T@sting the magic. Testing the hardware.

In November, the latest version of LibreRouter arrived in Buenos Aires. SAn was in charge of doing the hardware testing and in this chronicle we told how this important moment was.

First of all, when unpacking the boxes, a review of the shipping was made. This consists of a check to verify if there is any breakage for the trip from China and that nothing is missing. The shipment consisted of 4 routers, their antennas with their pigtails and PoE injectors. Fortunately everything was as Dragino promised us 🙂

Resisting the temptation to plug in the routers, they were opened with a screwdriver and a detailed ocular inspection of the boards and their components was carried out, at the same time as it was recorded with photographs. This part is very important because it allows us to document how the boards associated with a manufacturing lot were received, so that in the future we do not have to remember if a chip was soldered or not, the exact part number of a component, etc.

Except for some small details (there were missing some pins in the GPS module, some LEDs had desoldado in the shipping) everything looked very good.



At last the moment arrived!
Using a laboratory power source and with a lot of excitement, the first LibreRouter was connected to 12V: some LEDs came on, others flashed, all right! A serial to USB adapter was connected to the pins of the serial port of the LibreRouter and it was observed to boot: the network interfaces raised, then the mPCI wifi modules and the onboard WiFi 🙂

Now, what to test?
This version of the LibreRouter changed significantly the organization of the boards. Originally the whole design was a single board and now there are two: the CoreBoard with the QCA9558 processor, memmory (RAM and Flash); and the MegaBoard, a peripheral expansion and power supply board.

Based on the great previous work of measurements made by the community in the first prototype, mostly carried out by Terry Gilliett, we had a roadmap of what to do.

In principle, make sure that what was working was still working and especially to review the changes we had requested from Dragino based on the issues we found in the tests of the first prototype.

At that time we had 20 issues that were documented in github, some of vital importance, others were details or improvements.

To enter the world of hardware measurements, the first thing we needed was to update and install the bootloader and firmware. A few weeks before the shipment arrived, SAn had resumed work in the bootloader to support the new boards, make some improvements and make an open source release of it:

The firmware that was used in the laboratory tests is based on OpenWrt 18.06.01 with modifications to support the LibreRouter

Once the LibreRouter was flashed with this software, all the hardware components were tested: Ethernet, USB, GPIO ports, button, LEDs, JTAG, GPS, power supplies, PoE, PoE Passthrough, HW watchdog and wireless. In general, everything went very well.

The wireless measurements of reception and transmission gave super well in the 3 radios:

  • 1 embedded 2.4GHz radio with two Rx / Tx chains
  • 2 mPCI 5.8GHz radios with two Rx / Tx chains

The details of wireless measurements can be read here

In this test some hardware problems arose in PoE and PoE Passthrough. Thanks to the community that develops the LibreRouter we find very fast solutions for these issues. SAn made new tests with these small hardware changes that only required a soldering iron, some resistors and capacitors.

This solved all the known hardware problems of the previous version and the current one 🙂

Indoor field testing.
During those days, PyConAr 2018 was celebrated in Buenos Aires and what better opportunity for our router to perform an indoor field test.
We used a LibreRouter to provide Internet to the Django Girls Workshop with near 100 atendees for a whole day.

Our LibreRouter performed great, WiFi coverage in 5.8GHz and 2.4GHz was very good!


Field testing outdoors.
In addition, the field tests of the LibreRouter are being carried out on the QuitanaLibre network, where the it is routing traffic of 20 nodes of the San Isidro network. In this test we are testing how well the antennas work, the box of the LibreRouter, that does not get water, that can be mounted easily, that it is robust, the PoE power supply and the use for long periods in outdoors.


Some documentation of measurements is here: and here

Very soon we will have the LibreRouter available to everyone!

LibreRouter presentation at IGF 2018 in Paris

AlterMundi wrote a chapter on the LibreRouter for The community network manual : how to build the Internet yourself . This volume is jointly published by the Fundação Getulio Vargas (FGV), the International Telecommunication Union (ITU) and the Internet Society (ISOC). It is the result of the 2018 Call for Papers of the UN IGF Dynamic Coalition on Community Connectivity (DC3) and is the Official 2018 DC3 Outcome.

This video registers NicoEchániz’s presentation of the LibreRouter chapter during the Internet Governance Forum 2018 in Paris.

The LibreRouter is almost out: who wants one?


The LibreRouter is an Open Source Hardware WiFi Router designed from the ground up for Community Networks.

Technical Specification

Based on the the AR9558 SoC and AR8327 Gigabit Switch, the router features:
• 128 MB DDR RAM, allowing all the required services to run with no burden
• 16 MB Flash, plenty for most router needs
• Hardware Watchdog based on PIC10F200, to handle failed flashes or hardware failures
• 1 on-chip 2.4Ghz 802.11bgn MIMO 3×3 Atheros radio
• optional GPS module
• 2 mPCIe slots, to connect WiFi radios or GSM cards, and allowing expandability
• 2 populated Gigabit Ethernet Sockets
• 1 USB 2.0 internal connector, where additional storage, webcams, bluetooth, sensors, etc. can be plugged in
• 1 Serial console pinout allowing external debugging
• POE and POE Passthrough up to 24W meaning one cable can power all the devices in an installation, no matter if there are 1 or more routers.
• Exposed GPIO pins, which will allow tinkerers to connect other electronics to the device
The standard LibreRouter setup will come with 2 mPCIe Power Amplified 5 Ghz 802.11an MIMO 2×2 Atheros radio based on the AR9582.

