Dumphfdl is a multichannel HFDL (High Frequency Data Link) decoder

Thanks for the help Gents,

I am way out of my depth but working through it, all seemed well until I got to this stage.

dumphfdl --soapysdr driver=sdrplay --sample-rate 250000 8834 8885 8894 8912 8927 8939 8942 8948 8957 8977

Then I got this.

If I do a SoapySDRUtil --probe=driver=sdrplay the RSP1A is there and looks good.

Ok, I’m getting there, I now have ACARS messages on screen. :slight_smile: :slight_smile: :slight_smile:

I don’t understand how I do the outputs, where do I add this?

–output decoded:basestation:tcp:address=192.168.0.55,port=44444
–output decoded:basestation:tcp:address=192.168.0.74,port=44445

I think the sdrplay driver gets hung up. You can reboot or sudo systemctl restart sdrplay.
Those commands just get added to the start up string:

dumphfdl --soapysdr driver=sdrplay --sample-rate 250000 8834 8885 8894 8912 8927 8939 8942 8948 8957 8977 --output decoded:basestation:tcp:address=192.168.0.55,port=44444 --output decoded:basestation:tcp:address=192.168.0.74,port=44445

There are quite a few options! Here’s an example of a script I use for the 8-13MHz bands. It’s based on jaymot’s script posted earlier. In this case the \ at the end of the lines keep the script reading the commands and make it easier to understand. But, you can jumble them all together as needed.

#!/bin/bash
   #
   #scan 8-13MHz aeronautical bands with dumphfdl
   #
dumphfdl --soapysdr driver=sdrplay --gain-elements IFGR=40,RFGR=0 \
 --sample-rate 6000000 8825 8834 8843 8885 8886 8894 8912 8921 8927 8936 8939 8942 8948 8957 8977 10027 10060 10063 10066 10075 10081 10084 10087 10093 11306 11312 11318 11321 11327 11348 11354 11384 11387 11184 11306 13264 13270 13276 13303 13312 13315 13321 13324 13342 13351 13354 \
 --freq-as-squawk \
 --output decoded:basestation:tcp:address=127.0.0.1,port=20005 \
 --output decoded:text:file:path=/home/aholt/hfdl-logs/hfdl.log \
 --system-table /home/aholt/dumphfdl/etc/systable.conf

Enjoy!

You Sir are a legend, that has got me going, all connected to VRS and working well, just need some A/C on screen now. :slightly_smiling_face:

Last question, if I want to monitor the 5/6MHz band as well I get this error, is there a way to monitor the 5, 6 & 8MHz bands together?.

Great news! Thanks, I’m learning, too!
Just increase the sampling rate to 6MHz (6000000)
RSP-1a does a good job at this rate. 2.5MHz sampling will only cover two bands.
So, I have a bunch of shell scripts to cover the bands, just to make it easy to hop around.
jaymot’s scanning script is a great way to find the most active band!

Scripts won’t necessarily echo AC to the screen. So I have that line that outputs to a log and then use MultiTail to view it. It’s a great option. It’s outlined pretty well in the GitHub page. But, to get it to work I had to become more than just sudo, and I forget how I did it :slight_smile:
All it takes is some persistence!

The lower the sampling rate the better. Rates below 6MHz give you a 14 bit ADC rate (analog to digital converter, or conversion) which gives you better signal-processing capability, a lower noise floor and better dynamic range for detecting weaker signals. Note that at 9 and 10MHz you only have 8-bit ADC which is the same as the standard RTL-based SDR dongles. At these bitrates you would be wasting the RSP1A’s performance capabilities for the sake of monitoring more frequencies and would be losing a lot of messages. From the RSP1A specifications:

14bit 2MSPS – 6.048MSPS
12bit 6.048MSPS – 8.064MSPS
10bit 8.064MSPS – 9.216MSPS
8bit >9.216MSPS

You want to be able to monitor 5, 6 and 8MHz which are 5451 5502 5508 5514 5529 5538 5544 5547 5583 5589 5622 5652 5655 5720 6529 6535 6559 6565 6589 6619 6661 8825 8834 8843 8885 8886 8894 8912 8921 8927 8936 8939 8942 8948 8957 8977.

