LHG 60G experience

solelunauno · Thu Feb 21, 2019 5:29 pm

I installed a system composed by one wAP 60G and 6 LHG 60G @ distance of 300-600m from the AP.
I noticed an instability on the wAP: if one station is moved or the signal drops, the wAP disconnects all other stations and it locks untill someone disables and enables the w60G interface.
Does anyone experienced the same behavior?

djvolt · Thu Feb 21, 2019 5:40 pm

Check Client Isolation

server8 · Thu Feb 21, 2019 5:51 pm

All latest firmwares 'd solve the disconnection issue

solelunauno · Thu Feb 21, 2019 6:33 pm

All latest firmwares 'd solve the disconnection issue

is it possible to update Ap and one station a time, via wireless link?

djvolt · Fri Feb 22, 2019 1:43 am

First do upgrade on Station, second on AP and will be ok

anav · Sat Feb 23, 2019 1:28 am

Has anyone compared these two mounts both by Mikrotik?
https://mikrotik.com/product/solidmount
and
https://i.mt.lv/cdn/rb_files/quickMount ... 113333.pdf
The bottom model for LHG.

blue · Sat Feb 23, 2019 9:51 am

Has anyone compared these two mounts both by Mikrotik?
https://mikrotik.com/product/solidmount
and
https://i.mt.lv/cdn/rb_files/quickMount ... 113333.pdf
The bottom model for LHG.

Solid mount is way way way better for 60G! I had a problem with quickmount when winter came. It was fastened during the summer, and in the winter couple of LHG's was pointing to the ground! Plastic shrink in winter, and screws got loose. Since I have solidmounts for 60G, I used it's mounts (now called and sold as LHG mount https://mikrotik.com/product/lhg_mount) to fix those loose LHG.

BR...

djvolt · Sat Feb 23, 2019 10:07 am

This is slightly better, because is more stable, two U-bolts and bigger flat surface... http://24.multimediahd.pl/en/60ghz/110- ... r-alu.html

blue · Sat Feb 23, 2019 3:07 pm

This is slightly better, because is more stable, two U-bolts and bigger flat surface... http://24.multimediahd.pl/en/60ghz/110- ... r-alu.html

SolidMount is great product and way cheaper then Your product. Also, me and other people are waiting info from Your customers about performance of You Alu dishes for LHG60. For now we only have unchecked info from You, and very bad info's from guys in forum...

anav · Sat Feb 23, 2019 5:06 pm

Much thanks Blue!
For the solid mount, or any of them, is there a graduated controlled click during turning in horizontal or vertical directions?
(or does one loosen bolts and move free form)?

Of course I am still also trying to get my head around how you get something one mile let alone 5 miles apart aimed close enough in the first place to think about refining?
For example, is there target material you can put at an expected location (for pointing a laser at for instance).
Then attach laser somehow (parallel to beam of antenna) and aim it at target sight??

blue · Sat Feb 23, 2019 5:18 pm

There are no click. Use wrench 10mm and finetune it. Than fasten all fixing bolts and that's it. I never used laser for pointing. Years of experience and google earth helped me to point antenna in desired direction. But since LHG60 is only 1 degree You absolutely have to use solidmount on greater distance...

blue · Sat Feb 23, 2019 6:07 pm

60GHz - 4330 meters - my experience

First of all, if You don't want to read my whole post, conclusion is don't do it - maximum reliable is 3000 meters.

Finally we had some sunny and warm days in Serbia. So I reserved whole weekend for 60G links. First of all, You need a lot of patience. When this long link got disconnected for more than couple of minutes it takes anywhere from 30min to more than 2 hours to get back online! I don't know why? Reboot, unplug/plug din't give any results.

We combined LHG60 with SolidMount, and that is a MUST on any 60G link that is 1000m+.

When You adjust the link use bandwith test between two devices. It is powerfull enough to let UDP trafic spread it's wings. TCP got stuck arround 500Mbit with 100%CPU.

Don't look at the "rssi". You only need MCS (max is

, SIGNAL (100 is ideal so I assume that is in %), and try to get TX-SECTOR to center. Also Error rate is crucial!

Don't use "align" since it disconnect link more often. Use only "monitor" tool.

One person should be at the antenna and another should look at the notebook. First get 60G link to any working order with eye sight. I know that it could be difficult for someone, but LHG60 have mesh grid and it is see-through. Then You need wrench 10mm to finetune solidmount. It takes half of the turn to lose signal. So it should be fastened to maximum allowed to turn the wrench.

And then start to play. Don't be afraid to try different things and positions. Most annoying is when You got disconnected for more than a minute! Than it takes forever to get connected again. It is not a matter of misalignment at all since once we got disconnected for 2 hours and after that link got connected and we had MCS 7 and Signal 80. Bandwith test was 900Mbit.

Next day it got very windy, and link got worse. I was thinking what is somehow moved, but after wind has stopped link got OK again. Also rain is killing this link, but 2950 meter link is holding OK in wind/rain.

Now winter came back in Serbia for a few days, and on first warm day I will take down this 4,3km link since it is useless on that distance.

BR to all and good luck (You will need it - at least until Ubnt start to sell 60GHz)

aisme · Tue Feb 26, 2019 4:24 am

Hi,

Earlier today a network loop was introduced in our internal network. While investigating the source of the loop we disconnected the 4 links with other buildings that we have using LHG60s and a link with wAP60. While rebooting one of the LHG60 links the devices never really came back.

I thought it had to do with the network loop and did not pay attention. When the loop was discovered and elliminated (someone connected two adjacent network sockets by mistake) all links and devices came immediately back and I could access their interface except from a specific link.

When I try to ping their IP I get a couple of pings and then maybe 10-15 requests timed out and the same things repeats. At the beginning I thought that maybe one of the antennas has died but the strange thing is that BOTH master and slave behave the same way.

Even when pinging the devices from their respective side the same thing happens which leads me to suspect that something else is the underlying cause.

When checking the LED lights on the devices the pattern is equally strange. Upon rebooting the power LED switches on and then the ethernet LED switches on and blinks as expected however as soon as the wireless link is apparently established and the wifi LED is on then the ethernet LED starts blinking really slow instead of the typical fast rate blinking when traffic is going through.

Any suggestions? Or should I RMA them and replace them? I bought these like 6 weeks ago.

rkau045 · Tue Feb 26, 2019 6:58 am

Cycling timeouts like this are usually caused by STP blocking and reenabling forwarding on a port in my experience.

Sent from my Pixel 3 XL using Tapatalk

djvolt · Tue Feb 26, 2019 11:14 pm

This is slightly better, because is more stable, two U-bolts and bigger flat surface... http://24.multimediahd.pl/en/60ghz/110- ... r-alu.html
SolidMount is great product and way cheaper then Your product. Also, me and other people are waiting info from Your customers about performance of You Alu dishes for LHG60. For now we only have unchecked info from You, and very bad info's from guys in forum...

Buy and try itself like other products

Confirmation from other:
https://ispforum.cz/viewtopic.php?p=250027#p250027

pikpik · Wed Feb 27, 2019 12:41 pm

Guys, which devices are best for PtMP ?

2mailwelkom · Thu Feb 28, 2019 8:50 pm

Awesome product took me a lot off time to get this stable

djvolt · Thu Feb 28, 2019 9:23 pm

wAP60G + LHG60 up to 500 meters

server8 · Sun Mar 03, 2019 6:37 pm

@mikrotik Is it finally possible to know if the chipset allows narrower channel widths (like 1000 MHz or 500 MHz) and if the next future we 'll see this options?

fox76 · Wed Mar 13, 2019 3:30 pm

Does anyone use ALU 600?
Please
Print Screen

http://24.multimediahd.pl/pl/upgrade-s/ ... -kit-.html

blue · Wed Mar 13, 2019 4:17 pm

Does anyone use ALU 600?
Please
Print Screen

http://24.multimediahd.pl/pl/upgrade-s/ ... -kit-.html

autostoper76 is using those and he is not happy

Untill now, we don;t have any proof that those devices are working OK. Just words - no pictures/viideos

fox76 · Thu Mar 14, 2019 5:49 pm

Does anyone use ALU 600?
Please
Print Screen

http://24.multimediahd.pl/pl/upgrade-s/ ... -kit-.html

Does anyone use ALU 400?

nimbo78 · Fri Mar 15, 2019 6:15 pm

2MikroTik
guys, plz add debug logging for w60 wireless like "wireless" for 802.11ac\n\g\b\a wifi

anav · Sun Mar 17, 2019 4:26 pm

Hi,

Earlier today a network loop was introduced in our internal network. While investigating the source of the loop we disconnected the 4 links with other buildings that we have using LHG60s and a link with wAP60. While rebooting one of the LHG60 links the devices never really came back.

I thought it had to do with the network loop and did not pay attention. When the loop was discovered and elliminated (someone connected two adjacent network sockets by mistake) all links and devices came immediately back and I could access their interface except from a specific link.

When I try to ping their IP I get a couple of pings and then maybe 10-15 requests timed out and the same things repeats. At the beginning I thought that maybe one of the antennas has died but the strange thing is that BOTH master and slave behave the same way.

Even when pinging the devices from their respective side the same thing happens which leads me to suspect that something else is the underlying cause.

When checking the LED lights on the devices the pattern is equally strange. Upon rebooting the power LED switches on and then the ethernet LED switches on and blinks as expected however as soon as the wireless link is apparently established and the wifi LED is on then the ethernet LED starts blinking really slow instead of the typical fast rate blinking when traffic is going through.

Any suggestions? Or should I RMA them and replace them? I bought these like 6 weeks ago.

........
While you are at, suggest you use trained staff or create a checklist for plugging stuff in.......... or was this sabotage??

server8 · Sun Mar 17, 2019 6:41 pm

Bad news for mikrotik

60ghz.png

mistry7 · Sun Mar 17, 2019 6:54 pm

Bad news for mikrotik
60ghz.png

Yes and they do it right from the beginning, integrated 5 GHz Backup....
And Mikrotik engeneers are sleeping....

WebItG · Sun Mar 17, 2019 7:04 pm

Bad news for mikrotik
60ghz.png

please original link ???

mistry7 · Sun Mar 17, 2019 7:12 pm

Bad news for mikrotik
60ghz.png
please original link ???

This is early access shop, you need to be registered and allowed to see this

WebItG · Sun Mar 17, 2019 7:45 pm

Bad news for mikrotik
60ghz.png
please original link ???
This is early access shop, you need to be registered and allowed to see this

i am registered
but send URL

server8 · Sun Mar 17, 2019 7:46 pm

you have to be registered to early access https://store.ui.com/collections/early-access

Stril · Sun Mar 17, 2019 7:49 pm

Prior to thinking, these products will be better than the LHGs, just read the UB-forums and search for FCS-errors and corrupted packets...

Just added an LHG to take over the traffic of an AirFiber...

SpanMG · Mon Mar 18, 2019 8:59 am

Friends, help calculate the track and its reliability.
Going to do PTMP.
As the base I want to use Wap
The first client at a distance of 160 m. LHG
The second and third on the same building at a distance of 490 m. LHG

How much such a link will be reliable, whether there will be frequent problems with it. What signal do I expect from distant links?

aisme · Mon Mar 18, 2019 9:12 am

I am reporting some weird findings. The device is connected on an HP A5500 48port PoE+ switch worth 7,000+ euros. There are 3 other LHG60 on the same switch all working properly/normally. There are 35 VoIP phones on the same switch all working properly.

Even when switching off the PoE power from the switch port the antenna stays on! Or at least its LEDs are on.

It will only go off when disconnecting the ground cable!

The weird findings continue. When disconnecting the ground cable the problematic antenna comes back to life and the link works. When connecting back the ground cable it goes crazy again.

The same ground is used for all the other 4 LHG60s on that roof and the rest behave/work normally.

Should I use the ground or not? Is it better to use the Ubiquiti surge arrestors instead of the ground? Should I use both?

I had people check the earthing itself and it is fine.

I could not find documentation on the matter

Hi,

Earlier today a network loop was introduced in our internal network. While investigating the source of the loop we disconnected the 4 links with other buildings that we have using LHG60s and a link with wAP60. While rebooting one of the LHG60 links the devices never really came back.

I thought it had to do with the network loop and did not pay attention. When the loop was discovered and elliminated (someone connected two adjacent network sockets by mistake) all links and devices came immediately back and I could access their interface except from a specific link.

When I try to ping their IP I get a couple of pings and then maybe 10-15 requests timed out and the same things repeats. At the beginning I thought that maybe one of the antennas has died but the strange thing is that BOTH master and slave behave the same way.

Even when pinging the devices from their respective side the same thing happens which leads me to suspect that something else is the underlying cause.

When checking the LED lights on the devices the pattern is equally strange. Upon rebooting the power LED switches on and then the ethernet LED switches on and blinks as expected however as soon as the wireless link is apparently established and the wifi LED is on then the ethernet LED starts blinking really slow instead of the typical fast rate blinking when traffic is going through.

Any suggestions? Or should I RMA them and replace them? I bought these like 6 weeks ago.

mistry7 · Mon Mar 18, 2019 9:34 am

I am reporting some weird findings. The device is connected on an HP A5500 48port PoE+ switch worth 7,000+ euros. There are 3 other LHG60 on the same switch all working properly/normally. There are 35 VoIP phones on the same switch all working properly.

Even when switching off the PoE power from the switch port the antenna stays on! Or at least its LEDs are on.

It will only go off when disconnecting the ground cable!

The weird findings continue. When disconnecting the ground cable the problematic antenna comes back to life and the link works. When connecting back the ground cable it goes crazy again.

The same ground is used for all the other 4 LHG60s on that roof and the rest behave/work normally.

Should I use the ground or not? Is it better to use the Ubiquiti surge arrestors instead of the ground? Should I use both?

I had people check the earthing itself and it is fine.

I could not find documentation on the matter

Hi,

Earlier today a network loop was introduced in our internal network. While investigating the source of the loop we disconnected the 4 links with other buildings that we have using LHG60s and a link with wAP60. While rebooting one of the LHG60 links the devices never really came back.

I thought it had to do with the network loop and did not pay attention. When the loop was discovered and elliminated (someone connected two adjacent network sockets by mistake) all links and devices came immediately back and I could access their interface except from a specific link.

When I try to ping their IP I get a couple of pings and then maybe 10-15 requests timed out and the same things repeats. At the beginning I thought that maybe one of the antennas has died but the strange thing is that BOTH master and slave behave the same way.

Even when pinging the devices from their respective side the same thing happens which leads me to suspect that something else is the underlying cause.

When checking the LED lights on the devices the pattern is equally strange. Upon rebooting the power LED switches on and then the ethernet LED switches on and blinks as expected however as soon as the wireless link is apparently established and the wifi LED is on then the ethernet LED starts blinking really slow instead of the typical fast rate blinking when traffic is going through.

Any suggestions? Or should I RMA them and replace them? I bought these like 6 weeks ago.

This Device only needs 2-3w if this is RF Tower it could be induction to your cables, or different earth potential

server8 · Mon Mar 18, 2019 11:50 am

Mikrotik is wasting 18 months of advantage on 60 GHZ, is it possible to know if there is any hardware with 5 ghz backup in the close future?

mistry7 · Mon Mar 18, 2019 12:17 pm

Mikrotik is wasting 18 months of advantage on 60 GHZ, is it possible to know if there is any hardware with 5 ghz backup in the close future?

Why?
UBNT will deliver such product, and there would be an AP as well, so why ask them here thinks that they know from the beginning of there
60 GHz Productlinie.
Again, Mikrotik Wireless is dead end, they are sleeping!

server8 · Mon Mar 18, 2019 12:27 pm

Because is stupid don't put a 5 ghz feed in an hardware that has the pcb ready, I can't uderstand why this leak of 5 ghz backup.

Mon Mar 18, 2019 12:30 pm

Mikrotik is wasting 18 months of advantage on 60 GHZ, is it possible to know if there is any hardware with 5 ghz backup in the close future?
Why?
UBNT will deliver such product, and there would be an AP as well, so why ask them here thinks that they now from the beginning of there
60 GHz Productlinie.
Again, Mikrotik Wireless is dead end, they are sleeping!

LHG60 is great and available (for a while now). It runs ROS which is way better than AirOS. So wrong thread to blame them sleeping. This new gear looks very small. So I guess gain is between WAP 60G and LHG60G. If you need 5GHz Backup use SXT or combine LHG60G with af5xhd using a powerbox pro.

server8 · Mon Mar 18, 2019 12:52 pm

LHG60 is great and available (for a while now). It runs ROS which is way better than AirOS. So wrong thread to blame them sleeping. This new gear looks very small. So I guess gain is between WAP 60G and LHG60G. If you need 5GHz Backup use SXT or combine LHG60G with af5xhd using a powerbox pro.

Client side is 3 times hardware and a lot of configuration.... for the backhauling is the double and the colocation on the tower is not for free

Mon Mar 18, 2019 12:54 pm

LHG60 is great and available (for a while now). It runs ROS which is way better than AirOS. So wrong thread to blame them sleeping. This new gear looks very small. So I guess gain is between WAP 60G and LHG60G. If you need 5GHz Backup use SXT or combine LHG60G with af5xhd using a powerbox pro.
Client side is 3 times hardware and a lot of configuration.... for the backhauling is the double and the colocation on the tower is not for free

Doing backhaul + backup with one Antenna in one direction is not a good idea anyway.

server8 · Mon Mar 18, 2019 3:28 pm

Doing backhaul + backup with one Antenna in one direction is not a good idea anyway.

Mostly it's rain backup... OK I 'll wait for ubiquiti

Mon Mar 18, 2019 3:37 pm

Doing backhaul + backup with one Antenna in one direction is not a good idea anyway.
Mostly it's rain backup... OK I 'll wait for ubiquiti

May take a while ...

server8 · Mon Mar 18, 2019 4:33 pm

better late than never

r00t · Mon Mar 18, 2019 5:02 pm

On 60GHz WAP radio PCB, there is place for 5GHz radio chips, but it was never used. Mikrotik appears to be just sleeping, while competition is creeping closer and closer and may beat them at their own 60GHz game...
Hopefully this will be a wakeup call and we will see some new 60GHz gear soon. They must have known about Ubiquity releasing their 60GHz solution for some time.

Next months will tell us more.
- if we see new 60GHz radios with real new features and improvements, there is still hope MIkrotik R&D have been actually doing something for past months. Question is how long it will take to get new devices certified and released to the market
- if no new devices are announced or they just release minor updated version (let's make a dish metal and call it a day...) it's probably not very good sign. At that point all that Mikrotik can do is they can start lowering the prices of existing 60GHz gear when Ubiquity releases their 60GHz hardware, but that would mean they have nothing new and all they can do is to lower prices to stay in the game... not good.

Mon Mar 18, 2019 5:12 pm

On 60GHz WAP radio PCB, there is place for 5GHz radio chips, but it was never used. Mikrotik appears to be just sleeping, while competition is creeping closer and closer and may beat them at their own 60GHz game...
Hopefully this will be a wakeup call and we will see some new 60GHz gear soon. They must have known about Ubiquity releasing their 60GHz solution for some time.

Next months will tell us more.
- if we see new 60GHz radios with real new features and improvements, there is still hope MIkrotik R&D have been actually doing something for past months. Question is how long it will take to get new devices certified and released to the market
- if no new devices are announced or they just release minor updated version (let's make a dish metal and call it a day...) it's probably not very good sign. At that point all that Mikrotik can do is they can start lowering the prices of existing 60GHz gear when Ubiquity releases their 60GHz hardware, but that would mean they have nothing new and all they can do is to lower prices to stay in the game... not good.

There is more stuff on the road. MT announced to do Terragraph. This is a 60GHz Mesh solution and I guess it will be 802.11ay.
https://terragraph.com.

r00t · Mon Mar 18, 2019 6:37 pm

While Terragraph sounds interesting, it certainly can't replace P2P links (whitepaper says range is about 250m, using small antennas).

Mon Mar 18, 2019 8:56 pm

While Terragraph sounds interesting, it certainly can't replace P2P links (whitepaper says range is about 250m, using small antennas).

The range will depend on the used antennas. PTP links will be part of the mesh. I hope different Vendors will be compatible as Radwin and Cambium has announced terragraph, too.

nobbie · Mon Mar 18, 2019 10:10 pm

Hej guys, I have 750m PTP link to do and three small 230V wires by the road between these two points. Is it ok to use LHG60 kit ? I don't need 1Gbps throughput, 300Mbps will be ok. What about rain and snow, I want to sleep well

djvolt · Mon Mar 18, 2019 10:19 pm

Does anyone use ALU 600?
Please
Print Screen

http://24.multimediahd.pl/pl/upgrade-s/ ... -kit-.html
autostoper76 is using those and he is not happy Untill now, we don;t have any proof that those devices are working OK. Just words - no pictures/viideos

NO NO NO, autostoper have 400mm, NOT 600mm and please DO NOT put people in the wrong place ok?....

aisme · Mon Mar 18, 2019 10:41 pm

This Device only needs 2-3w if this is RF Tower it could be induction to your cables, or different earth potential

I understand however this is just the roof of the building without any RF tower on it. All the cables going up to the roof are 35-40m Ubiquiti Touch cable. 4 LHG60s doing PTP links. All connected to the same ground. All working fine except this one

mistry7 · Mon Mar 18, 2019 11:09 pm

Hej guys, I have 750m PTP link to do and three small 230V wires by the road between these two points. Is it ok to use LHG60 kit ? I don't need 1Gbps throughput, 300Mbps will be ok. What about rain and snow, I want to sleep well

You need to point very well, and the money for solid mount is a good investment,
And then such Link will work in most conditions
We had heavy snow and the 1,2km dropped for 10min, the 550m did not do anything

Mon Mar 18, 2019 11:26 pm

Hej guys, I have 750m PTP link to do and three small 230V wires by the road between these two points. Is it ok to use LHG60 kit ? I don't need 1Gbps throughput, 300Mbps will be ok. What about rain and snow, I want to sleep well

Snow is a problem. Expect to get problems in winter. Do a 5GHz Backup.

Stril · Thu Mar 21, 2019 2:15 pm

Hi!

Can you give me a hint on how to transfer ALL vlans through a LHG-Link? Do I have to configure the VLANs, or is it enough to leave "VLAN Filtering" disabled on the bridge?

Thank you for your help!

Thu Mar 21, 2019 2:30 pm

Hi!

Can you give me a hint on how to transfer ALL vlans through a LHG-Link? Do I have to configure the VLANs, or is it enough to leave "VLAN Filtering" disabled on the bridge?

Thank you for your help!

Yes. Do a plain bridge and do no tagging then all packets go through.

Stril · Fri Mar 22, 2019 9:22 am

Can you give me a hint on how to transfer ALL vlans through a LHG-Link? Do I have to configure the VLANs, or is it enough to leave "VLAN Filtering" disabled on the bridge?
Yes. Do a plain bridge and do no tagging then all packets go through.

