Troubleshooting a wireless link

Hello,

looking for my graphs… anyone knows what’s the posible problem of my link?

The link is installed in a electric mast (wood mast with wires that comes from the left to the right..)
In the same mast I have 2 APs and 3 radios. The other 2 radios are working fine.


http://screencast.com/t/6vw8CY49qCn
http://screencast.com/t/6CuoZrlc50

The links are linking to other towers with no issues.

This AP is having reboots each 3 or 4 days.


I never seen a graph like this.. installed in a place without more APs.. it’s in a very remote town… they don’t have mobile phone there!

We will go to change the AP, but I want to know why this malfunction.


thanks!

Hi Martin,
Looks like the link is passing through an high Humidity zone (Water or some rice farm) which create some link deviation after the hottest hours of the day.
Maybe the link has some Fresnel zones obstructed too.
There’s not much you can do beside moving up some meters one of the sides or using dualpol antennas (which not eliminate the issue but mitigates the effect)

It’s a common issue on my place, where links cross seas or lakes.

Hello Renato!

I told this to my customer, but it’s a 5 km link and is clear LOS with no river or water between the 2 points. Only humidity of the near mountains, but not at 16:00h.. it’s strange..

Also you can see in cacti graphs that noise floor graph is the SAME as the signal graph.. I think this is a non sense thing.. with better noise floor we get worse signal..

If you change the frequency when the signal is low, and get a good result, then i’m correct!
If the signal does not improve, then i’m wrong and there’s a pigeon that sits there at 16PM every day.

We will try.


thanks!

Hi Martin

Looks like temperature inversions.

Take a look at the second graph on first 24 hour period Monday to Tuesday.. Here you can see that the cold air which has been trapped beneath a warm front is starting to disperse.

It could be that you also installed the link on a day that there was a tropospheric duct and hence the antennas were never aligned properly in the first place, and that for all the days following, you are being misled by a tropospheric duct causing refraction.

Saludos

Simon

Hi,

It’s possible due to this values in the power supply?

http://screencast.com/t/GL43mYy61ig

It’s only 0,4V… but it’s the same graph that I have in the frequency. The strange thing it’s that the other link in the same AP is stable.


http://screencast.com/t/kzCili91F54




O_O

Hello,

It’s nothing to do with the power supply. The variations during the day are normal, hotter is the temperature bigger is the voltage. I have sun power in some places and the variation is bigger than 1,5V with no variation in the wireless link.

The fadding is function of frequency, humidity and warm. In my opinion it is not the effect that Renato descibes because is too cold now (april). In the other hand this effect, in Barcelona, remains minutes with a fadding up to 30dB, no hours.

I will bet for the pigeon or a witch (due it is in the North of Spain).

This morning we changed the POE and power supply and now it seams fixed in a small piece of graph. Let’s see the graph tomorrow..

http://screencast.com/t/tgyGH8EGIS
http://screencast.com/t/NmNjwZrFbd5