To find the correct sampling rate for the frequencies you want to monitor, subtract the lowest from the highest, in this case 8977-5451=3526. This looks like it will fit within the 4MHz sampling rate but the SDRPlay actually only covers around 80% of its capable bandwidth due to some attenuation at the edges due to its internal filters, which you can actually see in the FFT of standard SDR software such as SDRuno, GNUradio or SDR++. 80% of 4000 is only 3200 so you will need a 5MHz sample rate (80% of 5000 is 4000 so 3200 will fit within that sample rate with room to spare.) I use Percentage Calculator to find 80% of the sample rates for figuring out which one to use.

2 Likes

I had tried to adjust the angles of my horizontal wire loop antenna to make it a bit vertical over the past week to 10 days. Was just curious about where new planes might appear. I got some of that, but in general aircraft count was a bit lower. Yesterday I moved it back up to nearly level 60’ feet or so on all sides. That’s given me two sunsets with over 300 aircraft with a one-hour VRS map timeout. Antenna height seems to have more impact than anything else I tweak. HF is very forgiving.

I’ve been seeing the occasional 9,000+ nautical mile aircraft. I think 9,300 nmi is the farthest I’ve seen so far. That’s about 10,500 miles which is approaching the 12,450 max distance possible between any two points on earth. :heart_eyes:

Here’s tonight’s VRS map with a 3-hour timeout about an hour and half after sunset.

1 Like

Generally, for horizontal shortwave longwire/random wire antennas the higher the better. Transmitting antennas should ideally be at least 1/2 wavelength high at the lowest frequency you want to use, otherwise some of the signal gets sent up into the sky where it’s wasted and the lower the antenna the more that happens. 60 feet is a little over 18 meters so it’s a wavelength high at around 16MHz, 1/2 wave at 9MHz, 1/4 wave at 4.5MHz and so on.

This probably isn’t as important for receive-only antennas but they are more efficient toward the horizon when they’re high like yours. You more get more groundwave signals from relatively nearby airplanes than people with lower antennas, signals that haven’t been refracted by the ionosphere. OTOH lots of people have reported getting good results from using the top wire of a wire fence as a receive antenna, though for general shortwave listening, not specifically HFDL reception.

With the antenna high like yours you’ve also moved it farther away from local noise sources which I’m sure is helping your reception. BTW I suggest including a lightning arrestor connected to a good ground near the antenna end of your coax. That will also dissipate any static electricity caused by the wind blowing on the wire and keep your house from catching on fire if there is any nearby lightening. It’s not quite as important for someone with a shorter antenna than yours (they can just disconnect the coax if there’s a thunderstorm) but you have a lot of wire up there.

1 Like

For receive-only antenna, any wire with something cheap like the Balun One-Nine seems to produce decent results on multiple bands. And yes, even at low heights. I did deep dive and learn a bunch about antenna rules for TX too.

After the YouLoop disintegrated in wind, I tested all sorts of wire antenna configurations and shapes. Height rules no matter what the shape. I save the NanoVNA charts with each test. It is worth noting that different lengths and shapes do change the noise floor and other characteristics that can impact receive-only. It’s just, other than height, those impacts are limited for a receive-only use case like HFDL at least. What was more interesting to me was where the aircraft would appear, depending on the wire shape.

Oh ya, I’ve got lots of grounding and chokes going on with HFDL, ADSB, UAT, Airband, Scanner antennas. When initially testing the first pretty long wire dipole with HFDL, I quickly noticed some probable static effects on my server/network/UPS rack. Had not noticed that with small vertical antennas coming inside before. Now even when quick testing, I ground the coax lightning arrestor to a grounding rod I added outside. And now even the indoor server rack and splitter used with HFDL are grounded.

One thing haven’t solved is bonding the grounding rods I put in with the house grounding rods. I understand ground loops and why that is recommended, but it isn’t really feasible with distance and structures in the way.

My ADSB and Airband antennas and RPIs have their own ground rod about 130 feet from the house ground rods. RPIs are mounted on an actual yard shed (all powered by POE with ethernet cables grounded on both ends, one to the house ground rod…). HFDL COAX is grounded before entering the house on a ground rod shared by a scanner discone and digital TV antenna, but there would be about 40 feet and a hard turn and going under a concrete walkway to bond that rod with one of the house grounding rods.

Luckily in these parts lightening isn’t much of a worry, but that wire antenna certainly brings in tons of static.

The coax or the two boxes? How much wind and what happened? Because I have a YouLoop that I want to use with my RSP1A once my LaNA HF arrives. I’m just wondering what happened to yours so I can gauge how reliable mine will be or if it’ll fall apart.