Thank you!
This is working perfectly.

aisme · Thu Mar 28, 2019 12:14 am

Hi

Should I use the ground on the LHG60 or not? Is it better to use the Ubiquiti surge arrestors instead of the ground? Should I use both?

Does anybody have any documentation on it? I cannot really find any proper documentation for the LHG60

Thank you

Lakis · Mon Apr 08, 2019 1:29 pm

Hello little info and I need some advice.
We have 3km 60G Link already deployed almost 3 mounts ago
Ap Side
----------
connected: yes
frequency: 66000
remote-address:
tx-mode: dmg
tx-mcs: 8
tx-phy-rate: 2.3Gbps
signal: 80
rssi: -60
tx-sector: 6
tx-sector-info: right 1 degrees, down 1.4 degrees
distance: 3072.53m
tx-packet-error-rate: 0%

Client Side
_________
connected: yes
frequency: 66000
remote-address:
tx-mode: dmg
tx-mcs: 8
tx-phy-rate: 2.3Gbps
signal: 80
rssi: -60
tx-sector: 28
tx-sector-info: center
distance: 3072.53m
tx-packet-error-rate: 0%

Overall, we are satisfied, and yes on a moderate rain link drops occasionally

08:31:46 interface,info wlan1: link down
08:33:43 interface,info wlan1: link up
08:33:43 interface,info wlan1: link down
08:37:23 interface,info wlan1: link up
08:37:35 interface,info wlan1: link down
08:37:40 interface,info wlan1: link up
08:37:58 interface,info wlan1: link down
08:38:03 interface,info wlan1: link up
08:40:16 interface,info wlan1: link down
08:40:19 interface,info wlan1: link up
This is from today's rain, link was down periodically
After 8:40 Rain was light and link was stable.
Yearly we have very little snow, and heavy or tropical rain are common here but they last very short 5-10min.
Overall that is what we have expected, and the link is 99.5% of the time usable.

I saw some posts of dudes like Blue (love enthusiasm) have made links up to 4km and record 5km lol that was excellent.
What I need is some advice, we want to deploy 3.45km 60G Link (of course with 5Ghz backup) what should I expect? Is it going to be usable nearly as my 3.1km link, or it is going to be total waste of time?
Thanks.

kalkunte · Tue Apr 09, 2019 6:40 pm

Can Microtik interwork with other 11ad hardware?

How to setup Microtik 60G as STA that connects to another vendors who acts as AP?

So far we've been able to make Microtik60G to another Microtik 60G to work. Is this by design/intent?

Thanks

r00t · Wed Apr 10, 2019 10:36 pm

All Mikrotik 60G hardware is so far compatible, so you can mix LHGs and WAPs as you like. But don't expect it to work with other vendors, even if it uses common 802.11ad platform. Everyone is playing on it's own playground, there is no intercompatibility due to customized protocols and different vendors design devices to lock you in on their platform to buy their devices... nothing new.

komenski · Thu Apr 11, 2019 7:09 am

All Mikrotik 60G hardware is so far compatible, so you can mix LHGs and WAPs as you like. But don't expect it to work with other vendors, even if it uses common 802.11ad platform. Everyone is playing on it's own playground, there is no intercompatibility due to customized protocols and different vendors design devices to lock you in on their platform to buy their devices... nothing new.

I'm not quite sure about the full compatibility between LHG and WAP60g. WAP60g seems to have some internal distance limit between clients, which Mikrotik is gradually and secretly extending, without any informations for the users who purchased the device. All we have to do is to firmly believe that Mikrotik (before the competition comes up with something better) quickly releases the FW, which again allows (at least for the initial setup of the link) fixed settings not only for TX-sector, but also for TX power, MCS, ACK, etc .

asterisco · Thu Apr 11, 2019 2:17 pm

Hi

I'm starting to use WAP60 to connect customers in PtMP setup.

Is there any way to connect the wireless link to radius? this mean a LHG60 STA only connects to AP if radius autenticates.

Thanks!
Antonio

anav · Thu Apr 11, 2019 7:04 pm

All Mikrotik 60G hardware is so far compatible, so you can mix LHGs and WAPs as you like. But don't expect it to work with other vendors, even if it uses common 802.11ad platform. Everyone is playing on it's own playground, there is no intercompatibility due to customized protocols and different vendors design devices to lock you in on their platform to buy their devices... nothing new.

The smart vendor includes compatibility with other vendors equipment when possible, kinda like how Russian rifles were designed to use Russian bullets but also NATO bullets.
So smart!!

doneware · Thu Apr 11, 2019 8:15 pm

All Mikrotik 60G hardware is so far compatible, so you can mix LHGs and WAPs as you like. But don't expect it to work with other vendors, even if it uses common 802.11ad platform. Everyone is playing on it's own playground, there is no intercompatibility due to customized protocols and different vendors design devices to lock you in on their platform to buy their devices... nothing new.
The smart vendor includes compatibility with other vendors equipment when possible, kinda like how Russian rifles were designed to use Russian bullets but also NATO bullets.
So smart!!

guys please. Vanilla 802.11ad is just like vanilla 802.11b or n.
there are certain things in the standard. also, the standard is from 2012, and it was never ever intended to be used outdoors at a distance. because of oxygen absorption. yes, there are vanilla implementations out there, more or less for indoors and not using phased array antennas.
the other thing is MAC. if you want to use vanilla .11n or anything 802.11 in a multi-access scenario, it sucks. because vanilla 802.11 MAC sucks. that's why vendors implemented their own TDMA based stuff, to somewhat control the inefficient behaviour of the WiFi clients. Or synchronised APs.
this is no different here. if you want to use a purpose-built device in a certain environment what the device was optimised for, then you can't expect something very generic to be compatible with it w/o serious compromises to its original goals. Mikrotik did not intend to build a 'generic' .11ad AP. this thing is for outdoors, it uses technology that is 4-5 years younger than the original WiGig standard.
back at that time receiver side beam forming wasn't widespread, and the original standard did not rely on that. as the result the maximum distance it was able to operate with was literally meters or maybe 2-30 meters. you have here very precise beamforming down to 3-4 degree or so, both on transmitter and receiver side, just to concentrate all available RF energy into this tiny beam to bridge the massive free path loss in the air. so you can reach over 100s of meters if required. or coordination between the clients. none of those were part of the original standard, so each vendor, each silicon manufacturer had to figure this out for himself.
what exactly do you expect? to operate a 60GHz based (most probably proprietary) video headset with it?
i've seen the prices of other vendors, even the ones who use the same BB and RF as Mikrotik. we're doing great here.

on the other hand, the next standard - 802.11ay - already includes specific things that are required for proper outdoor operations, like synchronisation, TDD/TDMA, transmit power control, god knows what. there is a lot more chance to see actually compatible equipment - if not more than compatible at PHY and MAC layers - when vendors start adopting silicons from certified .11ay chip manufacturers.

SpanMG · Mon Apr 15, 2019 8:33 am

Colleagues, tell me, there is a line at 1,500 m., We were reducing antennas, but I noticed that after 30 minutes, the indicators just hang in one place and do not move, it only helps to reboot the devices or to tear the radio link.
[admin@client] > int w6 moni wl
connected: yes
frequency: 64800
remote-address: 24:18:**********
tx-mode: dmg
tx-mcs: 8
tx-phy-rate: 2.3Gbps
signal: 80
rssi: -68
tx-sector: 60
tx-sector-info: right 0.2 degrees, up 1.4 degrees
distance: 1593.31m
tx-packet-error-rate: 1%
-- [Q quit|D dump|C-z pause]

[admin@AP] > int w60g monitor wlan60-1
connected: yes
frequency: 64800
remote-address: 24:18:*****
tx-mode: dmg
tx-mcs: 7
tx-phy-rate: 1925.0Mbps
signal: 60
rssi: -69
tx-sector: 51
tx-sector-info: left 0.2 degrees, up 1 degrees
distance: 1593.32m
tx-packet-error-rate: 12%

doneware · Tue Apr 16, 2019 9:03 pm

tx-packet-error-rate: 12%

this error rate is way too high. also notice that the AP side uses lower modulation scheme (7 vs

.
most probably you get quite some noise there. i don't know your setup, but it you line up your relays back-to-back, and use the same frequency on both of them, you might get some interference from the next link.

|( - - - - - )|( - - - - - )|
|             |             |
|             |             |

brasileottanta · Tue Apr 16, 2019 9:24 pm

Colleagues, tell me, there is a line at 1,500 m., We were reducing antennas, but I noticed that after 30 minutes, the indicators just hang in one place and do not move, it only helps to reboot the devices or to tear the radio link.

Hello ,

I' don't know what's the probs, but you need better aiming.

brasileottanta

SpanMG · Fri Apr 19, 2019 2:39 pm

Colleagues, tell me, there is a line at 1,500 m., We were reducing antennas, but I noticed that after 30 minutes, the indicators just hang in one place and do not move, it only helps to reboot the devices or to tear the radio link.

Hello ,

I' don't know what's the probs, but you need better aiming.

brasileottanta

I understand that the question will sound silly, but even so, I will ask it.
How did you achieve these indicators, did you use special fixtures?
And as you reduce the antennas, I have a line at 2,300 m and it behaves very well, but sometimes it does not work and to reduce to 1500, the signal jumps too much and it fails to achieve good, and as I wrote above, after 30 minutes, the signal DEPENDS.
And you have to reload the antenna.

jobewankenobi · Fri Apr 19, 2019 7:47 pm

We currently use a pair of sxtsq 5.8 ghz point to point to drive content to football Video boards... Looking to get out of the 5ghz spectrum because of congestion during football games. Bought pair of sxtsq 60g to test with.... Cannot find directions on how to set up 2 devices in ptp.

hercules30 · Tue Apr 23, 2019 3:51 pm

Hi all.
I have read through the forum, that there are already installers, who have mounted the 60G lhg at a distance between 1.5Km and 3Km .. And I would like to ask, how does it behave in moderate rain to strong ...
I would also like to know if anyone has thought of using the LHG XL HP5 plate, for an Lhg 60 .. We are talking about a 550mm plate with a brake diameter of 390mm. I suppose that for the distance that I have mentioned, it will make it more stable.

Thank you

mman1982 · Wed Apr 24, 2019 10:27 pm

Hi,

So guys,it isn't possible to make a stable 60G link in rain for 1km?If possible which values need to be aligned??(mcs,quality rssi) Anyone from Mikrotik can explain which values needed for stable link and how to align antennas properly?

Stril · Mon May 06, 2019 9:23 am

Hi!

Did anybody of you have long-term-experience with any "weather-cover"?
Which ones did work fine for you?

Thank you for your thoughts

Stril · Mon May 06, 2019 9:55 am

Hi!

I have a strange problem:
- 2x LHG 60G link, 395m
- RSSI stable at -50 to -52 dB, Signal-Quality 80-95

Next to the LHG on one side, there is a large (3x3x3m), rotating advertisement-cube

When the rotation of the cube is off, the link is 100% stable. When I start the rotation, the MCS drops down from time to time from 8 to 6 or less.

What I tried is to disable auto-alignment (tx-sector=36).

Do you have any other idea on how to solve this?

