Installing a radio into my car

About a year ago I changed my car and it’s taken a while to get around to fitting the radio into it. I’ve missed having a wireless in the car and recently resorted to keeping a handheld in the glove box so at least I could natter on the local repeater while driving around town.

Although I could probably have done it myself, when it comes to installing radios in my cars, I always prefer to have someone do it for me, normally someone who does this as part of their job and will have all the tools and know all the tricks of hiding the wires out of the way and making it all nice and tidy. I just don’t have the patience!

Last week I travelled to Chelmsford and met up with Nigel, G6ZVV who installed my Icom ID-5100E into my Mazda 6. He did a great good job, it’s neat and tidy and the control head is easily useable and visible without blocking my view out of the windscreen at all. Icom don’t produce a proper mobile mounting bracket for the control head so I have their MBA-2 paired with a RAM Mount RAM-B-102U-A which I imported specially from the USA. For what it’s worth, I think it’s pretty crazy that the only bracket Icom produce is a suction mount.

Mazda 6 with Watson W-770HB
Mazda 6 with Watson W-770HB

The installation itself was pretty straightforward, Nigel ran the cables through and took a feed direct from the battery while I mounted the aerial on the boot. I can’t remember the make of the boot mount, it’s either Diamond, Watson or Sirio but it did me well on my previous car and although it looks a little tattered, it’s perfectly good.

The aerial I use is a Watson W-770HB which I’ve had for years. There are many similarly specced aerials on the market (it’s a half wave on 2m and 2 x 5/8 on 70cms) but out of all I’ve tried, this has been about the best. I tried replacing it with a Diamond aerial a couple of years ago but that didn’t perform as well so I returned it and put this old one back on.

The main body of the radio is mounted in the boot, underneath the parcel shelf. It’s accessible enough that I can get to the SD card slot but positioned well enough that it won’t get in the way of anything I put in the boot. It’s neat and tidy.

Icom ID-5100E mounted in the boot
Icom ID-5100E mounted in the boot

The important part is how well I can see it while I’m sitting in the driving seat and how accessible the radio is. I wanted it to be easily viewable and in a position where I can operate it without having to move my hands far from the wheel or my eyes far from the road, without actually blocking any of the view out of the window. I appreciate this is a tall order for modern cars but I think that I found a good place. This meets all my requirements and was surprisingly easy to fit.

Icom ID-5100E control head
Icom ID-5100E control head

I mentioned the mount before, here’s how it looks from behind.

The back of the control head mount
The back of the control head mount

When I get out of the car, it’s a very simple matter to remove the control head, disconnect the cable and take it away for security.

The mount without the control head
The mount without the control head

I was a little concerned as to whether I’d suffer from any electrical interference from the car or whether transmitting high power on 2m or 70cms would interfere with anything within the car but it all seems to be OK. I had a long drive from home to Manchester and back just a couple of days after installing the radio and it worked very well.

It’s nice to be mobile again after over a year with no radio in the car. I’ve missed it!

Additional – I’ve been asked how I ran the power from the battery through the bulkhead and into the car so here’s how it was done.

Power cable from the battery
Power cable from the battery

The power cable comes from the battery and is fused, then there’s a section of plastic held in place by a number of clips similar to the one marked in yellow. Those clips simply prize off and leave space for the cable to run underneath.

Power cable comes out from the plastic tray
Power cable comes out from the plastic tray

The cable comes out from the tray at the other end of the car (note, this is a right-hand drive car) and drops into the space down by the door – The picture below should help explain this.

Power cable dropping down into the space by the door
Power cable dropping down into the space by the door

Once it’s dropped down the gap, it’s a simple matter to bring it in at the bottom of the door, loop it over the door seal (held in place with a cable tie and drop it under the trim and into the cable run to the back of the car along with the cables for the control head, microphone and speaker. This is the only place that the power cable is visible. As you can see, it doesn’t actually go through the bulkhead as we couldn’t find an easy and available place to run it.

Power cable coming in through the bottom of the door
Power cable coming in through the bottom of the door

Posted in Amateur radio, D-STAR, FM, UHF, VHF | Tagged , , | Leave a comment

Essex Radar

I’ve been running an instance of Virtual Radar Server (VRS) for some time and recently decided that this is something I can easily share to anyone who is interested.