I’m not an electrician but I’m not sure they’d be electrically bonded anyway due to the length of wire you’d need to run to connect them. Anyway, where you live the soil should be damp most of the time which I think may eliminate or lessen the chance of ground loops.

By the way, what day of the week do you think summer might fall on this year? :grin:

Seriously though, what SDR are you using? Airspy Mini plus an upconverter or HF+ Discovery?

Thanks for the advice gents. :slight_smile:

Some success this morning on 8, 11 & 13MHz, I’m happy with that for the moment, it’s certainly a nice addition to my 4 ADS-B receivers.

You guys have some great results, maybe I need to have a play with antennas again, I’m currently running on my 14MHz EFHW, I might swap to 9:1 random and see how that goes, I also have a YouLoop sitting in the drawer.

I need to sit down and do the math with multiple bands, I’m getting errors at the moment when I try to monitor mulitple bands, has anyone tried to monitor all bands between 5-13MHz?, is it possible?, this is the first time I’ve used an RSP - I’m an Airspy guy!.

Next job is to make all of this start automatically if the Pi shuts down. :slight_smile:

2 Likes

A nice option is --freq-as-squawk and displaying it in VRS’ columns. That way you can get a feel for the most active band. Also, in VRS, displaying ā€˜Distance’ as davidinjp has done.
Cheers!

The connection point with the coax. The taped coax had broken free from the connector. Probably weakened by tight taping, 20 mph wind and rain. I did also try regular coax with SMA adapter, it worked okay but long wires worked better.

July. Actually, since the epic dry summer of 2015, we’ve had almost 90 days of warm without rain every summer. Every summer has turned into basically central California weather recently… Some native plants are dying (azaleas, rhododendrons), even quasi-native (birch) trees dying as bugs eat them with warmer winters.

  • The original two HF+ Discoveries (17 & 21 MHz)
  • RSP1A (8 - 13 MHz)
  • Airspy Mini + Up converter (5 & 6 MHz)

The SDRplay is technically on loan/swap from a ham friend who rarely uses it. Without it, I had a cron job and script change Airspy from low to mid frequencies in the day. But I prefer having 4 receivers with my fancy unbalanced input filter/LNA/4-way splitter self xmas present. That device does good things to the unbalanced signal which I’m perceiving even helps a bit with receive-only.
(Red is with, green without the SV1AFN splitter/LNA)


I swap frequencies with Airspy & sdrplay or run them side by side on same frequency to compare along with HF+. Some difference in messages received occasionally, but they all work out to roughly the same aircraft count on the map. HF+ may slightly outperform the others, but limited sample rates limit usage,

Interesting antenna
Tuning capacitor of the small loop antenna with tuning motor & copper interconnects.

1 Like

Looks like that would have no problem with wind. :grin:

Geez, dude! What are you doing??? You can’t let that get out!!! :smiley:

Correction - it rains here. All of the time. It is gray and dreary and cold and damp and nobody would ever want to live here, much less move here. It’s basically London, but without the amusing accents.

1 Like

And our favorite sport is throwing salmon around! :slight_smile:

Really nice program. I’ve been receiving HFDL for many years using a licensed copy of PC-HFDL, and have enjoyed the spotting, but this takes it to a new level, and allows me to automate and have a standalone Rpi do the work. I look forward to playing with the stats on this, as it’s useful for propagation studies given we have these airborne transmitters operating at relatively low power 24/7 all across the world.

The image is my tracking of flights on 17Mhz at 2220Z from my Virginia location.
DumpHFDL, Airspy HF Discovery / Inverted-L antenna.

3 Likes

Nice! If you have another HF receiver or SDR and a second antenna (or a low-loss splitter) you might be able to hear shortwave broadcast stations in Africa. They have some cool-sounding music. I used to be able to hear Africa Number One in Libreville, Gabon on summer mornings when I lived in Seattle. That was my introduction to the soukous genre from what was then Zaire, now Democratic Republic of the Congo: Kanda Bongo Man, Diblo Dibala, Loketo, etc. Soukous is, or was, popular throughout most of Africa and has influenced the musical styles of many countries there,

If you figure out the statsd and grafana stats setup mentioned on the dumphfdl readme, please post a rough guide on how to set that up. I gave it a rushed effort a few months ago and failed. Not familiar with Grafana setups. :laughing: