Friday, 22 April 2016

A little news

I volunteered to act as Chairman of the UK Microwave Group, following the Martlesham Microwave Round Table AGM. I've just heard from one of the current Committee members that my offer has been accepted
No one on the Committee wanted the role, which to be honest can be a bit stressful at times, so I made the offer to act as Chairman for the next year, at which time a prospective, new, Chairman may be willing to take over. 
I have previously been Chairman, taking that role for the first 3 years of the then new UK Microwave Group. 
As one of the original 'Gang of Three' who set up the group when it appeared that the RSGB spectrum committees were probably going to be disbanded, I was extremely pleased to be appointed as the first Chairman and saw the Group through the first, sometimes difficult, few years of its life.
The UKMicrowave Group quickly became the recognised elected voice of UK microwavers. That is a role it has performed well for nearly 17 years. 
OFCOM recognises the Group and has been known to seek our help in matters relating to amateur microwave operation, allocations and even technical help. 
I am pleased my offer to act as Chairman for the year has been accepted and I will do my best to live up to the role.










Sunday, 10 April 2016

Mast maintenance

With the help of G8BHC (Scatterpoint editor) I did some maintenance on my antennas this afternoon. My intention was to replace my failed (coax relay) 2m masthead preamp with a DG8 premp and return to single cable working ( leaving a free coax for 6/4m).
In the event we ended up removing one of my two 23cm 44 element yagis and moving the 67element 13cm yagi into the space vacated. The 3cm dish was moved up the mast (slightly). All of this in addition to fitting the DG8. However, we ran out of time before I could run in the new length of LMR600   coax between the preamp and the shack. I did manage to measure out the run and the new coax was was just long enough.
Listening on 23cm, using the single antenna, didn't reveal any great loss of system performance but time will tell. I have been suspicious of the 23cm antenna combiner and the removal of one antenna seems not to have resulted in a 2-3dB drop in performance, as best I can tell. Similarly the 3cm system was checked out and reception of GB3SEE at -13dB on JT4G was a bit above normal, but probably due more to conditions than the marginal move of the dish to its 'higher' location on the stub mast.
All in all a useful afternoon!


Saturday, 19 March 2016

Anglian transverter

Some of the readers of my blog will be familiar with my Anglian 144MHz transverter kit.
Around 200 of these have been sold with the majority being of the form of an assembled SMD PCB. The builder is then left to install the board in the supplied case, solder in the leaded tunable inductors and the crystal. 
This has worked well and although there has been the occasional dry joint problem, most seem to work first time and provide the builder with a transverter that is high in performance and low in cost whilst allowing the builder to boast that they built it themselves.
I am now coming to the end of the current batch of assembled boards, helped by a recent 'sale'. This has been more successful than expected and the boards have quickly depleted.
My dilemma is what to do next. Having boards assembled has not been without its problems and is by no means an economical way to go with relatively small production runs.
I am not inclined to have any further boards made this way.
Unfortunately the PCBs are designed for reflow solding and therefore have small SMD pads. These are awkward to solder to when using hand soldering.
As the Anglian has been so successful I am reluctant to stop production. Maybe the best way forward is to re-design the board, changing a few aspects of the circuit for even better performance and selling this as a kit where the builder has to solder around 120 SMD and leaded parts onto the board?
With about 12 boards left I am going to have to decide soon!
73 de Sam



Life on 23cm EME

Last weekend I got the 23cm EME system back on line. A careful check of my log shows I had worked 62 'initials' ( first contact with a new station or #) up to May 2012, when I last operated on 2cm EME.
Since last weekend I have brought the initial score up to 74
It is quite clear that a lot of new stations have appeared in the last few years. I recognise a lot of the callsigns as those belonging to customers who have bought an VLNA23 from me. I find that very gratifying!  I think it should be possible to raise the number of # to 100 by the end of the year. At least it gives me a target to aim for and means I will have to keep the 23cm preamp and feed on the dish.
The majority of new contacts have been worked on JT65C and running MAP65 has allowed me to see who is on without sitting in front of the PC laptop all the time.
I can see that not withstanding last weekend's 23cm DUBUS CW EME contest, when up to 16 stations at a time were visible on the spectrum display ( P3 and SVGA card to 23 inch monitor), the move to digital on the band is seriously underway. I realise this will unsettle a good few of the old CW stalwarts, but King Catute showed the fallacy of trying to hold back 'nature'. In this case human nature.
I'm not a good CW operator and am always relieved when a CW contact has finished and in the log, wheras I actually enjoy the surety of digital contacts. It's not fashionable to say this, but I still can't see the attraction of CW when digital allows so much greater reliability of contact exchange information. And for me personally, far less anguish about not copying callsigns correctly or dealing with 'off sequence' information.
Sorry guys!

Thursday, 25 February 2016

Initial results of LNA destructive testing

I've repaired a few VLNA preamps for customers where the noise figure has gone up noticeably whilst the gain remains at the original level.  Because the gain remains normal it can be difficult to diagnose without noise measurement, as everything appears nornal except signals are weaker.
As part of my tests of failure levels in VLNA preamps at 23cm I have now managed to duplicate this problem. This, with other test results, will appear in the proceedings of the EME2016 Conference, later this year.
The problem is due, not surprisingly, to RF overload, probably causing increased gate current to flow.
The signal level it took to cause the problem is the big surprise!


SDRPlay

CHaving read the review of the SDRPlay in Radcom I thought that I'd buy one. Now it has arrived I needed to connect it to my TS2000X in order to monitor 23cm. I am able to do this already via my SDR-IQ and a PAT board connected to the 10.695MHz second IF.  However, I find this most unsatisfactory as the 'span' is limited to about 30kHz and it reverse tunes (LO is high).
Connecting an SDR to the TS2000X first IF (135.495MHz) turned out to be surprisingly easy as there are a number of MMCX connector test points on the 23cm module board. CN8 is a light tap (1pF) between two 135MHz tuned circuits and after the first mixer (dual gate fet). CN7 is after the 135MHz monolithic filter and first IF AGC controlled stage. I really didn't want to have the SDR connected after an AGC controlled stage as it meant high level signal level measurements would likely be wrong as the AGC activated. I was not sure whether the IF at CN8 would be clean enough. It was.

I had a short UF/L to SMA cable end jack available, purchased originally to use with a small GPS module. The cable is about 15cm long and very thing, but as this is 135MHz it doesn't really matter.
This cable turned out to be long enough and thin enough to loop around the 23cm antenna flying lead as it emerges from the TS2000x. By carefully drilling a 7mm hole in the removeable (2 screws) cable clamp plate I was able to fit the SMA bulkhead jack in place next to the cable lead. There is not quite enough room to do up the SMA nut, but a dab of superglue soon fixed that!

With this IF tap cable in place I was able to do some tests through the front end. Previously ARRL Lab had measured the TS2000X noise figure as 6dB as quoted in a QST review. From 23cm RF cable input to CN8 cable IF output I measured a very acceptable 4.1dB, consistent across the whole 1240 to 1300MHz range with a centre band gain of 14dB falling by about 1dB at either end of the range. I felt this was adequate for connecting the SDRPlay without using one of the G4HUP PAT boards, although this might be fitted at a later date. 

Checking with my spectrum analyser there were no spurious mixing signals within the 135MHz IF range down to at least 70dB below my -40dBm 1296MHz input test signal.
Using HDSDR I was able to see noise from the SDRPlay up to about 3.5MHz either side of centre band (IF) and able to zoom right into signals such as the Martlesham 23cm beacon. There is no centre (DC) spike even using the zero IF option!

If ever the SDR application teams ever get round to adding Continuum mode to this or any of the other SDR programes they will become really useful for measurements rather than just looking at signal spectrum (Spectravue does have Continuum but doesn't run on the SDRplay!)

Back to destroying 23cm preamps in the name of determining safe input levels......

Sam


Thursday, 4 February 2016

G3PYB -SK

I am off to Havant in Hampshire tomorrow. Peter's funeral is on Saturday morning and to avoid an early start from over here in Suffolk I thought I would stay over Friday night at the hotel where the post funeral reception was to be held.  It turns out that a number of other microwavers and ATV enthusiasts had the same idea. I only just managed to get a room!
There will be, among memorial items, a QSL board where we will be able to place our QSL cards. 
This seems to be a popular idea and has been seen at several memorial receptions over the last few years. 

I haven't been in the Havant area for many years and I'm looking forward to seeing it again, albeit in somewhat sad circumstances.