Friday, 22 August 2014

116MHz frequency locking

At last I got round to connecting my ZLPLL 116MHz source to the Anglian 144MHz transverter.
I am pleased to report it works really well. I could detect no difference in the quality of received signals supporting the observed good phase noise 'measured' on the spectrum analyser.

I was initially a bit concerned at the high level of the source harmonic output. However, when connected to the Anglian it injection locks the internal 116MHz crystal oscillator, leading to a really clean injection signal. 

Monitoring the 144MHz output, with the input 28MHz drive signal supplied by my R&S SMG low noise signal generator, the output spectrum was just as clean as with the internal ( unlocked) oscillator.

Driving the ZLPLL with a clean 10MHz GPS locked reference signal from the SMG, the Anglian signal sounded excellent. I am really pleased with these results, which I will be reporting in my talk at the EME Conference in three day's time!

The ZLPLL as used to provide the locked 116MHz LO for the Anglian.

Google ZLPLL for details of the source. Note that Wayne can no longer supply the OCXO version of the Source, although TCXO and external-only versions are available and OCXOs may be back in due course.

Sam


Thursday, 21 August 2014

EME 2014

Early Sunday morning Dave, G4HUP, and I will be off to France via Weymouth and St Malo 
for the EME2014 conference, being held in Pleumeur Bodou, Brittany, on Monday and Tuesday next week.
Pleumeur Bodou is the site of the former French TransAtlantic satellite station and is credited with making the first successful TransAtlantic TV transmission with the USA. Goonhilly had previously failed due to a misunderstanding as to what defined the sense of circular polarisation to be used! I well remember Richard Dimbleby, on the BBC, sat waiting for the first pictures from the USA and seeing a terribly noisy picture, that was decidedly unusable. FranceTelecom got it right the following day. They weren't first, but were first with a good picture.........there, that will get me into trouble with our French hosts!

I am very much looking forward to the Conference. This will be my fifth EME Conference, the first being in Wurzburg, Germany, followed by Florence, Dallas and then Cambridge. It will be interesting to see where the next one will be. We will be voting for the next venue at Pleumeur Bodou. Venice is one of the known proposals for 2016

My presentation slides are now finished (apart from the usual final polishing that always takes place....) and they now contain new data on the performance of the Anglian 144MHz transverter. It has been interesting preparing the talk for a generally technically well-versed audience from across the globe.
I hope to use the same slides in my talk on the Anglian at the RSGB Convention in October.

Unfortunately, I have been inactive on EME for more than two years. I am currently looking at a smaller dish for use at 10GHz. It should be easier to steer onto the moon and will give me an opportunity to develop my 10GHz capability. Much of the 10Ghz gear is ready to go, but I have only just found a source of suitable dishes.

I hope to have more to say about 10GHz EME in due course.

Sam






Tuesday, 12 August 2014

Filters for beacons

Today I built two 432MHz beacon low pass filters. One is destined for GB3UHF in Kent and the other for GB3NGI in Northern Ireland.
The filters were designed using ELSIE and the implementation is silver plated wire coils and paralleled ATC fixed capacitors and 6pF Voltronics ( I think) ceramic trimmer capacitors. The filters are assembled in small Eddystone diecast boxes with 17mm flange N female connectors as shown.

The measured performance of the filters matches the ELSIE predictions quite well. Insertion loss is below 0.2dB with a return loss of well over 20dB ( depending slightly on tuning for minimum insertion loss).

Sunday, 10 August 2014

Frequency locking 116MHz

I spent part of the day playing with frequency locking my Anglian 144MHz transverter.
The transverter has facilities to either inject or extract a 116MHz LO signal.
Although I had conducted this test previously on my DDK2001 source board, I still needed to check that it also worked with the Anglian LO.
I was pleasantly surprised to see just how far off frequency the on-board 116MHz could be and still lock, with extremely clean output.
Not only that but the level of external 116MHz required to achieve lock proved to be extremely small, easing the requirements on the external source.
This has raised the possibilities of sub harmonic locking! I will be exploring this over the next few days.
All this is being done as content for my EME 2014 Conference presentation.
Sam

Monday, 4 August 2014

My grandfather in WW1

I am posting this as a tribute to my paternal grandfather, Edward Jewell.
'Ted' joined up to the Royal Sussex Regiment on the 19th February 1915 at the age of 18.
He was sent to France on 10th December 1915

On the 14 February 1916 (St Valentines day) he was wounded by shell fire and evacuated to Boulogne and thence returned to to England on the St David hospital ship.