Thank you for your help!

komenski · Tue May 07, 2019 7:00 am

Ask Mikrotik to finally publish the LHG60g radiation pattern, then everything will be clear. The situation has been unacceptable for so long time, we have a broadcasting device and we don't know exactly where it is broadcasting.

venthyl · Tue May 07, 2019 11:25 am

Hi

60ad mikrotik is Higly unstable when:

-raining above 600m

-snowing ( wet snow) kill even 300m links
using RADOME helps snow downtime : but decrease rssi about 5db
~~10hour link on 500m was dead without Radome
~~20minutes same link with radome after wet swnow

i have ~~ 12 links on lhg backup'ed with 5g and ospf.

olegl · Mon May 13, 2019 3:07 pm

Hello.
Recently we`ve installed two LHG Lite60
Periodically, we see log entries like:
10:34:28 interface,info wlan60-station-2: link down
10:34:31 interface,info wlan60-station-2: link up

10:36:45 interface,info wlan60-station-2: link down
10:36:46 interface,info wlan60-station-2: link up

10:39:23 interface,info wlan60-station-2: link down
10:39:25 interface,info wlan60-station-2: link up

10:46:42 interface,info wlan60-station-2: link down
10:46:44 interface,info wlan60-station-2: link up

10:50:17 interface,info wlan60-station-2: link down
10:50:18 interface,info wlan60-station-2: link up

11:44:10 interface,info wlan60-station-2: link down
11:44:12 interface,info wlan60-station-2: link up

12:14:44 interface,info wlan60-station-2: link down
12:15:00 interface,info wlan60-station-2: link up

12:27:11 interface,info wlan60-station-2: link down
12:27:17 interface,info wlan60-station-2: link up

12:27:49 interface,info wlan60-station-2: link down
12:27:51 interface,info wlan60-station-2: link up

13:19:23 interface,info wlan60-station-2: link down
13:19:25 interface,info wlan60-station-2: link up

13:31:05 interface,info wlan60-station-2: link down
13:31:11 interface,info wlan60-station-2: link up

13:33:27 interface,info wlan60-station-2: link down
13:33:29 interface,info wlan60-station-2: link up

Link is flapping, but we don`t knwo the reason
What I see from both sides:
SIDE A
int w60g monitor wlan60-1
connected: yes
frequency: 60480
remote-address: 04:D6:AA:A8:D5:18
tx-mode: dmg
tx-mcs: 8
tx-phy-rate: 2.3Gbps
signal: 80
rssi: -69
tx-sector: 36
tx-sector-info: center
distance: 84.61m
tx-packet-error-rate: 0%

SIDE B
int w60g monitor wlan60-1
connected: yes
frequency: 60480
remote-address: 24:18:1D:99:62:E1
tx-mode: dmg
tx-mcs: 8
tx-phy-rate: 2.3Gbps
signal: 80
rssi: -64
tx-sector: 58
tx-sector-info: left 0.6 degrees, up 1.4 degrees
distance: 84.63m
tx-packet-error-rate: 0%

Periodically "tx-packet-error-rate" grows from 0 up to 20% on side B.
I suppose error-rate is the reason of link flapping
Can you give me info - what can be the problem?

server8 · Mon May 13, 2019 4:38 pm

your rssi is very bad

-69@84m is too low at such distance on one side is better to use a wag60 or sxtsq60

WebItG · Wed May 22, 2019 7:45 am

RBLHGG-60adkit
fell from a height of 10 meters to the ground and never broke but
When connected to electricity
only power led it works (PWR led on)
but all another led not work (ETH ,USR , Wireless ,align left right ............)
i try to connect it to my computer
The computer has never seen the network cable
What is the problem and what is the solution ?????

mman1982 · Wed May 22, 2019 8:07 am

Hi,

We had the same problem.(without fell down) RMA..

RBLHGG-60adkit
fell from a height of 10 meters to the ground and never broke but
When connected to electricity
only power led it works (PWR led on)
but all another led not work (ETH ,USR , Wireless ,align left right ............)
i try to connect it to my computer
The computer has never seen the network cable
What is the problem and what is the solution ?????

WebItG · Wed May 22, 2019 8:46 am

Hi,

We had the same problem.(without fell down) RMA..

RBLHGG-60adkit
fell from a height of 10 meters to the ground and never broke but
When connected to electricity
only power led it works (PWR led on)
but all another led not work (ETH ,USR , Wireless ,align left right ............)
i try to connect it to my computer
The computer has never seen the network cable
What is the problem and what is the solution ?????

You have a solution to this problem
What do you suggest

mman1982 · Wed May 22, 2019 8:51 am

Sadly,no:( We had to warranty that LHG.

Hi,

We had the same problem.(without fell down) RMA..

RBLHGG-60adkit
fell from a height of 10 meters to the ground and never broke but
When connected to electricity
only power led it works (PWR led on)
but all another led not work (ETH ,USR , Wireless ,align left right ............)
i try to connect it to my computer
The computer has never seen the network cable
What is the problem and what is the solution ?????

You have a solution to this problem
What do you suggest

olypen · Thu May 23, 2019 6:17 pm

We have found some interesting experiences working with the LHG 60. We have a PTP link that is 373m. During initial setup we aligned the dishes and locked them in with an RSSI of -54 (MCS 8, Signal 80, PHY Rate 2.3 Gbps).

We have been tracking TX-Sector, RSSI, MCS and Signal using PRTG and have noted some strange anomalies. The link appears completely stable and polling the data above every one minute shows it is rock solid, but then out the clear blue, the TX-Sector will suddenly shift and everything changes. When this happens the signal will usually drop (from 80 to as low as 50), the RSSI will go up (anywhere from -60 to -70) but the MCS will usually stay steady at 8.

The first time this occurred, I thought perhaps a wind gust had shifted the dish, so attempted to revert the TX-Sector back to the last best TX-Sector (from 34 to 11), but that caused the Signal and RSSI to vary constantly, so I switched it back to auto and it settled down to a new sector (63) with a much better RSSI and a solid signal level of 80.

It ran solid that way for about 24 hours and then once again, for no apparent reason, the TX-Sector suddenly shifted causing the signal to drop to 50 and the RSSI to jump to -70. But, even with the poor RSSI, the MCS stayed at 8 (PHY 2.3). I let it run that way for a couple of hours and during that time there appeared to be no loss in throughput and no significant latency. The link remained rock solid despite the odd readings. Eventually, though I toggled it from auto TX-Sector to manual and back again forcing it to realign, after which it settled at about -64, with a signal of 80.

As I said previously,the first time this happened I suspected wind gusts as we noted some guts around the same time period, but in all of the subsequent shifts, there have been no significant weather changes.

One other interesting observation is that when it makes this shift, the link does not drop (which is good). I am polling the data once every minute (as well as logging it) and note that it just seems to suddenly jump from one sector to another, with no searching or hoping around.

jellex · Thu May 23, 2019 11:49 pm

...
It ran solid that way for about 24 hours and then once again, for no apparent reason, the TX-Sector suddenly shifted causing the signal to drop to 50 and the RSSI to jump to -70. But, even with the poor RSSI, the MCS stayed at 8 (PHY 2.3). I let it run that way for a couple of hours and during that time there appeared to be no loss in throughput and no significant latency. The link remained rock solid despite the odd readings. Eventually, though I toggled it from auto TX-Sector to manual and back again forcing it to realign, after which it settled at about -64, with a signal of 80.

As I said previously,the first time this happened I suspected wind gusts as we noted some guts around the same time period, but in all of the subsequent shifts, there have been no significant weather changes.

One other interesting observation is that when it makes this shift, the link does not drop (which is good). I am polling the data once every minute (as well as logging it) and note that it just seems to suddenly jump from one sector to another, with no searching or hoping around.

Ah, I'm experiencing exactly the same! I had this with my first link (500 meters) and spent a lot of time adjusting and aiming. In the end, I let it settle for about two days and it seemed to lock to a certain sector. Signal steady at 80 and rssi -51. I then picked a center sector (36) and locked to that on both antennas (I was just trying things out).

Now, it's up and running for 60 days and it hasn't changed one bit! Not even with rain. Just no change in signal, rssi or MCS. I ran a hackathon over it with 1200 devices without any problem.

I'm configuring a new, second link on the same roofs now and I have a lot of trouble getting it right. Of course I'm running on another frequency, but choosing the right one is strange. When picking 60480 (first link is on 58ghz) and letting it run for a couple of hours the signal and rssi jumped all over the place. There was half hour without any tx-errors and a steady signal of 80. Then it just jumped to another sector and then stayed at 60, MCS turns down to 1 and it stays that way... Then choosing 62640, the signal stays on 95, MCS 8 and rssi -54, solid for some hours now. This is without any change to alignment of the dishes. I still have tx-sector set to auto now, but when it works I'll set to a locked sector.

I also couldn't think of any reason why the sector suddenly jumped apart from wind. The pole has a little slack with higher winds. I'll keep monitoring when wind changes.

Dinskuns · Thu Jun 13, 2019 6:18 pm

Anybody knows if LHG 60 supports large packet sizes to transfer qinq through the radio link?

mistry7 · Thu Jun 13, 2019 10:34 pm

For QinQ you need MTU of 1522, LHG has L2 MTU 1598, so QinQ is No Problem

Fri Jun 14, 2019 4:36 pm

max L2MTU for w60g is 7882

Louisk777 · Tue Jun 25, 2019 5:36 pm

Which package version is currently deemed the most stable?
Currently running 6.44beta75 which seems to be doing ok.

Stril · Wed Jun 26, 2019 10:51 am

Hi!

Did you find any possiblity to get valid RSSI-values?
I see fluctuation over the first 5 minutes. After that, RSSI is unchanged for weeks - no matter, if there is heavy rain.

On thunderstorms, sometimes the RSSI dropped and did not get back to the "real" value without dropping the link manually or starting "align-mode" for some seconds...

Thank you and best wishes

Stril · Wed Jun 26, 2019 10:53 am

Hi!

Did you find any possiblity to get valid RSSI-values?
I see fluctuation over the first 5 minutes. After that, RSSI is unchanged for weeks - no matter, if there is heavy rain.

On thunderstorms, sometimes the RSSI dropped and did not get back to the "real" value without dropping the link manually or starting "align-mode" for some seconds...

Thank you and best wishes

server8 · Wed Jun 26, 2019 12:45 pm

ros version?

Baur67 · Fri Aug 02, 2019 4:22 pm

Good day from Kazakhstan!
There is a problem, the installation of the frequency 64800 crashes, I install it from the command line. After a while, the connection breaks, then the frequency 58 *** is established.
Who faced what could be the problem? The latest firmware, updated 3 days ago.

EggieRedmayne · Mon Aug 12, 2019 1:44 pm

Hello everyone,

I am curious as to whether it is possible to get higher MCS values than 8. Has anyone experienced this or is 8 the current supported maximum?

gamcash · Sat Aug 17, 2019 12:50 am

I NEED SOMEONE TO GUIDE ME INSTALL LHG 60G KIT 1,770 KM PTP FIRST TIME

server8 · Sat Aug 17, 2019 11:43 am

The latest version 'd have the viewfinder that helps a lot to make the first alignment after you may use via cli the command: int w60 align 0 using the average rssi as reference

use the solid mount adapter https://mikrotik.com/product/solidmount for fine alignment

gemb · Thu Aug 29, 2019 6:20 pm

2280m link @ 2.3Gbps @ LHG60G

2300m (1).png

djvolt · Fri Aug 30, 2019 1:20 am

Which ROS? Beamforming Auto or Manual?

gemb · Sat Sep 07, 2019 10:53 pm

Which ROS? Beamforming Auto or Manual?

6.46 beta 16. Beamforming on auto

dnyl · Wed Sep 18, 2019 10:50 pm

Hello.
We have a 60G link what works well for days. Random time the station device is unreachable, but passing traffic. I need to kick from the ap (bridge) side. When it comes back it works well again. One day after it happens again.