The full LibreRouter spec can be found at:

The current prototype

The current prototype is the second iteration of the LibreRouter, where we had the chance to fix most of the bugs and experiment with a two board design that will allow communities to explore local manufacturing, leaving the more densely populated (and more complex) electronics on the small blue board in the center, and a second green board with all the peripherals.

This will also allow faster iterations for alternative board designs, because we will only have to design the bigger and less densely populated green board to be manufactured.

We are already exploring such collaborations in Argentina and Mexico, and maintaining conversations with communities in India who want to do this.
This also means the device will be easier to repair.


Current tests

We have tested an initial batch to ensure that the device meets the expected performance and robustness required for Community Networks.
We have deployed 5 devices in existing networks in Argentina, where they have been operating non-stop for over three months as the main routers for 5 villages.
We have also tested the radios in Barcelona, Catalunya. The test results are here:
The hardware has proven to be reliable and performant.


How it works

In a nutshell, the LibreRouter is a weatherproof 3-radio wireless router, but it is much more than that.

It comes from the factory with LibreMesh, an Operating System for Geek-Free Wireless Mesh Networks, that makes it easy for a non-technical community to do the deployment, maintenance and expansion of the network.
Using its two 5 GHz radios and sector antennas, the LibreRouter automatically forms a mesh network with other LibreRouters within range. Using the 2.4 GHz radio, it creates a hotspot around it for clients to connect to the network, and the resulting mesh network enables communication between all the devices connected to it.

Moreover, if any router on the local network connects to other networks (such as Internet) all devices on the local network automatically have access to the external network through the mesh.


It also allows a multi-mesh environment, where each mesh is an administrative boundary (like a neighbourhood, a community, a building). Each mesh can relate with the others easily, including peering (allowing users of each network to reach the other network) and transit (to get to other networks through one of your neighbours).
A built-in mobile app facilitates the management of your LibreRouter, such as set up and troubleshooting.


We have also prepared a video to explain you how you can use it in your communities: (http://gg.gg/LR-video)


We have recently connected our software development repositories to translatewiki.net, a community of translators that support open source projects.
The same day the integration was done, the whole software got translated to 9 languages!

If your language is not there yet, or you want to contribute in a language you know, join at


Survey link

We have prepared a brief survey for those that are interested in purchasing the LibreRouter once it is available. This will allow us to forecast the initial demand and look for support to manufacture the device in bulk, thus lowering the price: (http://gg.gg/LibreRouter-whowantsone)

We are excited about all the interest expressed so far, and we are eager to see what the LibreRouter will help all of us do.

For any inquiry or consultation, use the emails or the contact form at (https://www.LibreRouter.org/)

First outdoor radio and antenna test

We did the first medium-distance test (500 mts) using the HPM5G: the 5Ghz high power radio that the LibreRouter will use, developed by Dragino.

While we were waiting for the LibreRouter mainboard to be ready, we used instead a Mikrotik RB953GS board to host the miniPCIe modules. The SoC is the same as the LibreRouter (AR9558), and while it is a very expensive piece of hardware, it was adequate for early prototyping. For running LibreMesh (LEDE/OpenWrt) on that board, some initial support was done by the team back in January, and this was used during the tests.

The antenna is also one of the candidate antennas to ship with the LibreRouter. It’s a MIMO 5ghz panel antenna, that worked pretty much as expected.

The remote node (up in the mountain) was a TPLink Archer C7 AC1750, also running LibreMesh, with its original antennas. Using the antenna orientation, I confirmed that the LibreRouter antenna (and radio chains) works as expected: With all the TPLink antennas in vertical position, the LibreRouter radio would receive -75 dBm in chain 1 and -85 dBm in chain 2, while putting all TPLink antennas in horizontal position would yield -87 dBm in the first chain, and -75dBm in the second. With two vertical antennas and one horizontal antenna (as pictured), i’d get a balanced -75 dBm on both chains of the LibreRouter radio.

With that final setup (balanced chains on -75 dBm), the TPLink would report the LibreRouter signal to be received at -42 dBm. The difference of received signals is huge, but some factors to consider: different drivers (ath10k vs ath9k), different TX power (23dB TPLink vs 27dB HPM5G), different antennas (3dBi omni vs 12dBi panel). I ran some netperf TCP throughput tests: the LibreRouter radio could transmit sustained ~120Mbps (+/-5Mbps) over 5 minutes without issues, which is great news!

Some context:

  • Spectral noise was low, but not null (you can see it’s a rural area in the outskirts of a capital city, seen in the background: Barcelona. In particular, the node was mounted in Can Masdeu, thanks a lot for the support!)
  • Distance was 550mts, with clear line-of-sight.
  • The max TCP throughput observed with the HPM5G is around ~150Mbps, both received or transmitted, to a client a few meters away, with optimal MCS15 rate.

I also took the time to experiment with lowering the HPM5G TX power via software, (uci set wireless.radioX.txpower), here are the resulting received signal strengths, as measured by the TPLink:

(all measurements +/- 1dBm)
txpower=27 … -42 dBm
txpower=25 … -43 dBm
txpower=23 … -45 dBm
txpower=20 … -47 dBm
txpower=15 … -53 dBm
txpower=10 … -59 dBm
txpower=5 …. -66 dBm
txpower=1 …. -68 dBm
txpower=0 …. -69 dBm

which shows an almost perfect correlation, and supports the notion that the radio is actually outputting 27 dB of power at its highest capacity, which is also outstanding.