After several months in the Norfolk Hunstanton VAD hospital he was turned to France on 29th September 1916. To the Somme, again!

He still had shrapnel in his upper body. It could not be removed without further damage.

On November 28 1916 he was digging a trench in the Somme when it caved in on him. He and others were buried or trapped.

My grandfather was buried up to his waist in mud and water for three days. When found and dug out he was evacuated  to an unidentified field hospital in France. There he was diagnosed as having trench foot and subsequently had both feet amputated on the 12th December.

He was returned to England and then had a double Syme's amputation of his lower legs. Separate to the removal of his feet.

He spent months at Roehampton hospital recuperating and for the fitting of artificial limbs.

He married Grace in November 1920. My father was born in 1921.

Ted lived until he was 72 when he died of a massive heart attack at the Royal Berkshire hospital in Reading. 

He suffered deteriorating health from the time of his amputation, but managed to live a full life.
I am very proud of my grandfather.

Sam


Monday, 21 July 2014

Rain water

Since my mast had been luffed over for several days due to thunder storms ( Sunday was particularly bad on the east coast) and heavy rain I thought I ought to check out the 10GHz transverter. Normally I put a cover over the transverter and over the masthead preamps to keep rain water off when it is in this state. When the mast is upright the boxes are reasonably waterproof, but not necessarily when they are on their side.

When I opened the 10GHz transverter diecast box I found some rainwater had indeed entered the box. Part of this was due to the fact I had not fully tightened down the four lid screws and part due to the fact that I hadn't applied sealer to the lid/box seam.
With the water removed, the box dried out and the dessicant bags replaced I set about upping the transmit power. I had suspected that the DB6NT transverter tx gain pot had been turned way down to accommodate my usual 2W of 144Mhz drive. With the pot readjusted I was easily able to get 10W output as indicated on the remote meter in the shack. This using 100mW on 144MHz in the shack.

After scraping the old sealant off the box seams I replaced the lid, screwing the four holding screws down tight. Then I used a mastic gun to squeeze Down Corning 745 non-corrosive sealant along all the box seams.

A quick test and up went the mast again.

All appears well, but with 6m open to North America, it didn't have any on air tests as I preferred to call  a VO. Unfortunately It seems the opening was not reaching thisart of the East Anglian coast as all I hear was the one VO and I couldn't make myself heard to him!

So I went away to do some measurements on a commercial 2m preamp ready for a published review in Radcom........

Sam


Wednesday, 9 July 2014

96Mhz LO

I've been meaning to upgrade the 96MHz LO in my homebrew 23cm transverter. My own design OXCO has proven to be less than ideal. Having rescued a 106.5MHz G8ACE from an old 3cm transverter I thought I would convert it to 96Mhz in order to drive the multiplier chain in my 'Kirton' 23cm transverter ( see RSGB Microwave Handbook Vol 2).
I spent much of today calibrating the crystal turn-over temperature and then refitting the OCXO in place of the original OCXO.

With the 96MHz oscillator running on frequency I was able to hear the GB3MHZ 23cm beacon using just a three element WA5VJB PCB yagi in the shack. It sounded much cleaner that with the old OCXO,  but some of that may be down to the new beacon hardware, commissioned since I last used the transverter.

Driving the 'Kirton' transverter from the 50-100mW 144MHz output of the 'Anglian' transverter, driven in turn by the ICOM IC756Pro3, I was able to see 15W out of the PA module connected to the Kirton.


Just right to drive the 200W SSPA!!!!

The picture above shows the inside of the HB transverter with the OCXO lower right and the Mitsubishi PA module inside the huge heat sink on the back of the transverter. The 'Kirton' module is inside the tinplate box in the centre of the photo with a nice connectorised multiple band pass filters over it. This is used on receive only and effectively removes all potential out of band interferers!


At 15W out the spectrum analyser shows the output spectrum from 10MHz to 2.9GHz. The second harmonic is a credible 50dBc, but even so a 1.3GHz LPF will be needed at the output if this is connected to an antenna.

Here, the spectrum is centred on 1296.2MHz, with a span of 100kHz. The output looks pretty clean, so I think I am going to be able to use the transverter again. Looks like the TS2000X may be finding itself redundant. At least for a while!

Sam