Not ok:
https://cloud.netkontakt.hu/index.php/s/FJWsgDMjTKdZ8AB
OK:
https://cloud.netkontakt.hu/index.php/s/jYH8Sj4j2Lbaosx

EggieRedmayne · Wed Sep 25, 2019 10:44 am

Hello everyone,

I have had an installation up for about two weeks now. It reaches the 950Mbps speed but not consistently. The link experiences some disconnections periodically. Is there a way to fix the MCS and transmit power? I have tried this command but its wrong apparently:

/interface w60g set wlan60-1 tx-mcs=7
/interface w60g set wlan60-1 tx-power=7

I used the same syntax to change the channel I was on so I am not sure why this doesn't work. Please help.

ThreeTreeDog · Tue Nov 05, 2019 9:27 pm

For all those questions regarding EIRP and TX power and all the other technical information..I, me, the one, have answers....Ok enough suspense (3 hours of searching)
BAM!
https://fccid.io/TV7LHGG60AD/Test-Repor ... rt-3767854

Your Welcome

Hitek · Sat Nov 09, 2019 3:11 am

For all those questions regarding EIRP and TX power and all the other technical information..I, me, the one, have answers....Ok enough suspense (3 hours of searching)
BAM!
https://fccid.io/TV7LHGG60AD/Test-Repor ... rt-3767854

Your Welcome

Thanks for spending you time trying to find this info. The link to the FCC's list of Mikrotik certifications was, however, posted up-thread already. It may also serve as a small window into future Mikrotik products that they have yet to announce(or do they announce them when the FCC posts the certs?).....

prozak · Wed Nov 13, 2019 2:01 pm

Hello all.

Pretty dissapointed after first rains here.
400m link between to LHG60ad's and cant seem to keep connection alive.
Keeps reconnecting almost every minute.
Tried 5 frequencies and lock tx-sector but with no success.
During rain Signal goes up and down from 80 down to 20 and then disconnects at some point.
I selected 60GHz as target area is heavily crowded on 5GHz band.
Right now im exposed and will swap to mimosa 5ghz backhaul equipment for reliability.

Current connection info without rain :

  connected: yes
             frequency: 58320
        remote-address: 24:18:1D:8D:D5:F0
                tx-mcs: 8
           tx-phy-rate: 2.3Gbps
                signal: 50
                  rssi: -68
             tx-sector: 29
        tx-sector-info: right 0.6 degrees, down 0.2 degrees
              distance: 393.6m
  tx-packet-error-rate: 0%

Also i have a second link which is 2km and while it has excellent performance at this range it also suffered from disconnects which i would understand better rather the shorter at 400m!

2KM link without rain:

 connected: yes
             frequency: 62640
        remote-address: 6C:C7:EC:35:59:70
               tx-mode: dmg
                tx-mcs: 6
           tx-phy-rate: 1540.0Mbps
                signal: 60
                  rssi: -66
             tx-sector: 9
        tx-sector-info: left 1 degrees, down 1 degrees
              distance: 2036.37m
  tx-packet-error-rate: 0%
  
  [b]Speed test TCP average 600Mbps[/b]

I believe that these devices should not be out to be sold as they are unreliable.
I sent a question to mikrotik support about whether this "issue" is to be characterized as a refund.

If those devices fail to stay connected on 400m with light/medium rain, then why are they even sold? I dont remember seeing any remarks that rain is excluded.

Wed Nov 13, 2019 2:59 pm

Hello all.

Pretty dissapointed after first rains here.
400m link between to LHG60ad's and cant seem to keep connection alive.
Keeps reconnecting almost every minute.
Tried 5 frequencies and lock tx-sector but with no success.
During rain Signal goes up and down from 80 down to 20 and then disconnects at some point.
I selected 60GHz as target area is heavily crowded on 5GHz band.
Right now im exposed and will swap to mimosa 5ghz backhaul equipment for reliability.

Current connection info without rain :
Code: Select all
  connected: yes
             frequency: 58320
        remote-address: 24:18:1D:8D:D5:F0
                tx-mcs: 8
           tx-phy-rate: 2.3Gbps
                signal: 50
                  rssi: -68
             tx-sector: 29
        tx-sector-info: right 0.6 degrees, down 0.2 degrees
              distance: 393.6m
  tx-packet-error-rate: 0%
Also i have a second link which is 2km and while it has excellent performance at this range it also suffered from disconnects which i would understand better rather the shorter at 400m!

2KM link without rain:
Code: Select all
 connected: yes
             frequency: 62640
        remote-address: 6C:C7:EC:35:59:70
               tx-mode: dmg
                tx-mcs: 6
           tx-phy-rate: 1540.0Mbps
                signal: 60
                  rssi: -66
             tx-sector: 9
        tx-sector-info: left 1 degrees, down 1 degrees
              distance: 2036.37m
  tx-packet-error-rate: 0%
  
  [b]Speed test TCP average 600Mbps[/b]
I believe that these devices should not be out to be sold as they are unreliable.
I sent a question to mikrotik support about whether this "issue" is to be characterized as a refund.

If those devices fail to stay connected on 400m with light/medium rain, then why are they even sold? I dont remember seeing any remarks that rain is excluded.

You should know what you are doing using 60GHz equipment. Do the math and select frequency/device. If you are low on 5GHz Spectrum Mimosa makes it worse. There is better equipment available. If you want to offer reliable service with crowded spectrum consider licensed frequencies.

prozak · Wed Nov 13, 2019 3:27 pm

Which part of my post showed that i was unaware of what I'm doing?
clear los 400m excellent signal/aim.
device has to be reliable without disconnects on light /medium rain.
I don't get the defence over this product by you?

blingblouw · Wed Nov 13, 2019 4:01 pm

I don't think you are aligned well enough. Here is a link of mine pretty much the same distance but much better signal and RSSI

             connected: yes
             frequency: 58320
        remote-address: ---
                tx-mcs: 8
           tx-phy-rate: 2.3Gbps
                signal: 95
                  rssi: -52
             tx-sector: 20
        tx-sector-info: right 0.2 degrees, down 0.6 degrees
              distance: 338.76m

Wed Nov 13, 2019 4:07 pm

Which part of my post showed that i was unaware of what I'm doing?
clear los 400m excellent signal/aim.
device has to be reliable without disconnects on light /medium rain.
I don't get the defence over this product by you?

What is the signal you calculated for this link?

alamgir79 · Mon Nov 25, 2019 10:13 pm

Just installed a LHG 60G. Here is the link details with throughput

Link Status:

connected: yes
frequency: 60480
remote-address: 24:18:1D:96:AF:BB
tx-mcs: 8
tx-phy-rate: 2.3Gbps
signal: 100
rssi: -56
tx-sector: 12
tx-sector-info: right 0.2 degrees, down 1 degrees
distance: 115.25m
tx-packet-error-rate: 0%

PTP througput test:

status: running
duration: 17s
tx-current: 292.1Mbps
tx-10-second-average: 281.8Mbps
tx-total-average: 276.7Mbps
rx-current: 280.3Mbps
rx-10-second-average: 280.5Mbps
rx-total-average: 275.0Mbps
random-data: no
direction: both
connection-count: 20
local-cpu-load: 100%
remote-cpu-load: 90%

When I connect the link to my production network then download is ok but upload bandwidth doesn't come more than 2Mbps.

Any suggestion pls?

Thanks
Alamgir

mistry7 · Thu Dec 05, 2019 11:05 pm

Please Show is the Link Status of the Second Antenna....

Elliot · Thu Jan 16, 2020 10:47 am

Hey Guys,
is there a way to calculate current TX power? In data sheet there's TX power= <10dBm, but there's no way to know this from device.

Dinskuns · Sat Feb 29, 2020 3:53 pm

In regards to snow protection, please comment on the following idea - what if we wrap the LHGG-60 in a several layers of stretch film, leaving the bottom part unwrapped, exit for the leaving water. Have anyone tried that or could place some reasonable critics?

karina · Sat Mar 07, 2020 8:27 pm

if you spend enough time you can achieve good resuts. look at this example.

w60g.JPG

honzam · Sun Mar 08, 2020 12:53 am

In regards to snow protection, please comment on the following idea - what if we wrap the LHGG-60 in a several layers of stretch film, leaving the bottom part unwrapped, exit for the leaving water. Have anyone tried that or could place some reasonable critics?

Why? We have radom - https://www.i4wifi.eu/en/211374-kryt-jrc-jh-lhg-2ks

mistry7 · Sun Mar 08, 2020 7:07 pm

https://www.ip-sa.com.pl/ubiquiti/AF60

No more need for MT Product with 5 Ghz Backup
and narrow channel support!

karina · Tue Mar 10, 2020 1:04 pm

https://www.ip-sa.com.pl/ubiquiti/AF60

No more need for MT Product with 5 Ghz Backup
and narrow channel support!

Not liking that design, a single leg holding the radio?? and no decent router os. I will stick with Mikrotik thanks.

Darmach · Sat May 23, 2020 11:40 am

1950 meters link

rrssi sta/ap / chanel

-63/-64 chanel 58g
-68/-68 chanel 60g
-68/-66 chanel 62g
-58/-57 chanel 64g

Heavy Rain comparision lgh60 1950m (58ghz), lhg60 650m (auto), siae 17g, airfiber 24g

deszcze.PNG

Hi Venthyl, I see that one of your link is similar to what I have to establish - I need a 350m link between house (about 9m) and a 4 story building. However, I have one line of trees (top of them actually) between the buildings, so the line of sight is not exactly clear. Can I expect LHG to get good results with so short distance, but trees on the way? Fresnel zone is not exactly clear too - there is one house behind the trees.

Darmach · Thu May 28, 2020 3:45 pm

Removed email address from this post

skoenman · Tue Jun 02, 2020 11:14 am

Hey guys been running 60GLHGG for about 7 months now ....not very stable when it rains but runs well otherwise. The link is 3KM long so it is a bit far. I loaded 6.47rc2 yesterday and the link kept dropping nothing changed on the link just the software. Question is I know they changed something on rates but i think it screwed with beam forming as it jumps from sector 1 to 55 to 9 and so on ...just on the tx sector. I loaded the older stable version and it seem to be stable now. 6.46.6.

Any ideas maybe??? I tried locking the TX sector but that made it drop every now and then or the rate dropped to 0.
any other tweaks i can maybe try to get it stable???

Thanks

sirbryan · Tue Jun 09, 2020 1:41 am

Hey guys been running 60GLHGG for about 7 months now ....not very stable when it rains but runs well otherwise. The link is 3KM long so it is a bit far. I loaded 6.47rc2 yesterday and the link kept dropping nothing changed on the link just the software. Question is I know they changed something on rates but i think it screwed with beam forming as it jumps from sector 1 to 55 to 9 and so on ...just on the tx sector. I loaded the older stable version and it seem to be stable now. 6.46.6.

Any ideas maybe??? I tried locking the TX sector but that made it drop every now and then or the rate dropped to 0.
any other tweaks i can maybe try to get it stable???

Thanks

There's something wrong with 6.47 and the 60G radios. I had three or four AP's loaded with it and my clients disconnected all over the place.

With more testing, I determined AP's with 6.47 set to channel 58320 was the recipe for disaster, including the 40' link from my backyard shop to the house. Moving AP's to other frequencies solved the bouncing problem.
Clients with 6.47 pointed to 6.46.6 AP's had no issues (but then, they were also not on 58320).

tatianab85 · Tue Jun 09, 2020 8:26 pm

Hi guys,

Installed today a PTP with LHG60 and is working as expected, the link is a 540 meters, lets see how it works when it rains. with btest send and receive at the same time i get around 350Mbs each, guess it is because CPUs reaches 90%.

shaunmol · Wed Jun 17, 2020 4:27 pm

Good day

I was wondering if someone has tried these units on longer distance than the 1500m rated maximum distance?
If anyone has had success establishing a successful link at longer distances, please share. I would appreciate it greatly.

