RSGB Xmas Cumulatives Contest 2015

Determined to take part in the RSGB Xmas Cumulatives this year, I worked from home on 3 days using completely inappropriate antennas.  On 26th December I spent 2 hours hunting around 6m, making only one contact using a dipole antenna.  The 27th and 28th faired better (~8 contacts a day) on 2/70 using a dual band colinear (vertical!).

Portable from Barbury Castle IO91CL

Portable from Barbury Castle IO91CL

Finally, on the 29th I decided to head up a hill, taking a more appropriate 70cm ZL special antenna.  James 2E0FUR also attended and on this occasion manned the station using his FT857 and own call, scoring 10 contacts in a 1.5 hour period.

Not too bad at all for 20W to a tiny beam mounted on some wooden poles!

I did make some token contacts using an FT817 on both 6 and 2m using the rubber helical antenna and 2.5W.

Matt M6ZIH and Russell M6??? also attended.  We all agreed we should be a little more organised in future, arming ourselves with larger antennas and proper masts we could maybe work some of the larger contests as a multi operator group.

Thanks to all who took part and Happy New Year!

Posted in Uncategorized | 1 Comment

Studio Mic (Balanced XLR) to FT847 (Unbalanced) Input

I’d previously simply connected half of the balanced XLR mic output to the mic input of the 847; I was aware this wasn’t ideal (and I’d lose the benefit of a balanced mic cable!) but it worked and I’m ultimately lazy.  See my post Upgrade From Yaesu MH-31 to Studio Mic for details.

1:1 600 Ohm Transformer

1:1 600 Ohm Transformers

Whilst trying to eliminate a buzz on TX (and exhausting all other options!) I figured I should maybe properly transform the mic output to an unbalanced connection.  Only, commercial transformers a) often add an impedance transformer and b) cost anywhere from £15 (ebay special) to £80 (quality brand), to £200+ for devices with active pre-amps and equalisers.

So, ebay 1:1 600 ohm transformers to the rescue!  I bought two for £5 including delivery. I then snipped off one end of my XLR cable, soldering the wires to a small section of stripboard. A 1:1 transformer was then added and a shielded mic cable exiting to a 1/4 inch jack (to match my customised adapter/wiring for the 847 mic input).  The mic side of the transformer is left floating and one side of the radio half is pulled to ground. No other components are required;  see below.

Balanced to Unbalanced Mic Schematic

Balanced to Unbalanced Mic Schematic

I’m unsure of the difference between chassis ground and mic ground on the 847 and can’t remember which I connected.  I know I tried both with no difference to my buzz.

The cause of my TX issue was eventually traced to internal 847 cabling running directly over the internal fan.  This caused a ‘buzz’ whilst the fan was spinning up to speed; re-routing cables has essentially solved the issue. However, I’m still unclear how internal routing of cables can interact with a specific external mic. I’d expect the buzz to be either present or not.  Whilst the problem is resolved the mystery is still ongoing.

At least I can now rule out poor balanced/unbalanced mic cabling from future issues 🙂

Posted in Uncategorized | Tagged , , , , , , | Leave a comment

Raspbian (Debian) Jessie and TNC-Pi

tnc-piSo, Raspbian has been updated to Debian Jessie;  we’re now in systemd land and the provided instructions from the tnc-pi instructions are no longer relevant.

After installing Raspbian I recommend extending your partition / filesystem to fill the space available on your SD card.  This can be done using the following menu driven configuration utility:-

sudo raspi-config

This utility can also configure the pi to boot into command line only mode rather than the full gui.  This is entirely up to the users preference.

In my case aptitude was already installed.  But if this is not the case for you:-

sudo apt-get install aptitude

Next we need to disable the console in /boot/cmdline.txt (remove the section ‘console=ttyAMA0, 115200’).  This can be done using your favourite editor (in my case vim, although the tnc-pi manual suggests using ‘leafpad’ if using the gui).

