Tuesday, 25 October 2016

Chance of Aurora and no visibility

Past few nights have had lovely clear, starry skies, tonight theres a class G3 geomagnetic storm

and its bloody overcast again!

There seems to be a correlation between likelyhood of visible Aurora, and how bloody cloudy it is in the UK!

Ive done a slight bit of work on the Ultimate 3S today, such as making up the power cable and the cable to link the GPS module. Ive also ordered a BNC panel socket with coax pigtail in order to make the RF connection. The pads to connect are too small to safely solder coax to, even the very thin stuff I have, so ive added another pair of header pins. I can either solder direct to them, or put a plug on the coax. It will probably be a couple of weeks before the cable arrives, but in the meantime I can test using the only other coax pigtail I have that will fit, which is currently attached to the prototype Wispy project that I couldnt get to work for me. The reason im not just using this on the U3S is that the coax connector on it is a TNC!

I also need to find a suitable case for it! The U3S itself needs a metal box, ideally one that I can modify to accommodate the battery housing from a PRC-349, but then a plastic box will need to be attached onto the metal one to house the GPS unit. The plan being to make the whole thing completely portable - just attach antenna and switch on!

Ive just returned to 40m WSPR - as expected, the geomagnetic storm has pretty much wiped out HF propagation, but, there are a few spots to be had!

Monday, 24 October 2016

PRC-349 batteries for Ultimate 3S?

A thought has occurred to me - Whatever case I decide to build the Ultimate 3S beacon into, could I possibly modify it to include a housing the same as on the PRC-349, perhaps by actually sacrificing one of the '349 cases, so that the Ultimate 3S, once boxed, will take a PRC-349 battery as its power supply?

This would be a very convenient way of powering it in the field, and give me something to use some of these batteries in!

Ultimate 3S Beacon - Almost Complete

As well as charging batteries, and reading more Mark Twain, ive also done a little more work on the Ultimate 3S today,

Having built the main unit, the synthesiser, and most of the GPS unit, I pondered slowly along today with the Low Pass Filters. One small thing I noticed, which I may mention to Hans for a possible future improvement, is that there is no easy way to see what band a filter is built for! The underside of the PCB is all groundplane, and so I have marked the band on mine using a Sharpie pen, but from the component side theres no place to mark. Im not sure what the solution would be, perhaps an extra 1 or 2mm along the side of the board to give a writing space? or perhaps scratch boxes on the silkscreen for each band to mark?

Anyway, the first job way to fit the capacitors. These filters use a tried and tested design detailed on the G-QRP club website and in SPRAT, being 7-pole filters.

Some way of holding the tiny PCB steady is essential! Below is the 10m LPF having its capacitors soldered in place.

One problem with this build, which is fully appreciated and warned about by Hans in the paperwork, is getting the header pins aligned properly! My solution to this problem is to slip the pins into a piece of Veroboard prior to soldering

A bit of Veroboard helps align the header pins when soldering
After this came the task of winding the toroids. L2, the middle one, went just fine, 24 turns. But somehow the other two I managed to get the wrong 'handedness' to them - meaning the wires were on the wrong side of the toroid from the holes! Both had to be unwound and done again! And then, on one of them, I managed to get the turns crossed over and knotted!

But, after unwinding as far as the crossed section and rewinding, followed by some hard work getting the damn thing soldered back in, I finally have the 40m filter complete. A bit of hot-melt glue was used to secure the toroids in place.

The above photo shows my 'almost' complete Ultimate 3S Beacon, ready for 40m operation. The last task will be to make up a header plug to connect power, and to connect the RF out to a coax socket. I should have a coax tailed socket of some sort that will work for this. Another header cable, this time 4-way ribbon, will be made up to connect the GPS unit.

Clansman Battery Not Charging? Try This...

When I was refurbishing my first PRC-350, I had a small problem, in that no matter what I did I just couldnt get any power from the battery cassette. Eventually, after a lot of head scratching, I discovered that something was causing a very bad voltage drop when on load. This was eventually found to be due to corrosion building up under the battery contacts. Removing them, cleaning them up, and in this case replacing a rusty washer, sorted the problem out.

Fast forward to today, and I am somewhat lumbered with a great stack of old PRC-349 batteries. I cant shift these, because no one wants to buy untested and potentially no use batteries, even when offered at under 50p a go!

So, ive started cycling them all to find which are usable. The first problem I had was making a reliable contact to the batteries, croc clips have a tendency to spring off! Here one of the spare PRC-349 chassis came in handy. Stripped of all other parts and connectors, save the battery terminals, this can be easily connected to the charger by, at present croc clips, but soon by its own dedicated pair of 4mm plugs. The battery then screws in and makes perfect contact just as if being fitted for use in a radio.

Chassis turned into charge adapter
One problem that ive been having though, is that some batteries that are showing decent terminal voltage when tested with a meter, fail when attached to the charger. The Accucel-8 reports 'connection break', suggesting that it thinks the battery isnt connected.

After a bit of thought, I realised that in order to check the presence of a battery, the charger must sense current flow to or from it. Now, as for the PRC-350 battery cassette mentioned above, if the terminals are dirty, perhaps the resulting volts drop over the added resistance is causing the problem?

So I took one of the batteries that this morning reported 'connection break', and cleaned the outer surface of the terminals. One connecting to the charger it still reported a fault. So, not just the mating surfaces! So, I removed the terminals

and cleaned away any corrosion and dirt on the fixed terminals of the battery casing, and the underside of the spring terminals

A very crudded up terminal
This time, when connecting to the charger, the battery was detected! It is now on the second of a pair of Discharge/Charge cycles, and so far showing every sign of being a good usable battery!

Sunday, 23 October 2016

Building the Ultimate 3S

Following on from the build of the Synth module yesterday, today ive done the majority of the construction of the QLG-1 GPS receiver module, and the Ultimate 3S Beacon transmitter itself,

First however, I had the faulty Clansman 10m Audio Extension to repair!

This was an absolute pig! The cores are tinsel wire, which gives the cable great flexibility, but makes repair really difficult. In order to rejoin the cores,I had to wire wrap them with solid tinned copper wire, but the size of the cable and the flexibility made it extremely hard to hold everything together while I did this, and then to solder it all together and sleeve the joins. But it is now repaired and tested as usable.

So, onto the Ultimate 3S. Building the QLG1 GPS module is remarkably simple, with very few leaded components to install. Unfortunately, my kit had two components missing. One, a 100nF dipped ceramic, I simply replaced with one from my own stock. But the other missing part is a 0.1uH axial inductor. I havent anything of this value nor even close to it. So ive had to drop the supplier a line asking for a replacement.

As the GPS module is not yet complete, I have deliberately not yet fitted the backup battery. It is just waiting on the inductor now and can then be powered and tested.

Building the 3S itself is a little more complex, as there are more parts to fit, and a higher density on the board, but still quite straightforward. The only thing that complicates it a bit is that the manual covers the controller build for three different projects, which leaves the builder to fathom out which links etc must be installed. I would have preferred the manual to have been more specific to the 3S build. But then again, theres still plenty of optional features in the 3S build to decide, such as whether to have software controlled LCD brightness (in my case yes), whether to soften the contrast control (yes), etc

The module can use its 5v supply to power the BS170 MOSFET PA stage, by adding a link, or the PA can be powered by a higher supply, for instance if fitting extra BS170s, without the link. As I want to use mine with the one installed transistor first and then add more later, and didnt fancy having to unsolder a wire link, ive been clever and fitted header pins and a removable link as used on PC motherboards. Ive also done this for the two links that hard wire the LPF module connections, as these links need to be removed if the relay switched band module is used, which I may wish to add later,

Im hoping alls well, but I wont be powering it up just yet. Im going to build one of the LPF modules first, so its all ready to test fully.

G-QRP Club Rishworth Convention

Today was the G-QRP club convention, an event which ive looked forward to all year, and that sadly has come to an end, this being the last one. For this, I managed to assign myself a healthy budget, far more than I can normally have for a rally, and set off very early, intending to make a good day of it. I even lugged a ton of PRC-349 batteries there with me to go on the bring and buy stall!

And what a great day it was! Three thoroughly enjoyable talks, cheap cups of tea and biscuits, a fun Buildathon, and lots of interesting junk!

Buildathon kit provided by QRPme

The Buildathon project was a novel no-solder RF probe kit. Over 100 of these were constructed, mine taking me around 2mins. W1REX of QRPme was the provider of these, and of the very enjoyable talk that went with them! Its actually such a nice little kit, and practical too, that im going to remove the SIL sockets and make mine up permanently. Ive got a couple of small analogue multimeters on order, I might sacrifice the available ranges on one to build the RF probe kit in with it!

I only managed to find one of the capacitors I require for the Pye PF8, the 18pF, and I found it in a £1 bag of random parts! Once home, it took me the best part of two hours to sort the bag out and pack all the bits away!

Some of the contents of a £1 goody bag
 My main purchase though was, as intended, the Ultimate 3S kit from QRP Labs. I opted for the additional GPS kit and an extra Low Pass Filter kit. Once complete I will have the option of 40m or 10m operation. I also got the extra BS170 MOSFETs to ramp the output power up a little.

I have already completed construction of the Si5351A Synthesiser module.

