PCM Kestrel HF Transceiver Part 2

In the previous post, the Kestrel had a two line LCD display on the lower area of the speaker grille. One of the nice things about this transceiver is the sound, and this was compromised by the LCD display obscuring part of the speaker aperture.

OLED displays are commonly available now, so a 0.91 inch OLED was interfaced to the VFO as an experiment.

A couple of VFO boards were obtained from Denys VK3ZYZ, and that was a welcome replacement for the dogs breakfast controller and clock generator in the previous post. Denys sent a couple of Nano VFO boards and a Tracker VFO board. The Tracker board has some spare space on it for custom interfacing etc. I ended up using the smaller Nano VFO board and adding a resistor ladder to interface to the 10 channel selector on the Kestrel. The output of the ladder feeds the Nano with an analogue stepped voltage representing the selected channel.
More information on the boards can be found here http://www.sadarc.org/index.php/projects/

Photos of the boards are below-

A replacement speaker area plate was cut from some 1.6mm Aluminum with a cutout matching the speaker aperture. A 3D printed bezel for the OLED is mounted to the lower left where it doesn’t get in the way of the speaker. The plastic grille was 3D printed and mounted behind the speaker cut out.

The OLED display is a touch small, but it is very readable due to the sharp bright characters. I am keeping an eye out for a larger display but it will do for the moment.

The Nano VFO in the radio has a small extension on it for some extra connectors and the resistor ladder.

That’s it for now. It is set up for 160m, 80m, 40m and some fixed HF frequencies and works pretty well. Power output is about 120 W pep.

PCM Kestrel HF Transceiver

The PCM Kestrel is a HF transceiver from the 1970s that was intended for marine applications and made here in Oakleigh, Victoria. There is also a Hawk version that was targeted at land base applications.

This particular one is a 131-12, which designates the 130 Watt, 12 Volt version. It was picked up at an EMDRC market day in March this year along with a rugged HF antenna. Despite being a Kestrel marine unit, it was channelised for Royal Flying Doctor Service (RFDS) frequencies along with the matching antenna.

The radio is very large by today’s standards, but this makes it very easy to work on with all the components readily accessible. There is also a fair bit of space for add ons, like a digital VFO, which is the main subject of this post.

I came across a nice well documented VFO conversion that was done by VK3ZYZ(ref 1). As I was faced with a weekend of lockdown due to Covid, building up a VFO seemed like a nice way to pass the time.

As traveling to Jaycar to get an Arduino Nano was out of the question, I found an old Ardupilot board from a drone project and pressed this into service as the controller. An Si5351 clock generator board was cannibalised from an old QRP project.

The whole thing was put together on a piece of blank PCB and some bits of stripboard where things needed to be stood off from ground. The radio itself is intended for operation between 2 and 12 MHz according to the manual. The plan is to have it working on 160m, 80m, 40m and 30m. It may work on 20m to be investigated later.

The radio has a rather impressive multi ganged rotary switch the almost extends the depth of the radio and is used for selection of crystals, receiver preselector/transmit amplifier, transmit PA low pass filter and some logic for selection of modes.

The Kestrel is a upper side band radio only which initially looked like a challenge for 80 and 40m, but it turns out that the radio has a lower side band crystal filter and the sideband is mirrored by the high side injection of the local oscillator. Operation on lower sideband is possible by using low side injection of the local oscillator. The only problem here is that on 160m, low side injection results in a very low local oscillator frequency and hence, image rejection problems. Not to worry, 160 will be AM only.

So far it’s working nicely on three bands but the VFO is rather untidy. I will probably make a PCB for the VFO to tidy it up.

There is not much space for the VFO display on the front panel, so for the moment it’s partially obscuring the speaker. The rotary encoder is between the clarifier and volume knob, it has the yellow knob top below. I am looking for a more compact display to replace the temporary one below.

Ref 1: http://www.sadarc.org/xenforo/upload/index.php?threads/pcm-hawk-arduino-vfo.36/

Stingray Marine Transceiver

Its been a while since the last post, a lot has happened. It looks like WordPress has changed quite a bit too!

