January 29, 2018

The Quest for HAARP

Last edited: 01.02.2018

The HAARP research facility in Gakona, Alaska under the lights of the Aurora borealis

As a radio hobbyist and space physics enthusiast I've always been highly fascinated by the HAARP research facility. HAARP stands for High Frequency Active Auroral Research Program. The facility is owned and operated by the University of Alaska Fairbanks and is located in Gakona, Alaska.
The most prominent instrument of the HAARP research facility is the so called "ionospheric heater", a huge array of transmitters and antennas (capable of transmitting up to a power of 4 GW ERP) with which temporarily a limited area of the ionosphere can be excited or "heated". For more technical, scientific, and historical information on HAARP I would like to refer you to the internet; a search on Google will give you plenty of hits. You'll also find a list of selected links below, and you might like to read my previous blog entry about HAARP here.

Already in early 1997 the first amateur-radio dedicated test of the then brand new HAARP facility (then still a US Military facility) was announced. The ARRL issued special bulletin ARLX005 which read as follows:

ARLX005 HAARP to air test transmission for hams

Special Bulletin 5  ARLX005
From ARRL Headquarters
Newington CT  February 21, 1997
To all radio amateurs

ARLX005 HAARP to air test transmission for hams

The High Frequency Active Auroral Research Program (HAARP) facility
in Gakona, Alaska, will transmit a test signal on March 8, 1997,
seeking reports from hams and SWLs in Alaska and in the ''Lower 48''
to determine how well the HAARP transmissions can be heard to the

During this test, the HAARP facility will transmit on or near the
frequencies 3.4 MHz and 6.99 MHz to give listeners the opportunity
to listen for the HAARP facility and to claim a QSL card. The test
will be conducted beginning at 0430 UTC on March 8 (2330 Eastern
Time on March 7).

The format of the test will be to transmit a constant carrier (no
modulation) signal followed by a CW message at 10 WPM, according to
the following schedule:


0430     6.99 MHz    Carrier only      Pointed up
0435     6.99 MHz    CW message        Pointed up
0440     6.99 MHz    Carrier only      Pointed to SE
0445     6.99 MHz    CW message        Pointed to SE
0450     3.4 MHz     Carrier only      Pointed up
0455     3.4 MHz     CW message        Pointed up

The transmission beginning at 0450 is primarily for Alaskan hams.
Reception reports should include signal strength during the constant
carrier transmission period along with the text of the CW message.

Complete details on this test and a QSL address will be listed on
the HAARP home page at

The HAARP facility--now in the developmental prototype stage--has
been undergoing testing since it was completed in late 1994.
Multiple transmitters feed multiple antenna elements (there are 48
antenna elements on the site, arranged in eight rows of six columns;
however, at this time, only 18 elements are active. The test will
use 17 antennas with a net transmitter power of 340 kW). The final
configuration could have 180 antenna elements and a combined
transmitter power of 3.6 MW. HAARP's stated scientific purpose is to
study ''the properties and behavior of the ionosphere, with
particular emphasis on being able to understand and use it to
enhance communications and surveillance systems for both civilian
and defense purposes.'' HAARP is managed by the US Air Force and the
US Navy.

For more information, see ''The High Frequency Active Auroral
Research Program'' (QST, Sep 1996, p 33), or check out the HAARP home
page, http://server5550.itd.nrl.navy.mil/haarp.html.

In those days I was a keen SWL, and I remember the announcements, but for some reason I missed the actual test. I believe at the time I was busy preparing for travelling to, or was already travelling in, northern Scandinavia, which might explain things. Or did the announcements reach me too late? I can't remember clearly. I can't recall either ever having read or heard anything about the test being successful or not.

To my knowledge no other special amateur-radio dedicated tests were conducted thereafter (please correct me if I'm wrong). Then the US Air Force in 2014 announced that the facility would be completely closed down and HAARP went silent for good.
Or so was thought! The facility was taken over by the University of Alaska Fairbanks, and in February and September of 2017 HAARP was again heating up the ionosphere. Since the equipment of the facility had not been used for a couple of years, it was decided to go easy on the generators and transmitters, and the ionospheric heater was used at only a part of its full capacity, both antenna and power wise.

