UBLOX MAX M8Q GPS Breakout Board


GPS Breakout Finished

For those that want to build their own GPS Breakout board, I sell a PCB that allows you to do just that. Primarily designed for my own HAB tracker boards it can be used for other trackers and has both the Serial and I2C interfaces on the 8 way pin header. A previous article described how to attach this GPS board to my tracker boards.

For HAB tracking you can miss out most of the components. You if you wish can fit a small Lithium battery with some DS sticky tape and the components to trickle charge it from the trackers supply or fit an LED for the time pulse pin. My own tracker boards can have the components fitted to turn off the power supply to the GPS, whereby the battery will backup the GPS and allow it to be used in hot fix mode.


On the rear of the PCB are the pads for a small inductor, this helps to protect the GPS antenna input against static damage, a problem to which the UBLOX GPS are very prone to. The inductor can be omitted but take great care when handling the assembled break out board. If your UBLOX GPS takes a long time to get a fix, or never does, a probable cause is static damage to the antenna pin. The GPS data sheet does mention this as an issue.

You can fit the JTI_ANTENNA-1575AT43A40 ceramic stick antenna, but the board can also be used used with a 1/4 wave or 3/4 wave wire antenna and 1/4 wave radials. This will significantly improve GPS reception over the ceramic antenna, which in turn can significantly reduce power consumption, especially if hot fix mode is in use. The fitting of the wire antennas is described here fitting wire antennas. For these latest GPS breakout boards you will need to use Ernie Ball Custom Gauge 9 wire. Inductor on PCB


Taking suitable antistatic precautions, fit the inductor to the rear of the PCB



Ceramic antenna on PCB



If your fitting the ceramic antenna add it now.



Now position the GPS carefully in place and solder all the pads. I added C1 for some extra on board decoupling.

Add the 0.1” pin headers as shown in the first picture or solder the breakout PCB direct to one of my tracker boards and your ready to go.

Reception distances at UHF – My 1000:1 rule.

I often get asked what range you can get in ‘typical conditions’ using LoRa. There is a simple answer;

“There is no such thing as typical conditions.”

On several occasions constructors have got in touch and said; “I am only getting 1KM, yet you are quoting hundreds of KM, what is wrong with my set-up?” The answer to that is also simple;

“There is probably nothing wrong”.

It is not commonly understood how much the range\distance of communications at UHF can vary. Of course most people expect a difference between an urban area and hilltop to hilltop, but the actual differences are often a surprise. Are the communications over flat or hilly terrain, in urban, rural or forest or perhaps ground to satellite or ground to high altitude balloon?

I was fortunate during the radio testing for the $50SAT project to be able to develop a real world rule of thumb. Whilst testing the Morse beacon (on 437Mhz) I wondered if cutting the transmit power between dits and dahs would make an audible difference, there was an advantage in doing so as it saved around 33% of the battery power.

So I set-up the $50SAT transmitter board running the Morse beacon in my garden and wandered away up the road with my Yaesu FT60 hand-held till the Morse beacon was only just audible above the background noise. This was at a distance of 1km. I live in an urban area and its relatively flat.

The very same $50SAT transmitter board was put into orbit in November 2013. Some months later I was walking into town (it was a nice day) and $50SAT passed at approx 1200km distance. I heard the Morse Beacon very clearly with my Yaesu FT60.

In an urban environment the limit of reception was 1km, yet with the exact same transmitter and receiver and clear line of sight, the reception distance was 1200km, probably more.

That is where my 1000:1 rule comes from.

Thus whilst you might get 400KM with LoRa from the ground to a high altitude balloon, do not be surprised if at ground level in a city you get 400M or less.

This difference is also why it is so difficult to compare reception at different locations. I might get 1KM in my locality, the same equipment might cover anywhere between 250M and 10KM+ elsewhere. The differences are a good reason to be clear about the conditions applying to a reception report, preferably with pictures of the area.

Small LoRa Receiver

LoRa Relay as Receiver


This is a Cute LoRa Receiver, at least I think its cute, for bench use that I built up to be sure I always had a LoRa receiver handy. This is one of the small relay boards built as a LoRa receiver. I added a serial connection for the PC and an audio uplink cable to allow payloads to be decoded by FLDIGI on the PC then heat shrinked it all up.

The receiver is small enough to attach direct to the N Type socket I have on my bench, its the feed from the antenna on the shed roof.

EMF Field 2016

EMF 2016 Badge

Had a great time at this event, lots of really interesting stuff and very well organised. We all had these badges, wonder if you could use it as a tracker ?

I was at EMF giving a talk on a previous project, $50SAT, you can see the presentation here on You Tube

$50SAT Presentation

EMF Field 2016

There is a small section on LoRa at the end. It would be very interesting to see someone try a LoRa device from low Earth orbit, and I have received some interest from a ‘Space Agency’. There might be issues at apogee with Doppler, but if not the digital communications ought to have far greater range than we had with the RFM22B on $50SAT. One day maybe.


LoRa Relay – Part 3

With the relay built and working we can turn our attention to getting it into an advantageous position, such as 60ft up a handy tree.  20

The relay is sized so that it fits with a battery neatly inside two rocket nose cones taped together. This is a neat streamlined package that we can pull up and down a tree with minimal risk of it getting caught. The beads are there so that when pulling the strings, there is no pressure on the relay itself. But how to get the relay up a tree ?



Tree climbers do use crossbows for firing lines, but I figured that if I went over the local park with a crossbow, there would likely be unwelcome attention from the men and women in blue.

Continue reading “Trees”

Building the Relay Board

LoRa Relay – Part 2

A LoRa relay is a powerful tool in the search for a tracker. The concept is simple, the relay is put in an elevated position and listens for and re-transmits the LoRa packets.

Height above ground, particularly in relatively flat areas, can significantly improve signal reception and range. The relay PCB is small and light, taped to the end of a 10M telescopic pole it can improve signal reception by 6-10dB, this represents a double or more increase in range.

The relay board can be built to take a battery charger, so you can put the relay in a waterproof box and keep the battery charged with a small solar panel, current consumption of the relay is around 15mA average.1


Separate the PCB by scouring the line of small holes in the PCB centre with a sharp strong blade, and snap the PCB in two.


First solder the components that are fitted to the bottom side of the PCB, these will be out of sight underneath the Pro Mini when assembly is finished. These components are R1,R2,R3,R4,D1,D4, see picture below.



There are two 5pin angled headers to fit. The one on the top of the PCB (the side with the DRF1278F and the PCB lettering) fits in the row of holes nearest the non-antenna end of the PCB. This header is for the connection of the battery and another supply such as 5V from a RC receiver. See the picture right below with the top 5 pin header and the Polyfuse fitted.

  Continue reading “Building the Relay Board”

How to Search 500 Square Kilometres in 10 minutes

LoRa Relay – Part 1

Of all the LoRa programs I have written, the relay program is by far the most significant, the impact it can have on long range locating is monumental, and it’s a very simple program.6

At ground level a ‘lost’ tracker, be it a Radio Controlled (RC) model or high altitude balloon, could be just too far way to receive good signals. What can very significantly improve search range is increasing the altitude above ground of your receiver, even a 10M pole can make a big difference, especially in urban areas.

Its often not practical to put a large handheld LoRa receiver at height. A good location in an urban area for a receiver may well be on the roof of a house but how do you get all the kit up there ?

The simple answer is that you don’t have to, you put the relay ‘up there’ instead. The relay listens for packets coming from the ‘lost’ model or balloon and then retransmits them so you can pick them up on your normal LoRa receiver. The relay is small light, and self contained.

Using a bit of string and a weight, a small rubber ball is good a stone is not, its only a few minutes work to get a line over a house or tree and pull the relay up. If you carry a long extensible pole, also useful for rescuing models from trees, put the relay on that, the extra 10M or so of height can make a real difference.Part 1 - 2

With a low cost radio control plane or quadcopter, it’s possible to get the relay to 100M plus above ground very quickly. Take a look at the pictures, with a LoRa GPS tracker left running in my garden, once the plane was flown to about 100M I was able to pick up the trackers transmissions, and hence location, across 12kM+ of urban environment and countryside, C on the map below. In about ten minutes I had covered a search area of 500 square kilometres, imagine how long that would take to search at ground level !!!

Part 1 - 3

An alternative to using a RC model to get the relay to altitude is to use a kite, the relay is light enough.

The software for the relay is in the HAB programs folder on the dropbox, set the frequency and LoRa constants to match your normal tracker and receiver settings and turn it on. 

The next article will cover the building of the relay.

Not quite according to plan

Conditions looked good for a test of my easy build Pico balloon tracker. Picos are typically 36” foil party balloons filled with just the right amount of helium so that they rise to around 8000M and continue to float along at that altitude. CUSF Flight Prediction

The CUSF landing predictor suggested the balloon would from a Cardiff launch travel South, turn back North, circle back on itself, head North again and come back South over the Iceland and the West coast of the UK. It was not going to be quick but the battery would last around 6 days.



The tracker build for this flight was about as simple as it can get, just an Arduino Pro Mini, DRF1278F LoRa transceiver, UBLOX Max8Q GPS, PCBs, batteries, pin headers and some wire, no other additional components needed. The tracker board is multi-purpose, designed to be used as lost model locator, balloon tracker, remote sensor or a portable receiver. Extra components can be added for the audio uplink into habitat, powering from primary and backup supplies, components to switch on\off the GPS power and extra decoupling capacitors, if needed. There are connections available on the expansion header for an external serial or I2C GPS, a serial LCD display or a I2C environment sensor such as the BME280 or BMP280, you can even fit an independant watchdog supervisor.

Tracker Front

Tracker Rear

Continue reading “Not quite according to plan”

Finding Lost Trackers

Once a balloon tracker has landed it will normally continue to send its GPS derived position. With your radio gear you can decode the FSK RTTY to a laptop or phone application. With a LoRa based tracker you can use a small hand held receiver which will tell you where the tracker is and how far and in what direction you need to travel to find it.

But what do you do to find your tracker if after landing the GPS has no reception or it has failed for some reason, or maybe its just not convenient to carry a rucsac of radio gear to pick up a FSK RTTY only tracker.

One possible answer is to use basic radio direction finding (RDF) techniques by listening for the trackers FSK RTTY transmissions. A UHF Frequency Modulation (FM) hand held such as the cheap Baofengs will detect these FSK signals but you don’t need to decode them as such, just listen to the audio.

If you turn off the squelch on the hand held so that you hear continuous background noise, you will notice that the background noise goes away when FSK RTTY is received. The FM hand held sees the signal, but there is no real FM content for it to decode so you ‘hear’ silence. The stronger the received signal, the quieter it sounds.

This quieting caused by the FSK RTTY is noticeable even when the tracker is a considerable distance away and received signals are very weak. With a directional antenna such as a Yagi or simple Moxon, you can get a good bearing on the tracker by moving the antenna around for minimum noise, when noise is at a minimum the antenna is pointing at the tracker.

Directional Antennas

Examples of directional antennas are shown below, the first is an Arrow yagi, this is robust, collapses well for transport and is quick to assemble.

Continue reading “Finding Lost Trackers”

The MAX6369-MAX6374 Watchdog

This is a marvellous device to fit to a balloon tracker. It is a timed watchdog that needs no extra components to operate, you just provide a pulse every so often and it sits there doing not much at all.

If it does not see a pulse for the watchdog period then it will reset the tracker, recovering it from a possible fatal program crash.

Data Sheet is here

The watchdog timeout can be a range of values from 1 to 3mS up to 60Secs to 180Secs, depending on the particular device and how you connect 3 pins.

I use the 60Sec to 180Sec timeout, this is plenty for a tracker and has the advantage that even if the pulse to the watchdog fails, but the program is otherwise running, you have enough time to read the GPS and send the tracker payload before its reset.

The code I write for my trackers has the routines to pulse the watchdog already imbedded so all you need to do if you want a separate independent of the processor watchdog function is to fit this single device.