WSJTX Day Two

The PTT COM Port problems were widespread, worldwide, and a fix was issued so next day I was able to try on air.
Others were getting decodes of PA0O using JT9-2 I wasn’t although signals were strong. G6AVK could not decode my JT9-2 and my signal was strong with him.

Finally I tried JT9-1 and G6AVK copied me.
I know I can rely on valid tests with G6AVK so I had PM’d him with my plans.

I then set up another computer to produce local copy of JT9-2 that didn’t decode, again I closed down in disgust. The first time was when the Com Ports locked in TX mode the night before.

Later I fired up 2 RX and computer set ups to try JT9-2 local copy it worked, why the may be state? did the computers need a re start before the software performs?
I tried on air it was decoding PA0O and G3XIZ. I called G3XIZ and we exchanged calls but not full QSO details.

What happened to the rest of the QSO I don’t know, reports from others show patchy decode result also. There is an odd phantom signal that occurs quite often, others have reported it, when the RX session ends the signal just received is repeated again in the waterfall, no wonder folk get confused.
I do wonder whether this software was ever tested before being hurriedly released. Sound level settings are different on 2 machines using XP one has no Windows Wave Slider control of the levels and is very difficult to get set correct for my Transverter.

Some patchy local decodes here (and of my signal at a distance) have proved to be due to frequency drift, it will be interesting to see just how much the system will tolerate.
I have wrapped the Local Oscillator of my Transverter in foam and put on the case cover. It is cold in the shack the PA does not get very warm but it did cause drift, only a few Hz.

Today, Monday, I had a proper QSO with some chat and fun also including I LUV MAL one of the many messages that can be passed. Guys when you are using the mode please note the Tx first box it is unticked by default, learn to use it and we wont see stations in QSO both sending at the same time and taking 30 mins to exchange reports.

New mode for MF and LF wsjtx from Joe Taylor

I hope to be testing wsjtx very soon. Provisionally running on this computer I can’t get a COM Port to open, the software latches into permanent TX if I clear the can’t open message. I have to run it as administrator otherwise the program crashes before I can complete the Setup / Options page.

With no COM Port set, TXing takes place fine and the test de-code works.

The CQ K1ABC FN42 de code is the JT9-1 test file signal the fastest JT9 and the one with the poorest S/N. Take a look in the waterfall the signal is circled in yellow.
Right Click on the picture and select View Image to see it all.

Next step for me is to set it up on a shack computer and try it out on 500KHz. It will be good to see the performance of the longer narrower JT9’s.

EDIT Shack computer does the same things, COM1 latches on permanently, Software crashes unless its an Admin Account. Additionally Audio Output levels are huge much greater than any other Ham Software, I would need an attenuator to run it into a rig.

472 / 500 KHz Transverter Antenna Match

An odd thing I have noted before is an apparent change of resonance of my loaded antenna when I changed rigs.
I was quite wary at first with the XBM Transverter feeding my Hi Q Antenna, I hate killing IRF510’s, they are not expensive, its the fear of becoming a serial killer.

I carefully tuned the loading coil using one of my QRPp TX’s, its on 501.7 KHz (OPERA freq), then changed to the Transverter, all was well but I noted that the fine tuning was at the end of its range, looking like it wanted more inductance.
Well I have the Transverter on 501.5 KHz, so the tuning really is that sharp, I had to make quite a bit of physical adjustment to resonate that 200Hz lower, it is difficult to be sure but I think I have increased my ERP by 3dB.

Here is the field at the top of my loading / tuning arrangement at the inv.L wire terminal, the lamp is a low energy florescent without the electronics, one of the many ‘failed when new out of the box’ individuals, due to dry joints, usually Philips brand.
Now, please don’t say why don’t I return them under guarantee, there are no receipts with free bulbs. I do however return the comments slip.;)

,Trefoil Load Test

I looked at the scroll of 84 resistors and decided that maybe the Trefoil construction would be better. Today I tested it and it certainly is, VSWR at 144MHz is 1.3 to 1, 50Mhz 1.05 to 1 below those frequencies it is 1 to 1.

10 Watt 500KHz XBM Transverter Distance Record

The G3XBM transverter design is doing well here, lots of distant stations are already on 475KHz so it is particularly good to have some still looking on 500KHz.