What is VRS though?

VRS is a server running on my network that can take a feed from multiple ADS-B receivers, consolidate them together and provide a decent graphical interface showing all the aircraft tracked. Unlike most of the commercial services, this is unfiltered and real-time. Of course, it only shows aircraft that I’m actually receiving signals from but I do have a pretty decent setup here so the coverage is good.

Essex Radar

A project like this needed its own website and so Essex Radar was born at http://essexradar.co.uk.

I have two receivers feeding aircraft to Essex Radar – One is in my loft and is using a Raspberry Pi 3 Model B, a FlightAware Pro Stick Plus SDR dongle and a homebrew two element J-Pole collinear. The second is outside on top of my mast and is a Pi 3 Model B+, an Airspy Mini SDR dongle, LNA and FlightAware 26″ aerial.

Essex Radar was used to create this 24 hour timelapse video of Boeing 747s.

Monitoring aircraft is something I find quite fascinating and it combines two of my interests, radio and aviation and is something you can do quite cheaply. Although I’ve briefly described my two receivers above, I’d like to give a few more details and options as to what you need.

Be aware, this is going to contain a lot of links to various products.

No matter what you choose to do, you’ll need to start with a single board computer and in my opinion the best one to use will be a Raspberry Pi. There are others on the market but the Pi is the simplest to get hold of and appears to have the best support.

At time of writing, the current Pi is the Raspberry Pi 3 Model B+ and if you don’t own a Pi, this is the version I recommend you buy. If you already have a Pi then it will almost certainly work as an ADS-B receiver although I’m not sure I’d want to use anything older than a 2B. There are many different suppliers and they’re all around the same price.

You’ll need a power supply for your Pi. This must be a good, stable PSU capable of at least a couple of amps at 5 volts. My recommendation is the official Raspberry Pi Universal Power Supply but another option is to run a less stable PSU at higher voltage and drop it down using an LM2596 DC-DC converter. I know from bitter experience that some power sources are particularly bad. My ADS-B receiver which is mounted on top of my mast was originally running from PoE but continually gave me under voltage warnings. To resolve this, I switched the output from my PoE device to 9 volts and used an LM2596 converter to drop it to 5 volts. I no longer have any power problems.

The other essential item is a memory card. Not all SD cards are built equally and some are significantly slower than others. I’ve had great success with SanDisk Ultra 16 GB MicroSDHC UHS-I cards and use them in all my Pis.

A Pi, PSU and memory card will cost around £40-£45 depending on where you buy them from. I’m not going to recommend anywhere specific for these three items because prices change regularly so just shop around for the best deal.

The next thing you need is a receiver. If you want to do this the cheapest way possible then you can get an RTL SDR dongle from eBay for less than a tenner and it will work but it won’t be very good.

There are other receivers available that cost a little more but in my opinion, the one that will give you the best value for money is the FlightAware Pro Stick Plus. This dongle comes with a built in amplifier (LNA) and 1090MHz filter and is what I’m currently using for the receiver located in my loft. It’s available from various UK suppliers for around £30, works really well and is perfectly good for most people.

If you want more performance then you’re starting to get a bit more expensive and the law of diminishing returns comes into effect. The next step up is to buy a separate receiver and amplifier. If you want to do this then my recommendation is the RTL-SDR V3 dongle and the RTL-SDR 1090MHz LNA at $21.95 and $26.95 respectively and available from the RTL-SDR shop. These will take a couple of weeks to be delivered to the UK and you’ll also need to buy a bias-tee module as well which will add another tenner or so to the cost and further complicate the build.

I think that the next upgrade from this setup would be to replace the receiver with an Airspy Mini SDR dongle but at around £120, that’s not going to appeal to many people. If you really want to squeeze the maximum performance then you should consider a Mode-S Beast receiver but that’s getting extravagant as one of those will cost you around €240 in kit form.

Next, you’ll need an aerial. There have been many aerial designs posted on the internet and I’m not going to link to all of them but I do suggest looking at this thread on the FlightAware forums for some simple and easy to build antennas. The first aerial I built was the spider which is effectively a quarter wave groundplane and you can see a picture of it here. It works but it doesn’t work as well as the next aerial I built which is a two element J-Pole collinear. That collinear is still in use on my receiver in the loft. It’s a superb aerial and it works far better than I ever expected. It was easy to build and as I already had the wire and the connector, it didn’t cost me a penny.

