Whilst enjoying my studies of Machine Learning I had the radio on in the background. I was very interested to see some ‘patterns’ on the 40m band
Interesting patterns in the waterfall
Not right at this time, but earlier in the day my fellow ‘neighbour’ ham G7VRD told me these are OTH. What is that ? Well this is Over the Horizon Radar, where radio signals are based off the ionosphere and then measured. They have been around a while, as in the 1950s, but technology has ‘evolved’ and ‘new’ arrays are being built ! These patterns were throughout the 40m band most of the evening ! Whilst annoying if you wanted to use the band for its intended use, nevertheless provided some interesting reading on what OTH was and how it worked.
FT8 and JS8 Call
Whilst I was busy with Juypter Notebooks, and as I now have the antennas plugged in thought it interesting to monitor JS8. Its nearby digital mode of FT8 gets plenty of traffic, JS8 is very sparse, which seems a shame as it looks a good way to ‘chat’ via FT8 digital mode.
I sent out HB and ACK’s and it was clear there are people listening, but not many sending CQ, nethertheless it was an interesting evening and this somewhat ‘dubious’ amateur transmission was observed, whilst interesting am wondering the legitimiacy of it.
Covid Stats transmitted into JS8 Call (no call signs only a reference of the source material)
I enjoyed seeing JS8 Call in operation though and didnt interrupt my studies, so i could do a bit of both.
late into the night… that’s data science for you and having a day job
So i enjoyed my FT8 and Data Science, having worked thru the night and with work-work the next day headed for bed eventually at 4AM.. 🙂
As i have been doing more ‘re-arranging’ in the QTH, I thought I would revisit the snippet from the previous weeks posting on NOAA & METEOR decoding. I have evolved from ‘home brew’ antenna and tools on OS-X to fully automated and a specialist NOAA/METEOR antenna.
Hopefully by the end of the post, you will get a feel and and idea of where you would like to start or explore further !
My first interest in weather and amateur radio came from receving Wefax images. I still do this as it complements and also gives me some idea of interference/reception issues as i can usually clearly see any issues in the fax.
My favourtie charts are the UK ones available from on 4608 Khz trasmitted by Northwood.
<insert pic>
I find the detail and various types of graph really satisfying to read and decode via HF.
Following on, i found on youtube about a simple antenna and using SDR receiver to decode ‘NOAA’ satellites
where it started
The audience here is very clearly windows users, and whilst I have a Windows 10 machine for HF Digital modes, i wanted to keep the SDR seperate from that system.
I built the antenna from bits I had around, although it did take up quite a bit of decent low-loss coax to get it up a reasonable height.
Original home-brew NOAA antenna
Get the antenna orientated well North-South effected the signals the most. Whilst heigh was important, gettign the polarization brought about the best results.
first NOAA decodes via standard SDR, gqrx and
My first reception pics, whilst not amazing, really pleased me as the technology was at least working correctly. On a mac the missing component was being able to decode the ‘wavs’ to images, for this i used https://github.com/artlav/meteor_decoder which was easy enough to build via homebrew.
I monitored https://github.com/csete/gpredict where i could manually and getting the direct overpasses with the V antenna produced very good results.
results when following a direct pass over, still some interference, but clearly visible UK and well defined cloud structures
I continued to do this for some weeks and built up my collection of NOAA pictures. I had still yet to sucssfully decode METEOR-2 as that was a digital signal and passing times were not in favour of a day-time working schedule.
We are still currently in a Covid-19 situation as time of writing and since Feburary 2020 here in the UK, so whilst many designs of NOAA antennas exist, I very much avoid supermarkets/large DIY outlets,etc unless absolutely essential, and usually for ‘click and collect; (Order online, pick u pin store, no wandering around).