Regards

romankof2002 · Thu Jun 18, 2020 5:13 pm

Hello friends, I am very happy with the performance of the equipment in 60 ghz, I currently have them at 1 km with an added speed of 2.3g as marked by the winbox, I do not know if they will have future plans to modify the firmware in a matter of width. channel, example ubiquiti 60ghz can modify the channel width parameter 1080 and 2060mhz achieving longer distances. In addition to having an extra channel at 60ghz. Now in the future, why not develop a team that has 60ghz and backs up a 5ghz radio but that this radio is connectorized (pigtail rsma), to be able to connect some MANT30 or better that it be a Radio in 6ghz, it is only as a suggestion . Greetings from Mexico

arendn · Fri Jun 26, 2020 7:15 pm

Good day

I was wondering if someone has tried these units on longer distance than the 1500m rated maximum distance?
If anyone has had success establishing a successful link at longer distances, please share. I would appreciate it greatly.

Regards

We have one that runs at 1.8Gbps in when wind doesnt blow, but right now it sits at 770Mbps. 3.1km. More than 20mm/hr rain it drops out and we failover to a set of Netmetals on Use solidmounts for precisionalignment. 2 sets of LHG60G

markscotford · Sun Jul 05, 2020 8:32 pm

I have never used MikroTik radios, even though all of our Routers (approx 20 of them) are MikroTik. So I have to confess I dont understand a lot of the technical stuff what is being said here (that's what Ubiquiti does to you). I have two PTP links which are causing me problems because of congestion on 5Ghz, they currently use Ubiquiti M5 AC ISO Powerbeam pairs. The MikroTik 60Mhz radios have a spectacular price point, my problem is the distance. the two links are 3.61Km's, what sort of throughput do you think I might get (if any)? (PS its Spain, so rain is not an issue). If not the LHG 60G's any other (small budget) suggestions?

nithinkumar2000 · Sun Jul 05, 2020 9:16 pm

Wonderful Piece of Work from Mikrotik.

Planning to Establish at Backhaul for Tower Distance: 1.5KM

millenium7 · Sun Aug 16, 2020 4:17 am

I've read through this entire thread and I still don't know exactly how to use these things 'properly'. There's too much confusion
We have a few of them out in the field and they are unpredictable. Some links drop in the rain at just 300m, yet people are claiming over 10x that distance
Speed doesn't matter, I really don't care if its 500Mbit or 2300Mbit, it's the reliability thats killing us

For instance what exactly does Tx and Rx sector mean? I know that this relates to the internal grid, but what exactly is say 54? what is 21? what is 36? what is 96? these just seem like arbitrary numbers. If it's on 54 does this mean I should point the antenna left more? right more? up? down? Is there a X,Y reference grid I can look at so I know that "oh, sector 54 is in the top left, so radio should be pointed down and right"
It might say 'right 0.4 degrees' but what does that even mean? that the antenna is pointing 0.4 degrees too much to the right, or that I should point the antenna another 0.4 degrees to the right?

I read about disabling beamforming, ok how do we do this in the real world? Do we only do it on the station side? only on the AP side? on both? one at a time or both at the same time? so does this mean we need 1 person at each side of the link because if enabling it and the link drops we need to disable to get it back
How do we get it close enough to start with?

So please can someone post some clear precise information on exactly how to set them up correctly for optimal performance? In both PTP and PTMP mode, because we just cannot achieve guaranteed reliability
Clear simple instructions like
- If link is above 500m, buy the Solid Mount
- Install, point roughly by eye
- Set frequency to X
- Disable beamforming on the SM side, move until RSSI is the best it can be, re-enable beamforming on SM
- Repeat above but for the AP
- Do both at the same time, link should remain up
- Sacrifice 1 chicken per 250M of range under a full moon otherwise link will still drop-out in very light rain

nithinkumar2000 · Mon Aug 17, 2020 8:23 pm

I've read through this entire thread and I still don't know exactly how to use these things 'properly'. There's too much confusion
We have a few of them out in the field and they are unpredictable. Some links drop in the rain at just 300m, yet people are claiming over 10x that distance
Speed doesn't matter, I really don't care if its 500Mbit or 2300Mbit, it's the reliability thats killing us

For instance what exactly does Tx and Rx sector mean? I know that this relates to the internal grid, but what exactly is say 54? what is 21? what is 36? what is 96? these just seem like arbitrary numbers. If it's on 54 does this mean I should point the antenna left more? right more? up? down? Is there a X,Y reference grid I can look at so I know that "oh, sector 54 is in the top left, so radio should be pointed down and right"
It might say 'right 0.4 degrees' but what does that even mean? that the antenna is pointing 0.4 degrees too much to the right, or that I should point the antenna another 0.4 degrees to the right?

I read about disabling beamforming, ok how do we do this in the real world? Do we only do it on the station side? only on the AP side? on both? one at a time or both at the same time? so does this mean we need 1 person at each side of the link because if enabling it and the link drops we need to disable to get it back
How do we get it close enough to start with?

So please can someone post some clear precise information on exactly how to set them up correctly for optimal performance? In both PTP and PTMP mode, because we just cannot achieve guaranteed reliability
Clear simple instructions like
- If link is above 500m, buy the Solid Mount
- Install, point roughly by eye
- Set frequency to X
- Disable beamforming on the SM side, move until RSSI is the best it can be, re-enable beamforming on SM
- Repeat above but for the AP
- Do both at the same time, link should remain up
- Sacrifice 1 chicken per 250M of range under a full moon otherwise link will still drop-out in very light rain

+1 from Me

millenium7 · Tue Aug 18, 2020 2:13 am

So to add some more experience and more confusion needing clarification....

Went out to the AP side of a site thats doing multipoint, spread is about 10 degrees between the 2 SM's. Distance is ~170m and ~340m respectively
All of them are LHG 60G
Found it was a little bit out of alignment, slightly up (about 3 degrees) and slightly too much right, but nothing crazy. Despite that it's been dropping out regularly, sometimes even in clear weather but if there's even a whiff of rain vapor in the air it'll drop
On the other hand we have another site with 1 LHG60G feeding 7 SM's (also LHG60G) and the horizontal spread is approx 120 degrees and vertical about 10. Distance is shorter at ~30-100m but its pretty solid, this is nowhere near 'aligned' on any of the radio's, they're just roughly pointed by eye. This site is reasonably reliable and we don't seem to get many complains. Yet we have point-to-point links that are far better aligned that drop out regularly, no makey sensey

So I started by re-aligning the radio roughly to the further one at ~340m away, then locking the TX sector to 36 (only on the AP) thinking that this means it'll need to be aligned pretty damn well or it won't link up at all, and the one at ~170m will definitely disconnect. Nope, BOTH came back up and were still doing fantastic speed, infact the packet error rate was lowest like that
Ok either way its been aligned as best we can. I figure then putting it back to auto it should use sector 36 for that SM right? nope. The sector number keeps changing every second, very rarely does it actually use the center one, it'll pick 2, 5, 22, 54 or any number it wants to. The RSSI changes a lot more by +/- 15db when it changes sector. And the error rate is also higher it spikes up to ~22% but most of the time is hovering on 0%

So some more questions here.....
- Whats all this talk about millimetre beams? Because surely the closer site at ~170m shouldn't have connected at all if thats the case? it's now ~10 degrees off to the side, yet it does. On the other hand a Siklu 60ghz radio will not link up or stay linked up unless its pretty much bang on, it's truly about a 2 degree beam. MikroTik seems to be a complete mystery
- Why does the sector flap and change around so damn much? the radio is very stable with just the teensy tiniest bit of movement in the wind but that's not the pole its mounted to its just flex in the plastic housing ergo nothing can be done about it
- Is it still desirable to leave beamforming/tx-sector=auto on if the sector keeps changing and the error seems higher? Or should I leave the AP locked on sector 36 as that seems to do better in terms of less errors
- Should we do it the other way around and leave the AP on auto, but go and set the SM's to only transmit on 36 and perfectly align them to the AP?
- How many more chickens do we need to sacrifice? this is just getting out of hand

I can't stand having no proper processes on how to use these things. I like to work with solid information, facts and figures. Not prayers, pixie dust and unicorn farts and just hope it stays up. Right now whilst every installation comes up and gives good speeds, we just have to flip a coin as to whether or not its actually going to be reliable. Good speeds is one thing but if it constantly drops out its useless

sirbryan · Tue Aug 18, 2020 2:53 am

Went out to the AP side of a site thats doing multipoint, spread is about 10 degrees between the 2 SM's. Distance is ~170m and ~340m respectively
All of them are LHG 60G
Found it was a little bit out of alignment, slightly up (about 3 degrees) and slightly too much right, but nothing crazy. Despite that it's been dropping out regularly, sometimes even in clear weather but if there's even a whiff of rain vapor in the air it'll drop
On the other hand we have another site with 1 LHG60G feeding 7 SM's (also LHG60G) and the horizontal spread is approx 120 degrees and vertical about 10. Distance is shorter at ~30-100m but its pretty solid, this is nowhere near 'aligned' on any of the radio's, they're just roughly pointed by eye

You're using LHG-60's as AP's? The antenna pattern is 3 degrees. So, across 120 degrees at ~100m, you're running off the RF equivalent of fumes.

I would strongly suggest replacing your "AP" LHG with one or two wAP-60. wAP-60's will talk to each other at 200m (60° beamwidth), and an LHG-60 pointed at a wAP-60 will go 800m+.

wAP -> wAP = ~200m (240 in my network)
wAP -> Cube = ~500m (510 in my network)
wAP -> LHG = ~800m (870m in my network)

Found a bunch of beamwidth data here:
https://wiki.mikrotik.com/wiki/Manual:Interface/W60G

I have 13 wAP-60's as AP's and dozens of them, along with Cubes and LHG's as clients (about 50).

- Whats all this talk about millimetre beams? Because surely the closer site at ~170m shouldn't have connected at all if thats the case? it's now ~10 degrees off to the side, yet it does.

You're probably hitting the feed horn's patch outside of the focal point of the dish. At close range (like across the street or next door one or two houses), I've found there's a "bubble" where the wAP-60's will work without any respect to directionality. There may also be reflection off of another building or object(?). (Anecdotally, my 870m wap60-LHG60 link probably only works because it's in a little "valley" on the customer's roof formed by the proximity of two gables next to each other in the front of the house.)

- Why does the sector flap and change around so damn much?

Which firmware version are you using? Which frequency are your radios set to? I set my AP's to 6.45.9 (latest stable) and have clients on 6.45.9, 6.46.6, 6.47.x. AP's on 6.47.x are unstable for me.

- Is it still desirable to leave beamforming/tx-sector=auto on if the sector keeps changing and the error seems higher? Or should I leave the AP locked on sector 36 as that seems to do better in terms of less errors

I leave everything on auto. I don't monitor the numbers too much. I run long ping tests and let my customers tell me they're down, with occasionally checking in on the radios' logs for disconnects. Not perhaps the best way, but I've found worrying and tweaking these things too much ends up causing more issues than it fixes.

Definitely set the specific frequency for each AP. Use 64800 for longer paths. I had the most issues with 58320 on 6.47.

millenium7 · Wed Aug 19, 2020 3:08 am

You're using LHG-60's as AP's? The antenna pattern is 3 degrees. So, across 120 degrees at ~100m, you're running off the RF equivalent of fumes.