As we’re using systemd we no longer have an inittab.  Instead we need to stop and disable the getty service for ttyAMA0:-

sudo systemctl stop serial-getty@ttyAMA0.service
sudo systemctl disable serial-getty@ttyAMA0.service

We’re now ready to install the ax25 tools:-

sudo aptitude install ax25-tools ax25-apps

Edit our /etc/ax25/axports file as per the original manual using our choice of editor; in my case it looks like this (be sure to have no blank lines):-

# /etc/ax25/axports
# The format of this file is:
# name callsign speed paclen window description
1 M0SPN-1 19200 236 2 TNC 1

Then finally kissattach to the serial port.  Using my axports file above I simply issue a :-

sudo kissattach /dev/ttyAMA0 1

‘sudo ifconfig’ should then list ax0 with the IP address configured.  We’re good to go!

root@raspberrypi:~# ifconfig
ax0 Link encap:AMPR AX.25 HWaddr M0SPN-1 
 inet addr: Bcast: Mask:
 RX packets:192 errors:0 dropped:0 overruns:0 frame:0
 TX packets:273 errors:0 dropped:0 overruns:0 carrier:0
 collisions:0 txqueuelen:10 
 RX bytes:17380 (16.9 KiB) TX bytes:25950 (25.3 KiB)

Finally, I want ‘kissattach’ to run at boot so the ax25 interface (and IP address) are ready to go without requiring operator interaction.  I do this by editing the /etc/rc.local file.  Mine now looks like this:-

#!/bin/sh -e
# rc.local
# This script is executed at the end of each multiuser runlevel.
# Make sure that the script will "exit 0" on success or any other
# value on error.
# In order to enable or disable this script just change the execution
# bits.
# By default this script does nothing.

# Print the IP address
_IP=$(hostname -I) || true
if [ "$_IP" ]; then
 printf "My IP address is %s\n" "$_IP"

echo "kissattach"
kissattach /dev/ttyAMA0 1

exit 0

Additionally, as I’m interested in running TCP/IP over ax25 rather than the standard ax25 tools, I install a telnet server on the pi and enable the service:-

sudo aptitude install telnetd
sudo systemctl enable inetd.service

Also, for debugging I install the telnet client:-

sudo aptitude install telnet

Now, from my main ‘radio’ PC I can simply telnet to my pi:-

steve@radio:~$ telnet
 Connected to
 Escape character is '^]'.
 Raspbian GNU/Linux 8
 raspberrypi login: pi
 Last login: Tue Nov 24 20:36:53 UTC 2015 from on pts/0
 Linux raspberrypi 4.1.13+ #826 PREEMPT Fri Nov 13 20:13:22 GMT 2015 armv6l

The programs included with the Debian GNU/Linux system are free software;
 the exact distribution terms for each program are described in the
 individual files in /usr/share/doc/*/copyright.

Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent
 permitted by applicable law.
 pi@raspberrypi:~ $ id
 uid=1000(pi) gid=1000(pi) groups=1000(pi),4(adm),20(dialout),24(cdrom),27(sudo),29(audio),44(video),46(plugdev),60(games),100(users),101(input),108(netdev),997(gpio),998(i2c),999(spi)
 pi@raspberrypi:~ $ df -h
 Filesystem Size Used Avail Use% Mounted on
 /dev/root 7.3G 3.3G 3.7G 47% /
 devtmpfs 87M 0 87M 0% /dev
 tmpfs 91M 0 91M 0% /dev/shm
 tmpfs 91M 4.5M 87M 5% /run
 tmpfs 5.0M 8.0K 5.0M 1% /run/lock
 tmpfs 91M 0 91M 0% /sys/fs/cgroup
 /dev/mmcblk0p1 60M 20M 41M 34% /boot
 tmpfs 19M 0 19M 0% /run/user/1000
 pi@raspberrypi:~ $ exit
 Connection closed by foreign host.

