Month: January 2017

5 MHz WSPR signal reports over 24 HRS

George M1GEO has recently been running a SDR decoding WSPR on several HF bands at the same time.

Very much work in progress, I thought it would be good to try WSPR spots over a 24 Hr period and then graph the results. George is located on JO01OK and thats 119 KM away from my JO02CN location. In the winter you would expect 5 MHz to support QSO through the day, with the band closing around sunset.

In reality the band is often open for just a few hrs in winter, and the graph provides a good demonstration. The data set is made up from 62 WSPR reports, collected by M1GEO. I set my QRP labs U3S to TX at 200mW every 10 minutes.

The total number of samples we could have collected was 6 (Per Hr) x 24 = 144, although in reality George was unable to receive my signal from around 17:00 Hrs till 07.40. This would account for 14 Hrs, approximately 80 samples.

The samples that are missing (no signal received) have been replaced with -30 so they are plotted on the axis, in effect -30 indicates no signal detected.

The yellow line on the left indicates sunrise, and on the right the black line is sunset.

Some points to note on the data have been highlighted. Its interesting to note the band opens shortly after sunrise and closes shortly after sunset. On this day while the band did open around sunrise, it sparked into life once around 08:10 but didn’t really support good signals till after noon.

From around midday, till around 14:00 hrs it was quite stable, but you will notice signals deteriorate from around 14:00 till around 1630. It has one little flourish of activity around 1710 before closing at 17.15.

This can sometimes be seen, as darkness approaches the band improves just before it closes.

As the lighter evenings approach you would expect the band to stay open for longer, and even support inter G all night.

Its interesting to see how not just daylight but time of day has an impact on the bands propagation.

QRP labs Ultimate 3S Part 2

Welcome to Part 2, the project started in Part 1 and this covered the initial build and setup of the Ultimate 3S. Changes to the TXM are possible but require you to manually change the LPF and input a new frequency via the software.

QRP labs have developed a relay switched LPF kit, and Part 2 covers the assembly and operating of this unit.

Building this kit twill allows you to increase the bands from 1 to 6. The relay kit has space for 5 further LPF, and you can continue to use the 1 from the original kit giving you a total of 6 band operation.

LPF are available for all bands from 136 Khz to 50 MHz.

I chose to order and make up the following 3.5 MHz, 5 MHz, 7 MHz, 10 MHz and 21 MHz. The process of assembling the  LPF is very easy, the circuit is standard and simply requires different values of capacitor or inductor to achieve the correct roll off characteristics. The  Low Pass Filter kit is based on the G-QRP technical notes, a design by Ed Wetherhold W3NQN. Ed has published an interesting article on band pass filters published in QST dated 1998.

Winding the toroids without doubt takes the longest, and with 15 to wind I spaced it out over the course of the day. Here you can see the 5 MHz and 10 MHz (left) LPF assembled and ready for installation.

The winding of the toroids for the LPF was interspaced with adding the relays to the band switching relay kit.


The kit uses Panasonic TX2-5V relays, the kit includes 6. One is to mount on the Ultimate 3S main board.

Here you can see the kit is completed, and awaits LPF. Its important to note the highest frequency LPF must be fitted in position 1. The positions are numbered 1 to 5, with position zero being on the main Ultimate 3S board.


The next stage is to mount the relay LPF on top of the main Ultimate 3S, the system has been designed to stack on top of each other, providing a very neat solution. In this example I have 3 LPF fitted, with room for a further 2. Its seems logical that you should arrange the LPF in band sequence, I don’t think this is strictly necessary but it helps the planning and programming later.


The completed item is installed, working from right to left 21, 10, 7, 5 and 3.5 Mhz. I also installed a SMA socket onto the relay board, as the RF is now routed from here after the expansion. RF can be coupled in one of 3 ways, the board has options for SMA upright, SMA side mount and pin connections. The SMA upright doesn’t provide enough clearance with the LPF directly above it, so you could try a right angled SMA socket. I didnt have one so opted for the SMA mounted on its side.

The RF is now routed to the side angled SMA, here you can see the modular design with the GPS module to the top.

QRP labs to stock cases for the completed project, and I have recently ordered one to provide the finishing touch.



Further thoughts and lessons learned.

I need to investigate why the o/p power has dropped on 14 MHz, as a single band unit I was getting 250 mW, but this has recently dropped to 100mW. I suspect this may have something to do with the LPF arrangement, as all RF must now pass via the LPF placed in position 1. This shouldn’t present an issue as LPF 1 should pass anything below 21 MHz, but without a spectrum analyser or a Vector Network Analyser I haven’t been able to investigate further at this stage. Other bands have an o/p between 200 and 300 mW.

Its worth mentioning again you need to take out the original U3S board and add a relay and diode before LPF zero. This isn’t required unless you have added the LPF relay kit.

You will need to cut jumpers previously installed on the U3S, you need to cut W0 – W1 and W2 – W3.

