Monday, 16 November 2015

Pye PF8 Conversion - Partial Success

With the new 70cm SU20 channel crystals in hand, I found time today to work on the conversion of the Pye PF8 to 70cm. Earlier efforts resulted in me effectively buggering it up, due to not following the procedure properly, and by not appreciating that modern crystals have just a pair of insulated lead outs rather than the whole base of the crystal being an insulator - with the result that I shorted the bloody thing out! I also inadvertently put a little more voltage on it than it likes, luckily it seems not to have killed anything, although I cant be certain yet!

With thin slivers of polythene cut and fitted as insulators under the crystals, I started the tune up. It was here that I hit a major problem. L13, part of the Tx chain, was broken. Both the ferrite core and the coil former were shattered. Replacing it required a complete strip down of the set, and some very careful desoldering and soldering. L13 is the 2nd can from the left below, shown after replacement.

And the original, knackered coil -

Since I had the whole thing apart, I also took the time to replace the lower microphone housing, which was split around the threaded insert.

I also removed C69. This capacitor is only needed when the Tx and Rx frequencies are more than 4MHz apart, but of course im converting this to simplex!

After following the official alignment procedure (mostly!), the set is now working on 70cm, admittedly though not as well as i'd like. For a 500mW set, if so far managed to coerce a mere 50mW from it. On receive, after a lot of fiddling ive got it to work at -85dBm, a far cry from the -118dBm im used to setting PMR squelch pots at!

Before I look at improvements based on trying to convert the U0 band (440-470MHz) components to T1 band (405-440MHz), I want to check that my lack of success is to be expected, and not the result of another fault somewhere.

Friday, 13 November 2015

Musings on LiPo protection

One of the big problems with Lithium Polymer batteries, and Lithium chemistry in general, is that if you get it wrong, it has a tenancy to bite! Overcharging, overdischarging, or piercing/scratching the cells can all lead to the lithium reacting violently and catching fire, and believe me, having used lithium, sodium, magnesium and caesium in the lab, reactive metal fires are NOT something you want happening in your radio!

Its one thing having a low voltage alarm monitoring module keeping an eye on your battery, but if your working an SSB transmitter, the current draw, and hence the cell voltages, can fluctuate rapidly, which could result in the monitor going into alarm fleetingly, and if your wearing a headset you might not hear it sounding. So, modifying the module to trip a cut-off relay is the way to go. But proper dual coil latching relays are expensive, especially with contacts rated for the sort of current I need to pass!

So, ive thought up a simple circuit that should allow me to do the cut-off with a normal non-latching relay. For this to work, the relay must be energized whilst ever the battery is at a good state of charge. Yes, this does mean that when not in use the protection circuit itself will slowly discharge the battery. To prevent this when the battery isnt being used, but still in the Clansman housing, a 'Trip' button will be provided that will de-energize the relay.

The concept is simple. A pair of NPN transistors control the relay coil. The supply to the coil comes through one half of the relays DPCO contacts. At start, the coil will have no supply. An 'Arm' button temporarily bypasses the contacts, putting bias on the transistor base and energizing the relay coil, this pulls in the contacts and the button can be released. A second transistor goes from the firsts base to ground, and is driven by the output of the alarm module. When the voltage drops to the cutoff alarm level, the alarm sounds, this transistor turns on, grounds the bias to the other transistor, which turns off, the relay de-energizes, the contacts open, and the battery output is disconnected. Attempting to re-arm the relay circuit will just result in the alarm module activating and tripping the protection again.

A 1N4148 or 1N40001 diode serves as back-EMF protection across the relay coil. A suitable fuse is also provided as close to the battery terminals as possible. A series resistor sets the voltage of the supply to the protection circuit suitable for the relay coil. It may be possible to eliminate this if a 24v relay is available.

I now need to find a suitable relay and to mock up the circuit to prove that it works as expected.

Wednesday, 11 November 2015

Clansman Battery Upgrade Progress

After waiting for what seemed like months, but was in fact about two weeks, I finally received an email telling me that the specialist Balancing Charger I require was back in stock. I jumped on the 'net and got it ordered, along with the Lithium Polymer battery. If i'd had a bit more spare money I could have gone with a much higher capacity (these are available at over 6Ahr!) but decided to save a few bob and stick with 4Ahr, the same as the original NiCd batteries.

The charger, a Turnigy Accucel-8 150W, is capable of charging and balancing up to 8 Lithium cells (8S), it can also handle up to 27 NiCd/NiMH cells, so will also do the original Clansman batteries, and it can do SLABs as well. The battery is a 7S 25C 4000mAh Zippy Compact Lithium Polymer (similar to the one pictured). This means it has a terminal voltage of around 26v. Its size is such that it will fit in the empty 4Ahr NiCd, but it only weighs about 650g - the originals are about 3kg!

