Help - I've tried everything I can find in the manuals and the forum and can't get NStreme2 to perform well. I read on the forum that other people can, so I'm confused. Any help is appreciated.

My configuration is as follows:

-Two sites, 4 miles apart. Clear Line of Site and fresnel zone between them.
-Two Ubiquiti XR5 cards on each side, one running 5.3GHz to a 28dBi vertical grid & the other running 5.8GHz to a 29dBi horizontal grid connected via 8 foot of LMR-400 coax, default power
-Started out using routerboard 532A, but am now using Intel PCs
-Alignment has been tweaked for best signal strengths, which are -87 dBm on the 5.8GHz & -79 dBm on the 5.3GHz. Nothing else is using the frequencies I'm using or anywhere near them.
-Wireless rate bounces between 6 & 9Mbps on 5.8GHz and between 9 & 24Mbps on 5.3GHz
-Pings are mostly < 1ms with no traffic, but occasionally jump up to 80 ms
-Btest.exe run from PCs on both ends through wireless link gets max TCP throughput of 2Mbps
-Pings jump to 300ms to 800ms or even fail while btest is running across the link. Wireless rate drops to 6Mbps.
-I've replaced the wireless cards, the PCs, the LMR-400 coax, tried all the frequencies, tweaked & tweaked on alignment & just can't get this any better
-On the "bench" 100 foot apart before putting them up on the tower I was able to use 5GHz Turbo and lock in at 54Mbps and get 57Mbps full duplex (send and receive at the same time)

My wireless config is as follows:

/interface wireless
set 0 ack-timeout=dynamic allow-sharedkey=no antenna-gain=0 antenna-mode=ant-a \
area="" arp=enabled band=5ghz basic-rates-a/g=6Mbps,54Mbps burst-time=1500 \
comment="" compression=yes country=no_country_set default-ap-tx-limit=0 \
default-authentication=yes default-client-tx-limit=0 \
default-forwarding=yes dfs-mode=none disable-running-check=no disabled=no \
disconnect-timeout=3s frame-lifetime=0 frequency=5825 \
frequency-mode=manual-txpower hide-ssid=no hw-retries=15 \
mac-address=00:15:6D:63:1A:24 max-station-count=2007 \
mode=nstreme-dual-slave mtu=1500 name="wlan2_RX_bott_Horz_5.8GHz" \
noise-floor-threshold=default on-fail-retry-time=100ms \
periodic-calibration=enabled periodic-calibration-interval=60 \
preamble-mode=both proprietary-extensions=post-2.9.25 \
radio-name="00156D631A24" rate-set=default scan-list=default \
security-profile=default ssid="HGP2" \
station-bridge-clone-mac=00:00:00:00:00:00 \
supported-rates-a/g=6Mbps,54Mbps tx-power-mode=default \
update-stats-interval=disabled wds-cost-range=50-150 \
wds-default-bridge=none wds-default-cost=100 wds-ignore-ssid=no \
wds-mode=disabled wmm-support=disabled
set 1 ack-timeout=dynamic allow-sharedkey=no antenna-gain=0 antenna-mode=ant-a \
area="" arp=enabled band=5ghz basic-rates-a/g=54Mbps burst-time=1500 \
comment="" compression=yes country=no_country_set default-ap-tx-limit=0 \
default-authentication=yes default-client-tx-limit=0 \
default-forwarding=yes dfs-mode=none disable-running-check=no disabled=no \
disconnect-timeout=3s frame-lifetime=0 frequency=5320 \
frequency-mode=manual-txpower hide-ssid=no hw-retries=15 \
mac-address=00:15:6D:63:43:6B max-station-count=2007 \
mode=nstreme-dual-slave mtu=1500 name="wlan1_TX_top_Vert_5.3GHz" \
noise-floor-threshold=default on-fail-retry-time=100ms \
periodic-calibration=enabled periodic-calibration-interval=60 \
preamble-mode=both proprietary-extensions=post-2.9.25 \
radio-name="00156D63436B" rate-set=default scan-list=default \
security-profile=default ssid="HGP1" \
station-bridge-clone-mac=00:00:00:00:00:00 supported-rates-a/g=54Mbps \
tx-power-mode=default update-stats-interval=disabled wds-cost-range=50-150 \
wds-default-bridge=none wds-default-cost=100 wds-ignore-ssid=no \
wds-mode=disabled wmm-support=disabled
/interface wireless nstreme-dual
add arp=enabled comment="" disable-csma=no disable-running-check=no \
disabled=no framer-limit=4000 framer-policy=exact-size \
ht-channel-width=20mhz ht-guard-interval=long ht-rates=1,2,3,4,5,6,7,8 \
ht-streams=single mtu=1500 name="nstreme1" \
rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps,24Mbps,36Mbps,48Mbps,54Mbps rates-b="" \
remote-mac=00:15:6D:63:1A:14 rx-band=5ghz rx-frequency=5825 \
rx-radio=wlan2_RX_bott_Horz_5.8GHz tx-band=5ghz tx-frequency=5320 \
tx-radio=wlan1_TX_top_Vert_5.3GHz

-Alignment has been tweaked for best signal strengths, which are -87 dBm on the 5.8GHz & -79 dBm on the 5.3GHz.

Those are god-awful signals. With antennae that size, I'd expect to be able to do a 25-mile shot easily.

oscarBravo, thanks for the comment. I expected much better out of this configuration too. Do you have any suggestions as to why I'd have such terrible signals? The antennas are tuned to the frequencies I'm using, I've tried all the frequencies available in those ranges, the cables should have very low loss & I've replaced the cables (had an initial problem of bad connector installation from the VAR). I haven't tried replacing the antennas as I couldn't think of anything likely that would cause 4 new antennas to be bad.

Not sure. Do you definitely have the polarisations correct?

I think that the polarity is correct. The top-most antenna is vertical on both sides, is the Tx on one side and RX on the other and is 5.3GHz. The bottom-most antenna is horizontal on both sides, is Rx on one side and Tx on the other and is 5.8GHz. Is there anything else?

Something really wrong with those signals.

I'd go back to basic's, take 1 card out of each radio and get a single link working first.

Sould be like -60 or something at that distance

If there were interference other than another Mikrotik network would Scan show it? If there were some other system operating in the same frequency range how could I "see" it?

hi dingram. did you check your rf cables?. I had a problem like this and the problem came up to be rf cable. test the rf. I also had a problem with damaged wireless card. I was getting -90 signal with 19 rootnenna 5ghz from 1 km. I change the wireless card the problem get fixed. also if you intsall the antenna next to a radio tower it affect theou can not network cable, so can not have a good bandwith. also if your motherboard pci slot is not enough to feed the wireless card it gives this low bandwith too. I had this issue with asus intel mx version mainboards. when I make a bandwith test the pc was hanging up or I was not able to pass 30 mbit. one I have replaced the mainboard with asrock, I have reached 60mbit for 60 km's. I wish these helps you.

The hardware achieved 54Mbps one way and 57Mbps the other concurrently when on the bench, so I don't think there is a problem with the PCI. I did have bad antenna cables, the connectors came loose during the installation on the towers, but I've replaced the coax cables with good ones. The signal strengths reported in my original e-mail were after I replaced the coax cables.

When you refer to being next to a "radio" tower, what kind of radio? The antennas are mounted on a Rohn style metal tower & we are within 1500' of a cell tower, but I didn't think cellular used 5GHz.

when the network cable is passing behind an hi power fm radio or gsm tower links it causes interference to the ethernet cat5 cable.

We wouldn't pick up interference through a shielded cable from over 1500 feet would we? There's no FM on the nearby tower but I don't know about GSM. Also, would interference picked up on the ethernet side cause poor signal on the wireless?