Configuration in the menu system for multi band operation is a little fiddly, but once you understand the steps needed its easy enough to follow.

You will need to change the menu option  TxS to increase the amount of transmit slots from the default 3 to 6 in order to make use of the enhanced LPF.

Typical display options explained. starting from top left

1] is the sequence number, this indicates its the second mode option (zero then one)

1 This indicates it will select LPF 1 (check this corresponds to the correct frequency LPF) For example number 5 here would select the LPF in slot 5

021,096,125 is the frequency for TXM. You can find a complete list of freq in part 1 of the ultimate 3S construction.

JO01 is my locator, shown when in txm

23 is my power in dBm

93 is the symbol number (WSPR is 162 for a complete message)

0 is the tone currently being sent.

If you want to add additional tx sequences, then this can be done by editing the menu system, just ensure you have the correct frequency and LPF slot set.

This is covered in great detail in the firmware manual.

GPS Track my Cat

Have you ever thought I would really like to know what the cat gets up to during the day? No, OK its just me then. Anyhow, I was surfing one of the many ChinaBay websites the other evening and spotted a rather cheap “Pet tracker”.

Marked as D69 Pet Locator, the advert suggested it could record GPS locations and pass these to your mobile phone via a downloadable app called “there”. As with many of these Chinese products the online description is somewhat unclear, so it comes down to price and if you want to take a chance.


The online description claims the following:-
Standby time 6 Days using the  MT3337 GPS chip, and a weight on 32 Grams. Size 16x56x38 mm. Search terms seem to be GPS tracker D69 GSM, and this should locate the product for around £25 to £30 UKP.

You will need a SIM card, this can be pay as you go, so long as you have GPS data and the ability to send and receive text messages. I purchased £10 of credit with GiffGaff, as they seem to offer good value and your credit doesn’t expire to 180 days (longer if you make / receive calls)

                                     Tracking options

Real time, you click on the app and it attempts to locate your pet and return a position on the screen. Its often not clear if this is live or historical, but it does work if you have a GPS fix.




Historical, you can set the tracker to upload spots every 1 or 10 minutes.

By sending a text, you can text your pet using url# and the device will return a text message and a google maps link showing the last known locations. It will also return “NO GPS” if you don’t have a GPS fix at the time of location request.



Example of text received

Locate date:2017-1-26
Locate time:4:52:15


Calling the tracker

You can phone the tracker and speak to your pet (no I dont know why) or call and mute your audio so you can hear the location your pet is in. This can be especially handy if you suspect moggy is visiting other houses in your street.

Does it work

Yes its adequate but you need to keep in mind the following. It has two position modes GPS providing an accurate fix to around +/- 10m. If GPS fails it will use Location Based Services, this provides an approximate fix but for the purposes of tracking your pet it’s useless.

Cats often sit, rest and stalk in locations that are not GPS friendly, under cars and in bushes and this can cause the GPS fix to be infrequent.

Its also possible to setup a white list number, so only this number can call and text the tracker and Geofence your pet!


LF and MF reception in an urban environment

SW listening and the ability to receive stations many thousands of miles away on HF has always fascinated me. Reception on HF, more especially LF and MF has always been a challenge due to the high urban noise floor.

The doublet antenna at home while working well on 7 to 30 MHz, is a poor performer below 3 MHz. This coupled with the high noise floor makes reception of all but the strongest signals very difficult. In part the answer is a dedicated antenna for low frequency reception. Options range from very large beverage / directions arrays to small active antenna. I have no experience with large antenna on sub 7 MHz, as the average urban garden will not support such ambitions projects. The answer for many of us comes from a surprisingly small antenna.

Receive only active antenna may provide a solution, the active part providing some amplification in the receive path. Designs are available for both commercial and homebrew, the price can vary from a £20 homebrew solution to £250 commercial antenna.

I was lucky enough to spot a second hand Wellbrook loop on ebay, reasonably cheap and this gave me a chance to compare the performance with my other HF antennas. It also provided me with an opportunity to monitor WSPR signals on 472 KHz and provide some data into the WSPR network.


Wellbrook Communication have a number of versions available, comparison of models here.

Cross Country Wireless  with a FAQ


PA0RDT has designed a LF / MF Mini Whip antenna, you can read the fundamentals of the mini whip.

M1GEO has recently repaired a Wellbrook loop antenna, his website includes a detailed teardown.

LZ1AQ has detailed design notes on his website for his wideband active loop antenna.

PA0LUX provides a video demonstrating HF reception on his wire antenna and Wellbrook loop. You can see and hear the difference for yourself.

How well does the Wellbrook work on 472 KHz

Well it outperforms my doublet antenna by some considerable margin. Signals that I cant hear just pop out of the noise and for interest I have included some WSPR data for that band.

Monitoring 472 KHz on the evening of 15th and morning of the 16th Jan 17  provided 19 unique calls.