At the Ballarat Hamfest recently, I picked up an old Marine transceiver for a princely sum of $10. What I liked about it was the cast housing and it seemed to have a bit of waterproofing which I guess is no surprise. It was made in Sydney by Findlay Communications. Australia used to have a lot of radio manufacturers once, what a shame most have now disappeared.

I had no idea what condition it was in, I could not go wrong at $10. After getting it home and prising the covers off (they were sealed with a silicone type stuff), it looked pretty good inside. It is crystal controlled with a claimed power output of 120 Watts. Frequency coverage is stated at 2-13 MHz, AM/USB with 10 channels.

Power was applied and a big fat nothing. It didn’t take long to see that the speaker wire was disconnected. After reconnecting, it sprung to life and crackled when a bit of wire was put in the BNC socket.

When the antenna was connected it seemed good and a signal generator confirmed that the sensitivity was good. Next thing was to try transmit. PTT resulted in a soft click from the relay but no change from receive to transmit. Inspection of the relay showed that it had sticky stuff on the armature. Removal of this fixed the problem and it went into transmit on PTT. No output though 😦

I posted a request for a manual on the local forum and was surprised that a guy, John in WA had a manual. It didn’t take long to find a burned out resistor in the PA from the output transformer to ground. The resistor was replaced and power output looked respectable. Something like 50 Watts in AM mode and more in SSB mode.

Crystals are petty expensive if you can get them these days, so I went about replacing the crystal oscillators with an Si5351a controlled by an Arduino. My main aim was to have a boat anchor radio for 160m AM and something on 30m SSB. The radio looked very tatty so I gave it a paint job and made some side grab handles for it out of 5mm thick Aluminium

It all worked out well, and I had the first call on it today to a VK2 on 30m. Good signals and the radio sounded pretty good. The radio has retained a few marine channels and I have added 1.825 MHz AM and 10.125 MHz USB. There is another output on the Si5351a clock generator, so I might add a 20 metre frequency. It would be easy to add a rotary encoder and display but that would probably be over capitalising the radio 🙂 The paint cost more than the radio.

PD Powerbank for SOTA Radios

This post describes how the newish Power Delivery (PD) mode mobile powerbanks can be used to power a SOTA transceiver as well as keeping your other hiking devices charged.

The PD protocol enables fast charging of mobile devices from a powerbank by negotiating a high voltage for the mobile device if fast charging is supported. Common fast charge voltages are 9 and 12 Volts. When the device to be charged is connected, serial data messaging transacts between the powerbank and the mobile device to establish a supported fast charge mode.

The 12 Volt capability of the powerbank got my interest as it would be a simple way of supplying power to a SOTA radio. The powerbank that I have can deliver 12 Volts at 1.5 Amp. This is more than enough for a 5 Watt CW rig. When powering the KX2 transceiver, I had to limit the power output to 5 Watts on SSB.

Typical powered devices that are in the backpack for hiking include a USB chargeable head torch, a mobile phone, USB chargeable Garmin and a USB chargeable 2 metre portable. All these devices can be charged by the one powerbank.

This PD power bank was purchased at Big W for $27 on special. It claims to have a capacity of 10 Amp Hours. Of course, the stated capacity is at the 3.6 Volt cell voltage. I ran a discharge test at 9 Volts (PD mode) while repeatedly charging a Galaxy phone. When the measured capacity was scaled back to the 3.6V battery voltage, and assuming an efficiency of 85% for the voltage converter, the battery capacity worked out at 9.4 Amp Hours. I didn’t really expect it to be this close.

Powerbank Rating

In order to trick the powerbank to think that it has a fast charge device connected, you need to connect a PD trigger device between the powerbank and the load. The trigger device in the photo was sourced from Ebay. It communicates with the powerbank to trigger a 9 V or 12 V output. The trigger device also can display voltage and current.

KX2 powered from the Powerbank – 5 Watt SSB maximum

I did not notice any significant RF noise from the powerbank or the trigger unit on 40 and 20 metres.

QCX Mini powered from the Powerbank – 5 Watt output

The PD Trigger unit probably would not stand up to life in the backpack without some more protection and connector strain reliefing. This particular unit has a model number of XY-WPDT and is available from a number of sources such as Ebay or Bangood. This unit also comes with a cable and connector that fits the QCX Mini and KX2 although it looks like the centre pin might not be quite the correct size.