If you would rather buy a cheap aerial then look for an ADS-B PCB aerial but be warned, it may arrive unsoldered and you’ll have to fix it yourself.

If you want a good solid weatherproof aerial to use outside, then I have experience of two different types. Search eBay for COL1090/5-H and you’ll find the same aerial we use for the receiver at the Martello Tower Group site. This isn’t always available but when it’s there, it’s a good buy at around £30 including shipping. These aerials are all built by hand so be aware that performance between them may vary slightly (I know this from experience).

A little more expensive at around £54 but of more consistent quality is the FlightAware 26″ aerial. I use one of these on my receiver here and I can’t fault it. The build quality is very good and mine has been mounted outside at 10m AGL in all weather conditions for nearly two years with no noticeable problems. Apart from being a bit faded, it’s just as good as when it was new.

I will mention one more commercial aerial but I have no experience of using it and can only go by the reports I’ve read. The ADS-B Vertical Outdoor Base Antenna is $149 from DPD Productions and is very highly rated.

Finally – Coax. Buy the best you can afford and use as short a run as possible. Cheap cable is very lossy and this is one area where you mustn’t cut corners. For example, 10m of RG58 will give you around 5.5dB loss, 10m of RG213 will give you roughly 2.8dB loss and 10m of Messi & Paoloni HyperFlex 10 has around 1.4dB loss. Don’t skimp on your coaxial cable.

Once you’ve collected all your hardware, you’ll need to set up the software. That’s pretty straightforward as there are a number of pre-built images available which do all the hard work for you so I’m not going to go into long winded instructions.
I suggest using the FlightAware image and full details can be found here. Because I’ve already gone through the hardware, you just need to jump forward to the second section “Install PiAware on your SD card”. You should be able to set the software up from start to finish in about half an hour. The guide linked above covers absolutely everything.

I appreciate that there’s a lot to take in here and I congratulate you if you’ve read this far! If you have an interest in radio and aviation then setting up your own aircraft tracker is neither difficult or expensive and it’s really fun to be able to see aircraft in the sky near you. If you do decide to do this and would like to be included as part of Essex Radar then please contact me.

The day after I put Essex Radar online was the day of the historic D-Day Flypast where a number of DC-3 aircraft flew from Duxford, over Colchester and Southend on their way to Normandy. Essex Ham were monitoring the flight and they used Essex Radar to track it. You can read more about this and see a video including Essex Radar here. I’m very pleased to have been able to provide this service.

Posted in ADS-B, Construction, Data, Not amateur radio, Raspberry Pi | Tagged , , , | Leave a comment

Moving to VNC Connect from TeamViewer

Technically this isn’t directly about amateur radio although there is a sort of radio connection.

As part of my involvement with the Martello Tower Group, I look after some equipment at two of our remote sites. We have a Linux server running our D-STAR repeater, a Windows 10 PC running our SSTV repeater and a Raspberry Pi as an ADS-B receiver. Up until recently, I used TeamViewer to connect to these computers because they’re linked to the internet via a cellular connection and therefore behind a CGNAT so port forwarding isn’t possible. It worked perfectly well until the following message appeared.

TeamViewer Commercial Use message
TeamViewer Commercial Use message

Clicking the More info button took me to the TeamViewer web site where I was able to complete a form explaining that we’re not using it commercially. The web site says that I should receive an answer within seven days but a search of the TeamViewer support forums produced a thread where people have been waiting over a month for a response and even if they accept the explanation, it’s not uncommon for it to get grumpy again and revert back to being time limited.

I don’t generally need to do much remote work for the two sites and I figured that I could probably manage to do what I needed in five minute sessions so I didn’t worry too much. However I was wrong as the “5 minutes” they claim is wrong. My sessions were being timed out after around thirty seconds which was barely enough time to log on and authenticate.

I looked into various different options to replace TeamViewer with and settled on VNC Connect which consists of two products, VNC Viewer and VNC Server. You install Server on the computer you want to control and Viewer on the computer you want to use to take control.

The process is quite straightforward – Head to the VNC site and create an account, install Server on the system you want to control and and log in using your account details. Install Viewer on your home computer and log in using your account details and the remote computer appears in your address book. This was easy enough on the Windows 10 PC but not quite so easy on the Raspberry Pi.

Generally when I set up a Pi, I build it using the ‘Lite’ version which doesn’t include various features, most noticeably the graphical user interface (GUI). I do everything on a Pi via the terminal rather than the GUI. In order to use VNC, there needs to be a GUI installed. Fortunately, it’s possible to add the GUI from the command line with the following commands:

sudo apt-get update
sudo apt install raspberrypi-ui-mods -y

This was quite a large download at over 600Mb so I just started it going and left it running for a few hours. I have no idea how long it took to install but it was done when I checked later. I had to install VNC Server as well which I did with this command:

sudo apt-get install realvnc-vnc-server -y

Once that’s installed, I had to enable VNC via the Raspberry Pi configuration utility:
sudo raspi-config

Choose option 5, Interfacing Options, then P3 VNC, select <Yes> to the question “Would you like the VNC Server to be enabled?” and then exit the configuration utility by selecting Finish.

Now enter the following command:

sudo nano /boot/config.txt

Scroll down to the two lines that start
#framebuffer_width=1280
#framebuffer_height=720

Remove the # from the start of each line, press ctrl-x, hit the Y key and then return. This sets the screen resolution of the session when you connect. I changed mine to 1680 and 1050 respectively. Once you’ve saved the file, reboot your Pi.

Finally I used my laptop on the same network with VNC Viewer installed to connect to the GUI of the Pi, open VNC Server by clicking on the icon in the top bar, click the burger menu in the top right corner, select Licensing… and log on with the account you created earlier.

All that was left was to uninstall TeamViewer from the remote systems.

The free version of VNC Connect will allow connection to up to five computers and I can invite two other members of the group to the account as “team members” so they can access these computers as well.

Bearing in mind how straightforward this is to set up and after having read so many reports of how bad TeamViewer has become with regard to flagging non commercial use as commercial, I really can’t recommend VNC Connect highly enough as an alternative. I have no connection with VNC apart from being a satisfied user of their free service having moved there from TeamViewer.

You can find out more about VNC Connect here.

Posted in Not amateur radio, Raspberry Pi | Tagged | 4 Comments

1090 MHz ADS-B PCB antenna from eBay

About a month ago I bought a 1090MHz PCB aerial on eBay and immediately forgot about it. It arrived a few days ago and I had a good look at it before I connected it to anything to try and work out what it actually was. I decided from the construction that it was a vertical half wave dipole.

As with all new aerials, I decided to put it on one of my analysers so see how it appeared to perform. I chose the straightforward vector impedance analyzer (sic) rather than the full VNA due to the simplicity.

I was very surprised to see the SWR showing as a disappointing 6:1 on the analyser.

PCB aerial showing 6:1 SWRPCB aerial showing 6:1 SWR
PCB aerial showing 6:1 SWR

I did a visual inspection of the aerial which showed that one side of it wasn’t even soldered – Effectively this was working as quarter wave vertical with no groundplane whatsoever.

Unsoldered 'connections'
Unsoldered ‘connections’

I fired up the soldering iron and it only took a few seconds to fix this.

The connections soldered in place
The connections soldered in place

You can see that both sides are now nicely soldered in place to the board. With that done, I put it back on the analyser again and it showed a much more pleasing SWR curve.

That's better - The repaired ADS-B PCB antenna
That’s better – The repaired ADS-B PCB antenna

For completeness, I also tested the aerial on my VNA. The marker here is at 1090 MHz and shows an SWR of 1.22:1. I do appreciate that SWR on a receive aerial isn’t anywhere near as important as when one is transmitting but at least it’s a good indication that the aerial is resonant on the required frequency.

ADS-B PCB antenna SWR
ADS-B PCB antenna SWR

I haven’t yet had a proper chance to test the effectiveness of the aerial but I see no reason it shouldn’t work as well as a simple quarter wave vertical with radials (spider). The main reason for this post is as a advisory – If you’ve bought one of these types of aerials then it’s worth checking to make sure that it’s actually soldered properly.