Is it possible for all the hardware to be perfect but RF interference alone to cause this poor of a signal or must there be a hardware problem for the signal to be this poor? Determining this will at least help me narrow my focus.

Normally Interference will only affect one side

eg -80 at one end - 60 at the other.

The signal strength with that equipment at 4 miles no obstruction should be in the -58 to -68 range. We have several AP to client links in the 5-7 mile range, 18 db omni to 24 db grid, 100 mw radios that run with those signal strengths. That same equipment set up 100' apart (your "bench" test) with no attenuation could damage your radios for sure. Can you do a spectrum analysis at your sites? Switch to 2.4? Try a different channel? Your high ping times are an indicator of bad connections, bad cable, interference, built up static, bad radio to board connections etc etc. Can also be caused by very heavy traffic load, but not running just B test.

I've replaced the TX cards on both sides since my bench testing, but haven't replaced the RX cards. Could RX be damaged by too much TX power on the other side? Can the grid antennas themselves be damaged by too much power from an XR5?

When you say try a different channel, you mean different frequency? I've tried all the frequencies & all the bands.

I've replaced all the lmr-400 coax cable and connections. The housing for the PC is metal, the PC is grounded to the housing, the coax is grounded to the housing, the housing is grounded to the tower, the housing is grounded to a ground rod, the tower is grounded to multiple 10' ground rods, so I don't see how static can be building up.

Radios are plugged into a 4-port miniPCI to PCI card & I'm not having any problems communicating with the radios.

Today i ordered two of the R52H cards and two 2.4GHz grid antennas, so I'll try 2.4 but I'm concerned about interference from the 24 2.4GHz access points I have around our 40 acre campus. They're inside metal buildings, but some RF still bleeds out.

No heavy traffic load during alignment.

The grids wont be damaged but the XR's may be.

Yes indeed the Rx cards can be damaged this way-I've accomplished that feat myself.
As long as you're into grounding, you could install ground kits on the coax. I had an AP running next to an FM transmitter and that was the only thing that got me steady and low pings (50' of LMR 400). I don't think that's your problem however. I would try new single cards both ends and see if your signal strengths come up to where they should be.

Using two brand-new XR5 cards on the 5.8GHz horizontal antennas & the power set to "default" I'm still getting RX & TX Quality in the -90s.

I've purchased two of the new R52H cards to try some of the additional frequency ranges beyond what the XR5 cards support. Is a different license required to use the additional frequencies?

Well, what's the odd that 4 radio cards would go deaf at the same time. Seeing that he had already tried changing the coax, my prime suspects would be the antennas. What kind of antennas are u using ?

the theory that was presented was that I had blown them out by setting the power to 30 with the 29 dB antennas @ 100 ft between them. That is why I replaced all the cards.

just swap antena-mode=ant-a with antena-mode=ant-b and you're done

Ubiquiti XR5s only have one antenna port as far as I know.

Question? Since I do not have experience with an XR5 but do have experience with SR5 radio's, is there a setting for Antenna A and B even though there is only one actual mmcx antenna port? On the SR5 you must use antenna B to access the mmcx port. Antenna A is the Ufl port.

Mikrotik does let you set it. I haven't seen any difference set either way, but this was before I replaced everything. I'll try it again tommorow with the new hardware. Thanks.

what is the lenght of your ethernet cable and what is the dc adpter you are using 18V or 48V?

The Ethernet cable is less than 150 feet. This system is based on Intel powered PCs and have PC power supplies in the PCs. The PCs are mounted on top of the tower & have 8 ft. LMR400 coax cable from the PC cabinet to the antennas.

Setting Mikrotik to Antenna A or B doesn't seem to make any difference on the RX or TX Quality.

Setting the cards to AP bridge & running scan from one side at a time, I'm seeing similar signal strengths of -88 to -92 with a Noise Floor of -91 to -112.