There are smaller PD Trigger units that are integrated into the cable connectors that might be more suitable for the backpack.

3D Printed Accessories for the FT-817

A few radio accessories printed out recently for the ‘817 transceiver (from Thingyverse, see links):

DC plug adaptor using an XT60 connector.  I have been using these connectors for a while now, they are small and capable of a lot of current.  On my model plane, they were passing 50 Amps for brief periods with no signs of overheating.

18650 Lithium battery 3 cell carrier for inside the FT-817.  I have put together 18650 carriers for the ‘817 in the past, but the original lid would not go on properly.  This carrier has a slot for a 3 cell battery management system board (BMS), so it is an improvement over the previous.  The printed lid is raised slightly so as to accommodate the larger cells.  It fits very well.   Still waiting for the BMS to arrive.

Three band end fed half wave antenna.  This design was posted previously in the blog.  It was made into a 3D print project and posted on thingyverse by mfhepp.
It came out a bit lighter than this one.  The matcher box is integrated into the winder.

This one is for the KX2. Its a KX2 paddle using RF finger stock. I had some finger stock left over from a HF screwdriver antenna build and it was just the right fit.

XT60 chassis mounted flange:

USX or uSDX QRP HF Transceiver

This build is based on the interesting design by Guido, PE1NNZ.  The transceiver originally used the very successful QCX as an initial platform, but it now has it’s own life and there are several scratch builds using dedicated PCBs on the forum.

An Atmel ATMEGA 328 is pushed close to the limit in an SDR application processing SSB IQ from a Tayloe IQ detector.  The speaker audio is supplied directly from the 328 pins as PWM.  So no audio amp is required.
The transmitter uses a class D or E PA and implements Envelope Elimination and Restoration (EER) to deliver a highly efficient effectively linear power amplifer.

The implementation does have several compromises, however, running in a simple 8 bit processor with 10 bit A/D convertors.
Refer to the Github page for design details.

This build presently uses two boards, the UI and processor and the RF board in a stacked arrangement connected by headers.  A test PCB was put together for the RF side, with the intention of experimenting with the design.  The processor board was built using proto board.

The unit all boxed up measures about 7.5 x 8 x 3.5 cm and hopefully could be used on a SOTA trip to keep the weight down.  It was painted bright orange so as not to lose it!

Completed unit

RF Test Board

Processor UI Proto

The PWM speaker audio worked fairly well on an efficient speaker but not so well on a small 50mm speaker.  An LM386 amp was added to boost the volume to useful outdoor SOTA levels.
The RF board ended up with lots of links and suffered from changing chips and experimenting.  I might do a final board, but there are others now doing a pretty good job of boards on the UCX forum.

To speed up the build, I used an Adafruit Si5351a breakout board, although this uses a 25MHz clock necessitating a def change in the software and some hardware changes.

So,how does it perform?

The RX audio sounds OK, I had some problems keeping the PWM frequency out of the extra audio stage despite additional filtering.  There is some residual hiss that is quite noticeable on low volume settings.   The Tayloe detector has spurious responses at odd harmonics, so it’s relying on the transmitter LPF to reduce these.  A BPF or tuned bifilar toroid like the QCX would improve selectivity.
There are regular “pops” on the receiver that can be eliminated by selecting 13dB attenuation in the menu at the cost of some sensitivity.
Others have had problems with OLED display noise getting into the receiver, although none noted on this build.  Good supply rail bypassing and grounding is essential.

The transmit audio is quite intelligible, although does sound a bit edgy.  The transmit spectrum looks just clean enough for amateur use with harmonics and spurs around 50dB.  I have made some SOTA chaser contacts successfully into NSW with it.  The Transmit power is around 7W, this should be reduced back to 5W as it’s marginal with the design at this level.

Considering all the above, on a summit in most cases it is probably ok, although you might get some comments on rough TX audio quality.

In summary, it’s a pretty amazing minimalistic SDR design and squeezes almost everything out of the processor.  I look forward to an implementation on a more powerful platform with higher A/D resolution with more speed and bits.