Posted in ADS-B, Amateur radio, Broken aerial, Construction | Tagged | 1 Comment

QRSS revisited

For the last couple of months, I’ve not done much in the way of traditional radio, largely down to the fact that I’ve been really busy at work. I’ve had lots to do and I’ve spent a month working away from home either on training courses or in meetings preparing for a very large project.

My Hans Summers Ultimate 3S transmitter has been running continually on 10m for a couple of years and earlier this month I decided to shift it around for a while and so it ran for a week on 20m and then a week on 17m while I was working away from home. At time of writing, it’s transmitting on 80m as there have been a few signals European transmissions received in North America recently and I thought it would be interesting to see if my 200mW QRSS signal would make the trip across the pond.

Here’s a reminder what QRSS actually is.

Imagine Morse code sent very slowly, so slowly that one simple ‘dit’ takes six seconds to send and a ‘dah’ takes eighteen seconds. Imagine also that it’s transmitted using audio frequency shift keying (AFSK) with just 4Hz differentiating the two tones and what you’ve got is QRSS. If you were to listen to it, you’d struggle to hear the difference between the two tones and that’s if you could even hear the signal because they’re transmitted with very low power, typically under 200mW.

Unlike WSPR which is automatically decoded and the callsign of the sender is uploaded to a central site, QRSS is decoded by eye and signals are so weak that it can take many repetitions stacked on top of each other to be able to see who is transmitting. There are very few stations in the world who transmit QRSS or run grabbers to receive QRSS signals.

I’ve been running my grabber again and have received some nice signals from Vernon, VE1VDM on 20m. Here’s a typical ten minute transmission from Vern.

Single QRSS frame from VE1VDM
Single QRSS frame from VE1VDM

If you click the image above, you’ll get a larger version where you can see the bare trace of a signal on the screen. Looking at the scale on the right hand side, the noise is at around -38dB and the signal is a fraction above that at about -36dB. If you were to tune your radio to the same frequency I’m monitoring on 20m, you wouldn’t hear anything because it’s that weak. FT8 can decode signals that are at -26dB and WSPR is good for about -30dB so using QRSS you can get decodes of signals significantly weaker than FT8 or WSPR.

“Ahah”, I hear you say. “But you can’t actually decode anything from the image above”. Of course, you’re correct and this is where stacking comes into play. Vernon’s signal was visible on my grabber between 11:00z and 17:50z on the 27th March 2019. QRSS operators normally transmit the same thing over and over again in ten minute frames allowing for stacking, that is combining multiple images on top of each other to build up an overall picture. If I combine the forty one frames captured on Wednesday, this is what I get.

VE1VDM stacked on 27th March 2019
VE1VDM stacked on 27th March 2019

This is literally Morse code painted on screen by the transmitted audio tones (AFSK). In order to read it you look at the top half of each character and you’ll see the code. This will make it easier.

VE1VDM drawn in QRSS

As you can tell, QRSS isn’t for the impatient. I’ve gone for weeks and weeks without anyone spotting my 10m QRSS transmitter but all of a sudden it can appear out of the noise for a few minutes if the band opens briefly. Over the last year or so, there have been quite a few spots of QRSS beacons on 10m that are purely down to meteor scatter. QRSS is also really good for aircraft scatter on 30m and there have been some fascinating studies written about aircraft scatter. By knowing the route and speed of an aircraft, the transmitting station can be located within a few miles due to the scatter on the received signal.

Frequency stability on both QRSS transmitters and receivers is absolutely vital to get a good clean signal. The difference between the two lines on the picture above is just 4Hz and the overall bandwidth on the image is just over 200Hz. If the transmitter or the receiver were to drift just one or two Hz the the signal would be nowhere near as clean as can be seen in this example from the day before. Vernon hadn’t quite got the temperature stability sorted in his transmitter and this can be seen in the approximate 2Hz variation in frequency.

Good frequency stability is the reason I installed a replacement temperature controlled crystal oscillator into my TS-590SG last year as it means my radio is absolutely bang on frequency with zero noticeable drift, even on 10m which was particularly bad before.