Incidentally, if you do ping machines to check for connectivity and/or audio levels, I suggest you increase the interval to 5 seconds to accommodate the 1200 baud packet TX/RX times.  Anything less than 5 seconds and you’ll get collisions and dropped packets.

steve@radio:~$ ping -c 10 -i 4
PING ( 56(84) bytes of data.
64 bytes from icmp_seq=1 ttl=64 time=2330 ms
64 bytes from icmp_seq=3 ttl=64 time=4629 ms
--- ping statistics ---
4 packets transmitted, 2 received, 50% packet loss, time 12001ms
rtt min/avg/max/mdev = 2330.581/3479.982/4629.384/1149.403 ms, pipe 2
steve@radio:~$ ping -c 10 -i 5
PING ( 56(84) bytes of data.
64 bytes from icmp_seq=1 ttl=64 time=2250 ms
64 bytes from icmp_seq=2 ttl=64 time=2242 ms
64 bytes from icmp_seq=3 ttl=64 time=2244 ms
64 bytes from icmp_seq=4 ttl=64 time=2251 ms
64 bytes from icmp_seq=5 ttl=64 time=2254 ms
64 bytes from icmp_seq=6 ttl=64 time=2247 ms
64 bytes from icmp_seq=7 ttl=64 time=2161 ms
64 bytes from icmp_seq=8 ttl=64 time=2253 ms
64 bytes from icmp_seq=9 ttl=64 time=2247 ms
64 bytes from icmp_seq=10 ttl=64 time=2248 ms
--- ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 45019ms
rtt min/avg/max/mdev = 2161.920/2240.195/2254.368/26.409 ms

I hope this has been useful!  Enjoy your 1200 baud packet 🙂

Posted in Uncategorized | 6 Comments

UK Foundation Licence Power Limit

When I progressed through the licence system over a decade ago I was taught that the foundation power limit referred to power measured at the radio, not the antenna.  This was different to the other licence classes which specified power fed to the antenna, after feedline losses.

SWR/Power Meter

An SWR & Power Meter

Recently, when assisting a newly licensed friend with his first station, I checked the current licence conditions and was surprised to see the same wording for all licence levels; ‘power supplied to the antenna’.

Speaking with other newly licensed foundation holders on-line it became apparent that some had also been taught the power limit applies at the transceiver output.

I thought perhaps license conditions had changed over the years, so found a BR68 from 2001 via Ofcom ( and that very same wording confronted me again.

Table A states the limit as ‘Maximum Peak Envelope Power’:-


2001 BR68, Table A

This is then further expanded in Notes to Schedule 1:-


2001 BR68, Notes to Schedule 1

Note: “Power supplied to the antenna”

If this is the case, why are we still teaching foundation licence holders otherwise?

It could be argued that feed line losses and measurements are not covered until later in the syllabus, so perhaps much of this is irrelevant.  However, I’m still curious how the misunderstanding has arisen.

Could this perhaps be a hang-over from the old Novice class?

Any comments or further insight appreciated!

Note: With 50m of RG58 potentially suffering 3dB loss at 10MHz, that’s half an S point;  perhaps enough to make or break a marginal contact.  At 50MHz that almost doubles to 6 dB or 75% loss.  Calculating (or measuring) power at the antenna can make a significant difference, especially when operating QRP.

Posted in Uncategorized | Tagged , , , , , , | Leave a comment

Amplifier PTT Buffer Schematic


The KW1000 (bottom)

I’ve been keying my KW1000 linear using a relay to buffer the FT847.  The FT847 closed a 12V circuit across the relay and the relay in turn keyed whatever voltage/current the linear requires.  The relay in the KW1000 is old, large and whilst it had been modified from the original circuit, I still don’t have confidence with it being within the limits of the 847.

All was well until a flyback diode I placed across the ‘buffer’ relay to protect the 847 failed short.  This resulted in the 847 keying 12V @ 1A ultimately limited by my bench PSU.  The surface mount 847 PTT transistor promptly cooked itself.

I now plan to redesign the PTT interface including additional transistors to protect the radio.  My planned schematic is below:-


Provisional HF Linear PTT Buffer (v1)

The box to the left is my simulated 847.  The 847 closes the HF STBY pin to ground on TX.   In the above design R1 now limits the current seen by the 847 to 12/4700=~3mA.  By default Q1 is biased ‘on’ by R1, pulling the collector to ground, resulting in Q2 switched ‘off’ and the relay unactivated.

When the 847 keys the PTT line, the base of Q1 is pulled to ground, the emitter then floats to a positive voltage (>0.6V) and Q2 switches ‘on’.  This allows current to flow from Q2’s emitter->collector and the relay energises.


Current plot, orange=FT847 PTT circuit, blue=relay

Flyback diode D1 protects the transistor from reverse EMF generated from the relay coil. Only this time, if D1 fails short it’s Q2 that becomes toast, not the microscopic surface mount transistor in a £700 radio.

Both Q1 and Q2 are set to draw about 3mA across their base-emitter junctions.  Assuming an Hfe of 100, in the case of Q2 this should be more than enough to allow 120mA across emitter-collector (assuming relay coil is 100ohm).  In the case of Q1 it’s probably overkill; I wonder what the benefit is of lowering this besides power consumption?

Note: Circuit designed for 100ohm relay coil resistance whereas I believe the actual relay at hand is 720 ohm (so 20mA); current plot above has been updated to reflect this.

Any comments appreciated.

Edit: 16.06.15

As always, I seem to miss the simple things.  Rather than the FT847 radio pulling the first transistor low (requiring a second transistor to invert the output), the PTT could directly pull the second transistor high, simplifying the design and losing a transistor.


Simplified Schematic (v2)

However, my original quest was to not only isolate the radio and linear but provide additional protection against the flyback diode failing short.  In the above scenario, should D1 fail in this way, Q1 would probably quickly follow.  The radio would only be protected until this point, depending on exact mode of Q1 failure.

The original two-transistor schematic seems to offer an additional layer of protection.  In this case, should Q2 also fail short, Q1 would be outside the path of destructive current flow.

For this reason, I intend to continue with construction and testing of the first (v1) schematic.

Posted in Uncategorized | Tagged , , , , , | 1 Comment

Simple Fan Dipole

After starting life as a fan dipole, my home antenna was later converted into a single 40m dipole due to unwanted interaction between the bands (at the time 40/30m).  Recently, the need to operate multiple bands made me reconsider my options.

This weekend I added additional elements for 17m (18 MHz).  I thought I’d try a WARC band as it’s both contest free and a band I have little experience with.  I’ve not noticed any de-tuning of the main 40m elements and both bands perform very well indeed.

Fan Dipole Centre

Fan Dipole Centre Support

I sometimes have trouble explaining just how simple the antenna is on air, so thought I’d post some photos here.  The fan dipole now consists of 40m (horizontal) and 17m (inverted V) elements, supported at the centre and far end by 9ft bamboo canes; The near end is supported via fishing line to an upstairs window.  I used a low loss coax feed (aircell 5) from shack to dipole centre, which feeds the elements (solid copper) via Comet CBL-1000 (click for review).

Total height at peak is aprox 5m (16ft?), so practically NVIS; fantastic for inter-G but useless for DX due to the high angle of radiation.  Reports on both bands are very good indeed.  You can’t beat resonant antennas!  Obviously no ATU needed, although an ATU can be used to operate on 15m (21MHz) and possibly 6m (50MHz) which are the odd multiples of 7/18MHz.

Fan Dipole Highlighted

Fan dipole. Red = 40m elements, Green = 17m elements, Blue = Supports. CLICK TO ZOOM

Despite using bamboo canes which will need regular replacement, the antenna has now been up 6 months without issues, surviving several storms.  The bamboo canes, copper wire and fishing line blend in with the environment so are hard to spot; very neighbour friendly!

