new hardware Wireless Wire nRAY 60 ghz

server8 · Thu May 21, 2020 8:39 pm

https://www.cdr.pl/p7471,mikrotik-wirel ... 0-ghz.html

krafg · Fri May 22, 2020 4:43 am

Looks great!

Regards.

uberdome · Fri May 22, 2020 5:17 pm

I saw this link this morning for it: https://mt.lv/WWnRAY

I'm excited about the complete assembly and aiming mount.

It appears to be an attractive size, but seems long. Most of our situations where we install these will be on the sides of buildings, so the larger the distance from the wall, the more significant the installation appears.

It seems to me the configuration as shown will be just fine much of the time, other times I will be wanting to get it closer to a wall. This could possibly be achieved with 2x design changes:

Possibly a sideways or alternative cable exit can allow the back of the radio closer to a wall
A mount that can be oriented longitudinally instead of axially

I do have some potential concerns about function:

How are these screws accessed for aiming (It appears the gland seal is blocking access)?

Annotation 2020-05-22 100705.png

It seems there could be an additional hole to allow rotating the mount and reducing the front-to-back dimension. It appears this would have more clearance for those screws from the previous question. Any thoughts?

Annotation 2020-05-22 101038.png

Thank you, Chris

eworm · Mon May 25, 2020 6:50 pm

Interesting device...

Also nice to see that more devices are equipped with ARM 64bit CPUs (just like new CCR).

RogerWilco · Mon May 25, 2020 11:52 pm

It'd also be nice to see the LHG 2 in this design as well as the LHG LTE (with a better antenna for low gain).

Tue May 26, 2020 10:47 am

uberdome, the ethernet protection gland should not be plugged in yet, it can be on the cable, while you do the adjustment. After done, just slide it forward and screw it on

mistry7 · Tue May 26, 2020 2:16 pm

nice update but a 5 Ghz Interface is still missing why?

raceboy · Tue May 26, 2020 5:56 pm

estimate price?

Sun May 31, 2020 3:06 pm

Channels? Channel Size? TX Power? Wireless Chipset?

dnyl · Sun May 31, 2020 5:37 pm

I saw this link this morning for it: https://mt.lv/WWnRAY

I'm excited about the complete assembly and aiming mount.

It appears to be an attractive size, but seems long. Most of our situations where we install these will be on the sides of buildings, so the larger the distance from the wall, the more significant the installation appears.

It seems to me the configuration as shown will be just fine much of the time, other times I will be wanting to get it closer to a wall. This could possibly be achieved with 2x design changes:

Possibly a sideways or alternative cable exit can allow the back of the radio closer to a wall

A mount that can be oriented longitudinally instead of axially

I do have some potential concerns about function:

How are these screws accessed for aiming (It appears the gland seal is blocking access)?
Annotation 2020-05-22 100705.png

It seems there could be an additional hole to allow rotating the mount and reducing the front-to-back dimension. It appears this would have more clearance for those screws from the previous question. Any thoughts?
Annotation 2020-05-22 101038.png

Thank you, Chris

Dear Mikrotik!

Please modify the ethernet cable place. You do the same mistake, when you did with the grooves. If the device on a high tower and looks down, the water goes into the device. Please modify the connector down by 45° or 90°.

uberdome · Wed Jun 03, 2020 5:40 pm

nice update but a 5 Ghz Interface is still missing why?

I've debated about this for a while, and originally avoided MikroTik 60 GHz because of it for 6+ months.

Adding 5 GHz failover to 60 GHz just "increases" the range of the 60 GHz hardware. The problem is - if you didn't have the 5 GHz spectrum available in the first place, then you will be operating at limited speeds during failover. As a result, you aren't actually increasing the effective range of the hardware, you are using it in a circumstance where it can't meet bandwidth requirements and uptime requirements. And, if you have the 5 GHz spectrum at the longer ranges, it probably makes sense to just use 5 GHz instead of a mix.

I think it makes more sense to use 60 GHz within the tested limitations instead of using 5 GHz failover to try to buy more range. If you need more range, it is probably more reliable to add hops or switch to a band that already supports it.

[All that said, I think most microPoPs can be supported with 60 GHz for LOS and 5 GHz for short NLOS links. I would not be opposed to an AP that did both and could have mixed clients or have clients that could do either.]

marekm · Sat Jun 06, 2020 7:18 pm

Adding 5 GHz failover to 60 GHz just "increases" the range of the 60 GHz hardware. The problem is - if you didn't have the 5 GHz spectrum available in the first place, then you will be operating at limited speeds during failover. As a result, you aren't actually increasing the effective range of the hardware, you are using it in a circumstance where it can't meet bandwidth requirements and uptime requirements. And, if you have the 5 GHz spectrum at the longer ranges, it probably makes sense to just use 5 GHz instead of a mix.

Sometimes the bandwidth and uptime requirements can be flexible. For a few hours or days of bad weather in a year, it can make a big difference - slow service / no service at all.
Having the backup integrated in the same device could also have an advantage where the backup could be enabled automagically only when needed (when 60GHz station interface is connected, disable the 5GHz interface) so both can be put in a bridge without causing a loop (and without using any easy to misconfigure protocols like RSTP or OSPF).

server8 · Sun Jun 07, 2020 12:47 pm

zdiv · Mon Jun 08, 2020 10:50 pm

Technical specifications are still absent.
Block Diagram, chipset, tx/rx specs, antenna gain, etc...
Those are really needed, not just nice to have.

something · Wed Jun 10, 2020 11:02 pm

Is this still 802.11ad based, or is it 11ay? Would be nice to know what the frequency range is too...

r00t · Thu Jun 11, 2020 6:30 pm

It's still 802.11ad, compatible with all other 60GHz products. Same frequency range. For $298.00.

karina · Fri Jul 10, 2020 1:18 pm

uberdome, the ethernet protection gland should not be plugged in yet, it can be on the cable, while you do the adjustment. After done, just slide it forward and screw it on

Im amazed you had to explain that lol

uberdome · Fri Jul 10, 2020 3:33 pm

uberdome, the ethernet protection gland should not be plugged in yet, it can be on the cable, while you do the adjustment. After done, just slide it forward and screw it on
Im amazed you had to explain that lol

I accept that this can be part the installation procedure, but I have not used this style of MikroTik gland seal so I do not know how far it extends. It seemed like a reasonable question. Further, we don't have any radios in our system currently that require partial disassembly of the radio or seals to be able to aim them. We normally complete the radio assembly entirely and then make adjustments.

marekm · Fri Jul 10, 2020 3:46 pm

Any news about real availability date?

server8 · Sat Jul 11, 2020 11:04 am

karina · Tue Jul 14, 2020 1:00 pm

nice update but a 5 Ghz Interface is still missing why?

Why do you need it incorporated? just sling up a couple of lhg5's in bridge mode and run a failover script somewhere. This way if a PSU fails you still have link, thats what I do anyway.
I am a great believer in one device one job.

cero · Wed Jul 15, 2020 4:30 pm

Good afternoon gentlemen ,

I want to link 2 sites (~700m line of sight) and was looking for to buy Wireless Wire Dish when I noticed there is an "improved" version ->
Wireless Wire nRAY.
Since I couldn't find any (proper) review of the "nRAY", can you please tell me if you have any experience with it / can compare it with "Dish" version ?
I saw the main difference in
CPU
nRAY: 64 bit 2x core ARM @1000 Mhz(88F3720) 6W vs
Dish: 32 bit 4x core ARM @ 716 Mhz(IPQ-4019) 5W
Size /shape:
nRAY: 261 x 166 mm
Dish: 391 x 222 mm

While the shape/size of the dish is not that important for me and neither the snow,do you know which one has the :
1. Highest output (up/down)?
2. Signal quality / rx/tx power ?
3. easiest calibration process ?
4. longest lifespan ?
Also please let me know if you consider other things important between the 2.

Thank you very much !
Best regards !
C

marekm · Sat Jul 18, 2020 8:02 pm

Why do you need it incorporated? just sling up a couple of lhg5's in bridge mode and run a failover script somewhere. This way if a PSU fails you still have link, thats what I do anyway.
I am a great believer in one device one job.

It's not about hardware failure but to keep services available (even if degraded quality) during bad weather. Having it all in one device would allow very fast fali-over. Just do the client side (add a feature to the bridge where link up on one port [60] disables another port [5] and vice versa), AP side is best done with separate 5GHz and 60GHz sectors which would also handle single-band clients (60GHz at <400m where rain is not an issue, 5GHz at >800m where 60GHz wouldn't work at all, use dual-band clients in 400-800m range where 60GHz works in good weather but sometimes fails in heavy rain). Single device is also better due to size/power/aesthetics constraints.

uberdome · Thu Jul 23, 2020 4:07 am

It's not about hardware failure but to keep services available (even if degraded quality) during bad weather. Having it all in one device would allow very fast fali-over. Just do the client side (add a feature to the bridge where link up on one port [60] disables another port [5] and vice versa), AP side is best done with separate 5GHz and 60GHz sectors which would also handle single-band clients (60GHz at <400m where rain is not an issue, 5GHz at >800m where 60GHz wouldn't work at all, use dual-band clients in 400-800m range where 60GHz works in good weather but sometimes fails in heavy rain). Single device is also better due to size/power/aesthetics constraints.

I think you make a reasonable argument here, but I'm not sure I am in agreement. I like the explanations that the problem is not about hardware failure and that a single device is more acceptable by the customer than multiple devices. That being said, I'm not sure I see a compelling argument for a device intended only for a narrow range (e.g., 400-800m). In my mind, if 5 GHz is enough to serve that range, it should just be used. If it can't serve that range, then it shouldn't be used for failover. I can't picture knowingly going through the effort to hookup a customer that would drop to an unsatisfactory connection during rain storms. It seems the better choices would be to add microPoPs, or use an alternative band.

marekm · Wed Jul 29, 2020 2:56 pm

Well, it depends on a specific situation, but here I happen to have a lot of existing and even more potential customers exactly in that narrow distance range (large numbers of many-houses-in-a-row), 5GHz spectrum is hopelessly overcrowded (about 5 competitors in the area, and available channels limited by nearby weather radar at 5650MHz), 60GHz offers about 10x higher speeds, most people want low cost, and heavy rain rarely happens exactly during peak usage hours (though the peak has become much wider due to the pandemic and lots of people suddenly working remotely - previously it was mostly Netflix in the evenings).

andreamark543 · Tue Aug 11, 2020 1:04 pm

Wow its great.

saaremaa · Tue Aug 11, 2020 4:23 pm

What is the maximum distance for a link outdoors with STABLE modulation 8? Not in the laboratory and not calculated.

pacman · Tue Aug 18, 2020 5:25 pm

I wonder if Mikrotik actually intends to make any of these during 2020? Am starting to have my doubts!

server8 · Tue Aug 18, 2020 9:16 pm

I wonder if Mikrotik actually intends to make any of these during 2020? Am starting to have my doubts!

+1

rzirzi · Mon Aug 24, 2020 11:22 am

WHEN it will be available at Poland? MikroTik Wireless Wire nRAY.

rzirzi · Mon Aug 31, 2020 9:38 pm

Anyone from MikroTik Team?
WHEN?

mistry7 · Mon Aug 31, 2020 9:45 pm

My Distri says end of September

roc-noc.com · Tue Oct 20, 2020 7:22 pm

They are shipping. I received my first allotment of nRAYs yesterday. Shipped them to customers. Kept one for my own install this week. The rest of my nRAY order should arrive around the end of November.

pacman · Tue Oct 20, 2020 8:29 pm

I received my first few too today. Happy Days! :-)

server8 · Thu Oct 22, 2020 9:32 am

I have seen that NRAY can be mounted rotated 90° changing antenna polarization this can help with rain and colocation?