I really enjoy tinkering with QRSS and seeing what can be done with these incredibly weak signals. There aren’t many radio amateurs playing with it around the world so it’s very satisfying to see these signals appearing on grabbers and watching the grabbers to see where you’re being received.

If you’re interested and want to know more, our small community has an email reflector at groups.io. With just under 160 members, you can see that it’s fairly specialised.

To see some of the ‘grabber’ stations around the world, take a look at Scott’s page here which is an automatically updating list of grabbers with the current latest image. It takes more work than WSPR to find yourself but it’s worth the effort.


Posted in Amateur radio, Construction, HF, QRSS, WSPR | Tagged , | Leave a comment

Receiving weather fax pictures on HF using fldigi

A week or so ago I saw a weather fax (WEFAX) picture someone had posted that they’d captured on HF. This grabbed my attention because I like maps and I like to see things to do with the weather. I’d tried a couple of years ago to receive satellite weather pictures with little success (due to a very poor aerial) and the idea that I can receive weather maps on HF intrigued me. I know that strictly this isn’t amateur radio in the true sense because these transmissions aren’t within our frequency allocations but it’s still interesting.

After very little searching, I discovered that fldigi can receive fax pictures. I use fldigi for operating data contests so I’m very familiar with it. I soon had it configured correctly and so the next stage was to find what frequency the WEFAX pictures are transmitted on. A quick search around and an ask via Twitter pointed me towards this document which lists various transmitting stations around the world. There are two in Europe that I decided to try listening for, one in the UK at Northwood and the other in Hamburg, Germany.

I first started listening on the UK frequencies but heard nothing so I switched over to Germany and over the course of the week, I received a lot of pictures. The 3,855.000 kHz signal seemed about the best for me here. This morning I switched to the Northwood frequency of 4,610.000 kHz and very soon the pictures started rolling in.

fldigi receiving a WEFAX image from the Northwood transmitter
fldigi receiving a WEFAX image from the Northwood transmitter

If you want to use fldigi to receive WEFAX, you need to do the following:

Op Mode / WEFAX / WEFAX-IOC576
If you are receiving signals from Germany then the shift is 850Hz instead of the default 800Hz. This is changed through Configure / Operating / Active Modem, select the Other tab on the second row and then the Wax tab below it. For Germany, put 850 in the Frequency shift box or leave it at 800 for UK reception. If you switch between the two transmitters, remember to change this figure. Make sure you hit the Save box at the bottom.

Set LPM (lines per minute) to 120 in the box below the receive window.

For UK reception, I’ve found the Slant figure needs to be at 0.0071 and for Germany, set it to 0.0118. This is to ensure the picture is vertical and doesn’t slant.

Most sites I’ve found suggest tuning your radio to 1.9kHz below the transmit frequency but it really doesn’t matter, as long as you have a good signal in the waterfall and you set the two red vertical lines on top of it. My radio ‘sweet spot’ is at 1,500Hz so I tune 1.5kHz below the transmit frequency and set the radio filter down to 1,000Hz as can be seen above.

That’s it! Tune your wireless in to one of the frequencies in the list I posted above and wait. You’ll soon have an image folder full of stuff like this.

You can click on any of the images above for a higher quality version.

Receiving WEFAX maps is very straightforward, quite simple to do and can produce some good results. Not all pictures I’ve received are as good as above but on the whole, I’m very pleased with how easy this has been.

Posted in Data, HF, Weather | Leave a comment

An unexpected new DXCC

This weekend was the ARRL RTTY Roundup contest and for the first time, FT8 was allowed. I spent a couple of hours operating just to give away a few points and get some experience using the new version of WSJT-X in contest mode.

I’d been working mainly European stations and all of a sudden, I noticed an unfamiliar callsign on my screen – 9L1YXJ which WSJT-X identified as Sierra Leone. I was fairly sure I’d not worked this country before and it only took a few seconds to double check. I was right, this was a potential new one.

A couple of minutes later I managed to complete a QSO with Mark and I’m very pleased that he’s already confirmed the contact on LoTW. This takes me to 277 DXCC entities worked and confirmed.

All new DXCCs are good, they’re especially good when they’re unexpected.

Posted in Amateur radio, Contesting, FT8, HF, New DXCC, RTTY | Tagged , , , | 1 Comment