In your description of the link you said :

"-Two Ubiquiti XR5 cards on each side, one running 5.3GHz to a 28dBi vertical grid & the other running 5.8GHz to a 29dBi horizontal grid connected via 8 foot of LMR-400 coax, default power"

You need to make sure that your polarization of the "link" itself is correct. By that I mean that since you said you are using a vertical and horizontal antenna, one will need to be turned to match the other so that the link is working on the same polarization. For example, turn you vertical antenna 90 degrees to make it horizontal, which will match the other end of the link. You should see about a 20db difference when changing polarizations on an antenna.

Hope this helps you understand why you were seeing such low signals, and is not too late to help.

sldnkarm, please clarify as I'm confused. I currently have the antennas on both ends of each link turned the same way. I.e. the 5.3GHz link is on top, vertical on both sides & the 5.8GHz link is on the bottom, horizontal on both ends. Are you saying that the antennas that go together should be rotated 90 degrees from each other-one end vertical & the end other horizontal?

sorry to confuse you. after reviewing the initial post, you have it setup correctly. Can you post the antenna make and models?

I have two each of the Pacific Wireless GD58-29 and GD53-28 grid antennas.

About your suspicion of blowing the antennas - unlikely as they normally are rated anywhere from 50-100W input power rating.

About setting your XR5's to manual TX power - quite possible you've blown these, as there is a tx-power offset table that should be consulted that is on Ubiquity's website.
I think it's set 10dBm lower than you require. e.g. - if you want full power from the XR5 (which is 600mW/27dBm max btw - and not 30dBm) then you set 17dBm as tx-power.

Either way, if you have set them to 30dBm for any appreciable length of time, then you have most likely damaged them by overdriving the amplifiers.

Have you tested your pigtails?

Dingram, why have you screwed up all your rates? Set to default and don't touch those things.

area="" arp=enabled band=5ghz basic-rates-a/g=54Mbps burst-time=1500 \
comment="" compression=yes country=no_country_set default-ap-tx-limit=0 \
default-authentication=yes default-client-tx-limit=0 \

Dingram, why have you screwed up all your rates? Set to default and don't touch those things.

area="" arp=enabled band=5ghz basic-rates-a/g=54Mbps burst-time=1500 \
comment="" compression=yes country=no_country_set default-ap-tx-limit=0 \
default-authentication=yes default-client-tx-limit=0 \

Not so sure if that's such good advice there... and not really a factor of the signal strengths you are getting.

once you have made the connection with a decent signal strength, I would run a bandwidth test between the sites using the bandwidth test tool. Set your rates to 'configured' and drop the 'supported rates' by one degree until you see the bandwidth 'flatlining' - i.e. no variance in the throughput. 'Basic rates' can be set to minimum (1Mbps for B, 6Mbps for a/g). This will stabilise your ping times, and improve the consistency of the link.

Either way, if you have set them to 30dBm for any appreciable length of time, then you have most likely damaged them by overdriving the amplifiers.

Have you tested your pigtails?

I assumed that I had blown the cards, so I bought 2 more but they don't get any better signals (I'm leaving them on default power).

The pigtails and the antennas are the only things I haven't replaced yet. As soon as the weather clears I'm going back up to replace the pigtails & try the new R52H cards I got.

Anybody know if the frequencies that the R52H supports above 5825 are legal in the US? Will they just show up in the list of frequencies in the GUI when I install the cards or do I have to do something special to be able to use them?

I noticed on the spec sheet for those antennas that both the vertical and horizontal beamwidth's fairly tight.

Horizontal Beamwidth for GD5x-29/28 = 4 degrees
Vertical Beamwidth for the GD5x-29/28 = 5 degrees

Have you checked and re-checked the alignment of the links? I would suggest turning up one link at a time. Sometimes we over look the simple things.

If you need a path calculation completed let me know, I would need accurate tower GPS locations and antenna heights.

Please forgive my ignorance, but what is a "path calculation"? I can physically see the opposite tower from either end with binoculars and estimate we have 30+ feet of the tower visible. The highest obstructions are at either end, the middle is fairly open.