But, its of course not that simple! These batteries if handled or charged incorrectly can be very dangerous. There is a risk of fire. The fact it will be installed in a metal housing goes some way to protecting against this, but further specialist electronics are needed to make this into a safe and effective power system.

Firstly, I need adaptors. The chargers main connection is a different connector type to the battery, so I need an adaptor to connect to the charger, and one to connect to the case terminals. I also need one that converts the 5+4 balancing connections to a single 8 pin connector for the chargers balance port. Then, I need a protection device to prevent over discharge. The PRC320 and PRC351 would quite happily keep taking power beyond the safe discharge cell voltage of these batteries, which would lead to a risk of rupture and fire. So, it needs one of these

This is a low voltage alarm module suitable for up to 8S batteries. When the cells reach the minimum allowable voltage it sounds a loud alarm. I intend to modify this to drive a cut-off relay instead. All this kit will be fitted into the battery housing. Whether I then add panel connectors for the charger, and a window for the meter display, I havent decided yet. It would make charging convenient. I would also need a reset button to reset the latched cut-off relay.
The charger is also capable of monitoring the battery pack temperature with a probe based on the LM35DZ temperature IC. It makes sense to include one of these bonded to the pack inside the housing, and provide a connection point to it as well, as an extra level of safety when charging.
So im now waiting on the adaptors and the monitor board before I can proceed futher with this. In the meantime, I have a couple of bids on for other Clansman items that im missing, notably the 1.2m whip and the carrier frame for the PRC351.
I also have the Simoco PRP70 series radios to play with. I have had to buy a USB 3.5inch floppy drive to get the programming software onto the DOS machine from my main PC, which doesnt have a floppy! Next step there is to build a programming interface and then find a facilities plug for the radios!

Sunday, 8 November 2015

Antennas, APRS, and storage

Its not often in amateur radio that several diverse problems can be all sorted out in a mutual way, but a few of the problems I currently have, that is, the poor reception of APRS signals at my QTH, my lack of directivity for 2m SSB operation, and my lack of segregated storage for components.

But it seems all these can be alleviated by grouping them all together into one task - create storage space.

The first part of this combined task will also remove one more thing that Julie has told me to sort - the growing number of empty butter tubs under the sink!

So, with as many of these as I can use now labelled up with their future contents, ive mostly used them up. But that leaves me facing another problem - lack of workshop shelf space. That problem can be relieved by a simple expedient - stop clogging the shelf up with a currently unused 144MHz Yagi in a box! And, at the same time, make some space by moving a 2m collinear out of the workshop!

So, If I do that to create space to put the butter tubs on the shelf, I need to put those antennas somewhere. And clearly, the sensible place to put them is up in the air! So I will change the layout of my 'test' mast (the only one currently available) to accommodate these two antennas. By putting them on the mast, I will improve my SSB capability to the SE, and at the same time my 2m FM and APRS reception!

It would help, of course, if I completed a few projects that are on the bench!

Much of the time at present however is taken up with testing and charging batteries, to get the Lions radios ready for use with the Christmas Sleigh. On the subject of batteries, ive finally had an alert that the special multpurpose charger, which will do the 7s LiPo's for the Clansman, is in stock. That is now ordered, along with the LiPo battery!

Saturday, 7 November 2015


Its been quite a while since I played with packet! In fact, I probably havent used packet radio since the affair of Bobs laptop, way back in my teenage years. Now, Bobs laptop was in fact somewhat more - it was a complete portable packet radio station in a briefcase! Comprising of small handheld radio, SLA battery and charger, Antenna (mounted beside case handle), TNC (Baycom) and laptop. The affair refers to myself and Ian having to visit Bob to retrieve either the machine or payment for it!

But tonight I decided to see about setting up a packet radio system, in particular, an APRS system.

APRS is Automatic Packet Reporting System. Using the old packet radio methods, but with modern internet linking, it allows message services, transmitter tracking, etc etc. Ive become interested in this because I use the APRS systems ability to give propagation information for VHF/UHF at work to help track co-channel interference due to tropo. Ive noticed that many areas of the UK are not well covered for this!

Its been hard work though! Much of the online information is not well presented, many links dont work, and the software needed - clients, gateways, engines, all rather confusing. After a lot of asking about, downloading, and fiddling, ive now managed to get a packet soundcard TNC engine called Direwolf working, along with an IP and GUI front end called APRSISCE/32. On this I am seeing APRS stations around the north, but sadly only via the IP connection - despite having a working VHF radio feed into the system, theres not been a single packet decoded off-air, why, because theres no one within range transmitting it!!!

Thursday, 5 November 2015

A bit more surplus

A few days ago, one of our engineers left me a note to say he'd found a few oddments of surplus PMR kit, that I could have if I could make use of it. Gimme Gimme Gimme!

