I noticed someone else also talking about this in another thread:
I was driving around the past few days and was disappointed that the gateways in my city received so few packets from the One running the default tracking software. I then wrote my own code to make sure the RN2483 is driven correctly (code on my github page).
That didn’t help, so I decided to program a serial bridge into the One so that I can manually send commands to the RN2483 from my computer and then see what the responses are. The same set of commands were also sent to a bare RN2483 module. A gateway 3km away picked up the bare RN2483 (with a random piece of wire as antenna), while the Sodaq One with it’s provided antenna wire was not picked up.
This led me into investigating the RF path on the One. I noticed the RF Low and RF High outputs are connected to each other with 0 ohm resistors. This can’t be good as the different matching circuits for the two RF chains will interact and cause a mismatch. I therefore removed the 0 ohm resistor closest to the RF Low output on the RN2483 as I am only using the RF High output (868MHz). After this the One was still not received by the gateway that received the bare RN2483.
Adding extra “stub” lines to the RF path also changes the impedance, causing less energy to flow to the antenna. On the One there are 0 ohm resistors connecting both the UFL connector and the pad for an SMA connector. As both are connected to the RF path, the one will form a stub while the other is in use. As I’m only using the UFL connector, I removed the 0 ohm resistor going to the SMA footprint. After this modification the One was received by the gateway.
Having all these 0 ohm resistors on the One connected can’t be correct. Is this a bug in the final design, or are users expected to remove the unnecessary 0 ohm resistors themselves? Removing these 0 ohm resistors are not that easy.
I went right ahead and picked up my hot air soldering station and tried out your proposed solution. After that I went on a test drive and noticed a small increase in range. I was able to connect to my TTN gateway (for the moment is lying in the house, waiting to be installed up high outside) from 1,2 km away. (Rural environment, not a lot of buildings in between)
I ordered pigtail to sma cable and a bigger antenna to test out if this influences the SodaqOne range. When I receive this I will be doing some more field testing.
EDIT: Got my pigtail to sma connector and connected this antenna to the SodaqOne. Did another field test but did not notice an improvement in range… Would the pigtail to sma- connector be a bottleneck? Soldering on a SMA connector to the SodaqOne maybe is a better idea?
I’d like to offer a suggestion - but note that I don’t have a SODAQ One to test this idea …
We’re talking about monopole antennas here. For a monopole to follow its theoretical model, it should be sitting above a large ground plane. “Large” is relative to a quarter wavelength which is about 9cm @ 868MHz.
The SODAQ One is nice and small, so much less than a quarter wave length, and this means that the fields around the antenna have nothing to react against.
I’d suggest tacking on a second piece of wire of equal length, but connected to ground and positioned to go the opposite direction away from the antenna. This will make a dipole effect.
Generally loop antennas are better for small devices as they don’t need the ground area.
I sadly have to agree that the range on the Sodaq ONE is very very poor. Even with Aircell7 cabling and a 14dBi antenne with a gateway in a high tower (nearly line of sight) i wasnt able to send packets from 1.2km away.
did you contact sodaq on this and what was their reply?
Ill be doing more tests the coming week to see the range i can actually make, but so far i am quite disappointed.
Can you give an estimate how much the range of the One has improved by removing the 0 ohm resistors? And given your statement “removal is not that easy”, exactly how DID you remove them?
When using a hot air soldering iron, it is not that difficult. Just heat up the resistors and hold/pull them off with a smd pliers. I can even put them back on with this method. Using a traditional soldering iron it is more difficult: One method is to apply a lot of solder so both ends of the resistor is melted, then pull off or gently knock the pcb to your desk, the resistor will drop off. Remove excess solder with desoldering bread.
I did @jpmeijers 's mod with a 16w soldering iron which was quite easy to do. Just heat up the resistor with slight side pressure and it comes off after a little waiting. I also did not destroy the board, it is sending packets. Now see if the performance is better.
I also removed the resistors, but I don’t see a significant change in RSSI or SNR reported by the gateway. The RSSI is between -110 and -117. The SNR between -10 and -14. I connect to a gateway at 300 meter and there are several houses in between.
The 0 ohm resistors, which in real life looks like small black boxes with a white 0 on them, and like zigzag lines in my sketch, are just bridges between two points. They are like those jumper pins you used in your computer 20 years ago. Removing the 0 ohm resistor is like pulling the jumper off and breaking the connection.
If you are using TTN, or any other LoraWan network, you have to remove the resistor marked in RED.
Then you need to choose if you are going to use the u.fl connector, the one where the antenna wire that came with the One connects to. Or are you going to use the copper pads to which you can solder an SMA connector or wire.
Going to use U.FL: Remove the resistor on the left (marked in BLUE) which goes to the SMA pads.
Going to use SMA: Remove the resistor on the right (marked in BLUE) going to the u.fl connector.
Look for cellphone repair shops to help you if you can not solder this yourself. Removing these resistors can be easy if you have the right equipment. I observed that the solder used on the Sodaq One needed more heat than normal solder, meaning it is likely lead free solder. More heat means a high chance of damaging some of the components or tracks on the board. Be really careful!
So i did some range testing and with some awesome help from @jpmeijers i concluded that the Sodaq ONE is pretty sad in performance of RF. After remove the solar panel and have the original antenna go straight vertically i was able to get about 950meters at SF7, some sporadic blips at higher range, but nothing solid. Even by adding a pigtail and a 14dBi antenna there is no increase of performance at all.