We have adjusted vertically and horizontally to get the best signals. At the time we were getting -86 or so but they are worse now for some reason. Even at -86 I couldn't get a link stable at anything more than 18Mbps

A path calculation will give you accurate azimuths and down-tilt or up-tilt information. Also will calculate what the received signal level should be.

With that tight of beam on the antennas, it is very easy to never truly get the alignment on the "main lobe" of the antenna pattern. Can you describe how you were making adjustments to find the "best alignment"?

One end is 32 deg 29' 38.20" N x 94 deg 18' 52.20" W & antennas are 410' above sea level. The other is 32 deg 32' 34.00" N x 94 deg 16' 57.90" W & antennas are 460' above sea level.

Here are the details.

Site A (Located close to Interstate 20)
Azimuth = 28.84 degrees
Ant height = 120ft
Path Inclination = .1309 degrees (uptilt of antenna)

Site B (Located close to Scottsville Rd)
Azimuth 208.85 degrees
Ant height = 106ft
Path Inclination = -.1365 degrees (downtilt of antenna)

Total System Loss = 129.64db
Total System Gain = 77.10 (Includes 20dbm out of radio card, not sure what setting that would be in MikroTik software)
Unfaded Received Signal Level = -52.54dbm

Does this help at all? Can you verify my antenna height calculations?

The I20 location tower is actually 110' and the Scottsville Rd. tower is actually 120' but you're pretty close.

Your calculations confirm I should be getting a great signal. I think I'm going to have to bring someone in with more expertise & experience. I've been working on this too long and need to move on to other projects. I'll try the new pigtails, one link at a time & grounding the N connections where they connect to the cabinet & update with the results, but I don't have high expectations.

thanks for everyone's help.

Do you have the intel PC's mounted that high on the tower, since you only have 8foot of LMR 400?

Would you be able to setup remote access to the radios for someone to review them?

The PCs are mounted at the top of the towers. Yes, I could set up remote access to them, but from what everyone has said I have a physical problem not a software configuration problem. How would remote access to the software configuration help?

Point taken. Just wanted to verify the settings on the other end of the link. Could you post those also? I only saw two interfaces in the initial post from the one side.

I can't remember how i got the output originally, but here's what I could get off the other side:

0 name="wlan1_RX_top_vert_5.3GHz" mtu=1500 mac-address=00:15:6D:63:1A:14
arp=enabled interface-type=Atheros AR5413 mode=nstreme-dual-slave
ssid="HGP1" frequency=5320 band=5ghz scan-list=default
antenna-mode=ant-a wds-mode=dynamic wds-default-bridge=bridge1
wds-ignore-ssid=no default-authentication=yes default-forwarding=yes
default-ap-tx-limit=0 default-client-tx-limit=0 hide-ssid=no
security-profile=default compression=no

1 name="wlan2_TX_bott_Horz_5.8GHz" mtu=1500 mac-address=00:15:6D:63:43:6A
arp=enabled interface-type=Atheros AR5413 mode=nstreme-dual-slave
ssid="HGP2" frequency=5745 band=5ghz scan-list=default
antenna-mode=ant-a wds-mode=disabled wds-default-bridge=none
wds-ignore-ssid=no default-authentication=yes default-forwarding=yes
default-ap-tx-limit=0 default-client-tx-limit=0 hide-ssid=no
security-profile=default compression=no

I've changed lots of things since I made my original post and don't know at this point that it is configured correctly. I tried to set everything back real quick to give you something to look at.

Have you try to set lower TX power ?
Try to use 20dB and use configured datarate set at 6Mbps first (support and basic).
I think using default XR5 tx power for 4 miles link is just make the noise greater.

CMIIW

OK, I set the TX Power to "all rates fixed" 20 dBm & the data rate to 6Mbps but couldn't get any reading in the alignment tool. I kept incrementing each side up, restarting the alignment after each change, until I started getting data through. I had to get to 26dBm before alignment could read the other side and the quality was still -90 or worse.

I hope to put the R52H cards in today, which are only 350mw if I remember right. So, if too much power is the problem maybe they'll be better.

OK, I set the TX Power to "all rates fixed" 20 dBm & the data rate to 6Mbps but couldn't get any reading in the alignment tool. I kept incrementing each side up, restarting the alignment after each change, until I started getting data through. I had to get to 26dBm before alignment could read the other side and the quality was still -90 or worse.

I hope to put the R52H cards in today, which are only 350mw if I remember right. So, if too much power is the problem maybe they'll be better.

I'm not joking when I say there is an offset table to be consulted for XR5's - http://www.ubnt.com/downloads/SRXR_power_offset.pdf

The MAX programmed power setting for an XR5 in MT is 18dBm!! If you persist in setting the values higher than the maximum rated output of the card, you WILL damage them!

When I first set up my 5.8 link, 4 miles, I too was getting really very bad signals. I later found out that the pigtails were my problems after I had changed everything else. Hope this helps.

Thats a good point

Are the piggies rates to 6ghz or just 3ghz like some of the cheaper ones.

Also, not sure if you have lightning arrestors also. If so, make sure they are rated for 0-6ghz. That got me once on a 5.2 link.

Hmm... I didn't know the pigtails were rated for specific frequencies. The reseller site doesn't say what frequency range they support. It says it is RG316 cable, whatever it is rated at. They are from wisp-router & are part # MMCX-NFB-316-1 (http://www.wisp-router.com/itemdesc.asp ... D1&eq=&Tp=).

No lightening protection on the coax. Only a lightening rod above the antennas and all the tower equipment is isolated from the rest of our network.

I use

http://www.titanwirelessonline.com/Prod ... MCX%2DNFBH

Put in the 2 new R52H cards and 2 new pigtails & initially got RX & TX quality of -84 to -85 on both sides at 5180MHz. Stepped through the 5GHz band frequencies, restarting alignment on both sides after applying the change on both sides, and got similar RX & TX quality up until I got to 5300MHz, at which point I stopped getting any reading in the alignment tool on one side and the other side only had an RX quality reading, no TX quality reading. Went back to the bottom of the frequency range & still couldn't get one side to get any reading in the alignment tool. Restarted both sides without any change. No matter what I've tried so far, I can't get the Alignment tool working again. Is this a problem with the alignment tool or is this indicative of my hardware problem? Also, even with these cards & new pigtails I didn't get the quality in the 60s like is should be. My next step, once I get alignment working again, is to try to adjust the antennas again & see if i can improve alignment over what I could before with the XR5 cards.

Thanks!

Those signals are BAD! I suspect that there is some interference that is causing you the problem. Trango and Canopy didnt show in my system. I bought a spectrum analyzer and found a lot of stuff. Here is some of my signals
3' dish---22 miles---SR5....-62db 5foot LMR400 one end... 40ft Hard Line other end...5.8ghz
3' dish---12 miles---CM9....-53db 4foot LMR400... 5.8ghz
2' dish---8 miles----CM9.....-64db 4 foot LMR400... 5.3ghz
Pacwireless Grid dish 27?db---4 miles---CM9....-62db 4foot LMR400... 5.3ghz
If you have replaced everything and you still are having problems, I think it is interference. Changing out to better/high power cards will not solve the problem. One last thought....Try and change antenna polarity Vert vs Horz. See it that makes a difference. Canopy is usally Vertical polarized. Hope this helps.

Montana, my orignal suspision was that my problem was interference from other non-Mikrotik equipment on these frequences but earlier in this post there were various discussions about interference, one stating that interference wouldn't cause bad signals on both ends, only on one end. Another stating that interference wouldn't cause bad signals, only intermittent communication & widely varying throughput. Is your experience that interference will cause bad signals on both ends? If so, where do I find a spectrum analyzer? Also, since I've tried all the frequencies Mikrotik offers, how do I utilize the other frequencies that the cards support?

Thanks.

If you have changed everything out and you still have a problem, it is my opinion that something is causing interference. It is possiable to have one antenna Vert Pol and the other Horz Pol and they will see each other but there will be about 20db drop in signal. At least this is what I have seen on my stuff. Check and try different Pol is the only suggestion that I can offer right now. A spectrum Analyzer would be a great way to check everything out. That way you can rule out the unknown. Check for a rental. I got a Avcom off Ebay and I bought the upconverter. They are not cheap, about 1/3 to 1/2 new but they keep ya from pulling out your hair.

http://85.255.179.2/mt/

"pakiety" means packets (pl)

p3-1ghz and p3-933mhz, 192mb ram, cpu 100% during the bt-test

I tested also mt3.0beta10 with 2x core2duo, 512mb ram about 77-88mbps (tcp/udp with preamble compression)

Montana (or anyone else) - are grid antennas as good as dish antennas for a 4-mile point to point link? I noticed that you are using dish antennas and most of the resellers list dish antennas for "backhaul" links but list grid antennas for "client" links. I bought grid because they were recommended based on lower wind load. Could the style of antennas be the cause of my poor signal quality?

Yes I have used a grid when I didnt have a dish.
Here is one of my links.
Pacwireless Grid dish 27?db---4 miles---CM9....-62db 4foot LMR400... 5.3ghz

Thanks Montana. I'm going to try swapping the polarity, but don't have high hopes that will resolve it. I've also got a pair of 2.4GHz grids to try but we've got 32 2.4GHz interior access points with dual omni antennas & I'm concerned about RF interference with them.

Anyone know how to get Mikrotik ROS to support the other frequencies the card supports (4920MHz to 5180MHz and 5830MHz to 6100MHz) & are they legal to use in the US?

FYI
I just had to pull a RB153 out of service that I just put in a few weeks ago. It was one that was setting on the shelf that I bought awhile back. Anyway when I replaced the board with a RB532A I left the cards and pigtails connected (no strain on the connectors) . When I started it up the signals were in the trash. -80db from about a mile away. I was easy on the pigtails but they had to twist just a little to fit the card in the socket. Just goes to show that these things dont handle well. I hate uFL connectors.

The original XR5 cards used MMCX connectors & the new R52H cards use the uFL, but that doesn't seem to be the problem since I'm still getting bad signal quality. I hope to try rotating the top antennas (I'm only using one card on each side now) to horizontal today to see if that makes any difference.

I rotated the antennas horizontal & got a 10 dB improvement one direction, but I'm unable to get the alignment to show any results the other direction still. One side has an RX Quality but no TX Quality and the other side doesn't show an RX or TX Quality. How frustrating.

dingram,

have you try changing channel?
It might be the interference from other wireless antenna around you.

I've tried all the different frequencies offered. Also, I believe the prevailing wisdom is that interference doesn't cause bad signal quality, only bad connections.

I put up 19dBi panel antennas with 16 degree beam width & I'm getting a much better signal one way, -76dB RX Quality but just as before no TX Quality on that side and no RX or TX quality on the other side. Anyone know what this means? I put in a new coax on the side with no signals just to see if that was it, but no change.

does your antennas support 5.3gHz band that you are using ?

Do you have trees in front but to the side?

I have links when I use grids if a tree is about 100 meters to one side (even tho I have LOS) teh link is bad.

I put panels and it was fine.

I think in my case it was the side lobes reflecting off the tree.

No Trees at the elevation the antennas are at. There are trees beside and below the antennas, but outside the fresnal zone.

The panels I have up now and the grids I had up before do support 5.3GHz & I have tried all the frequencies in that range with no improvement. I believe that the panels are going to solve my problem if I can just get communication both ways. There's got to be something major wrong but I can't figure out what.

Its look like a dead spot of a multipath.
Move your antenna up or down on one side....
Just 2 or 3 feets.