Excellent article on HAARP in the October 1996 issue of the legendary Monitoring Times magazine. Great article debunking the early myths and conspiracy theories about HAARP*. MT magazine from the collection of PA7MDJ.

The campaigns at HAARP in 2017 were primarily scientific research projects of course, but they had a ham radio flavour to it!
The ionospheric research campaigns done in 2017 were led by Assistant research professor Chris Fallen of the University of Alaska Fairbanks Geophysical Institute. And Chris Fallen also happens to be radio amateur KL3WX! Chris announced the tests on his website and on his Twitter account, and radio amateurs and shortwave listeners were asked to participate by monitoring for the HAARP signals. During the actual campaigns, Chris announced on and off times, frequencies, and mode on his Twitter account, helping radio hobbyists to tune in to the signals, and feedback was given to reports and questions from listeners.
The signals mainly consisted of unmodulated carriers or carrier pulses, but during the September 2017 campaign especially for the radio amateurs and SWLs listening in, also some SSTV images were transmitted! Also some AM transmissions were done with tones and musical tunes to generate the Luxembourg or cross-modulation effect. But both the SSTV and AM signals were transmitted on lower frequencies and were difficult or impossible to receive outside of North America.

During the February 2017 campaign I did not manage to receive the HAARP signals. During the September 2017 campaign I again did not manage to pick up the HAARP signals on the lower frequencies, but this time also a higher frequency was used: 9.500 MHz. And this higher frequency proved to be suitable to propagate the unmodulated carrier of the HAARP experiments to me in the Netherlands several times. I used my Yaesu FT-991 transceiver and a HyEndFed 10/20/40 wire antenna to pick up the signals.

On 9.500 MHz I do have a lot of interference from PLC devices, and at times there was also some interference from broadcast stations on nearby frequencies. But by using a narrow digital filter of 300 to 50 Hz in CW mode, wide enough for just an unmodulated carrier, much of this could be eliminated. And nowadays we have the advantage of being able to detect received carriers and make them visible with the computer and spectrum analyzer software like for instance Spectran or Argo. And that's what I did.

I was surprised and still am by the choice of the higher frequency; 9.500 MHz is in the middle of the 31m broadcast band, and there's no way of determining if the received carrier is from a weak broadcast station or from HAARP except by watching closely for the on and off times of the carrier, as these were announced by Chris Fallen on Twitter.

Only one time I was able to determine both the exact on and off time of the received carrier, which enabled me to identify the signal with almost 100% certainty as coming from HAARP. On September 24th, 2017 a carrier was received starting at 0100 UTC and ending at 0200 UTC, exactly as announced by Chris Fallen on his Twitter account.

I checked this website https://www.shortwaveschedule.com/index.php?freq=9500 for the schedules of broadcast stations on 9.500 MHz. Only one station was transmitting on this frequency during the time of reception and that is China National Radio 1. But this station broadcasts continuously from 1825 to 0600 UTC and does not have a sign-on and sign-off time of respectively 0100 and 0200 hours UTC.

But still the on and off times exactly at the hour are of course common for broadcast stations, and there's always a slight chance that the carrier was originating from some broadcast station that no schedule is known of. To be really sure an odd on or off time of for instance a couple of minutes after the hour would have been helpful. At one occasion when I was able to detect the carrier an odd off time was announced but unfortunately the carrier had faded out by the time it was turned off. Also if the carrier would have been pulsed there wouldn't have been any doubt left, but I did not receive any of the pulsed carrier transmissions.

Furthermore, I'd like to note that at times the carrier was strong enough to actually be audible through the speaker of my transceiver, beside also being detected in Spectran.

Carrier detected on 9.500MHz with Spectran with on time of 0100 UTC (Spectran shows local time UTC+2). My receiver's CW sidetone pitch is set to 880 Hz, explaining the frequency of the carrier in the screenshot. The ticks are placed at one second intervals.

The same carrier with off time 0200 UTC

I know the campaigns at HAARP are primarily scientific research experiments, but from an amateur radio point of view I would like to be so free to make some suggestions, or requests as you will, to Assistant research professor Chris Fallen for future HAARP campaigns. In case of implementation these suggestions will not directly add scientific value but might cater to the radio amateur, and in particular the ones at DX locations. This is the point where I'd like to consider this blog entry doubling as an open letter to Assistant research prof. Chris Fallen.

