Realistic signal to noise ratio

General
1

I have Two halowlink-1 units on towers that are 7 miles apart total line of sight. Both units are on 8.5 db gain antennas with about 100 ft of LMR 400 cable. I am using a 1 Mhz channel showing 88/98 on both ends. I am showing 0.7Mbp throughput. Trying to open the config page remotely fails most of the time. Is this expected with a 10db signal to noise ratio?

2

@kilosierra A few of things about your setup:

  1. You might be getting close to the FCC regulatory limit for max EIRP of 30dBm for 915MHz ISM. Something to keep in mind.

  2. If you do have the head room through things like insertion loss and the length of the cable itself (ChatGPT says it’s about a 4dB / 100ft), you can gain another ~2dB of TX power on HL1’s by enabling the 4.3V VFEM under the advanced settings of the HaLow interface.

  3. According to the datasheet for the Azurewave AW-HM593 which I believe is the module with our MM6108 in the HL1’s, it has a receive sensitivity for MCS 0 / 1MHz of -104dBm, typical. So, your RSSI of -88dBm is in spec. However, only a 10dB of SNR is probably what is getting you as far as achieving better performance with http traffic.

  4. Based on your throughput of 0.7Mbps, assuming that’s a UDP iPerf test, that probably means you’re hitting MCS2 or MCS3. (https://mcsindex.net/ sub-1GHz tab at the bottom). If you can gain a couple more dB of SNR by trying different channels, moving them higher on the towers, and/or the 4.3V VFEM, you might bump things up enough to have a noticeable improvement.

  5. We’ll see if any of my colleagues have more to add, but your http performance at the distance seems expected. How does ssh perform? What is your use case for this distance?

Finally, we’ve announced the HaLowLink 2 which has a more improved radio. They should be shipping soon from Mouser and would be an easy drop in for your setup. That may get you slightly better http performance.

3

I forgot to mention that I have a cavity band pass filter on both sites with an insertion loss of 2 db. The units were totally non functional until I added the filters. There are cell sites on both towers so I assume it was just swamping the front end of the halowlink-1’s. I am aware of the 1 watt /4 watt eirp rules and will do the math.

How much better will the new radios be?

I have moved around the band and saw little change. Doing the band sweep on the Halowlink-1 always shows no signals other than the device. I have looked with a real spectrum analyzer and there as some traffic on 900 but not much. The local power company is doing meter reading and they hop the whole band. I will try bumping the power up today and see if there is any change.

FYI I am replacing a meshtastic network I had in place, Just for conversation, meshtastic had a perfect link on the exact same antennas.

4

I would argue you have better than meshtastic performance if you’re getting 0.7 Mbps throughput in your current setup since mestastic runs on LoRa which is super low bandwidth. You’re probably not going to be able to find lower noise anymore than you’ve been able to up to this point. So, your best bet is to make sure you’re getting as much gain on either end as possible up to regulatory EIRP limits. But, you’re right at the edge of HaLow’s range.

We did some YouTube videos on range tests and long range examples. The 16km (~10miles) one was in an absolutely perfect environment. Here are some links about them: