Alignment Process Anomaly(?)

This link has a B5 (master) at one end and a B5c (slave) at the other end with a 2’ dish. Snapshot is taken from the B5c end of the link. I currently don’t have a snapshot of the B5 end of the link during this phase. The firmware is which I know needs to be changed (1.5.4 or 2.8.1) as of 1/22/21. Target RSL is way off and actual RSL is approximately 8dBm more than anticipated. Thoughts?

Probably some sort of issue with either how you have the B5c configured as far as it’s operating antenna. Could just be a bug in the B5/B5c with how they report their antennas to each other…

I would work it out on a link planner to figure out what kind of signal you should be getting and then go from there. Looks like a pretty baller link though, congrats and good luck.

Usually when you see this, it is because you aren’t following EIRP – example: Antenna gain set to 0. If you correct that, the correct target should display, but you lose that beautful -34dBm signal.

Thanks so much Dustin. I will review the configuration to ensure it is set accordingly.

Because the B5 is the master of this link, I don’t have control over the remote setup which is a B5c.

Subsequently, the B5 is set for 46/21 which is not indicative of the B5c side which is 49/21/

Couldn’t take the config that high for the remote as the master B5 radio assumed it was a B5 on the other side of the link apparently.

Both are well within EIRP limitations for the frequencies I’m operating on.

I specifically asked Mimosa if this setup would be a problem and there was a decisive no at that time.


Thanks William. I have designed these paths using PL5.1 as I don’t like Mimosa’s planning tool and the Cambium planning tool although improving is not there yet.

I need and will verify configuration but the B5 being master with a B5c as the slave may expose a bug or limitation in the way the radio operates.

These links were designed to operate at 41dBm RSL as I’m collocated and want to match the RSLs of the other links on that site.


You could swap which radio is the master and just use the B side of the timing (reverse of A) which would allow you to have the B5c as the master and configure things properly… This is probably why I have not seen this issue on the forums before, most people will probably just setup the B5c as the AP… Really interesting bug you found here. Thanks for reporting it!

Since there is a B5 and B5c in play, its better to let the B5 be master since it has an integrated antenna. If you let B5c be master, things might bug out especially on gain settings.

Huh, weird. I guess that makes some sense. So the B5 is programmed to “assume” that there is a B5 on the other end of the link instead of just being a hard programmed B5c that detects a B5 or B5c on the other end…

It appears to be a ‘bug/feature’ in the B5 that it won’t permit one to configure the remote/slave side EIRP/TX output power. This was verified in and 2.8.1.

I did not get a chance to swap ends (master/slave) to see if the alignment (-99.5dBm) target RSL changed but will at some point nor did I get a chance to try 1.5.4 firmware.

The RSL still shows ~-33dBm on the dashboard but the chains show anywhere from -39 to -41dBm which is about where I calculated this link to come in at (still a little hot but not too far off).

Did a simultaneous throughput test in the current state – 938Mbps (MCS on all chains 6-7) – 2x80 with one channel in 5.1 and one in 5.7GHz. CINRs are well above 30dBm at both ends on all chains.

If you can get some clean channels you probably could get MCS 9, which is what I would be aiming for, though I don’t think even in my rural setting I could set aside 160 MHz of spectrum for a single PTP link…

Consequently I am running 20-40 MHz wide channels on my Mimosa PTP links, far easier to max out the MCS and if I end up needing more then a couple 100 Mbps I’ll re-engineer the link.

Thanks William. These are rural environments as well but quite a few wirelessn ISPs in the area. Directional dishes with high RSLs are the only way I will get above the noise floor.

1 Like