1. The higher frequency of 9.500 MHz proved to be suitable for easy worldwide reception of the HAARP transmissions, but in the future could maybe a frequency be chosen in the same range but outside of the 31m broadcast band, for instance 9.300 MHz?

2. Is there a possibility to have HAARP transmitting WSPR signals, or even better JT9 (as in JT9 a free-text message can be transmitted; in WSPR you'll have to transmit a callsign)? WSPR or JT9 would enable radio amateurs to pick up and identify the signals up to a signal to noise ratio of about -30 dB. So even extremely weak signals will be detected. If desired, announce a WSPR or JT9 test to the weak signal community, and you'll have a huge audience worldwide. WSPR and JT9 are FSK type emissions and possibly might be generated by the HAARP computers/software that drive the transmitters.

3. In order to make identification easier, could some intelligence be added to each continuous unmodulated carrier transmission, like for instance a morse code identification? QRSS (very slow CW) could be used with a "dit" length of for instance 6 seconds. Even when the signal is inaudible, with for instance Argo a QRSS morse code identifier could still be made visible. Argo is already widely used by hams and SWLs listening for HAARP.
The SSTV was very nice, but is useless when the signals are weak, or at the low frequencies that were used aren't present at all, like at my station in the Netherlands.

Last but not least I'd like to thank Assistant research prof. Chris Fallen for giving hams and SWLs the opportunity to participate in the fascinating research done at HAARP. I'm looking forward to the next research campaign!

Addendum 01.02.2018
* The full Monitoring Times article can be found here:

See also:


January 28, 2018

DKARS HF Weak Signal Day

Last edited: 28.01.2018

On March 10 the Dutch Kingdom Amateur Radio Society DKARS will be hosting a "HF Weak Signal Day" seminar. There will be many interesting lectures, and it will be a great opportunity to meet with other weak signal enthusiasts. I'm planning on being there.

The programme (in Dutch) looks as follows (from the DKARS website):

Op zaterdag 10 maart 2018 wordt de eerste DKARS HF Weak Signal Dag gehouden in Kootwijkerbroek.

Deze dag is speciaal bedoeld voor alle amateurs die graag met HF banden bezig zijn en dan vooral interesse hebben in DX werk, contesten, verschillende landen werken maar ook amateurs die daarvoor graag aan het bouwen zijn. Bouwen van apparatuur, antennes en een compleet station. Amateurs die graag een uitdaging zoeken en de radioamateur hobby als een deel van hun leven beschouwen.

Op deze dag worden diverse lezingen gegeven door gedreven en enthousiaste radioamateurs die op hun gebied ruim ervaring hebben. Van contesten op 160 mtr tot en met het werken met WSJT op 60 mtr, van een QRP baken op 80 mtr en WSPR experimenten met de Red Pitaya en een verhaal over het zelf bouwen van apparatuur en antennes. Het hele scala komt op deze dag aan de orde.

Uiteraard is er ook gelegenheid tot het leren kennen van gelijkgezinden maar ook heel belangrijk het opdoen van ideeën om de hobby verder te kunnen verdiepen. De dag wordt gehouden in De Essenburcht, Schoonbeekhof 1, 3774 DA Kootwijkerbroek. In de Essenburcht kan u koffie en lunch kopen tegen heel nette prijzen. Keuze is ondermeer mogelijk uit: Koffie €1,50; Broodje kroket €2,50; Broodje kaas €2,00; ETC. Bier € 1,50 Ook sterker is verkrijgbaar. We beschikken over twee ruimtes de zaal voor de presentaties en een tweede zaal waar de Surplus Radio Society en de QRP club wat apparatuur laat zien. In de tweede zaal is ook gelegenheid voor onderling QSO en gelegenheid van Bar en Eten.

Het programma ziet er als volgt uit:

Zaal is open vanaf 10 uur
  • 10.30 uur - Opening
  • 10.45-12.00 - Lezing door Henk, PA1A: bouwen met passie
  • 12.00-13.00 – lunch
  • 13.00-14.00 - Lezing door Remco, PA3FYM: contesten op 160 mtr
  • 14.00-14.30 - Lezing door Robert, PA0RYL: het CW QRP baken op 80 mtr
  • 14.30-15.30 - Lezing door Henk, PA2S: 60 mtr WSJT ervaringen en propagatie onderzoek
  • 15.30-16.00 – korte pauze voor thee of koffie
  • 16.00-17.00 - Lezing door Hans, PA0EHG: 5 banden WSPR zenden met de Red Pitaya
  • 17.00 uur sluiting van de dag

January 27, 2018


Last edited: 27.01.2018

This is my latest high tech creation, hi. It's a little 4-leg "spider" antenna for 1090 MHz (1.090 GHz) constructed around a nice old Amphenol SO-239 chassis connector that I had lying around. Its 4 legs and top element are made from copper wire and are about 68 mm in length.

Beside being a ham radio operator, I've always been a radio hobbyist in general, and every once in a while I like to explore radio related stuff outside of the ham bands. The little spider antenna is used to receive the ADS-B signals broadcasted from aircraft on 1090 MHz. ADS-B stands for Automatic Dependent Surveillance - Broadcast. ADS-B signals are periodically (twice every second according to the YouTube movie below) broadcasted by aircraft and contain among other things the aircraft's unique ICAO identifier, the aircraft's callsign, and data provided by the GPS navigation system of the aircraft, like GPS location, altitude, heading, speed, etc. ADS-B signals enable an aircraft to be tracked in real-time by for instance air traffic control stations. Since other aircraft can also receive the ADS-B signals from ADS-B equipped aircraft surrounding them, it also provides pilots with improved situational awareness.

I use my RSP1A SDR receiver (see my last blog entry) connected to the spider antenna to receive the ADS-B signals. Special software extracts the data from the signals. This data is sent to another computer program called Virtual Radar Server (VRS) which then plots the received aircraft on a map. With the aircraft identifiers received on 1090 MHz, VRS is able to retrieve additional information for the aircraft from various online databases, including the aircraft registration or "tail number" and the current route (destination and departure airport). When you select an aircraft, VRS also shows the aircraft type and model, operator, country of registration, and when available (most of the time) even photos of the aircraft (not just a random photo of the type and model, but the actual aircraft with the registration of the selected aircraft).

It's really fun to see in real-time what's up in the airspace above you. Of course you can also use some online service like www.flightradar24.com for this, but for the radio hobbyist it's a thrill to know that the aircraft that are plotted on the map are plotted there because their ADS-B signals have been received with his own radio equipment directly over the air.
With the antenna inside the house, depending on the height I place the antenna at (ground level or attic), I've been receiving aircraft up to distances of 100 to 200 km. You can even do some DXing, trying to catch aircraft from as far away as possible and trying to beat your personal record.

The ADS-B data extracted from the signals received on 1090 MHz
The aircraft plotted in Virtual Radar Server. For the selected aircraft (the one in yellow) additional information is displayed, including route, operator, country of registration, tail number, aircraft type and model, and any available photos of the aircraft.

Below some YouTube movies explaining how to set up ADS-B with your SDRPlay RSPA1/1A/2 receiver and explaining how ADS-B works.

See also:


January 26, 2018

WSPR monitoring and the SDRPlay RSP1A 1 kHz - 2 GHz SDR-receiver

Last edited:

The SDRPlay RSP1A "DC to Daylight" SDR receiver

The downside of the popularity of stand alone WSPR transmitters like for instance the QRP Labs U3S and the Sotabeams WSPRlite is that more and more hams in the field of WSPR have become transmit-only stations and do little WSPR monitoring or none at all. On the other hand though, if it wasn't for these neat little stand alone WSPR transmitters many radio amateurs would probably never have been active in WSPR at all. I'm one of them; it was my beloved U3S that has made a keen and active WSPR operator out of me.

But we all depend on eachother; without WSPR monitoring stations there's no use in transmitting WSPR beacons, and without transmitting stations there's no use in monitoring. It occurred to me that if you're an active user of a U3S or a WSPRlite, and you're enjoying seeing your WSPR beacons being spotted all over the world, you more or less have the obligation to do the occasional monitoring session, as a favour in return and to keep the WSPR network alive and interesting! To me, being a transmitting-only station has always felt a little selfish.

