SM with great upload by terrible download link

Trying to figure out a problem with one SM on an AP with 15 others users on it (Force 180 connected to 2000 AP).

The downlink RSSI is -67 but the uplink is -58

Downlink SNR is 20, uplink is 33

On a 25 Mbps MIR, they do 9 Mbps down and 26 up.

My first and obvious thought is interference. eDedect comes back clean. There is another SM 7 houses away on the same AP with -57/-57 and 33/33 for RSSI and SNR. I can't find anything locally obviously causing the issue - unless it is not showing up on an eDetect?  It would have to be their house or an immediate neighbor to get that kind of noise. But the other SM right behind them is fine.

Am I missing something in my trouble shooting? Other than perhaps a bad radio?

Path obstruction, and bad Fresnel incursion can cause missmatch rssi.

Check alignment and if that doesn’t help, that radio is likely messed up

1 Like

eDetect will only show you 802.11 interferers running the same center channel and channel width.  You would need to run the spectrum analyzer on the SM to know for certain about interference.  The low reported SNR on the downlink certainly implies interference though Chris' points are quite possible too.

1 Like

I saw those kind of prblem if the're nanostations around a SM. Exemple, if customer have cheap PTP links done with nanostation, on the same pole as the SM, even if the're s huge channel difference using DFS channels on nanostations. We just shut down nanostations and we saw normal througput on SM.

Saw this problems 2 or 3 times. More focused Ubnt radio like nanobridge or powerbridge don't seem to affect that much.

I just check this out again and I think Force180 and Force 200 are the antenna really affected by Ubnt device, or maybe other brand, since the're no sheilding on those radio.

I'm trying to find a  a way to shield Force 200 but dish size is not the same as Ubnt, 400 or 500 mm for Ubnt while Force 200 is 18.5 inch so around 457 mm, so RF armor shield kits are not a option.

Somebody know a way to shield force 200 ? I saw a co-located nanostation with around 300 mhz of channel spacing, using DFS, affect a Force 200 a lot. Else, I will have to try a connectorized radio.


@Mike99 wrote:

I just check this out again and I think Force180 and Force 200 are the antenna really affected by Ubnt device, or maybe other brand, since the're no sheilding on those radio.

I'm trying to find a  a way to shield Force 200 but dish size is not the same as Ubnt, 400 or 500 mm for Ubnt while Force 200 is 18.5 inch so around 457 mm, so RF armor shield kits are not a option.

Somebody know a way to shield force 200 ? I saw a co-located nanostation with around 300 mhz of channel spacing, using DFS, affect a Force 200 a lot. Else, I will have to try a connectorized radio.


It might be easier just to shield the Ubiquiti equipment using RF Armor, and possibly moving the equipment farther away from each other. I've never heard of anyone having to shield ePMP equipment, as it's already, internally, shielded pretty well.

1 Like

For sheilding, It seem to be the case with 1st generation equipement but I doubt Force 180 or 200 have lot of shielding since radios don't weigh much.

As for shielding Ubnt device, that would not be possible in my case but I will try a connecterized radio instead.

How did your endeavor go?
The force 200 is sheilded quite well. But as any rf guy will tell you. Any rf ‘noise’ that is not 30db below your intended signal will cause a lot of issues. The ubiquity radio puts out 1w power on a fairly focused antenna. But if you have a force200 10ft above it then the side lobe of the antenna is filling the rf receiver causing it to become desensitized. A good rule is to keep 100ft of vertical separation. If that is not possible then use anything that can absorb a radio signal and put it between the two radios. Metal screen for a screen door in a metal square frame that is well grounded and bonded to the force200 dish and the ubiquity grounding point will work as an rf sheild when placed between the two radios. Ymmv and craftsmanship will determine how it looks. You will need 2.5 times the largest antenna area for this to even become effective.