Update: The fan dipole has now been up over a year with no bamboo casualties, even surviving the storms.

Posted in Uncategorized | Tagged , , , , | Leave a comment

Simple & Cheap Speaker Upgrade


The Gale Centre Speaker

This will be a very brief post but I thought it too good not to share.  For some time now I’ve had my FT847 in a less than ideal position meaning I’ve either had muffled audio from the internal speaker or poor/thin audio from a cheap external ‘CB’ type speaker.

Many years ago I tried using a standard hifi speaker for communications but found the frequency response to be less than ideal for voice; boomy bass and great treble but an apparent total lack of mid range ‘voice’ frequency.

Recently, I was given a cheap (when new, even cheaper to me being free!) Gale centre speaker from an old surround sound setup.  It took a couple of days of this speaker sat on a shelf before the brain gears started whirring.  A centre speaker is *designed* for speech, it’s job is to make vocal passages stand out in amongst the booming noise of the other 4 speakers and ground shaking sub.

5 Minutes later, a 3.5mm jack soldered to a cable and the difference is incredible.  This is being driven directly from the 847 output with no external amplifier.  The output is incredibly crisp with plenty of clarity-giving mid range, whilst also having the wider frequency response for that ‘BBC’ broadcast sound.  It’s fantastic – don’t throw that old centre speaker away, it’s probably just as good (if not better) than an expensive dedicated communications speaker!

Posted in Uncategorized | Tagged , , | 1 Comment

Upgrade: Yaesu MH-31 to Studio Mic


The Chord Mic, Desk Mount and Foam Pop Shield

Some time ago I started receiving reports that the audio on my FT847 was very narrow. After much investigation and eventually swapping mics with my FT817 I realised it was the mic that was faulty. Oddly, swapping the mic element for another also failed to rectify the problem; the capacitor, resistor and switch in the mic tone control all checked out OK too.

With none of the above making much sense I decided to use the FT817 mic as a temporary solution whilst considering long term upgrades.  I thought of this as an opportunity to upgrade to outboard gear; a decent mic, preamp, compressor, gate and possibly EQ. However, after reviewing my options, the cost and possible problems converting line level output back down to mic level – I decided I should perhaps start with the basics.


Chord vs Yaesu

I ordered a cheap Chord ‘karaoke’ microphone (£10), desk stand (£10) and foam pop shield (£1) from eBay.  I then added a discounted latching foot switch from Maplin (reduced to £10) and wired some 1/4 jack sockets to the FT847 mic connector; one for the mic, one for the PTT.

Initial reports have been positive, although testing TX audio via Hack Green I think I may need some carrier adjustment. The Yaesu MH-31 does sound quite narrow whereas the studio mic audio has significantly more bass. Personally, whilst I agree the studio mic sounds ‘better’, some carrier adjustment would likely reduce the bass and reach a happy balance between fidelity and clarity.


The Maplin Footswitch / PTT

In summary, for ~£30, a very worthwhile investment. Plus, it means the working MH-31 can be returned to my FT817!

I’ll attempt to record some samples of the audio from both microphones once I have the carrier adjusted to the optimum level for both mics.

Update: I’ve been running with this setup for several months now and continue to get unsolicited fantastic audio reports.

Posted in Uncategorized | Tagged , , , , , , , , | 3 Comments

Pixie CW Transceiver

Here’s a quick review of the Pixie CW Transceiver, built for 40m.  I ordered two identical kits from China via ebay, one for myself and another for a friend.  The kits arrived and appeared to be well put together;  the PCB (5cm sq) seemed of a good quality with clear printing and the components (even a dummy load resistor) were supplied in individual bags.


The Pixie Kit

The Pixie took approximately 45 minutes to assemble.  During this time I noticed some pads were reluctant to hold solder. I’m unsure why this was – visually they looked great, but solder just wouldn’t flow and ‘bond’ in a satisfactory manner.  I haven’t seen others mention this issue so perhaps it’s my fault for not cleaning the PCB first.