Si5351A Synth module from QRP Labs
I also paid up my G-QRP club subs whilst there, and picked up a few non-electronic items, a copy of the ARRL Handbook for Bob, so he might finally stop nagging me about things and leave us free to discuss VHF rhombics more, an old IBA engineers pocket book for 20p, and another mechanical calculator disk for my collection. I also somehow missed Mike Walker who visited but briefly. And, sadly, I also came away with all the batteries I had taken!

Ive made a start then on testing the remaining batteries. Ive taken one of the spare PRC-349 cases to make a battery charging dock for use with the Accucel-8. Perhaps if I do test them and find only the ones that hold a decent charge, i'll manage to sell a few! Ive also made a start baring back the cores of the damaged audio extension cable to repair it, this will be no easy job, as the damn wire is Litz! some very fine wire winding work might be needed.

Wednesday, 19 October 2016

Further Restoration of the Pye PF8

With G-QRP Clubs Rishworth convention just around the corner, ive decided some long standing projects need to be brought back to the fore.

One of these is the Pye PF8 restoration. As of my last post on this subject, some time ago, the PF8 is working and crystalled for 433.500MHz, but, it is working poorly.

The reasons for this are that 433.5MHz falls within the T1 band 405-440MHz, but this radio was built for the U0 band 440-470MHz. Many of the tuning components are at their extremes in trying to bring the radio down to 433MHz.

One thing that can be done though, is to change the band critical fixed capacitors for the correct values for T1. By a stroke of good fortune, ive been able to obtain three of these from my spare PF8 'donor' chassis. If I cant find the other two amongst my junk boxes, then hopefully these can be found on saturday at Rishworth.

Changing these capacitors is only half the modification really needed, but im not able to swap the variable inductors, as I dont have a T1 chassis to find them on. But 433MHz is very close to the lower edge of the U0 range, so im hoping that by fitting T1 capacitor values, it will kind of fudge the set into being a T1-U0 hybrid. The PF8 is only capable of 500mW output and has an internal antenna - so every little helps!

I also could do with finding some modern batteries for it!

The Clansman GSA - More thoughts

I have been in contact with Martin G8JNJ, regarding the GSA antenna, he has provided some very interesting insight into the construction of the base unit, which im hoping can be expanded on to find a way to modify the antenna for best performance on the amateur bands.

Below I have copied Martins comments -

The transformer is a variation of a standard UNUN. The winding ratio is approx 3 which gives an impedance transformation ratio of 3 squared = 9.

Its actually slightly less than this as one winding has fewer turns.

The input impedance of the transformer will reduce to a very low value of impedance at the lower end of the frequency range. So the input resistor and parallel capacitor are intended to add extra series resistance which will increase in value as the input frequency is decreased. The intention is to try and make the input match more like 50R across the whole operating range. However this does mean that the efficiency on the low frequencies will be quite poor, as the resistor will be dissipating a large proportion of the applied power.
 As expected the power resistor is intended to fool the radio into thinking it has a good match and driving maximum power. Now, if the antenna was designed for just one small frequency band, say 70-71MHz, then this lossy resistor would not be required. Whether an arrangement of element lengths can give a suitable match without the resistor but also without any serious re-engineering of the UNUN, I dont know yet, but its worth considering.

Below, reproduced by Martins kind permission, is his better diagram of the internals of the GSA - 

Monday, 17 October 2016

PRC-320 Remote VFO - Why not a DDS?

Ive been asked this several times recently - why go to all the trouble of taking control of the PRC-320s synthesiser, with all the attended difficulties of isolating the decade switches when in remote etc, rather than simply replacing the sythesiser in its entirety with an AD9850 or Si5351 DDS module and Arduino controller?

So i'll explain. Yes, I could remove the synthesier module 9, and VFO unit 3h and replace it with a DDS, using the Arduino to read the decade switches directly....

... and the radio will never work properly again!

Sadly, study of the service manuals reveals that the synthesiser not only controls the VFO, but it also provides the variable control voltages that tune the RF stages, generate the out of lock tones, phase error outputs etc. Recreating those control voltages would likely be a very difficult task. I suppose it could be done, but its not something I have the time to do im afraid.

Im hoping the Arduino stuff arrives soon, as its about time I made a start with the control code anyway!


With my Clansman manpack collection now very nearly complete, the time has come to do a bit of consolidation, and find sensible storage!

I find myself now with an excess of PRC-349s! and their batteries! So the first task is to find new homes for these,

three sets are already earmarked for movement across the pond to the former colonies, subject to the formalities of receiving payment and setting up tracked shipping. Three more remain in search of a new master, whilst three sit quietly in the workshop awaiting refurbishment. Some excess crypto is also on its way shortly to a new home.

More importantly though, I need to find a better way to store it all out of the way when not in use!