The best DX was LA1TN at 1346 KM

Its also interesting to look at the reception reports over time of day / night.




Oak Hills Research QRP RF Power Meter (WM-2 QRP)

The Oak Hills research produce a very nice QRP power meter kit, the RF power meter WM-2 QRP is produced with a high quality case and silkscreened and masked PCB and a large moving coil meter.

Meter Highlights

Operates from 300 KHz to 54 MHz

Measured power range from 100 mW to 10W FSD with an accuracy of 5%

I built one a few years ago, and found it very helpful when measuring QRP power from the FT817 and Ultimate 3 QRP kits.

The kit comes with everything you need, instructions are provided in the pack and takes a few hrs to build. The kits are available online and currently cost $129.95 and can be provided with BNC or SO239 sockets.

You can calibrate the meter without the need for any special test equipment.

Ive included some pictures of the completed meter, and think you will agree it looks very professional. Power is provided by a 9v battery or external supply.

Oakhills QRP meter

Oakhills QRP meter

Ultimate 3S QRSS/WSPR project (Part 1)

Welcome to Part 1, this covers the initial build of the U3S, Part 2 covers the addition of the LPF relay board.

Some month ago I purchased a Ultimate 3S QRP transmitter from QRP Labs. The latest version is the ultimate 3S and it comes with a number of options.

Over the christmas holidays I decided to make the project, inspired by Dave G7UVW and George M1GEO activity on QRSS over the holidays. I have used WSPR before from home, but really wanted a QRP transmitter that I could leave on over night, without running the main radio. The answer is one of these kits, as they are very low power consumption, and flexible enough to run multiple modes / bands.

The transmitter can be configured to run on any band between 2200m to 2m (136 KHz to 144 MHz)  with a suitable low pass filter. I have linked a table showing actual TX frequencies online.

I had purchased the Ultimate 3SSi5351A synthesiser module, Low Pass Filter for 14 MHz and the QLG1 GPS unit. Other kits / options are available, and the project is modular so you can pick and choose the options that work for you.

I found the instructions really easy to follow, and coupled with the pictures component identification was easy. I did use a magnifying glass to check some of the component values, and the “quality” of my solder joints. Starting with the GPS module, and then the synthesiser I found the build easy to complete. I then moved onto the main 3S transmitter, this again has various options, but I decided to get the basic version running with one transistor before considering other higher power options.

I found having an LC meter helped when winding the torriods, but the values didn’t seem to be too critical so this was more of a confidence check.

You can see the completed LPF on the left, very compact and the instructions suggest this is good for 10w.


Here you can see the completed main board, with synthesiser and LPF installed. You can see on the bottom of the board the single transistor installed, with space for more.

The instructions suggest a stable 5v power supply is required for the single transistor version, and looking around the junk box I found an obsolete USB blackberry phone charger. This seemed ideal, so off came the USB plug, replaced with some wire and heat shrink. This also provided a current limited supply, with the USB charger rated at 1A being more than adequate for the TXM.

Once built, assembled and powered the kit goes into diagnostic mode (assuming no blue smoke). Mine powered up and worked without any further modifications, but any issues are likely to be mine as the kit is very well designed.

This provide the initial indication all is well, next you need to adjust the display contrast and the the PA bias.

The instructions suggest you should expect around 250mW on 10 MHz, with the power falling away as you move higher in frequency.


I found I could achieve 250mW with ease on 14 MHz as can be seem here on my QRP power meter.

The next stage was to get it on air, this took a little longer than expected, as getting a GPS fix indoors required some careful positioning of the patch antenna. I also found the menu system a little difficult to navigate, its not actually difficult but you have so many options its a little overwhelming to start.

I think the priority should be getting the GPS feed configure, once this is correct the display shows a heart beat. The default option is this is not configured in the software, and so even connecting up the GPS unit will not start this process. The next is frequency correction. If you have the GPS unit fitted this can be completed automatically, but you still need to setup the calibration time, and frequency correction steps in the menu.

Calibration completed showing 27.004.382 (4.382 Hz correction) and the hart beat symbol for GPS lock.

On my first attempt I had not set this up correctly, and while it was slowly correcting the frequency the number of iterations would have been extremely high before I had anything like an aligned txm.




I found the FAQ helpful, as this told you the specific steps to take and values to input.  I set the operating frequency to 14.0971 and after a few calibration cycles it worked!

Here you can see the unit is transmitting WSPR from JO01. The first spot I received was from KK1D at 5474 KM, not bad for 250 mW.


Its always nice when you build something and it works 🙂

Other suggested frequencies

Update 17/1

The TXM has continued to work flawlessly without issue. Ive move the antenna over to a dedicated 1/4 ground mounted roach pole with 4 radials. The power has been checked again and its 250 mW, the best spot I have is from 6500 KM away into Asiatic Russia to the East, and West around the same distance into the USA.