LoRa Range problems

[Andrew Lindsay]

Hi,

I have a Conduit AEP model and a number of mDot modules that I am evaluating. I have a mDot with SMA connector and the 3dBi antenna on the mDot dev board along with a simple LCD display and a serial GPS module. The idea is that I can move around logging the position and sending this in the message to the Conduit. The Conduit is using the supplied 3dBi antenna and is mounted in my attic at a height of around 7 to 8m above ground level.

On the conduit, the Node-RED flows are setup to receive the message and log the payload to a database where the positions and RSSI values can be displayed on a map. From the Node-RED debug output, the payload is around 72 bytes.

The problem I am having is that the range is not great. At most I have seen around 200m which is a lot less than what I was expecting.
Are there any recommendations for positioning the Conduit, types of antenna, mDot antenna and general LoRa parameters in order to get the range up to advertised levels?

Another thing I’m seeing is that the communications seems to be mainly 1 way, i.e. it is received on the Conduit, but the response is not received every time by the mDot.

Thanks in advance

Andrew

[Brandon Bayer]

Hey Andrew,

In general, to get maximum range, you’ll want to set Conduit TX power to 26, mDot transmit power to 20, and data rate (spreading factor) to 12 (10 for North America). However, setting the data rate to 12 will limit your payload to 11 bytes. So you might need to use 10 or 11 if you can’t slim down your payload.

Also, there is a bug in AEP version 1.0.25 that affects EU mDot packet reception (join accepts and downlink packets). If you open a case in our support portal, we can send you an updated beta version that fixes this.

-Brandon

[Andrew Lindsay]

Thanks Brandon, the Conduit is running 1.0.28 as updated by deviceHQ if that helps.

I also changed the mDot spreading factor using dot->setTxDataRate( mDot::SF_9 ); but this doesn’t seem to have changed as Node-RED still reports a spreading factor of 7 in the debug window.

Slimming the payload down is the next job as this is too large at present due to sending gps lat/lon values.

Regards

Andrew

[Brandon Bayer]

Andrew,

1.0.28 still has the 868 bug. I just double checked my setup. Changing the spreading factor on the mDot correspondes to the datr field in the debug window.

"datr": "SF10BW125" for SF10

-Brandon

[Andrew Lindsay]

Brandon,

Thanks, I’ll request an updated firmware. Not sure why I’m not able to set the spreading factor. Could be a code issue as it works on the very basic library example.

Regards

Andrew

[Brian Wyld]

Hi,

I’m trying pretty much the same setup, and had exactly the same range issues (I would expect a factor of 20x what I actually see..)

With the gateway and the mdot on ajacent desks, the RSSIs (as reported by AT+PING and AT+RSSI) are -37 on the gateway and between -40 and -50 on the mDot.
Gateway tx power is 26, and mDot is 20….. ie the max for both…

this is with SPF 9 on the mDot.
mDot is version 0.1.4 and conduit is 1.0.33 ie the latest versions from the web site.

Andrew, did the change to your setup increase range?

thanks

Brian

[Brandon Bayer]

Brian,

Those signal levels seem pretty normal. What range are you getting? And what environment are you testing the range in?

The marketed 10+ mile range is only line-of-sight. Any obstacles will greatly reduce your range. What we’ve been seeing is 1-2 mile range with obstacles (buildings/foilage, etc) and with the Conduit antenna placed strategically (above as many obstructions as possible). This sort of range is very good compared to Wi-Fi & ZigBee.

-Brandon

[Andrew Lindsay]

With my current setup I have been able to achieve 1200m range between the Conduit AEP mounted in the highest point in my roof space, antenna pointing vertically down (as the mounting holes indicate this way) and a mDot.
Both are using the supplied Multitech 3dBi antenna. Spreading factor was 9, bw 125KHz.

This range was achieved through a large housing estate, trees etc so not line of sight. Other parts of the town I am only able to get about 500m but this may have been down to the mDot and antenna being inside a car. More testing is still being done.

Regards

Andrew

[Brandon Bayer]

Andrew,

SF_10 should get you a little longer range, but your payload will be limited to 11 bytes as noted on the Introduction to Lora page.

[Andrew Lindsay]

I’m trying SF_12 now as I’m in the EU there isn’t the restriction of 11 bytes payload size, payload is currently about 16 bytes but hoping to get this down by re-coding it as a binary payload instead of minimal json.

btw, almost every time I look at this site, there is new information!

Thanks

Andrew

[Brian Wyld]

Tried again, with both the standard 3Dbi antenna and with a wide band 7DBi one..

The problem is that the rssi reported by AT+PING varies between -13 and -28, for approx 1m of separation and no change in the environment….

Any ideas why the rssi would not be stable?
thanks

Brian

[Brandon Bayer]

Brian,

I took another look at your AT+PING and AT+RSSI results in your last post. It looks like you have a defective MTAC-LORA card as the AT+RSSI result should be higher than AT+PING with those transmit powers.

We actually just discovered this MTAC-LORA defect last week which is caused by a defective solder joint. They pass our production tests, and then fail after some amount of use. We have a plan in place to remedy this so it doesn’t happen in the future!

