Community discussions

MikroTik App
 
gotsprings
Forum Guru
Forum Guru
Topic Author
Posts: 2335
Joined: Mon May 14, 2012 9:30 pm

60GHZ link doesn't agree.

Wed Nov 13, 2024 12:12 am

PXL_20241031_211249398.jpg
Screenshot 2024-11-12 17.07.22.png
wAP 60G AP is at the Barn.
Cube 60 Pro AC at Cabin.

How come they don't agree on distance or link speed?
You do not have the required permissions to view the files attached to this post.
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 13711
Joined: Thu Mar 03, 2016 10:23 pm

Re: 60GHZ link doesn't agree.

Wed Nov 13, 2024 9:11 am

Distance is measured as round-trip time and difference of 9 m roughly translates to 30 nano seconds. I slightly doubt that timing resolution in WiFi chip is any better than this. RTT is measured by each radio independently ... so there can be slight differences.

Other link properties are dynamic and different for each direction. When transmitter has nothing to transmit, it may switch down to slower (and more robust) modulation and coding schemes (i.e. MCS). So to see what link can do, you have to observe parameters while monitored side is transmitting at maximum possible speed.

Difference can be also due to different hardware type of link partners: antenna gain (when adhering to country limits) doesn't help much for Tx (things are EIRP-limited), it actually helps for reception (so Rx signal strength will be higher on side with higher-gain antenna) and better Rx sensitivity (lower, more negative, number) helps improve SINR if Rx signal strength is not very strong (and thus improves CCQ).

BTW, beware of those tree branches ... link is fine now as trees are leafless, but link may start to struggle (a lot) in Spring after those trees grow leaves.
 
gotsprings
Forum Guru
Forum Guru
Topic Author
Posts: 2335
Joined: Mon May 14, 2012 9:30 pm

Re: 60GHZ link doesn't agree.

Wed Nov 13, 2024 11:25 am

It's actually shooting under / between the branches.
PXL_20240917_145006702.jpg
You do not have the required permissions to view the files attached to this post.
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 13711
Joined: Thu Mar 03, 2016 10:23 pm

Re: 60GHZ link doesn't agree.

Wed Nov 13, 2024 4:32 pm

It's actually shooting under / between the branches.

Good. So the problem won't happen in Spring of 2025, it'll happen in Spring of 2026 :wink:
 
gotsprings
Forum Guru
Forum Guru
Topic Author
Posts: 2335
Joined: Mon May 14, 2012 9:30 pm

Re: 60GHZ link doesn't agree.

Wed Nov 13, 2024 7:30 pm

Good. So the problem won't happen in Spring of 2025, it'll happen in Spring of 2026 :wink:
They offered to cut down trees. I said you may need to trim.
 
gotsprings
Forum Guru
Forum Guru
Topic Author
Posts: 2335
Joined: Mon May 14, 2012 9:30 pm

Re: 60GHZ link doesn't agree.

Wed Mar 19, 2025 2:08 pm

Bringing this one back up.

If I go into the Sender and Hit Align after about 20 seconds the link at the far end shows 2.3Gig again.

Do I need to set up a script to run the align tool at a regular interval?

Or Did I miss something in the setup?
Screenshot 2025-03-19 08.02.20.png
Screenshot 2025-03-19 08.05.17.png
You do not have the required permissions to view the files attached to this post.
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 13711
Joined: Thu Mar 03, 2016 10:23 pm

Re: 60GHZ link doesn't agree.

Wed Mar 19, 2025 10:37 pm

Is it possible that devices got slightly rotated on their mounting supports? E.g. wood swells/shrinks if humidity changes for longer period of time (during winter with low temperatures absolute humidity is low and wood dries). Or strong wind can move device, clamped to a pole .. either due to sheer force or with help of vibrations. So you nay want to check mounts of antennas if they are less than perfect.

IIRC sectors on those 60GHz active antennas are a few degrees large, so antenna doesn't have to be moved a lot to become "badly aligned".
 
gotsprings
Forum Guru
Forum Guru
Topic Author
Posts: 2335
Joined: Mon May 14, 2012 9:30 pm

Re: 60GHZ link doesn't agree.

Sat Mar 22, 2025 12:03 am

I don't think anything has moved.

The Transmitter (wAP 60G AP) always shows the connection BACK as 2.3.

But every time I log in the receiver at the cabin (Cube 60Pro ac), shows 380 until I hit align.
 
User avatar
antonsb
MikroTik Support
MikroTik Support
Posts: 431
Joined: Sun Jul 24, 2016 3:12 pm
Location: Riga, Latvia

Re: 60GHZ link doesn't agree.

Wed Mar 26, 2025 4:29 pm

On Cube PRO devices data is gathered during beamforming events, if there hasn't been beamforming event - it will use less reliable data for calculation, this is less reliable distance assumption that was used for older (wAP,LHG, nRAY series)
 
gotsprings
Forum Guru
Forum Guru
Topic Author
Posts: 2335
Joined: Mon May 14, 2012 9:30 pm

Re: 60GHZ link doesn't agree.

Wed Mar 26, 2025 8:21 pm

On Cube PRO devices data is gathered during beamforming events, if there hasn't been beamforming event - it will use less reliable data for calculation, this is less reliable distance assumption that was used for older (wAP,LHG, nRAY series)
Any thoughts on the phy rate dropping down and then going back up after hitting align?
 
flynno
Member
Member
Posts: 333
Joined: Wed Aug 27, 2014 8:11 pm

Re: 60GHZ link doesn't agree.

Wed Mar 26, 2025 8:50 pm

Why not use SXT AC units?
 
gotsprings
Forum Guru
Forum Guru
Topic Author
Posts: 2335
Joined: Mon May 14, 2012 9:30 pm

Re: 60GHZ link doesn't agree.

Wed Mar 26, 2025 11:49 pm

Why not use SXT AC units?
Because the barn has a wAP 60G AP so it can serve MULTIPLE antennas that are all less than 500 Meters away.

60GHZ doesn't interfere with the Client Serving Access Points at the other buildings.
 
User avatar
antonsb
MikroTik Support
MikroTik Support
Posts: 431
Joined: Sun Jul 24, 2016 3:12 pm
Location: Riga, Latvia

Re: 60GHZ link doesn't agree.

Mon Mar 31, 2025 11:34 am

Any thoughts on the phy rate dropping down and then going back up after hitting align?
Beamforming happens on first minutes, data is updated only on those events for stability reasons.
Align causes device to disconnect and sporadically run beamforming with high update interval - updating actual connection data.