Its mostly Philips/Simoco PRP70 series kit, which Ive not made much use of in the past, since most of what came my way was U0 band UHF stuff. But of interest here is the later model PRP76 on the right, which is E0 band - VHF Low Band! That means it would go on 4m. All the radios work, but theres only two batteries, and they are very dead. Im in the process of trying to rejuvenate one of them at present through a combination of high current burst, trickle charging, and pulse charging over several charge/discharge cycles.

A few years ago I did have proper charging equipment for them! But I passed it onto someone else. Even once I get a working battery, the next problem will be programming them! I will need to track down a facilities connector, then the required software, and then probably breadboard up a programming interface.

Theres also a rather nice little Yaesu VX2000V VHF 40ch mobile rig. This will probably go quite easily onto 2m, and might make a neat little set for special events, or maybe for APRS. Ive been considering setting up for APRS for a while. I use the Mountain Lake APRS propagation website to track active and potential UHF co-channel interference conditions at work, and would like to set up a fixed APRS unit up at Emley Moor to contribute to the propagation monitoring.

More pressing, is the need to build a crystal controlled oscillator for Sams school science fair project. Based around the big old 10X slab of quartz I got from Rishworth, we are going to perform an experiment where we will test the effect of grinding the crystal on its frequency. I of course already know the likely results, but Sam will discover this for himself. I just need an oscillator that will work with a crystal of this size, and to build some form of holder for it!

Saturday, 31 October 2015

Chinese Airband Radio nears completion

Two issues with the LCD frequency counter module needed resolving before I could finish the electronics on this project. First of course was the detuning affect of body proximity. The second was a desire to ensure the radios voltage regulator was not working too hard.

The second issue was the easiest fix. Rather than draw an 8v supply for the LCD module from the airband radios own 78L08 regulator, I modified the LCD module by removing the power connector, and soldering in another 78L08 onto the board, its output pin going to the +ve connection pad and direct to the counters input diode and 5v regulator (which, it seems is only spec'd to 9 or 10v anyway!), its GND pin to the -Ve pad, along with the black 0v wire, and the red +ve wire direct to the new regulators input pin, with a bit of heat-shrink sleeving. The black blob of the regulator and yellow sleeve can just about be seen in the photo below, above the 'SQL' pot connections. The whole LCD unit now gets a direct 12v supply from the fuseholder.

Issue number one, the loading of the oscillator, was a bit trickier. To help solve this, I knocked up a little emitter follower buffer amp, using a 2N3904, on a bit of spare PCB material, with pads milled out of the copper using a Dremel. This was mounted directly to the LCD module by removing the signal input connector and replacing it with PCB pins, the buffer PCB soldering to these. It can be seen above, sitting above the main PCB at a slight diagonal. The pink wire is the input, which taps the oscillator of the NE602 IC from a pad of one of the capacitors on pin 7.

Body proximity effects are now reduced to about 20kHz shift when close, which I think could only have been improved by using a metal case. This is still well within the IF passband! So even when detuned slightly, a selected station is still receivable.

The LCD frequency counter module was then secured in place with hot melt glue. The final tasks are to drill a grid of speaker holes in the lid and mount the loudspeaker, and to find a 10k switched pot to replace the volume control, so I have an on/off/volume function. In fact, it took a while to find a speaker that didnt cause too much detuning when in close proximity to the VFO!

Another task I need to get on with is a controller for the dew heater strap for my camera when taking astro photos. I got this cheap LED dimmer unit from the far east

 which claims to be a PWM circuit. I have to say I had my doubts, so opened it up. Inside, as well as a potentiometer, there is a small PCB, but no visible components!

But, on removing the PCB it does turn out there is some electronics on the other side. Ive tried to see the thing in action on the oscilloscope, but its playing up again (must be and earth thing!?) so instead I proved it does indeed dim an LED by putting one and a 4k7 resistor across its output.

Next step with this is to attach the heating strap, and monitor the temperature.

Sunday, 25 October 2015

Shack Refurb - Antenna Patch Panel

When Julie decided a new carpet was required, I decided to rebuild the shack. This involved some drastic redesign of the desk, which is still in the process of being repopulated with equipment. This will take some time, as new power runs and cable control is needed.

As part of this, and in order to get away from the interminable jungle of thick coax behind the desk, which makes connecting anything up a nightmare, I decided that all coax coming into the shack will go to a patch panel, and from there, flexible short jumpers will be used to the equipment.

The panel is cut from 3mm sheet aluminium, and will be mounted on a wooden frame. The wall cutout behind will need expanding somewhat, forming a cone shaped entry. Four N-type bulkheads, plus one BNC bulkhead, are provided for the main antennas, plus a little bit of expansion capability. A 30mm diameter grommeted hole allows other cables, such as temporary antenna feeds, DC cables, control cables etc, to be fed through.

