couple of basic questions....

RouterOS Beginner Basics
1

I was doing some basic testing of a bridged wireless link between two buildings (2 mile distance) this weekend over 900Mhz (SR9 cards) with a RB433 and 15dBi antenna. I have a couple of questions…

  1. The roof of Site-B is eye level with the upper parts of many trees. However, we found that we got a better signal when the antenna was just sitting on the roof, as opposed to when we put it 20’ into the air with an adjustable pole. The pole was NOT anchored or steady, it was just held by hand so it’s safe to say the antenna was moving around slightly in the air. Is it likely that this is the reason the signal was worse the higher it got? The noise floor seemed to also worsen as it went higher according to what Winbox was reporting for the link. I can’t think of a good reason the signal got worse when it more easily cleared the trees, unless it was because the antenna was moving/swaying in the air compared to be steady/solid just resting on the roof itself.

  2. We were seeing SNR values of around 0-4db when it was sitting on the roof with CCQ values around 30%. Received signal was around -89dBm on average. The dishes were not aligned at all on either sites, it was basically just a best-guess. Is it likely that I would see a worthwhile improvement from a good alignment on them? What’s the best way for an alignment? a GPS? How can I determine the best direction the antennas should be facing other than ‘best guess’?

  3. We also tested by having the Mikrotik boxes about 300 yards away (instead of using the 2 mile link). We saw the SNR shoot up to 46dB and the receive signal shoot up to -43dBm. MUCH better values obviously, but the CCQ only rose to around 60%. How concerning is this? I’m not sure I understand CCQ very well.

Thanks.

2

My thoughts?

  1. The higher the antenna became, the more noise and interference it was able to see from other sources. When you lowered the antenna, you shielded it from these other noise sources, thus the signal to noise ratio improved. This is a common trick to use, using natural terrain or obstructions (or trees in this case) to reduce noise and interference.

  2. A GPS won’t really help you align the dishes as you need to know the direction to point them in, not find out where you are. Hopefully you already know where you are!? :slight_smile:
    Setting up a link is dependent on a few factors. If you have sufficient persons to man both ends, then it becomes easier. If you have a means of communication (such as radios or cellphones) between the two teams then this helps too!
    If you can see the other end, i.e. it is LOS, then this is clearly much easier. One of the benefits of 900MHz is you don’t need perfect LOS, so I am guessing you can’t see each end. In Which case you may find a map and a compass will work better. Once you have both ends seeing each other, you can then start to move the dishes (one at a time!) for a better signal. It is important that only one end makes changes to the dish alignment at a time!
    If the two heights are different, don’t forget you will also need to finely adjust the elevation as well.
    Before leaving the two sites you should try checking the link quality for both horizontal and vertical polarisation. With fixed point to point links, horizontal is often better due to the higher quantity of noise and interference being vertically polarised.

  3. CCQ is a measure of packet throughput. If 100% of all packets is being received, this is perfect and leads to a 100% CCQ figure. If packets are being re-transmitted (due to noise/interference/distortion/reflections) then the throughput falls and the CCQ figure falls.
    Check your TX power when doing local tests with the antennas close together, if you are transmitting with too much power you may be overloading the Receiver’s RF Pre-Amplifier stages of the cards and thus causing IMDs (InterModulation Distortion) within them, this will cause packet loss and a lowering of the CCQ.

Ron.