Once assembled, I connected a speaker and CW key, applied 12V and … nothing. I could see some minimal current being drawn on the PSU so disconnected the speaker and attached some headphones.  Success!  It seems the LM386N isn’t up to the job of driving a small loudspeaker but it’s more than happy with headphones.


The Assembled Pixie

Attaching the Pixie to my 40m dipole I could hear several very clear CW transmissions.  Of course, without filters on a busy band – this could be a problem. Still, I hit the key and monitored my transmission on the FT847;  current draw increased on TX and a clear tone was heard on the 847.  Plus, an added bonus – despite learning and practising on a paddle key, I managed to send some CQs with good (IMO!) timing on a straight key.

Future plans:-

1. Measure power output
2. Hold an actual QSO!
3. Mount the Pixie in an enclosure (with 9V battery?)
4. Order and build a ‘Frog’

In summary, well worth the money.  I not only enjoyed the construction (despite the problem pads!) but have a working CW XCVR 🙂

Posted in Uncategorized | Tagged , , , | 4 Comments

Review: Atom 40S Mobile Antenna

My previous experiences with HF mobile antennas have not been so great.  On a previous car a pro-am/hamstick mounted on the boot required bonding of the boot to car body with copper in order to ensure a good enough ‘ground’; removing paint and welding copper braid to metal on a nearly-new Mercedes was not my idea of fun (but of course, radio wins – so I did it regardless!).

Atom 40S - Image from Moonraker

Atom 40S – Image from Moonraker

On my current vehicle I decided to use a mag mount.  Only, the 40m ham-stick I already had caused too much wind resistance, resulting in the mag mount becoming a missile with an unhealthy attraction to the rear window when travelling over 30mph.

In order to add further compromise to an already compromised solution, I ordered an Atom 40S from Moonraker.  This was the only ‘shortened’ 40m mobile antenna I could find.  Due to the short length (1.6m?) and base loading I wasn’t expecting much more than a dummy load on 40m.

The result?  Whilst mobile on my commute between Swindon and Knutsford (20m South of Manchester) I had constant QSOs on 40m.  I joined a net on 7.180 and talked the entire journey with great reports both ways from all stations.  Sure, I expect I was an S point down compared to my hamstick which itself is probably a couple (or more?) S points down in comparison to a full size antenna.  However, with 100W from an Icom 706 (possibly at a guess 10W ERP after antenna losses?) I wasn’t struggling to make contacts.

As the band closed to inter-G I finished the journey with QSOs into the Netherlands, also with great reports.

Due to the shortened size of the Atom 40S the SWR does climb steeply either side of the resonant frequency. I adjusted for 7.15 MHz and find it usable (SWR < 2:1) between 7.1 and 7.2 MHz.

Overall, I’m impressed.  My only criticism would be the grub screws which appear to be made from butter.  From the factory one loosened without issue.  The other required substantial swearing, application of heat (discolouring the chrome as a result) and several attempts at packing the grub screw with foil in order to gain any sort of grip on the now rounded cavity.  If this wasn’t a bank holiday weekend I’d have probably just called Moonraker and asked for a replacement!

The specifications from Moonraker are as follows:-

Type: Base loaded HF mobile antenna
Frequency: 7 MHz
Length: 155cm (whip only 48″)
VSWR: 1.5:1 or better
Power: 200 watts
Fitting: PL259 type
Bandwidth: 60kHz

I’m not linked with Moonraker in any way but a (mostly) happy customer.  The Atom 40S far exceeded my expectations and for < £30 has given me the ability to operate mobile on 40m.

I hope to perform some a/b tests between the Atom 40S and a traditional pro-am/hamstick.  This post will later be updated with results.

Posted in Uncategorized | Tagged , , , , , | 2 Comments