To get it replaced, open a support case stating I diagnosed you with a defective MTAC-LORA and reference this thread.

-Brandon

[Brandon Bayer]

Andrew,

You can test your MTAC-LORA using these steps:

1) Place Conduit & mDot right beside each other
2) Take antennas off mDot & Conduit
3) Set Conduit tx power to 26
4) Set mDot tx power to 20
5) Have mDot join network
6) On mDot issue, AT+PING then AT+RSSI and send us the output of these two commands.

-Brandon

• This reply was modified 9 years, 1 month ago by Brandon Bayer.

[Andrew Lindsay]

Hi Brandon,

I’ll give this a try and report back.

thanks

Andrew

[Andrew Lindsay]

Hi Brandon,

Results, No Antennas approx 6 inches between SMA connectors:

at+ping
-111,4.5
OK
at+rssi
-119, -119, -65, -87
OK

No Antennas, approx 3 inches between SMA connectors:

at+ping
-108,4.0
OK
at+rssi
-104, -119, -65, -90
OK

Thanks

Andrew

[Brandon Bayer]

Andrew,

Unfortunately it looks like your MTAC-LORA is probably defective also.

To get it replaced, open a support case stating I diagnosed you with a defective MTAC-LORA and reference this thread.

-Brandon

[Andrew Lindsay]

Brandon,

Thanks. Is this a simple solder fix or defective components? If solder joint as you’ve previously pointed out, I have enough years soldering experience to attempt a fix.

I’ll also raise a support call too.

Out of interest, what should the ping and RSSI values be for the ranges I used?

thanks for your help

Andrew

[Andrew Lindsay]

just looking at the MTAC-LoRa photos….metal shield so guess not an easy fix.

Andrew

[Brandon Bayer]

Andrew,

I can’t give a range as the rssi varies by orientation, proximity, etc, especially without antennas. The main thing is the AT+RSSI result should be higher than AT+PING in this situation because the Conduit is transmitting at 26 dBm and the mDot at 20 dBm (or less).

-Brandon

[Lawrence Griffiths]

Brandon run the same test my results

at+ping
-113,5.0
OK

at+rssi
-123, -123, -102, -115
OK

What should we be looking for here?
As it might help in the future when trying to diag issues.

BTW I can never get an mDot to join without an Antenna even at a few inches!

Lawrence

[Andrew Lindsay]

Looks like I’ll be without a LoRa card in my Conduit for as long as it takes Multitech to send a replacement after sending mine back to them.

Guess thats going to slow down the implementation of The Things Network.

Andrew

[Brandon Bayer]

Lawrence,

As long as you had both antenna’s off (or both on), Conduit tx power at 26, and mDot tx power at 20, then your MTAC-LORA also needs replaced. You can open a support ticket to get it replaced.

We are really sorry about all these defective parts! We have fixed the problem and are doing more post-production tests to catch these types of things in the future!

-Brandon

[Lawrence Griffiths]

Brandon both antenna’s off

Andrew are you saying you have to return the defective mCard before a replacement is shipped? If so I will have a word with UK office as to get us some replacements now.

[Andrew Lindsay]

Hi Lawrence,

I got an RMA number and form from the support site and a UK address to return the MTAC-LORA card. I forgot how small it is!!

I’m guessing if we had a support contract then we’d have replacement parts without having to send the faulty part back first. I’m hoping they can turn it around in less than a week but its all downtime and for a fully operational network this wouldn’t be acceptable.

If there is anything that can be done to speed this up then that would be great.

thanks

Andrew

[James Coleman]

Brandon,

I changed my mDot LoRA transmit power to 20 and the Conduit transmit power to 26.
Spreading factor is 10 which is the maximum for US.

I am now able to achieve a distance between 0.4 miles to 0.5 miles depending on location. What is required in order to increase the distance to 1 to 2 miles?

Can this be achieved with the current mDot LoRA radio and Conduit?

Thanks,

James

[Brandon Bayer]

James,

Was the ping & rssi results you posted with Tx power 26/20? If not, please run that test again and post the results. If it was, then your MTAC-LORA needs replaced (open a support case).

-Brandon

[Jose Olvera]

I just got a new box ( Multitech Conduit), the LoRa is set for band 1 (915 Mhz)
the TXpower = 20 dBm

question/problem : I cannot ‘see’ the RF signal

setup : I have an Spectrum Analyzer and using an antenna expecting to pick
the signal from the LoRa. The antenna can pick some other signal in the air
near 915 Mhz, but …. it seem ths box is not transmitting

[Brandon Bayer]

Jose,

Have you established mDot to Conduit communications and vice versa? The Conduit will only transmit after downstream data is queued for a mDot and a packet is received from that mDot.

-Brandon

[wojtek t]

Brandon,
My result without antennas:

AT+PING
-101,11.5

OK
AT+RSSI
-93, -97, -93, -95

Can I assume that my module is alright?
Wojtek

[Brandon Bayer]

Wojtek,

Yep, those numbers look good!

-Brandon

[Author]

Posts