Thu Oct 22, 2020 1:57 pm

I have seen that NRAY can be mounted rotated 90° changing antenna polarization this can help with rain and colocation?

Yes for experimental purposes, that's why we used horizontal Ethernet gland. Please keep in mind that dish design has drain hole, that needs to be sealed if used like that.

roc-noc.com · Sat Oct 24, 2020 6:57 am

Installed an nRAY pair today for building to building with distance of about 1/3 mile. The weather was wet and cold and we were losing daylight so I didn't spend much time there. Installation was quick and easy. I tried using the nRayAIM-DH1 scope mount with an $80 CenterPoint 6-20x50mm TAG/BDC Scope that I picked up at Walmart. I really like that option but the nRayAIM-DH1 mount is wobbly. They need to add some material to make it a tight fit. I slipped in my credit card as a spacer to tighten it up. I need to fix the mount for consistency and "sight in" the scope for it to be really useful.

The radios ship with a plastic clip-on gun type sight that I ended up using for a rough alignment since the scope was not adjusted for accuracy. That was good enough for this install. Signal was 90 to 100 on both ends.

I think nRAY is a winner and an improvement over the RBLHGG-60adkit. I just wish they made an accurate aluminum nRayAIM instead of the plastic nRayAIM-DH1 so I could trust my scope for alignment.

Tom

pacman · Sat Oct 24, 2020 11:12 am

Have to admit, I wish there was a facility in winbox to (temporarily) turn off beam steering on the nRay units. It is a bit frustrating aligning the unit and have beam steering messing around at the same time. With 70/80 GHz links, we can, and do, align them without any use of scopes - even over distances of 5 to 10 km. We do that just by monitoring how the signal level changes as we slowly/steadily sweep from left to right and down to up etc.

I suspect it is probably a hardware limitation (and thus there is nothing Mikrotik can do), but it is a shame the designers did not think of this. Perhaps the chip designers did not consider that companies such as Mikrotik would really want to increase the distances covered beyond 100-200m? Bit short sighted, in my humble opinion, if that is the case.

server8 · Sat Oct 24, 2020 12:59 pm

I have seen that NRAY can be mounted rotated 90° changing antenna polarization this can help with rain and colocation?
Yes for experimental purposes, that's why we used horizontal Ethernet gland. Please keep in mind that dish design has drain hole, that needs to be sealed if used like that.

Antonsb may you confirm that we can change antenna polarization mounting NRAY turned of 90°? Teorically the horizontal polarization 'd be stronger with rain event and maybe we can archive longer distance

roc-noc.com · Mon Oct 26, 2020 7:25 pm

I have seen that NRAY can be mounted rotated 90° changing antenna polarization this can help with rain and colocation?
Yes for experimental purposes, that's why we used horizontal Ethernet gland. Please keep in mind that dish design has drain hole, that needs to be sealed if used like that.
Antonsb may you confirm that we can change antenna polarization mounting NRAY turned of 90°? Teorically the horizontal polarization 'd be stronger with rain event and maybe we can archive longer distance

Yes, as antonsb already said. Look at the mounting. You can rotate 90 degrees but need to drill a new drain/vent hole on the new bottom and seal the old drain hole which is now on the side.

server8 · Mon Oct 26, 2020 11:39 pm

The question we are sure that rotating the antenna by 90° we 'll have an horizontal polarization?

wpeople · Fri Nov 06, 2020 7:18 pm

Have to admit, I wish there was a facility in winbox to (temporarily) turn off beam steering on the nRay units. It is a bit frustrating aligning the unit and have beam steering messing around at the same time. With 70/80 GHz links, we can, and do, align them without any use of scopes - even over distances of 5 to 10 km. We do that just by monitoring how the signal level changes as we slowly/steadily sweep from left to right and down to up etc.