With that in mind i reached out to Dr Google to find pre-made NOAA antennas. This thread on reddit https://www.reddit.com/r/RTLSDR/comments/8biful/is_there_any_place_to_buy_a_decent_137_mhz_noaa/ gave me the link to the National RF antena http://www.nationalrf.com/satellite-tenna.htm. These are built to order, so there is a wait time, but it is well worth it. I was kept fully informed of progress and still had my V antenna to keep me going. Packaging from the US was fantastic, very well protected and assembly was very easy. I did have to get some PVC pipe to use a ‘mount’ to the mast and a BNC to SO39 adaptor, but these are very easy to come by and have no real loss to the to the receive-only antenna.
National RF antenna up approx 2.5m
I used my VHF mast, replacing my 2m/70cm antenna with the NOAA antenna. Whilst the mast can go up 30ft, it requires careful observation and maintenance to keep that high. Instead, i opoted for just over roof height of 2m and a clear line of sight to the horizion.
The antenna greatly improved the reception of signals as later pictures show, albeit I am still using the mac and GQRX and a simple SDR at this time.
the helical antenna greatly reduced interference and improved the length of received signals
I continued with the Mac, SDR, GQRX, Gpredict and meteor decode for quite some time, but as i got busier with work i had no time even to manually decode, as much as I enjoyed it !
The next step was to further improve reception and automate. This was accomplished by purchasing filters and amplifiers specifically for the NOAA RCPT frequencys in VHF and then re-cycling a Raspberry PI3.
The original video gives the necessary filters, but for a shortlist here they are
NooElec SAWbird+ NOAA
Flamingo+ FM – Broadcast FM Bandstop Filter v2
NooElec NESDR SMArTee XTR SDR
As I wanted to automate this and provide a simple way to just look at the received images i used https://github.com/reynico/raspberry-noaa repo, which was by far the easiest package to setup and use on my Raspberry pi.
If your not familar with linux/unix this could be a bit of a blocker, as you do need to manually edit some configuration files, so being familary with standard OS commands and a text editor like ‘vi’ i would say are the ‘essentials’ to being able to use this. I dont mind saying i had to refresh my memory on how ‘at’ the scheduling tool worked having always be a ‘cron’ person.
In practice once setup, there is very little to do, but what i have done is to alias ‘atq’ to make the list of tasks in date order.
This makes reading the scans far easier and i can still use gpredict on my mac to see which ones are the most interest to me. The software will automate, and try, every scan, but if rtl_sdre is already running, will not be able to run. In that case its always useful to ‘prune’ out the less interesting or scans which will suffer the greatest interference. I always priortize Meteor passes over NOAA passes as Meteor decodes seem to be. a maximum of 2 per day. This is easily managed by the combination of ‘at -c <job>’ to ensure no duplicate tasks are running.
I have been running on a PI now for just over 2 weeks, and as part of the QTH tidy up am moving the Pi from a ‘desk’ to a shelf where it will be safe and less prone to me knocking the amplifier and notch filter out.
I am amazed by all the scans i can receive and thoroughly enjoy understanding how it works, and being a recipient to the amazing images that both the NOAA and Meteor satellite send for free !
Example NOAA IR via the Linux Raspberry PI setup (some interference due to building work near QTH)Meteor decode – note the interference here is to a digital signal, so rather than static, its loss of pixels in the decode.
Overall i can say that investigating the weather satellites has really complemented by amateur radio activities. I’ve learned how important good antenna design is by the progression from a simple V dipole to a professionally built helical antenna, and the use of amplifiers, band filters and the right SDR unit to use for the hardware available.
I have since purchased 2*Pi4 as I would like to use a Raspberry PI more with the other SDR’s i have (Airspy, HackRF) to learn to program and enjoy the decoding more. Building good antennas will only help what i need for future amateur broadcasts as well.
Overall, i can really strongly recommend decoding weather satellites, if anything you get amazing images of your QTH and combined with WeFAX you can make your own weather predictions !
Have had a great weekend on the radio logging up QSO’s via FT8. I’m now also using LoTW and importing those into QRZ – this has helped with getting more awards. I dont think i will do any more ‘logging’ tho as managing two (and wanting to automate them) is getting enough !
Was really glad that I got another award via QRZ – the grid squared award !
Grid2 accomplished !
For the non amateur radio users what this means is that the globe is segmented into ‘grid locations’, the reference codes being ‘maidenhead locators’ – for this award there has to be 100 unique and confirmed grids. Was very happy to get this done on my foundation licence.
Also was very lucky and got a QSO in the Canary Islands – which is officially classified as being in Africa, which means I’m now only one continent away from the continents of the world award !
Whilst i have heard south america, i’ve never yet managed a confirmed QSO – so it will be quite a challenge but hopefully with the summer hear propergation will improve and i can make that QSO !
Its been a lovely warm weekend down in Dorset, and that seems to have affected the propagation. I had a great time on FT8 and JS8 Call, reaching amazing distances and really good signals being put out and received. I had many logged contacts and good signal reports.
40m got *busy*
By the evening 40M got as busy as I have *ever* seen it, as shown in the picture above i switched about the bands and was making contacts on 15M as well, it was the best conditions I had seen so far !
Resume work on the Kanga Kit !
After a break of sometime, I have resumed work on my Kanga Kit – this fits in with the news I’m putting below, but I have a big thanks to John Clements who helped me fix a fault with the kit I was unable to locate. John found the issue (soldered pin-header together – hidden by the plastic) and returned the now working Kanga kit to me ! I have resumed work and am taking my time – I’ve completed more work on the resistors and capacitors on the Active BPF.
kanga kit with added capacitors for BPF stage
I’m taking my time, esp as the board is now quite ‘busy’ – the next step is to add diodes and IC’s. Its really close to completing the build, but taking my time to ensure no more mistakes !
RSGB Announce Intermediate Exams on-line !
As shown in the picture above from the RSGB site Intermediate exams will be available from the 13th ! (in 8 minutes time as of writing) i’m very excited that i will be able to book my exam and get use more wattage but also gain more insight to a very interesting hobby as I continue my amateur radio development.
Whilst the practial assessment is waived currently, i will continue to build my Kanga Kit, so if the issue did come up of what Kit I did build, this is well recorded and documented.
To close the evening off I had a splendid evening enjoying the good propergation and just listening to the QSO’s on 40m. The ICOM’s filter and noise blanker are amazing, SSB has never sounded so good.
The waterfall is incredibly useful for visualising contacts
What i really like about the 7300 is the ability to visualize all or in very close range the frequency. I can easily find conversations happening and enjoy what is being talked about and the signal reports. Almost everyone i hear on 40m is using 100W but the distances are still several hundreds of miles – in particular hearing about life on Guernsey was very interesting – seems they are Covid free and back to some ‘normality’.
Well, heres waiting on being able to book my Intermediate Exam and enjoy more contacts around the world !
UPDATE – Have booked Intermediate exam today (13 June) for in August for more time to do a structured revision plan and passing !
Am still planning to finish my Kanga Kit ! making daily progress !
So I have been using Wefax for getting fax weather transmissions. I really enjoy them and also find it useful in seeing how much QRM I am getting on the ‘wire’ so to speak. Recently a gentleman on youtube by the name of “Tech Minds” published this excellent video
Build a V-Pole for weather sats
Now I loved the instructions on this video and it has been something I’ve wanted to do. I asked “Tech Minds” if i could put the V-Pole adjacent to a 2m/70cm vertical, to which he said yes, so i was then on a mission to build my V-Dipole.
I already had decent electrical terminator blocks, so didnt need to order those.
I followed the instructions and set about mounting the antenna on the mast – i have to say i found this quite challenging on my own on how to attach the pvc pole to the mast, but sure enough, and with a few ‘oh dear’ (swearing may of been harsher) on dropping nuts/clamps i got it attached.
I then fed the coax from the mast back into the ‘shack’ where I put a PL-259 socket on .
I do love this video, and the guys no nonese approach
the method i use for puting on a PL259 onto coax.
With that i put the antenna via a PL-259 – SO-239 to SMA cable attached it to the RTL SDR Dongle.
Sure enough, i could pick up radio sounds no problem, so all the effort was worth it !
I then set about installing all the necessary software on my HAM computer, following yet another excellent tutorial from Tech Minds
Setting the software up on Windows
I initally tried with the HackRF, but for some reason it wasnt playing ball. Ironically as I knew that a sat should be overhead, i hooked up my other SDR to a basic scanner antenna, and i could hear the sat overhead ! Immediatly i set about setting up my mac for the receiver.
On the mac I used gqrx. The satellite track software is avaiable via macports, but i couldnt find a stable 64 bit version of translation software so used noaa-apt, which whilst not as feature rich, works incredbly well on a mac, as in being actively developed.
I tested with a ‘sample’ NOAA recording, and that produced the required map output ! iwas all set.
I waited patiently for the sats come and tracking and sure enough within an hour i had my first sat pass over ! Whilst the intial result wasnt amazing, it at least proved that antenna, SDR and decoder worked.
My first NOAA decode !
I then looked for future NOAA passes and a good candiate for going directly over was on for 6pm, so i waited and recorded, sure enough, i got a signal and a decode, and whilst the results arent amazing, its a good start in my book.
Second attempt at 6pm
As it was some time before the next pass i set about doing some fettling on the antenna, and found i hadnt got it North-South Aligned that well, so i rotated the antenna and made sure that all the connections were in good form after a full day up in the air.
checking connections
looking good, but those trees !
nicely on the mast, but requires orientation
checking after one day
I’ve not really done a write up of just how good the mast is I use for VHF/UHF, but you can see in the pictures below how high quality it is. All the stages are locking and push up/down without problem, even after rain, they is no fricion on the joints. The locking mechanisim is tough, but gives real confidence the pole isnt going to slip down. I use a ‘light’ guying system as i’ve only got 2 very light weight antennas on the mast (the 2m/70cm and now the NOAA) – any more would be overkill. The guy ropes and ground stakes are incredbly strong and go into the ground some 2ft at a guesstimate, the antenna only ever comes down for the worst weather forecasts.
If you want to invest in a good mast system i can really recommend these
With the antenna back up I now wait for the next good pass to try out how good the scan comes out. Its very exciting (for me) and thanks again to Tech Minds YouTube channel who puts really great content !
I’ll upload my pics to my STOATOPIC account as I get more decodes
gqrx receiving noaa 18gpredict communcating with gqrxtracking NOAA18 as it flys over my QTH
Tracking pattern
tracking of the pass, i can pick it up from 21:43Decode in NOAA-APTIts another fuzzy one!only the code-strip is visble, but keep trying !
So last night (11/5 – 12/5) Turned into a bit of a QRM mission. Not exactly sure how, but found myself really getting trying to find the source of a specific hum, as I am gradually reducingt the amount of interference/QRM I am picking up. I started with taking my little AM radio (D-328) around the house buzzing very present. What I wanted to rule out was the utilitys nearby and the garage. Being 1AM and wandering around with a pocket radio should make for amusing footage on the overnight security video from the garage today 🙂 Anyhow, I could pick none up.
I returned home,sure enough the ‘buzz’ was back. So as it was quiet in the house (trust me, we are 24/7 shop here!) i started going thru the fuse box, its an older one, but still thankfully a trip/RCD (?) type box where I can easily flip the circuits. Sadly the circuits are not labelled, which I find quite unprofessional, apart from the ‘light’ fuse. I started flipping fuses and on the 5th one the buzzing on the radio stopped, as did all the mains power to my study and I think the front part of the house.
I set about unplugging everything in my study with the mains back on, hum still there. So I started to research/googling hum between 50 and 500Khz – it seemed so precise, I figured it must be some form of ‘man made’ interferance.
Here is a video of what it sounds like :-
So you can hear it all the way up from 50kHz to a very loud abrupt end at 500kHz. I set about googling as to what that could be, and sure enough found th GM4FVM page on ‘power line adaptors’. I’ve been very careful to remove and limit the use of ethernet in the house now, so suprised that this had caught me out. I immediatly removed the BT Ethernet over Power adaptors I had. This is the result
A longer video with manual AM tuning and a ‘sweep’ of USB
If you are patient enough to have watched the whole video, i congratulate you 🙂 But you can hear the big difference between what a power line adaptor can do to HF/RF in a shack, i.e. completely destroy all but the strongest signals.
This morning I tried out WSPR, the results on receive and transmit speak for themselves :-
wspr map is somewhat busier…
So now its time for work but I think i have taken another big step in reducing QRM in my shack
The DX Commander comes with 100m of very good wire, which is enough for making the 4 element (40,30,20,17m) vertical and some for radials. Having already bought 50 meters of wire for 80m I was still down on the count for both radials and the ‘all band’ – having complete the ‘all-band’ construction it was now time to sort the radials !
With the orignial wire and radials I had around 28-30 radials, which whilst much better than only 1 and improvement on 15, is still slighty short for 80m and also more radials=higher Db out at other wave lengths.
So what is the science about radials ? The Calum has a good video to help, and the references are very good, so I’ll put that here first.
https://www.youtube.com/watch?v=qVG1jevrXaI
Radial videos from Calum
So basically adding radials helps (I did try to find a video of a person fallilng off a boat getting onto a pier, but was getting to distracted…) I took 50 meters of wire (the same type I had use on the verticals) and set about making into 3.5 meter lengths, which some great help from my son Paul which saved my back alot in terms of getting up and measuring and cutting !
pull and cut thanks Paul 🙂
it started as 50m, it became radials
cutting radials
With the cutting done, it was time to return indoors for strip, crimp and solder !
new crimper doing a grand job
solder together, solder the crimp
strip, crimp and solder (*2)
I then returned outdoors to install the addtional elements. I doubled up on one existing element, but had a pretty decent fan of 3.5m wire elements going on now.
a nice spread
view from the base
keeping it tidy
gaps to fill
now with added elements
adding elements
I’d need to an exact recount (will do so later) but I think I now have enough radials to accomodate the 80m vertical.
With having added the radials it was time to test ! For this I use FT8 as it has a great map and includes a signal report when a QSO is complete.
Reaching Kuwait on 17m
I was now reaching Kuwait on 17M and even better was that I could see the call sign in my WSJTX screen – i had to give it a try ! And sure enough within a few minutes (this guy was was getting alot of QSO’s in) I had made contact !
Contact with 9K2OW in Kuwait – a staggering 2,888 miles on 10 Watts of power !
So i think i could continue to add more radials, but for now I’m happy with how the DX Commander is performing and the radials are performing their function. I will investigate more ‘science’ based ways of radial performance, but his helps for now !
So its been a bank holiday here in the UK and its allowed me to work on quite a few different things (including re-wiring the Mazda Bongo door looms). When I got the DX Commander it came with enough wire for 4 verticals and the radials to support it, but I purchased some more wire from Radio World, namely the Watson Radio Products EQ Equipment Wire which goes for 40 per meter. I used Radio World as I had other bits and bobs coming from them as well, and my word, they are so QUICK to deliver (UPS Everything !).
I checked from the list what additonal wires I would need to make up., in this case only the 10m and 12m. Having learned from trying to measure the wire indoors I now measure all my wire outside. I have 3 meter workmans metal tape which so far has done me brilliantly in getting accurate measurements.
So the first part was to obviously lower the mast, and then remove the existing wires, Going from 4 to 6 requires adjustment as per DX Commanders docs (pg 2) here.
I also reprinted all new labels, removing the existing ones. One thing I didnt do orginally was to put labels on the bottom plate, so this time I would ! First I organized the cables so they was easy to pick up add, not having things on the floor and in a mess really adds to health and saftey when doing this kind of work.
wires in respective order of frequency (10,12,17,20,40,80 – the long green monster!)
With the mast lowered I followed the instructions and relocated the SO239 feed wire and the 80m connection
movement of feed wire and labelling up
What really helps is the DX Commander stickers. On my spreaders i do them so that they should be aligned when looking down the mast, this gives me a good point of reference.
To start with I just feed the 80m wire up to where it would be for 30m and taped it down. I then worked around the mast feeder positions adding labels as I worked in each new wire.
its not just a great little radio
its a box for all the stickers!
keeping the stickers safe
Now the DX Commander/Calum does say that the clips used on the paracord are hard to undo once set, as I found. Having to try and extract one, manage to break a clip, and I had no spares. So, i thought, well superglue it is then ! With using the YOCTOSUN Hands Free Magnifier, i was able to get it to glue pretty much back on. I was then with the existing cut paracorde and elastic cable to add the other elements which required some vertical tensile strength.
repaired clip holding out
loopbacks doing their job
some additonal ‘temp’ fixes
12m, tied directly to the lower guide/spreader
I continued to add all the elements and was happy with the tension and how tidy they are. I also took some time to put some tape around the clips where sometimes the wind will blow and wires get caught in the juberliee clips. As I’m not taking the mast down and collapsing, this wont effect me in the short-term.
nicely labeled plate
tidy going up
some ‘fettling’ with tape for now
wires in place
With assistance from my son, we got the mast vertical and I could start adding the existing radials I have. I add the radials in a N/E/S/W layout starting from the feed point, and adding where i have capacity. I also done a ground-visual of the 2m/70cm with my iphone (10x max zoom) and could see it was in good shape up there after I had raised it on Friday.
approx 28-30 radials
spreading them out i go N/E/S/W, then fill the gaps
A quick visual on 2m/70cm antenna
radials and a visual of the 2m/70cm mast
Testing and Results
So the next step would be to test. For this i used the two main data modes I currently use, being FT8 and WSPR. I used WSJTX v2.1.2 and finding a ‘gap’ to press the ‘tune’ button. This would give me repeatable results. I am using the TS-690S internal SWR reading and a 2nd hand HF SWR Transceiver (YW-3) for the external readings.
Band
SWR Transceiver
SWR Meter
Notes
10
1.1
1.3
12
2.2
80
17
1.4
1.6
20
OFFSCALE
2.5
RX is very clear 5/9
40
2.2
1.1
80
2.1
1
160
OFFSCALE
To be expected
30
OFFSCALE
80m wire ?
15
1.6
3
Does the VV-3 work at this frequency ?
12
2.1
3
FT8 Testings with new elements on 10+12m and resonant frequencies/others in WSJTx
Band
SWR Transceiver
SWR Meter
Notes
10
1
1.2
5 Watts is the min setting on the transmit
12
2
2.5
17
1.3
1.3
20
OFFSCALE
40
2.9
1.1
80
1.5
1.1
160
N/A
30
N/A
15
1.4
1.5
12
2
2.1
I am really impressed with the S.W.R. on all the bands, whilst 20M gave a high reading, i suspect that the curve on this band is quite specific. Reception is very strong, so the wire is doing its job in selecting the fequency nicely.
Whilst I wasnt expecting 160m and 30m to be available, for thorughness and future recording, added them (should of done 6m in hindsight). I was happy to see 15m and 12m resonating nicely with low S.W.Rs on of 1.6 and 2.1 respectively.
I will upload a gallery as there is so much evidiential data to show,so browse thru at your lesuire.
