Not specifically a Mikrotik question but reaching out for those with experience setting up point to point connections with the LHG 5 ax or similar setups over long distances. I've done some previous work with directional wifi but only over relatively short ranges (25-100m) so getting the initial alignment was fairly easy to eyeball and then I could tune it once I had a connection.
This time I'm trying to setup a connection over about 1km so eyeballing it is a little rough. Can anyone recommend tools or techniques for precision aiming over these kinds of distances? According to GPS data I need to be 2° off N (or south from the other end), but I can't find any easy way to attach a compass to the LHG to see what the actual direction its pointing.
The overall principle seems clear, and I have the iOS compass built-into the iPhone which is (I think) of similar quality as the NOAA one linked, but the trick is attaching it somehow to the LHG since there are no flat areas that I can hold the phone against to ensure that what’s showing on the screen reflects accurately the azimuth of the antenna.
I would venture to say that a LHG5 has an almost flat back side, where you can (optionally) screw on the QMP-LHG: https://mikrotik.com/product/QMP-LHG
and that most probably that flat surface is oriented at 90° from the beam.
So you need a high tech tool such as a STRAIGHT flat bar (and add or subtract 90° on the compass) applied to that surface and capable to prolong it enough to be able to push your phone to it.
Since the iphone (or was it a Samsung ) is a "rectangle with rounded corners" if you keep it in adherence to the bar it will be correctly oriented.
Or get the next generation tool , which consists in a flat bar BUT folded at 90°, forming a "try square" (but a flat one) you insert one side between the LHG and the pole, keeping it in contact with the backside of the LHG and align your phone against the other side.
Since the LHG5 is Ø 391 mm in diameter you need at least one side to be around or more than 25 cm long, these tools are usually sold as "shelf supports" and the problem is that they tend to be only sold in pairs, example: https://m.media-amazon.com/images/I/61v1qXubxyL._AC_SX679_.jpg
Hmmm - In my head I was focussed on the front side, but you're right that it makes more sense to try and attach something on the back where there is that flatter area. One of them isn't yet up on the roof so I'll go pick that one up from the other site and see what I can cobble together from the various bits of bent metal I have kicking around here.
Hmmm - spare rack wings might do the trick and I have few of those...
With "only" 1 km distance they should be more than enough to get the direction accurate enough to be able to use the software for alignment.
I have no experience with the LHG, but from what I understand from pictures, it doesn't come with a "proper" pole mount, i.e. you can rotate it in the horizontal plane but there are no provisions to orient it vertically accurately (approximately it can probably be done using spacers between the pole and the clamp) so I believe you will need a "better" mounting bracket, such as the mentioned QMP-LHG or similar.
According to the elevation displayed on the compass app, I’m going to be within 2m of difference so that aspect (thankfully) should be moot. For the mounting I just realized that this is a perfect use case for the 3D printer. A flat piece with either holes for passing bolts or integrated printed bolts to go into the mounting harness and then a flange with borders to hold the phone against coming out beside the pole…
Multiplied by (don't know the exact size of the back square (actually the distance between the two grips) but:
50 mm x 0.002=0.10 mm
60 mm x 0.002=0.12 mm
70 mm x 0.002=0.14 mm
80 mm x 0.002=0.16 mm
So you can do with a single (small) piece of (alluminum or brass) tin of adequate thickness.
In practice I would get a 0.05 mm thick strip and put as many pieces as needed, but most probably there is no need of vertical correction.
That is, in the actually impossible case that the actual verticality of the pole is perfectly correct.
Curvature of the earth should be irrelevant:
h=d^2/2x6,371,000
1,000,000/2x12,742,000=0.08 m
Document for the (original?) LHG5 contains antenna patterns. And it shows that that device has antenna beamwidth of around 5° both in vertical and horizontal direction. So aiming to around 2° precission should be more or less fine. I guess that antenna of LHG 5 ax is pretty much identical to it.
@jaclaz You won't believe, but it is possible to have almost perfectly horizontal link in hilly/mountainous area ... if aiming across a valley. Theoretically link would be horizontal when aiming through a mountain as well, but I'm pretty sure it would work much worse
It's a little terrifying holding my phone on the plate while up on the pole and hoping it won't slide off if I sway a bit while tightening the straps. A little bit of double-sided tape might be a useful.
not sure what the problem is - 1km with 2 x LHG is nothing, they will link up even if aimed into the sky. while working for WISP, we would get a link of 15km just by barely aiming a lhg to the base station, and then fine tuning it for max signal. it'd take max 5 minutes.
if we had to aim a ptp link then we'd spend a bit longer to perfectly aim them both.
when i did a 60ghz link last year, we aimed each side at once (not both at once as we didn't have enough people) and they linked up to 60ghz on the 1st try, we just had to correctly allign the other side. it was a 1.7km link. 5 ghz linked up still with the antenna hanging on the pole on its cable.
Hmmm - I configured the two locally and then installed one at the main site and when I tried to connect from the ground at the destination site (waiting for a ladder) I got absolutely no connection.
Interesting. Hopefully I'll be able to get a ladder on the other site to get higher up as I was holding the LHG over a wall trying to get a signal - not ideal.
I'm not sure if LHG can be "aimed at the sky" ... its antenna, (according to document I linked in my previous post) is classic high-gain narrow beam one. Unlike 60GHz solutions it doesn't have antenna array built in which allows for "software aiming".
So, as I wrote before, one has to get down to around +- 2° correct azimuth and elevation for optimal operation of such PtP link. Depending on distance it might be enough to point with precision of +-10° to get link going (which then allows for fine alignments ... to maximize signal level and, even more importantly, CCQ, by observing actual values).
Sometimes it may make sense to aim PtP link antennas "to the sky" ... in order to avoid interference which might originate in direction close to other PtP end point. The idea is to move move main lobe of antenna diagram away from interferer while keeping other link endpoint within (or near) main lobe. Sometimes pointing farther away from ideal direction makes sense if antennas have pronounced "side lobes" (which is not really the case with LHG antenna) and one uses those side lobes to actually make the link. But all of these cases are marginal (niche) cases, in most cases one wants to do the "main lobe" thing.
) is a "rectangle with rounded corners" if you keep it in adherence to the bar it will be correctly oriented.
, which consists in a flat bar BUT folded at 90°, forming a "try square" (but a flat one) you insert one side between the LHG and the pole, keeping it in contact with the backside of the LHG and align your phone against the other side.