I suspect it is probably a hardware limitation (and thus there is nothing Mikrotik can do), but it is a shame the designers did not think of this. Perhaps the chip designers did not consider that companies such as Mikrotik would really want to increase the distances covered beyond 100-200m? Bit short sighted, in my humble opinion, if that is the case.

you have the ability to disable auto and set specific sector:
interface w60g set wlan60-1 tx-sector=auto

modsx · Thu Nov 26, 2020 6:40 pm

Can anyone tell, what is the beam angle nRAY?

Thu Nov 26, 2020 6:49 pm

https://wiki.mikrotik.com/wiki/Manual:I ... cteristics

pacman · Thu Nov 26, 2020 7:25 pm

Interesting read.

In the Regions section it states that USA has channels 1 through 6 available. However, if you select USA you can still only "see" the usual 4 channels. This is of interest to us (in the UK) since we are now allowed to use the same expanded spectrum as they are in the USA.

mistry7 · Thu Nov 26, 2020 10:51 pm

@pacman
Probably only available at the us version

pacman · Fri Nov 27, 2020 11:22 am

@mistry7

It's possible, though have to admit I've never known Mikrotik do something like that before. More likely, I would have thought, is that either (a) it is a simple mistake on the wiki, or (b) they bought a job lot of 4 channel chips and need to get rid of those first, or (c) they have yet to implement the upper 2 channels in RouterOS.

I hope it is "just" a RouterOS issue that will be sorted (soon), as having access to those upper channels would be very useful.

xvo · Fri Nov 27, 2020 12:43 pm

If I remember correctly higher channels are available via CLI only.

server8 · Fri Nov 27, 2020 7:59 pm

I think that the chipset don't allow frequency over 67 GHz, CEPT in june 2019 has opened the usage here in europe up to 71 GHz so there is no difference from US regulatory.

Petzl · Tue Dec 01, 2020 11:13 am

The scope mount is not really stable when mounted correctly ... it still moves a couple of millimeters ....

erebusodora · Thu Dec 03, 2020 2:12 pm

What is MikroTik hardware answer to this https://eu.store.ui.com/collections/ope ... iber-60-lr
12 km Distances !!!

psannz · Thu Dec 03, 2020 2:36 pm

What is MikroTik hardware answer to this https://eu.store.ui.com/collections/ope ... iber-60-lr
12 km Distances !!!

I'd question the throughput and stability at 12km link length, considering you follow the local regulatory domains (e.g. max EIRP).
Considering how water influences the link stability, i also question whether those 12km can be achieved outside of a desert or similarly dry area while maintaining a reasonable SLA.

pepulis8008 · Mon Dec 07, 2020 9:52 pm

Hello,
can you tell me, what is tx-packet-error-rate: 1%? How can i influence value tx-packet-error-rate? Thank you for answer.

[admin@MikroTik] > /interface w60g monitor 0
connected: yes
frequency: 64800
remote-address: XX:XX:XX:XX:XX:XX
tx-mcs: 8
tx-phy-rate: 2.3Gbps
signal: 80
rssi: -54
tx-sector: 54
tx-sector-info: left 0.8 degrees, up 0.6 degrees
distance: 849.98m
tx-packet-error-rate: 1%

3freet · Sat Dec 12, 2020 5:39 pm

Hello,
can you tell me, what is tx-packet-error-rate: 1%? How can i influence value tx-packet-error-rate? Thank you for answer.

I would say the units are not perfectly aligned.

I have recently finished installing an nRAY pair. It was my first time dealing with 60GHz. At first I couldn't even make the units connect outside 100m range. I was very disappointed. Then I realized my mistake, the beam angle of these units is less than 1 degree! so every slight movement matters. Now I have a near perfect alignment at ~1200m.

> interface w60g align wlan60-1 
             connected: yes
             frequency: 64800
        remote-address: 08:55:31:XX:XX:XX
                tx-mcs: 8
           tx-phy-rate: 2.3Gbps
                signal: 95
                  rssi: -52
      10s-average-rssi: -53.5
             tx-sector: 33
        tx-sector-info: right 0.4 degrees, down 0.6 degrees
              distance: 1159.44m
  tx-packet-error-rate: 0%

EI3HG · Fri Jan 15, 2021 1:03 pm

I wonder can an nRay talk to an LHG60?
The reason I ask is I have spare LHG 60s ( I generally like to have at least one spare of any radio on the shelf)
I Guess I'll find out soon as I have ordered a pair to try!

r00t · Fri Jan 15, 2021 2:33 pm

All 60GHz stuff mikrotik released so far is compatible and you can connect them together as you like.

modsx · Fri Jan 15, 2021 8:49 pm

Wireless Wire nRAY (nRAYG-60adpair) disassembly in photos https://bit.ly/35HMQ3k

velk1001 · Fri Jan 22, 2021 1:04 pm

Thank you modsx for stripping the Nray, i am shocked just to see the very small phased array antenna at the heart of the Dish!

I was expecting multiple phased array board not just one.

Regards

Paul

Fri Jan 22, 2021 1:19 pm

Thank you modsx for stripping the Nray, i am shocked just to see the very small phased array antenna at the heart of the Dish!

I was expecting multiple phased array board not just one.

Regards

Paul

With Ignitenet you'll find a 60GHz USB Stick within the Antenna ...

velk1001 · Fri Jan 22, 2021 2:22 pm

Is the array vertical polarisation only as it looks like its horizontal too?

regards
Paul

modsx · Mon Mar 01, 2021 9:12 am

Be careful when adjusting the antenna with the existing solid mount bracket.
Its locking screws (hex head screws for bracket tightening), when rotated to lock, make a positioning shift of a few degrees. Tested with Optics and Signals.
Vertical fixation is weak. It has only two fixing screws that cannot be fastened enough to hold the antenna in place.
I have previously written to Mikrotik about the unfortunate design of this bracket and even sent it back for warranty, because the fixing screws tend to turn around.
P.S. With the optics I managed to adjust precisely. The nRayAIM-DH1 is not a versatile enough design for attaching various optics, but its design is precise.

velk1001 · Mon Mar 01, 2021 11:42 am

Watch out for the locking nuts stripping if you move it alot as well.

wpeople · Mon Mar 01, 2021 3:44 pm

We tested AF60LR units -they includes a new precise aiming mount - and a way faster to align the units.

Mon Mar 01, 2021 4:09 pm

We tested AF60LR units -they includes a new precise aiming mount - and a way faster to align the units.

The main advantage are additional channels which make a difference in 60GHz...

wpeople · Mon Mar 01, 2021 4:32 pm

The main advantage are additional channels which make a difference in 60GHz...

Not only only the additional channels, but they can "shrink" the used bandwidth to half, and use the same power on half channel - with less, but more robust thruput.

Also, it's a big difference if you can aim a link in half our or somthing 2x2 hours thanks to the mount...

r00t · Mon Mar 01, 2021 7:02 pm

If Mikrotik had radios with multiple GigE ports, 2.5GbE or SFP+ slot, then it would be possible to really use entire 802.11ad channel. But with just single GbE it's a waste of channel bandwidth. So far 60GHz interference is not a big problem, but for how long? What will happen when Terragraph is widely used?

syadnom · Tue Mar 02, 2021 8:27 pm

If Mikrotik had radios with multiple GigE ports, 2.5GbE or SFP+ slot, then it would be possible to really use entire 802.11ad channel. But with just single GbE it's a waste of channel bandwidth. So far 60GHz interference is not a big problem, but for how long? What will happen when Terragraph is widely used?

Nothing will happen. 60Ghz has such a small fresnel zone and tiny side lobes that the only interference you need to worry about is the AP you are pointed directly at. Even if you are doing a PTP shot in the middle of a terragraph deployment, as long as you are not pointing directly at one of those APs you're fine. Heck, you can co-locate on the same channel just a few feet away even if you are barely off azimuth.

EI3HG · Fri Mar 05, 2021 2:58 pm

I saw this link this morning for it: https://mt.lv/WWnRAY

I'm excited about the complete assembly and aiming mount.

It appears to be an attractive size, but seems long. Most of our situations where we install these will be on the sides of buildings, so the larger the distance from the wall, the more significant the installation appears.

It seems to me the configuration as shown will be just fine much of the time, other times I will be wanting to get it closer to a wall. This could possibly be achieved with 2x design changes:

Possibly a sideways or alternative cable exit can allow the back of the radio closer to a wall

A mount that can be oriented longitudinally instead of axially

I do have some potential concerns about function:

How are these screws accessed for aiming (It appears the gland seal is blocking access)?
Annotation 2020-05-22 100705.png

It seems there could be an additional hole to allow rotating the mount and reducing the front-to-back dimension. It appears this would have more clearance for those screws from the previous question. Any thoughts?
Annotation 2020-05-22 101038.png

Thank you, Chris
Dear Mikrotik!

Please modify the ethernet cable place. You do the same mistake, when you did with the grooves. If the device on a high tower and looks down, the water goes into the device. Please modify the connector down by 45° or 90°.

Yes we just found out this happens after only a couple of days! tilting down 1.7 degrees, water got in. it needs an angled shroud

Fri Mar 05, 2021 3:07 pm

Every WISP had water inside horizontal cable entries. It is just a matter of time until water finds its way inside.
We suffered from this years ago. Cable entries *always* has to be at the bottom.

doneware · Sun Mar 07, 2021 1:19 pm

Even if you are doing a PTP shot in the middle of a terragraph deployment, as long as you are not pointing directly at one of those APs you're fine.

even with p2p setups in terragraph - trust me, i've done that - you can have issues with p2p links in a multi hop setup. you have a separate radio dealing with the next hop, you might even have almost 80mm of aluminium shielding on one b2b radio setup, hell, you even have an entire link working in the opposite tx/rx schedule (terragraph is TDMA-TDD) and you can have interference in the "urban canyon" scenario. well, it doesn't necessarily needs to be that urban.

consider the setup below:

A - - - - - - - - B C - - - - - - - D E - - - - - - - F

each dashed line represents a link with 60-100m length. terragraph's official reach is 200m, but we successfully built 250m links, the longes one operational link to a client was 276m long.
each letter is a radio in a 'P2P' setup, albeit TG doesn't really support strict P2P setups - it's more like a MP setup with just 2 radios. anyways. TG is a single frequency network, so every single link is operating in the same 2.16GHz channel, but you have TDMA-TDD with GPS sync in the primary nodes and you can set a radio to operate at a certain 'polarity' of even or odd. that means when 'even' radios transmit in the network, the 'odd' radios receive.