I've done the occasional monitoring sessions, but only if some special WSPR project was in progress; maritime mobile, floaters, high altitude balloons, expeditions, etc. Still a little selfish and I decided I wanted to become more active with general WSPR monitoring as well. I don't like having my transceiver powered up for longer, extended periods of time though (it occurred to me that manufacturers should build in to their transceivers an on/off switch or a sleep mode for the display to be turned off during extended monitoring sessions).
So, for this reason (and a million others) I decided to purchase the new SDRPlay RSP1A SDR receiver. The RSP1A is manufactured in the UK and receives from 1 kHz to 2 GHz, or to use a popular term, it's a "DC to Daylight" receiver. It's my first experience with a Software Defined Radio, and I must say that I'm really thrilled with all the possibilities the RSP1A brings. It's really nice to also be able to explore the VLF frequency range for instance; for the first time in my life I managed to hear the signals of time signal station DCF77 on 77.5 kHz. I'm planning on making a PA0RDT mini whip antenna to do some more serious monitoring on VLF, LF and MF, including 2200m and 630m WSPR.

My first WSPR monitoring session with the RSP1A was done on 40m and instantly the SDR receiver connected to my HyEndFed 10/20/40m wire antenna managed to pull in the 5 Watt WSPR signals of the DP0GVN beacon on Antarctica for me (earlier, reciprocally the DP0GVN receiving station already picked up my 200mW WSPR signals on 20m, see the addendum to my blog entry of January 13th). I'm really pleased with the reception of the RSP1A.

A nice interview with Jon Hudson from SDRPlay about the new RSP1A can be found here on YouTube.

I also don't like leaving my laptop powered up for extended periods of time, so in the future I might look into the possibility to do WSPR monitoring sessions with the RSP1A connected to a Raspberry Pi.

January 23, 2018

My Pixie QRP CW transceiver

Last edited:

A pixie is an imaginary little creature like a fairy. Pixies have pointed ears and wear pointed hats.
elf, fairy, sprite, imp, brownie, puck, leprechaun, hobgoblin, peri

I finally got round to building my Pixie QRP CW transceiver kit. The result is shown in the pictures above and below. The kit was purchased for about € 10 at HAM RADIO 2017 in Friedrichshafen last summer. If you search for it on the various online selling sites you will find it for sale for even lower prices! For about the price of a Big Mac you can be the owner of this neat little transceiver kit, hi.

Traditionally, homebrewing hams have been building QRP transceivers in to "Altoids" type peppermint tins, so I've found this nice Amarelli tin for my Pixie to be placed in.

The Pixie transmits on 7.023 MHz in the 40m band. It also receives on that frequency of course but there's not much filtering and everything within a bandwith of many kHz is heard. I haven't figured out yet how many kHz, but I wouldn't be surprised if it's more than 5 kHz up and down of the receiving frequency. This could be a problem when operating the transceiver when there's a lot of CW activity, and it could mean that a lot of tone filtering has to be done by the brains. Apart from that I'm really surprised by the nice performance of the receiver. Below you can listen to a recording I made of the Pixie reception last weekend when a contest was going on. Listening to the Pixie with small earphones in gives me this nice sensation I got from experimenting with radio receivers when I was a kid.

The transmitter also works. I've been sending some "VVV VVV TEST DE PA7MDJ TEST DE PA7MDJ AR" messages with my Palm Pico paddle on its side using it as a straight key, and I managed to be picked up in Finland by a station of the Reverse Beacon Network (see screenshot below). Not bad considering that my straight-key keying really needs some improvement (some would say that all my keying does, hi). I've measured the output power to be about 250 mW, a little on the low side (specs say it should be 800 mW at 9 V), but I'm not sure if the battery I've been using has been used before or not (it was the only one I had lying around, and I couldn't wait powering up the Pixie).

I will do further experiments and will do a SOTA activation with the Pixie as soon as I've received and finished building the QRPGuys Mini Keyer V2 kit that I have on order. With this electronic keyer I will be able to use my Palm Pico Paddle with the Pixie in the way it's intended to be used, i.e. as a double paddle. The keyer will also add a sidetone; the Pixie doesn't have one (upon key-down it just mutes the receiver).

QRPGuys Mini Keyer V2
A lot can be found about the Pixie transceiver on the internet (Google is your friend!), but I really liked how the pdf-document "How The Pixie Transceiver Works", which can be found here, explains in a simplified way the working of the Pixie. For more about the Pixie, see also my other blog entries on this page.

January 20, 2018


Last edited:

On the last day of 2017 and the first one of 2018 I participated in the QRSS New Year's Eve Operation Celebration of the Las Cruces QRSS Mafia (see also my blog entries of December 12th and November 22nd). From 0000 UTC on December 31 until 2359 UTC on January 1st I had my 200mW U3S transmitting my callsign on 40m in FSKCW QRSS with a "dit" length of 6 seconds and a frequency shift of 4 Hz.
This was my first experience with QRSS. QRSS is extremely slow CW usually transmitted at QRPp power levels. Extremely slow as the length of a dit usually is several seconds. Very weak QRSS signals that are close to or below the noise floor and that are often not audible to the ears of the human ham operator can often still be detected and made visible with a computer and special software. This is the basic idea behind QRSS. This was (and still is) a great way of experimenting with low power signals, basically the same way you can do now with WSPR except that the way to find out where your signal is heard with QRSS is way more of a hassle. With QRSS you have to keep an eye on the online "grabbers" or someone must send you a reception report to let you know that they received your QRSS signals. Unfortunately there aren't that many online grabbers around, so I understand that most hams prefer to operate WSPR instead, and I'm afraid the QRSS operator is becoming an endangered species.
It's a real thrill though to find your own callsign on a grabber screen, so here's a little plea to get you into QRSS and to keep this mode alive. If you're the owner of for instance a QRP Labs U3S or some other transmitter that's capable of doing QRSS, why don't you just try it and send some QRSS beacons every once in a while. You can easily combine it with sending WSPR beacons. If there are more QRSS operators, hopefully this will also lead to more people getting interested in setting up a grabber.

Below you can see my signals during the New Year's Eve operation as found on the 40m QRSS grabbers of G3VYZ in England and SA6BSS in Sweden. My signal is the one all the way at the bottom. A compendium of active grabbers can be found on the QRSS Plus site. A list of QRSS frequencies can be found here.

January 13, 2018

Antarctic WSPR beacon / receiver soon to be operational

Last edited: 15.01.2018

This is the latest exciting news about the dedicated WSPR beacon / receiver to be set up at the Neumayer III station on Antarctica. For more information, see also my blog entry of July 29th, 2017.

From the DARC Facebook page, published on January 11th, 2018.
Die Hardware für die geplante WSPR-Funkbake in der Antarktis als gemeinsames Projekt von Technischer Universität München, Hochschule Bremen und dem DARC e.V. befindet sich zur Zeit auf dem Weg ins ewige Eis. Die Installation besteht aus einem Bakensender für die Bänder 160m bis 6m mit einer Ausgangsleistung von 5 Watt sowie einem WSPR-Multiband-Empfänger auf Basis eines Red Pitaya, der simultan alle Bänder von 160m bis 15m beobachtet und bis zu 700 Empfangsberichte stündlich in das WSPR-Net einspeisen kann. Die Inbetriebnahme des Systems an der deutschen Forschungsstation "Neumayer III" ist noch für den Januar 2018 geplant.
Addendum 15.01.2018
Felix Riess DL5XL reports on wsprnet.org that testing of the RX setup has begun today (January 15th, 2018). The beacon callsign is DP0GVN and the first RX spots have already found their way to the wsprnet.org database! They're not using the final antenna setup yet and there might be extended off times while they're working on the system.

According to info on fellow blogger PE4BAS' blogsite (http://pe4bas.blogspot.com) for TX they will be using a Procom vertical antenna at the main building. For RX they will use two loops, one of 170 m and one of 20 m size, located at the chemistry laboratory 3 km away from the main building. The planned up time of the beacon will be one complete 11 year solar cycle!

Addendum #2 15.01.2108
Wow, the DP0GVN beacon is already doing an excellent job! I started WSPRing on 20m around 18:50 UTC and in an hours time I've already been spotted in Antarctica three times! I'm using my 200 mW U3S transmitter and a sloper EFHW wire antenna.

And this is what it looked like at Neumayer III Station around the time I was spotted (from the Neumayer III webcam):