yagi antenna. They are sometimes used when proper sector antennas are not in the budget

Looks like a yagi antenna. They are sometimes used when proper sector antennas are not in the budget

Well, actually only "it looks like".
See the arrow on the top. ( it's says UP )
It's a sector antenna. Actually, it@mikrotik, you can open the top cover without too much trouble, and see inside a kind of an omni antenna with a long sheet of table, in V shape on it, along the vertical axis. That gives the opening for the signal, 90-120 degrees. (please do that and share it with us. You can seal it back with silicon gel, and it's like new.)
It's made in taiwan somewhere i think, and if i'm right, it should be a procell antenna.

I am using myself a few of these, and they seem to have close to the power, but instead of the 120 degrees opening they are supposed to be, i have strong signal in the back also. Sidelobes are terrible on these.

here is the different angle snaps of my sector antenna. How is this look like.......I hope this pics will clear the structure of this antenna. If you need more pics feel free to ask. I expect some help to solve my issue........

Pictures are not very clear, it's not the same as I have, but it's indeed a sector. The second picture shows a printed board, like on some panels.

One thing to avoid at cost are "omnidirectional sector" antennas, eg 3 sectors in an omnidirectional antenna package (one antenna). We made this mistake with the Pacific Wireless sectorized omnidirectional antenna, and performance is terrible. Sector-to-sector isolation is pathetic, even with low power there is significant interference in the other sectors when one is transmitting, even on opposite ends of the frequency. If I max out my downstream, clients on the other side of the antenna on a separate sector on a different frequency almost lose connectivity completely, it's so bad. Costly mistake .

One thing to avoid at cost are "omnidirectional sector" antennas, eg 3 sectors in an omnidirectional antenna package (one antenna). We made this mistake with the Pacific Wireless sectorized omnidirectional antenna, and performance is terrible. Sector-to-sector isolation is pathetic, even with low power there is significant interference in the other sectors when one is transmitting, even on opposite ends of the frequency. If I max out my downstream, clients on the other side of the antenna on a separate sector on a different frequency almost lose connectivity completely, it's so bad. Costly mistake .

Thank you for informing, i always wondered how those worked... -Jordan

Truth be known......

All Antennas are omni directional.
So called directional antennas distort (bend and/or focus) the field to accomplish directionality.
Some designs are more efficient than others, but really, they are all omni directional (hence the front to back ratios).

Hi Mperdue

This is how we do it on 25 towers. If we have 4 sectors 90 Degree we keep them apart for 1 meter, we have this setup in the top of each tower that we own.

And we use always 1 routerbord with 1 R52 card in it for each antenna.
so we dont have anny problems with inteference at all.

//Rickard

Rick, I like your design and am looking at adopting it for a site I need to convert from omni to sector. Could you please tell me:

What material are the pipes?
What method did you use to join the pipes?
What size are those antennas, and what's the 100mph 1/2" ice load? Edit: Datasheet for your antennas.
What wind zone are you in?
How heavy of storms are they able to survive?

We're located in Tornado Alley, so wind loading and survivability are *very* important to me.

Right now I'm thinking about a similar design for hanging a set of four Pacific Wireless SAH4-16 (datasheet).

I'm thinking about using 3.5" galvanized steel for the vertical, then 2.375" galvanized steel for the cross members and antenna verticals. The only change to the design would be to have a top X and bottom X such that the antennas would be supported from the top and bottom, rather than only the middle.

There's both good and bad about this design, that I can think of.

Good:

• It'll take a tornado to knock it down
• Shouldn't have a problem with the somtimes 3" of radial ice accumulation we can get, generally accompanied by 30-40Mph winds.
• Will last forever
• Provides 34" sector seperation

Bad:

• Weighs 3 metric nuts (over 300lb)

Hi nick.

Check this version
Its made of Stainless steel and it light weight and werry strong.it have survive a hurricane

The other version is made of standard parabolic pipes. 50 mm 2.5 mm thick.
and its velded togheter. and it light weight. the antennas are Stella doradus http://www.stelladoradus.com/pdfs/2.4Ba ... 03-06).pdf

I think it will survive your wind load

The reson for mee to build like this is to get so low inteferens as possible.
The nocie value is always -95 to -103 on 2,4 Ghz band.

//Rickard

Thanks to all that has posted in this thread... Just reading through helped me in solving a very crucial design problem.

Hi nick.

Check this version
Its made of Stainless steel and it light weight and werry strong.it have survive a hurricane

The other version is made of standard parabolic pipes. 50 mm 2.5 mm thick.
and its velded togheter. and it light weight. the antennas are Stella doradus http://www.stelladoradus.com/pdfs/2.4Ba ... 03-06).pdf

I think it will survive your wind load

The reson for mee to build like this is to get so low inteferens as possible.
The nocie value is always -95 to -103 on 2,4 Ghz band.

//Rickard

Would it be possible to get all of the dimensions? I would like to build one of these.