On the right, the newer build 150grams

Newer build RF board

Processor board

They are getting smaller, uSDX top RHS

Jpole for 70cm

While experimenting with LoRa, I needed a better antenna at home for range testing the LoRa tracker.  I came across this Jpole antenna calculator and used it as a starting point for dimensions.
As we are presently locked down and and travel is only allowed for essential reasons, I had to use materials that were on hand.  I found some 6mm copper pipe, and although its a bit soft, it would be easy to solder to and should stand up ok outside.

In the past, it’s been weatherproofing the Jpole match that has been the challenge.  I found a weatherproof looking plastic enclosure in the junk box that looked like it would protect it.
A test was done with the coax feed solder tacked onto the pipe at the calculated point using the calculated lengths for the radiator and stub.  A VNA was connected and the match was perfect without any need to make an adjustment.  The return loss read off at just over 30dB with the antenna in the clear, away from any metal.

Return Loss

Jpole Match

Stabilising the radiator and stub inside the box are some 6mm cable clips.  The nail was removed and the hole left was a perfect self tap diameter for 2mm stainless screws.  The coax is decoupled with a ferrite about 25mm long.  Not sure of it’s properties as it came from the junk box.  It seems to do the job as the return loss is not affected by touching the coax.

Copious amounts of neutral cure silcone sealant were added around the radiator and stub entry to the box.  A drain hole is left at the bottom of the box incase water does get in.
After fitting into the box and adding the sealant, the match did not change too much which was surprising.  It’s now on a pole outside at about roof high, ready for some LoRa testing

Archer Lookout VK3/VC-038

One of the newer Victorian summits added to the register late in 2019 was Archer Lookout, VK3/VC-038.  Ron, VK3AFW was the first to activate it and after some discussions on access, I planned an activation on an ebike.

Archer Lookout is just off Monda Rd., near Narbethong.  I accessed it from Myers Creek Rd., Toolangi and took Monda Rd to a point about 3km past the Mt. St. Leonard car park, where  the road started to deteriorate for my low ground clearance AWD.  I could have probably gone further, but the track was wet and boggy in a couple of spots.

At this point, I got out the ebike, put on the pack and rode off down the road.  The ride was pretty steep in parts going over Mt. Monda.  There were a lot of fist size rough rocks on the ascent and descent of Mt. Monda that made it pretty bumpy on the bike, especially going down under braking.  Apart from that it was a nice ride ~8km each way, the bike had no problem with ascending the steep parts of the track.

Muddy Track

After the activation on Mt. Monda, I rode north down Nursery Spur Rd., and then down Lookout Spur Rd., which looked ok for a normal AWD.  I turned back and returned to the car making a mental note to access Archer Lookout from the Narbethong direction next time via Anderson Lane, Plantation Road and then Lookout Spur Rd.  This should be straight forward and avoid the muddy rough sections on Monda Rd.

 

On the way

On the summit clearing

 

 

 

 

 

 

 

A bit later, I returned via Myers Creek Rd., and then over the Black Spur to Narbethong and took Andersons Lane, then Plantation Road  and Black Range Road to Mt. Mitchell, VK3/VN-012.  There are a lot of roads cut through this area where there are plantations.  It is pretty easy to lose your bearings if not careful, many of the tracks were new and not on my map.  Access was good AWD.

Since the inital activation in November 2019, I returned again on 2nd January and did the same ride again.

 

LoRa GPS Tracker

LoRa is a newish Internet Of Things communications technology.  It promises long range (LoRa), transmits a small amount of power and is designed for sensors that may need to operate from batteries for very long periods of time.  The air protocol works at very low received signal levels, with down to -148 dBm claimed with low data rate transfers.

There are several LoRa modules available for very reasonable prices from Ebay.  I purchased 3 x AIThinker RA-02 modules for $20.  These modules use the Semtech SX1278 chip which packs in an amazing amount of functionality.  The module is capable of +17dBm transmit, with a boost mode for +20dBm (100mW).

One of the LoRa bands is 433MHz, in the 70cm amateur band.  With an interest in a low power GPS tracker that could be used as the “last mile” for APRS and hiking, a test transmitter and receiver were designed and assembled.