Then came @jpmeijers with his Nano soldered with RN2483 module soldered onto the TTN Enschede breakout board. With a simple wire antenna of like 10cm i was able to get an amazing 3,4km range at -112 RSSI being indoor, next to my computer and having 2 racks of heavy duty GPU clusters between the module and the gateway. That’s 4 times the range and more in par of LoRa expectations. And that only at the first attempt, without even driving around yet.
Personally i am quite disappointed on the Sodaq ONE and wanna give some big KUDO’s to @jpmeijers and if i knew this in advance i rather spent my money on good equipment like JP’s
Adv: If your interested in this item, @jpmeijers has no stock at this moment, but you might be able to persuide him to get more and it would only cost you between 35 and 50 $ depending on models
I had already joined the group who noticed how the range of the sodaq one is dwarfed by home built kits but today I made some interesting observations that make me believe that this device can in fact deliver. There is one big if and that is: I only get good range if I do not move. And really when I was moving it was only 10 to 15 km/hr on a bicycle.
Look at these results:
Maximum range: Between 7 and 10 km on two different gateways! I’m pleasantly surprised! http://ttnmapper.org/special.php?node=00000000B7D205D0&allnodes=off&date=2016-09-13&gateways=on
My setup:
We still have 2 zero ohms resistors in the RF path, could they be the problem? They are not really zero ohms but probably something like 0.050 ohms, and will also have parasitic inductance and capacitance.
Should we remove them and patch a wire from the RN2483 side of the first resistor to the antenna side of the second resistor? Then we bypass the island that connected the 4 resistors together.
Paul, what SF are you using? I had forgotten to mention that I used SF10. Although the RSSI didn’t really increase, I do see less packets get lost, after removing the 2 resistors going to the left. So I guess it did help
I have no way to control the SF yet as I am using the standard software that came with the device.
If you look at the map in the above link you can inspect the SF by clicking on the plot point.
Here are some:
We are thoroughly investigating this issue. The 4 resistors were indeed meant to allow you to choose between 433 and 868 MHz and between U.Fl. and SMA. Somewhere in the automated manufacturing all 4 were accicdentaly placed. Sorry for the inconvencience this is causing. We only found this recently (after all boards were shipped) because most of the testing we did was close to the gateways we have on the roof of our building. Initially we didn’t see much difference between having these resistors placed or not, but after testing further away from the office we now can confirm that removing the resistors improves the RSSI by about 7 - 10 dB.
We wil make a video on how to remove these resistors and place it online in the next couple of days.
We are also testing a variety of different antenna’s to see what gives the best performance. Also various tests are running now with different spreading factors.
We will keep you posted.
@Paul_B well thats funny your saying this. Today i was cycling arround and notice the same thing, being far away from a gateway and standing still actually allows you to communicate on better ranges. I noticed the same thing while driving in a car and had to wait for a red light.
I managed to ‘ping’ my own gatway at just over 2km and another gateway with nearly LoS at 8km. That is again being still AND having them resitors removed.
@EelkeVisser i can’t get SF12 to work with SODAQ one, even when loading the passhtru serial and passing the command’s manually when my GW received a packet it still received at SF7. How did you manage to get SF10?
@Paul_B my attempt to try SF12 was to load passthru and then pass these commands:
mac set adr off
mac set pwridx 1
mac set dr 0
mac set appeui xxxxxxxxxxxxxxx
mac set deveui xxxxxxxxxxxxxxx
mac set devaddr xxxxxxxxxxxxxxx
mac set appskey xxxxxxxxxxxxxxx
mac set nwkskey xxxxxxxxxxxxxxx
radio set pwr 15
radio set sf sf12
mac join abp
mac tx uncnf 1 0102030406
When the TX is starting i do notice transmits are taking a LOT longer (as expected with SF12 ad DR0) so its a work in progress, but if anyone has the final clue i love to hear it.
@janwillem thank you for the update, could you confirm the problem while moving the node? So far all the antenna’s i tried has zero difference. Also even with a video, you cant ask everyone to do this … some people (like me) should stay far away from soldering iron’s as i’d prolly set the building on fire. So for those people you’d really need to have another solution (mine is fixed by met collegue)
Ok i have 2 nodes back 2 back on my desk, SodaqOne and the HomeBrew from @jpmeijers and i get these readings (only showing a small part, but ~1.5 hours all the same)
Other RSSI -64 SNR 7
Other RSSI -64 SNR 7.8
Other RSSI -64 SNR 7.8
Other RSSI -64 SNR 8
Other RSSI -64 SNR 9
Other RSSI -65 SNR 11
Other RSSI -65 SNR 8
Other RSSI -65 SNR 8.2
Other RSSI -66 SNR 8.8
Other RSSI -66 SNR 9.5
So even with the resistors gone and (i think the same chip even?) i am stll loosing 13 to 23 dB. But it was a lot worse before.
New experiment, I attached my sodaq set from the above posting to a drone and let it hover at a couple hundred feet for 10 minutes or so to make sure it was stationary. Here are the results: http://ttnmapper.org/experiments/map.php?name=PaulB%20drone
Max range around 50 km so not too shabby