In theory yes, theoretically it shouldn't even work at all, but so far nothings made sense and matched up with the theoretical data. We used the LHG 60G as an AP there to test the beamforming and stretch its capabilities. That site is reasonably reliable and infact far more reliable than some of the point-to-point setups we have that 'should' perform far better but actually don't. That site I just posted as an example, i'm not going to change anything there as it works good enough, we don't get complaints

I would strongly suggest replacing your "AP" LHG with one or two wAP-60. wAP-60's will talk to each other at 200m (60° beamwidth), and an LHG-60 pointed at a wAP-60 will go 800m+.

'In theory' I agree with you, however in practice we setup a couple of wAP 60's at our office just in a point-to-point scenario to test them, the range is 152 metres and they are utterly hopeless in rain, as soon as it even gets cloudy they drop out, let alone any sort of real rain. They constantly flap up and down, completely unusable unless its a perfectly clear day
Now in my mind, alignment with these shouldn't matter, and if being just slightly off (which they probably are, but nothing significant) means they drop out then 'in the real world' they'll be totally useless anyway as its impossible to have alignment to all SM's in a multipoint scenario

You're probably hitting the feed horn's patch outside of the focal point of the dish. At close range (like across the street or next door one or two houses), I've found there's a "bubble" where the wAP-60's will work without any respect to directionality. There may also be reflection off of another building or object(?). (Anecdotally, my 870m wap60-LHG60 link probably only works because it's in a little "valley" on the customer's roof formed by the proximity of two gables next to each other in the front of the house.)

The SM's in this case are ~170m and ~340m from the AP, so definitely more than just across the street. These are all the LHG 60G dishes

I want to believe and take it everything you'd said as fact but our real world testing doesn't reflect that at all. If I use Siklu 60ghz gear it performs exactly as expected in relation to the antenna alignment, if it goes 3 degrees to either side it drops out entirely, or it'll only remain connected for a few seconds at best. It's 'real world' matches the theoretical data of having millimetre beams that require very precise alignment

MikroTik's implementation - whilst i'm happy to have a radio that you can just basically chuck it up and it does stupidly high speeds - it's unreliable and unpredictable as hell. This is REALLY bad when selling services to customers and trying to meet their expectations. Having something predictable and knowing how it works so you can properly plan and implement is crucial. We can't sell a service based on "lets just put it up and see how we go, if it has dropout issues then well too bad, neither us nor the manufacturer seems to know what the hell is going on or how any of it works"

millenium7 · Thu Aug 20, 2020 7:32 am

Ok so 'frequency=auto' is seemingly a really bad idea. And using the terminal with 'int w60g mon 0' is essential as tx-error rate shows a more accurate picture than signal/MCS/rssi
Pretty sure the way auto works is it just picks 58320 regardless, as i've only ever seen it on that frequency. It definitely doesn't pick the 'best' channel

I looked at another link in our network, its only 350m apart. Radio's are roughly but not perfectly pointed at each other (certainly should be well enough to function), both are LHG 60G radio's
It's had 4400 link drops in 21 days. While it theoretically shouldn't matter as the outage period is way below timeout values (it reconnects in less than a second), it was killing the PPPoE session
It will drop even in clear weather. Radio's are stable though 1 is slightly rotated as the mast is not perfectly straight
It mostly stays on MCS 7/8. Signal 80. RSSI is pretty solidly staying at -50. TX sector is mostly stable on 48 but does change periodically. Throughput is approx 1900mbit/s both directions. At first glance it looks ok but it just drops the link every so often

Running '/int w60g mon 0' in terminal however shows the TX error rate is always 4% or higher, it goes to 99% just before the link drops out
Doing nothing else but changing the frequency manually to 64800 has resulted in a stable link (so far)
RSSI is worse at -52 to -55, MCS at 8, Signal 80-90. Throughput is the same but the TX error rate is now 0% the vast majority of the time and occasionally blips as high as 22%

Thu Aug 20, 2020 12:27 pm

frequency=auto picks first free channel (checks available SSIDs, in channel before selection)
frequent tx-sector changes indicate, that there is issue related to devices moving in wind or alignment issues.
In theory tx-sector should only change on very rare occasions - during first minutes of established connection, rain or on line of sight loss

blue · Thu Aug 20, 2020 3:05 pm

Hello,

I would like to add that OS version is very important. After a lot of problems with newest versions, I downgraded all my 60G links to long-term (6.45.9), and now links are working perfect. I don't have PTMP - only PTP 60GHz.

BR...

millenium7 · Fri Aug 21, 2020 3:57 am

frequent tx-sector changes indicate, that there is issue related to devices moving in wind or alignment issues.
In theory tx-sector should only change on very rare occasions - during first minutes of established connection, rain or on line of sight loss

And what about if its rapidly changing but the mount is solid?
The above example its staying reasonably solid on 48, thats fine

Another link it changes every second or 2. The radio is mounted as solidly as it possibly can be, the only movement is from the housing itself flexing very slightly since it is plastic. There's nothing I can do about that.
Can reflection from a device behind it be the cause? Since its mounted directly behind a AF24HD
And if that is the case can you provide some more information on the best practice here?

As mentioned before, manually locking the AP to TX-Sector 36 showed the least amount of errors (didn't try the SM), would that be a viable solution or should it be left on Auto? (despite higher error rate)
Alternatively, should be AP be left on auto and the SM's locked to 36?
Would either situation work in a multi-point setup since its not physically possible to align the AP to multiple different SM's at once, but the SM's can be aligned to the AP for the most part

I'm still waiting for any actual solid information to help demystify these things. Why is MikroTik so quiet on them? did all the engineers who worked on the device leave the company and nobody has a clue how it actually works? Can we ideally get one of them to spend 30 minutes writing a post clarifying their operation and they to properly utilize them?

Fri Aug 21, 2020 10:02 am

And what about if its rapidly changing but the mount is solid?

This should only happen after reboot or lost connection.

Can reflection from a device behind it be the cause? Since its mounted directly behind a AF24HD

Internal frequencies may overlap. For test you may try to move it a bit further if mounted place allows that.
Is there any structures that may cause unwanted reflections that may mess up beamforming pattern selection?

Alternatively, should be AP be left on auto and the SM's locked to 36?

Best practice would be to leave all on auto, as beamforming is there to compensate sudden environment changes.
Locking will also work in AP situation, but for most part should be avoided.

I'm still waiting for any actual solid information to help demystify these things. Why is MikroTik so quiet on them? did all the engineers who worked on the device leave the company and nobody has a clue how it actually works? Can we ideally get one of them to spend 30 minutes writing a post clarifying their operation and they to properly utilize them?

Those are not simple things and each installation is different and there are much more factors that may affect success with link. We can't simplify those details as they are much complicated compared to regular wireless. Simple explanation is available in our wiki.

millenium7 · Sat Aug 22, 2020 11:51 am

Those are not simple things and each installation is different and there are much more factors that may affect success with link. We can't simplify those details as they are much complicated compared to regular wireless. Simple explanation is available in our wiki.

I disagree. MikroTik engineers don't need to talk about customers individual deployments and environmental conditions. Just clarify how exactly it works

For instance nobody seems to know anything about the sector numbers
Are the sector numbers arrange in a grid like
1 2 3
4 5 6
7 8 9
Or maybe its
1 4 7
2 5 8
3 6 9

Or something completely different? Only answer I know is sector 36 is 'middle'. Knowing this could help with alignment
Or what 'right 0.4 degrees' even means? does that mean its 0.4 degrees (approximately) from the middle (sector 36) or is that 0.4 degrees from the CURRENT sector if i.e. its using sector 22 (which would mean aligning it according to these numbers is not correct, and it should be locked to sector 36, then aligned until the numbers say center)
And is that 'the remote radio is 0.4 degrees off' or is it 'this radio is 0.4 degrees off'? And by right does it mean the radio is pointed too much to the right, or its pointed too much to the left and should go right?

How does the beamforming actually work. Can a client be on multiple sectors simultaneously and it just picks the best one, or does it dynamically shift to only using a particular sector number and ignore incoming signals on all others? If its the former, then yes theoretically auto should be the best because it will receive signals from multiple sectors. But if its the latter then its potentially very bad because if it automatically picks a bad sector, its ignoring signals coming in from a potentially better sector. Etc etc etc, there are plenty of answers MikroTik could give that better illustrate how these things actually work. This has nothing to do with individual deployments. I perfectly accept that some deployments aren't 'correct' but the thing that matters is if we understand 'how' the radios work then we can make them work better. It's not possible to put up a 50m tall mast with perfectly clear visibility and perfect conditions in every location. It is possible to however adjust radio parameters to work with what we have, or at the very least have a better understanding if something will work or not.
Saying "best practice is leave everything on auto, should be fine, just try it and see" is not good enough, as shown by all the posts in this thread of people having issues..........

Mon Aug 24, 2020 1:08 pm

I disagree. MikroTik engineers don't need to talk about customers individual deployments and environmental conditions. Just clarify how exactly it works

For instance nobody seems to know anything about the sector numbers
Are the sector numbers arrange in a grid like
1 2 3
4 5 6
7 8 9
Or maybe its
1 4 7
2 5 8
3 6 9

https://wiki.mikrotik.com/wiki/Manual:I ... 0G#Monitor

""tx-sector-info" (feature in testing stage) provides information from currently used beamforming pattern and shows direction to center - theoretical highest power output point. "

How does the beamforming actually work. Can a client be on multiple sectors simultaneously and it just picks the best one, or does it dynamically shift to only using a particular sector number and ignore incoming signals on all others?

Think about beamforming as fast switch between different radiation patterns, this information is exchanged between devices and pattern is adjusted accordingly to get best signal. In PTMP setups it can also be a fast switching between patterns for each connected device.

millenium7 · Tue Aug 25, 2020 10:14 am

Thanks for responding but that just isn't true..... or at least its woefully inaccurate, or isn't explained correctly. I still don't know if its referring to the AP or the SM (assuming logged into the AP)

I was physically up onsite at this location, I know for a fact that this particular location the AP is pointed within 1 degree of 1 of the SM's. I know its rock solid and doesn't move anymore than the physical flex in the LHG60G housing. What I do know is the data i'm seeing is wrong in relation to the 'how far from center' information

SM1
TX Sector 46 (this changes every so often though) 'right 1 degrees, up 0.6 degrees'

Ok I can believe that its probably 1 degree off
However I know for a fact that SM2 is at least 10 degrees (might be as much as 20 degrees) off to the right, yet it shows
TX Sector 3 (again, changes every few seconds) 'right 0.2 degrees, down 0.6 degrees'

According to that, its even more accurately aligned than SM1, it's also more than twice as far away....... Rubbish. I know for a fact its AT LEAST 10 degrees to the right, possibly as much as 25 degrees. I don't know for sure, I didn't have my protractor on me at the time......

I'm convinced even MikroTik has absolutely no idea how this radio works, maybe its a product purchased from another company and licenced under MikroTik and runs RouterOS on it. Maybe its' just a bad implementation. At the same times its also quite good in that I have a site with a ~120 degree spread on a LHG 60G and all SM's are reasonably reliable, according to spec sheets it should be impossible, yet it works............ And I also have sites that are very close, closely aligned, yet still drop in even the slightest hint of rain. A couple even drop in clear weather
If I can't even get accurate answers to seemingly simple questions then I don't hold much hope for these things ever becoming 'reliable'. They're good when they work, but it's a total gamble, there's absolutely no consistency or predictability with them. Sometimes maybe good, sometimes maybe shit..........