now assume the following polarities:
A - odd
B - even
C - even
D - odd
E - odd
F - even

that means, A,D and E will transmit at the same time. and we can get interference on F, as it will see transmission from E (intended) and A (unintended). this obviously would not be relevant with 1km long links that we can build with the nRAY, but clearly shows how easy is to get interference in the same channel. i rebuilt this setup with mikrotik 60GHz devices and used alternating and even 4 coloured channels to get around this phenomenon.
plus since 60GHz behaves 'like light' you will have lots of reflections and even refractions from various unexpected things. this is pretty clearly described here in a white paper we've created.
https://telecominfraproject.com/wp-cont ... t-2019.pdf

you can somewhat get around this by using different different golay sequences, but that is out of our control with Mikrotik's current implementation, and essentially the MAC and PHY layer of terragraph is significantly different from vanilla .11ad. it was quite some effort to get the 'terragraph pre-ay' extensions with TDD and stuff into the .11ay standard.
also, the TG units i've worked with had a 60GHz hw based on some since then discontinued broadcom silicon, and had a largely different RF stage compared to the QC derived one we see in mikrotik. bottom line: (almost) no elevation control and +/- 1.5 degree vertical FOV. a separate TX and RX array, each the size of a twix bar.
and the official installation guide - backed by field practices of ~2 years - suggested to 'bubble' (level) the links regardless whether it's end wasn't at the same elevation.
so yes, the fresnel zone is tiny compared to the sub6 bands, but physics is still there. with TG you had 2.5-3 degree wide beams (pencil-beams) in the overall +/-35 degree horizontal FoV. but then it is kind of understandable why there wasn't much vertical FoV either.

as for the side-lobes: they are there, in many cases the side lobes were used to connect customers in 8-10m distance as they were outside the main beam's FoV.

uberdome · Mon Mar 08, 2021 1:44 am

Yes we just found out this happens after only a couple of days! tilting down 1.7 degrees, water got in. it needs an angled shroud

Is there any chance you have pictures of when it was installed before it stopped working?

When you took it down, did water pour out of it? If so, how much?

doneware · Tue Mar 09, 2021 12:09 am

Thank you modsx for stripping the Nray, i am shocked just to see the very small phased array antenna at the heart of the Dish!

I was expecting multiple phased array board not just one.

the tiny board was FCC'd back in 2019: https://fccid.io/TV7WAPG60ADM
that's a very compact customised QCA6310 RF stage which is used in all mikrotik .11ad products.
the same as we have in LHG60G. can't say much about the baseband, but most likely a QCA6335 as in LHG60G.
i guess the magic is with the dish and the internal 'reflectors' under the radome.

with regards to frequencies so far no devices support channel 6 in the mikrotik range.

velk1001 · Tue Mar 09, 2021 10:13 pm

Its a shame that there is no official information how the beam patterns are set for different TX-sector setting?

Regards

Paul

modsx · Thu Apr 22, 2021 3:23 pm

I have concluded that the optics on the nRayAIM-DH1 are much more accurate than the signal adjustment via / interface w60g align wlan60-1. If adjusted through the optical cross, the antenna works but the software wants angle correction, but if it is adjusted by at least 0.5 degrees, then the link breaks. I have not yet gone into how to explain it, but the fact.

modsx · Sun May 02, 2021 3:19 pm

I have seen that NRAY can be mounted rotated 90° changing antenna polarization this can help with rain and colocation?
Yes for experimental purposes, that's why we used horizontal Ethernet gland. Please keep in mind that dish design has drain hole, that needs to be sealed if used like that.
Antonsb may you confirm that we can change antenna polarization mounting NRAY turned of 90°? Teorically the horizontal polarization 'd be stronger with rain event and maybe we can archive longer distance
Yes, as antonsb already said. Look at the mounting. You can rotate 90 degrees but need to drill a new drain/vent hole on the new bottom and seal the old drain hole which is now on the side.

Please provide a reference stating that horizontal polarization is better in rainy weather? As far as I know, vertical polarization will be better in the rain and this means that nRAY does not need to change anything.

velk1001 · Sun May 02, 2021 4:12 pm

Does anyone know why you cant use the alignment tab on the routeros with the Nray, i have only managed to do it though the Terminal command line?

velk1001 · Wed May 26, 2021 1:34 pm

Due to new power restrictions in the UK does any one know how the lower the power output of the NRAY from 55dBm to 40dBm?

Regards

Paul

Who is online