metal 9hpn drops registration for 1-2 minutes when noise goes from ~-113 to -97. Peculiar to note that it’s -97 every time; no more or less.
Occurs as often as every 5 minutes. sometimes not for +10 minutes.
Normal signal is ~-77. chnl width 20mhz.
Bridge.9hpn))) nv2 (((9hpn.sta bridge
Ptp only
And.. why does config list 2GHz band on the 9hpn?
thanks
in which country are you using it?
Maybe you have a strong interference in 900mhz?
Band shows as 2.4ghz because it is down-converting 2.4ghz band to 900mhz. But the frequencies values should show you 900mhz frequencies.
-US
-Could interference result in the noise reading of precisely -97 dBm?
When disconnect occurs, the noise on the Bridge unit goes to -97 dBm (exactly, and every time; no variation)
Tx/Rx is ~-78dBm on both units (i.e., sufficient overhead for some noise)
Also, attached recent log and screenshot showing disconnects.
I tried reducing the power output to -17 per a note on nv2 at the bottom of this page:
http://wiki.mikrotik.com/wiki/Manual:Wireless_FAQ
log.txt (33.2 KB)
Actual noise readings:
20mhz channel ~ -113dBm noise
10mhz channel ~ -116dBm noise
5mhz channel ~ -119dBm noise
On all channel widths noted, noise is observed at exactly -97 with no variation, for 1 to 2 minutes; connection is dropped during this noise event.
-Is this noise from environment? i.e., coming through the antenna? I discount this options due to the exact noise reading during such event.
-Is this noise from with the radio? or a malfunction of the radio? or a response by the radio to some external problem?
-Could utility power problem cause the radio register a noise of -97 exactly?
Power levels were checked at Radio, and down the cable at Antenna using @900mhz:
http://www.praxsym.com/t-meter.html
Fwd, reflected and VSWR were within expected readings.
VSWR 1.2:1
With the thought the radio could be at fault, the radio was replaced with another 9HPn with same symptoms.
please contact support@mikrotik.com with the support output files attached.
Maybe in your area some other equipment is using 900mhz for data-communications. Maybe some utility meters are suing 900mhz to transmit data to the main central? Try to change to a different available frequency.
supout sent.
I didn’t mention before, but the exact -97 noise reading will occur across all measurable channel widths using “Freq Usage…”
Should the noise vary depending on channel width, even with a changed noise floor?
In my experience a -78 is hard to keep stable at 900MHz. You’ll need a -75 to keep it reliable.
It is probably misreading the noise floor.
Just general experience with a handful of 900MHz products. I haven’t gotten the 9hpn yet. I need time and money to play with it. Thunderstorms have made sure I have little of either.