Saturday, 24 October 2015

Boxing the Chinese Airband Receiver

Today was the annual G-QRP club Rishworth convention. Although I didnt have a particularly long list of things I wanted, I did have on the list a box of some sort to build the Chinese Airband receiver into. This, as it turned out, was one of the few items on my list I actually managed to get!

I got a fairly generic, basic black ABS box from Bowood's stall. Another stall furnished me with a little ABS box ready to build a regulator unit that Bob has asked me to make for him. I thought it felt a bit heavier than expected, but it was taped up so thought no more of it. Once home I found that as well as the screws for the lid, it also had a number of new connectors inside! Clearly someone had started acquiring bits for a project that never happened. I also found a small, ancient card box labelled as containing a 3.6MHz crystal. On opening, inside was a wad of tissue paper, within which I found a big square quartz crystal slab! But, it was marked as being 900kHz. I bought it anyway for 50p - a nice teaching aid I thought. On looking closer once home, I found two smaller bits of tissue underneath, each containing another crystal blank!

I also grabbed a pair of channel crystals on S16 (2m - 145.400MHz) marked as for Burndept. Ive no idea which Burndept radio, but couldnt resist for another 50p!

Once home, after a detour for myself and Sam to devour copious quantities of fried chicken, I set to on boxing up the airband receiver. This took some thinking about. Sam decided on the front panel layout, and I set to marking out and drilling

The big problem when doing these sorts of projects is the need for the square cutout for the display. This inevitably comes down to drilling lots of little holes, punching out the excess, and then spending a long time filing and checking, until the desired hole is created.

In the photo below, the front has been mocked-up to check everything looks right. The big knob on the right is the tuning control, on the left, the top knob is the volume, and the bottom one squelch.

Despite the space in the box, getting both the receiver board and the frequency counter module fitted took some doing. For a start, all the sockets and controls had to come off the board, and be replaced with panel mount parts. In most cases, the PCB mount component could be reused. Ive added a fuseholder as well, since its likely this will be powered off a 12v SLAB. In order to get the main PCB right to the back, I had to cut the base of two of the PCB mounts in half.

The frequency counter module was then taken out again to be modified (it needed some hex studs removing and sockets replacing with direct wiring) and the wiring added for the front panel controls.

With the various controls wired, and the wiring looms twisted in bundles, I clipped on a temporary 12v supply and powered the receiver up. The first problem to sort was that the volume pot AND the squelch pot were working backwards! I'd got them wired up wrong, and had to swap the wires over. With that done, I found I couldnt tune in a test signal on 125MHz. I guessed that the tuning pot wiper was wired wrong. Checking with my meter, this proved to be the case. That corrected, I could find the test signal, but on increasing it by ten MHz to 135, found I had to tune backwards to find it! I changed the wiring yet again, and now have all the controls working properly. Even finding a few planes whilst I was at it!

Final act for tonight was to add a fuse. I found a 2A fastblow 20mm, which will do for now. I'd rather it was smaller, maybe 500mA, but its all I have at present. Next task is to add a buffer amp for the counter, and wire that in. I will probably give it its own 5v regulator as well, so as not to draw too much on the receivers regulator.

Wednesday, 21 October 2015

Ex-PMR collinears - Preparing to test

It was a bit inconvenient of the riggers to just cut the connectors off, i'll have to say! But, you cant look a gift horse in the mouth, no matter how bad you think its teeth may be.

So, the first task was to get a connector onto one of these antennas. I happened to have some in-line female N-types ideal for the job

With only about three inches of cable to work on, I had to make sure it was right first time! The corrosion hadnt had time to get very far, thanks to our handyman informing me of these being available very shortly after they were decommissioned. Baring the cable back just 3/4 of an inch gave me good clean braid and inner conductors. The connector I had chosen has an insert that pushed into the braid to make the outer connection. With this in place, I trimmed off the excess braid strands, then cut away the excess dielectric.

By now the soldering iron was nice and hot. Tinning the center conductor allowed me to cut it to length for the center pin, which with its PTFE insulator was added next

Before going further, I hoovered up the cut off strands of the braid, and made sure there were none still present that could cause a short. Then, with the aid of a pair of adjustable spanners, the outer body was added and the securing nut tightened on.

So, I now have an ex-PMR UHF 440-470MHz collinear, with a short but usable connector attached, ready to be tested.

I would suggest to the riggers in future that they dont cut the connectors off, but leave the antenna attached, pull the whole feeder off down to the equipment cabin, coil the feeder and antenna up, and let me have the lot!

I need the weather to improve slightly now in order to rig up to test the antenna. Theres a lot of jet aircraft noise today in my area, which unfortunately is overcast, so it seems a good day to scan the UHF airband! Ive already found one active frequency with American accents asking for weather info for various UK airfields. I suspect from the S9 signal strength that this may be an AWACS controlling the exercise.