LoRa is intended to work in a network topology with nodes that talk to field units (LoRaWAN), but there is a peer to peer mode using the LoRa protocol that has been made available by the Radiohead driver collection.  I used Arduinos for the control side of things as there are many drivers and sample applications available.  The GPS is a UBlox 6M unit, not the smallest, but works well.

The system was set up for a 125kHz bandwidth and a spreading factor of 11, which gives a data rate of 500 bits/sec.  Initially, I tried a spreading factor of 7, giving a 5000bit/sec. data rate and this worked for a couple of kilometres with the receiver sitting on the car dash.

For a more controlled test, I set up the transmitting unit in the back yard connected to a mobile whip that was about 3 metres above the ground.   The receiver was packaged up in a small speaker enclosure and fitted with an LCD display so I could see the decoded packets.

I went for a ride on the bike with the receiver and chose a distance 5km from the transmitter that was non line of sight to see what would happen.  To my surprise, packets were decoded, a few were missed depending on antenna orientation (a 70cm portable rubber duck).

Here is an elevation plot of the radio path:

 

 

 

As you can see, it is not line of site.

Here is the GPS tracker:

GPS Tracker

LoRa Receiver

Here is the receiver:

Power for the tracker is supplied from a single 18650 LiPo cell.  A 3.3 volt supply is used for the Arduino Pro Mini, Neo 6M GPS and RA-02 LoRa module.

To save power, the GPS is powered up every minute at the moment for testing.  A REG103 LDO regulator with shutdown is used for turning off the GPS receiver and RA-02 between transmissions.  The initial cold fix takes about 30 seconds inside the house.  Subsequent warm fixes are very fast at about 2 seconds.

With the Arduino sleeping between the 2 minute GPS reports, the cell should last about 15 days.  The idle Arduino Pro Mini sleep current @ 8 MHz is about 2 mA.  The GPS uses about 60 mA and the LoRa transmit, about 120 mA.   It should run for more than two weeks hours on a 2500 mAh cell (providing GPS signal available, if not it will spend longer searching).   After a warm start, the GPS only takes a couple of seconds to acquire.

With the LED removed and the regulator bypassed, the sleep current should be even lower.

Tracker packaging

For the moment, the tracker is packaged up in a section of PVC pressure pipe with the cell.  This is quite robust and water resistant although not very efficient in terms of space.

If I can source a robust enclosure of the right dimensions, it could be much more compact.

PVC Pipe Housing

The housing turned out to be far to big, so I found a standard ABS enclosure that was about the right size and rebuilt the board to fit.  Unfortunately the 2500 mAh cell was too big for the enclosure, so I had to settle for a 1000 mAh cell with a rectangular footprint.

This cell fits very well, but the run time is now reduced to about 1.5 weeks, assuming a report time of 3 minutes.  The idle current has dropped again by reducing the bleed current from the Arduino port pins to the LoRa module.  The port pins are now all set low before entering sleep mode.  The idle current is now sitting just over 1mA.

New smaller case

Here is the path profile of a GPS report from the tracker when the antenna was a mobile whip on the car and the receiver antenna, a J-pole mounted at roof height.  This one must be an exception as the path is not line of sight.

Link Path of approximately 10km

NanoVNA and EFHW

The NanoVNA arrived the other week, these units are very good value and work remarkably well.  With small size and weight (63 grams)including battery, it can easily be taken out in the field, providing it is protected from the elements.  The NanoVNA is a PCB sandwich essentially, with open sides.

The first test was to sweep the filters in the compact 20 metre transceiver.  The filters came out pretty well spot on.

Next test was the 3 band SOTA portable end fed half wave antenna to see if it was still in resonance after many activations over the last few years.  It had been repaired a couple of times after tree snags, and maybe lost a cm or two.

The antenna was tested as a sloper in the back yard with one end supported by a 6 metre squid pole, and the feed end supported by a walking pole.  It was also tested in the same configuration but using an 8 metre squid pole.  The results only changed slightly, lowering the resonant frequency.
A short counterpoise was connected but this made very little difference to resonance.
The shots below have some light reflections, it would have been better to connect a PC, but convenience won.

40 metres using 6 metre pole

 

20m using 6 metre pole

10 metres on 6 metre pole

 

40 metres using 8 metre pole

20 metres using 8 metre pole

10 metres using 8 metre pole