Tue Aug 25, 2020 10:41 am

Device goes through pre-calibrated data over 3 degree span showing offset from center. Expecting that this will work outside calibrated span is ridiculous.
If device is turned 20 degrees of, there is no way it's using main lobe, but is connected by side lobe that disturbs normal beamforming operation.
This feature is experimental and should be used together with align mode for FINE TUNING as explained in wiki.

GuntOn · Thu Oct 15, 2020 10:35 am

Which is the best RoS version for Wireless Wire Dish for now?

Thu Oct 15, 2020 4:37 pm

Latest testing build includes some stability fixes, that will be put in upcoming current version.

hercules30 · Tue Nov 24, 2020 4:53 pm

Hello everyone. I have a question about the frequencies of 60Ghzm, and their mikrotik behavior over the 66000Ghz frequency. My idea is to make a 1.87km link with the frequency 66000Ghz, but I have to make another link stuck to these with the frequency 64800Ghz.. And my question is, am I going to have problems with both frequency, if the mikrotik works at 2160Ghz in channel width?

server8 · Tue Nov 24, 2020 10:30 pm

With nray 60 ghz you can try with LHG 60 you may have problem

Ivoshiee · Thu Nov 26, 2020 7:20 pm

What is the valid frequecy range for the MT 60G gear? The named 66000 MHz is unofficially supported, but what is the whole range of center channel frequencies radio is capable of using itself and how one may use these frequencies?

doukas · Tue Dec 08, 2020 7:05 pm

Thanks for responding but that just isn't true..... or at least its woefully inaccurate, or isn't explained correctly. I still don't know if its referring to the AP or the SM (assuming logged into the AP)

I was physically up onsite at this location, I know for a fact that this particular location the AP is pointed within 1 degree of 1 of the SM's. I know its rock solid and doesn't move anymore than the physical flex in the LHG60G housing. What I do know is the data i'm seeing is wrong in relation to the 'how far from center' information

SM1
TX Sector 46 (this changes every so often though) 'right 1 degrees, up 0.6 degrees'

Ok I can believe that its probably 1 degree off
However I know for a fact that SM2 is at least 10 degrees (might be as much as 20 degrees) off to the right, yet it shows
TX Sector 3 (again, changes every few seconds) 'right 0.2 degrees, down 0.6 degrees'

According to that, its even more accurately aligned than SM1, it's also more than twice as far away....... Rubbish. I know for a fact its AT LEAST 10 degrees to the right, possibly as much as 25 degrees. I don't know for sure, I didn't have my protractor on me at the time......

I'm convinced even MikroTik has absolutely no idea how this radio works, maybe its a product purchased from another company and licenced under MikroTik and runs RouterOS on it. Maybe its' just a bad implementation. At the same times its also quite good in that I have a site with a ~120 degree spread on a LHG 60G and all SM's are reasonably reliable, according to spec sheets it should be impossible, yet it works............ And I also have sites that are very close, closely aligned, yet still drop in even the slightest hint of rain. A couple even drop in clear weather
If I can't even get accurate answers to seemingly simple questions then I don't hold much hope for these things ever becoming 'reliable'. They're good when they work, but it's a total gamble, there's absolutely no consistency or predictability with them. Sometimes maybe good, sometimes maybe shit..........

doukas · Tue Dec 08, 2020 7:14 pm

Hi , about the center sector fer nRay 60G is sector 31 . start from sector 0, 1, 2, 3, 4, 5, 6, 7 ,8
9,10,11,12,13,14,15,16,17
18,19,20,21,22,23,24,25,26
........until sector 62
for LHG is not the same center is 35 & 36& 27&28
start from sector 0 ,1 2 3 4 5 6 7
8 9 10 11 12 13 14 15
16 17 18 19 20 21 22 23
until sector 63
I hope to Help

doukas · Tue Dec 08, 2020 7:27 pm

nRay 60G sector pattern

0 .................... 8
9......................17
18....................26
27........31........35
36....................44
45....................53
54....................62

LHG 60G sector pattern
0.........................7
8.........................15
16.......................23
24.......27.28.......31
32.......35.36.......39
40.......................47
48.......................55
56.......................63

beeonline · Thu Jan 28, 2021 9:09 pm

Wery unusual problem I have here. Just replaced an AF5x 800m link last evening to backup my Siklu eh600 link during rains.
And sure, the RSSI is good during high rain right now, I get -53db on both ends, 2.3Gb PHY rate..

But the ping is soooo high - 300ms AVG under 100mbps load.

Interfaces are ok , 1G no errors. Just getting phone calls all the way. I would better stay with af5 and buy a siklu 1200..

The tricky part is that i already have many links with good ping, sooo just nothing I can suggest.

Even upgraded FW to 6.48, same thing

wispwest · Sat Jan 30, 2021 7:57 am

I've heard that Mikrotik has released the new LHG 60 Ghz R2 review
for range up to 6000m and embedded wifi 5GHZ
Please, Ask Mikrotik Directly about the more details
The new LHG 60GHz R2
Have a distance of about 6,000 metar and More...
MIKROTIK is in the testing phase of the new R2 version

there are people on the forum who know more details

The biggest change is in the antenna - the construction
RX and TX
Resistance to rain, snow ...
prices will be twice as expensive
So, I have to ask, what ever happened with this?!?! Over 2 years later....

velk1001 · Sat Jan 30, 2021 10:41 am

I thought the new unit was the Nray?

Regards

Paul

JJT211 · Mon Feb 01, 2021 8:46 pm

I see everyone saying that links >1km should use 58320Mhz and when put into Auto mode, it always sets itself to 58320. (My link is ~400m)

But what about in heavier rain zones like the Southern US? Has anyone tried using 64800 due to the higher frequency being more rain friendly?

millenium7 · Wed Feb 17, 2021 7:46 am

I always use 64800 if the links are above about 250m and often even less than thag
58320 is complete garbage. It theoretically should go much further but we've had radios drop out daily at only ~170m (LHG60G units) when it rains. Others have dropped when it's just overcast it doesn't even need to rain

Moving to 64800 fixes them most of the time, it should even be the default frequency
When co-locating radios I often still leave them all on 64800 as I find they really don't interfere much at all, radios 1m apart and I can run bandwidth test and still see ~1800mbit/s with the same error rate if both radios are running on same or different frequency. As long as the link never drops I can afford any potential reduction in speed anyway since almost all our 60ghz setups exceed the bandwidth on the 1gbit/s port, but so far havnt seen interference be a problem

If I don't even want to risk it I'll run next frequency down, but never use 58320, it has so many issues, not worth it

mp3turbo · Sun Mar 21, 2021 1:34 pm

dare to share any worst collocation setup you have ? I'm interested in couple of short 60GHz and it really seems like that lower channels are serious garbage due to oxygen/rain. This is of course not LHG 60G limited, I'm generally interested in any 60GHz experience. Sure I know crappy plastic LHG is something different to professional grade all-metal antennas/devices etc. but Im interested in any experience. Found two other topics related to this (viewtopic.php?f=7&t=143618&p=811170) and viewtopic.php?f=7&t=148687) and will read them carefully.

I lack the personal experience of how much co-location is possible. Your post has been very convincing and put a lot of hope - those radios "1m apart" you mention, are they pointing in the same direction ? Even if not, my head somehow can't accept that coming from 5GHz world :)

Possible to share any photo(s) and more details, please ?

millenium7 · Mon Mar 22, 2021 1:59 am

This is of course not LHG 60G limited, I'm generally interested in any 60GHz experience. Sure I know crappy plastic LHG is something different to professional grade all-metal antennas/devices etc. but Im interested in any experience

Actually it is specifically a problem with the MikroTik 60ghz. The signal isn't so bad it would actually not work, the signal is definitely getting through. However I presume the beamforming or other mikrotik implementation causes periodic dropouts and it just so happens to be significantly worse at the lower frequency (auto picking the 'first' free channel means it also picks the worst one)

Plastic has nothing to do with it, infact its one of my favourite dishes for installation. You can partially see through the dish which makes alignment by eye much, much, much better. You don't even need the included little plastic scope thing, just use it like iron-sights and align holes in the dish with the front arms and center of the radio, gets you within a bees dick if the other side is visible. Still worth using a solidmount however, mostly because the crappy stock mount WILL move when you tighten it down to some degree

co-location is fantastic, I can have them on the same mast only separated by ~50cm or so and see no drop in performance even on the same channel
They do work well when used exclusively as a point-to-point and you get it dead nuts aligned (in which case you can leave beamforming turned off entirely if it causes problems) and follow a certain procedure. The problem is when you just slap them in, then you get all kinds of black magic, they work, they don't work, there's no explanation for why, and you are given incorrect readings (like the distance to center is woefully wrong, its just garbage, mikrotik please just disable it entirely it doesn't do the 1 job its supposed to)

sirbryan · Tue Mar 23, 2021 8:47 am

This is of course not LHG 60G limited, I'm generally interested in any 60GHz experience.

I lack the personal experience of how much co-location is possible. Your post has been very convincing and put a lot of hope - those radios "1m apart" you mention, are they pointing in the same direction ? Even if not, my head somehow can't accept that coming from 5GHz world :)

I have both MikroTik and Ubiquiti 60GHz gear at several locations. I had a GigaBeam Plus working just fine at 1.2km on 58320 for a long time before I realized it was on that channel. But few if any of my wAP60's are using that channel for anything. As millenium7 said, it's worthless on MTik's radios.

I also have AP's and backhauls sitting right close to each other, using the upper three channels, and I try to reserve 68400 for the APs with customers the furthest out. I have seen the modulation rates jump all over the place at one site with multiple radios configured for the same frequency, but its hard to say whether that was related to the hardware+channel combination, the number of devices on that channel, or external environmental factors affecting the channel.

I mainly use LHG60's as CPE, but have a couple deployed as PTP links. I haven't gone through the gyrations millenium7 mentioned to get them aligned; I use my phone's camera to align wAP60's and Cubes visually, and use the alignment tool that comes with the LHG60's, verifying any tweaks on the web interface and/or with the LED's. All MikroTik-based PTMP links are under 800m, and PTP are 1km or less.

If I want to do PTP over 1km, I use GigaBeams or AirFiber 60. The alignment tools and visual feedback of the UI radios' link performance are superior to the confusion of bouncing numbers and unstable beamforming details in RouterOS.

Ivoshiee · Sun May 29, 2022 2:48 am

A tear down of a MT 60G system:
https://www.youtube.com/watch?v=JI9fvAcNpdk

GuntOn · Tue Jul 19, 2022 8:57 pm

Does anybody knows center sector of Cube Lite60?

Wed Jul 20, 2022 12:25 am

A tear down of a MT 60G system:
https://www.youtube.com/watch?v=JI9fvAcNpdk

beautiful

DeDMorozzzz · Wed Jul 20, 2022 7:25 am

Does anybody knows center sector of Cube Lite60?

As far as i know there are four in the centre. try 36

prague · Mon Aug 15, 2022 6:33 pm

I installed a system composed by one wAP 60G and 6 LHG 60G @ distance of 300-600m from the AP.
I noticed an instability on the wAP: if one station is moved or the signal drops, the wAP disconnects all other stations and it locks untill someone disables and enables the w60G interface.
Does anyone experienced the same behavior?

I've the same issue. Did you find a solution ?

vchrizz · Fri Mar 24, 2023 9:52 am

nRay 60G sector pattern

0 .................... 8
9......................17
18....................26
27........31........35
36....................44
45....................53
54....................62

LHG 60G sector pattern
0.........................7
8.........................15
16.......................23
24.......27.28.......31
32.......35.36.......39
40.......................47
48.......................55
56.......................63

How does the 60G sector pattern look like for CubeG-5ac60ad and CubeG-5ac60ay please?
And how am I able to find out that information? Sources?