On the 16th Oct DG3LV reported seeing what he thought may be my signal,just too weak to decode the very next night was better, several decodes, he sent me the WSPR screenshot.

Tests will continue with my ERP of around 6mW at best, from my tiny antenna an inverted L 6m vertical and 6m top.

Trefoil 50 Ohm Load

Out with the 4K3 Ohm resistors again 84/3 = 28, instead of scrolling a ribbon of resistors we can have three circles. I created a complete ribbon, neater this time as I had done it before, then divided it into three and coiled the sections. Like the scroll this is as yet untested, except that it is 51 Ohms precisely. The three sections are fed via a central ring with very little distance to each resistor and little support work.

I want this one to go in oil, it should fit neatly with almost no wire needed on a SO239 or BNC chassis socket, now to find a tin and the oil.

In the USA its called Mineral Oil and seems to be sold cheaply by the gallon to keep in the cupboard as a handy, drink it, rub it all over, clean it, polish it what ever.

In the UK we call it Liquid Paraffin, apparently it is cheaper to produce than to dispose of it, we can buy it in small quantities at high cost with a smell added as Baby Oil. The UK is rapidly becoming the blame / compensation culture capital of the world, if I want 150 ml of unadulterated stuff fit for drinking (to help with constipation), and Dummy Loads, I have to be interviewed by a Pharmacist, fill in a questionnaire and sign a form. Just so that if I drink it all at once, **** myself to death my widow can’t sue the Chemists. What are we coming to?

Maybe a pound shop will have a big bottle of cheap brand Baby Oil and I will go for a sweet smelling load. I doubt if the smell ingredient will be conductive.

Millenium 50 Ohm Load Tests

I posted this design of a 50 Ohm Load yesterday, it was completed but not tested at all except for confirming the DC resistance. Now there is confidence, I had a feeling it was going to be OK. It is so good I just had to add a vertical photo and give it a name The Millenium Load.

One other factor I took into account when building was the fact that the resistors could be inductive, the metal oxide being trimmed to value by cutting a spiral. I assume that during production all resistor come out of the manufacturing process with the spiral in the same direction and and the colour code applied without turning the resistor. So I fitted one resistor one way and one the other. A waste of time, well maybe I don’t know but I am hoping that they cancel each other by so doing.

So here are the results, subject to my test gear accuracy. The Load is useable up to 145MHz where the VSWR is 1.5 to 1. At 50 MHz the VSWR is 1.4 to 1, I applied 100 Watts of FM for a short time perhaps 30 seconds with no smoke, the copper spider work got hot enough to be untouchable, well where do we expect 100 Watts to go.

VSWR performance is much better at 14 MHz than my original 2 x 100 Ohm solid carbon resistors in parallel. EDIT I thought that was a bit odd so checked again next day, the diode had gone funny (probably the high power tests pushed it too far) on the 2 x 100 Ohm load it is actually better than the Millenium Spiral.

Home Built 50 Ohm Load

When I let some smoke out of my QRP 50 Ohm load the other day I thought about making one with a higher power rating. We all know the 20 x 1K Ohm resistors in a paint tin design, I don’t have the resistors so it would be either send away or try Maplins.
I consider myself fortunate having a branch near by but oh dear they are getting worse. They list 1K Ohm 2 Watt resistors on the web, thankfully you can check your local branch stock online before you yet again waste your time on a trip to store. Locally they had one in stock, there were not 20 in the whole midlands, do they not understand that that will rarely be of any use to customers.

So I turned to what I have, in the loft I found a bandolier of 4K3 0.3W, hundreds of them. A mere 84 of these in parallel gives me 51 Ohms. I then put some thought into the construction, maybe having smaller resistors could reduce inductance, particularly with zero lead length. Assemblies for the 20 x 1K usually involve fairly long resistor leads, inductance, circular tin discs, capacitance. I devised what should be a low capacitance assembly.

The result is a compact solid pack with plenty of air space around each resistor, if inserted in oil it would have good oil circulation particularly if mounted horizontally.

It measures exactly 51 Ohms, next come the test to check the SWR and see what frequency it is useful to. I have added a low forward voltage RF Diode and decoupling capacitor for power measurement.

600m WSPR OPERA QRSS Transverter

Roger G3XBM recently produced a simple design for an 80m to 500KHz Transverter. The output is way above my power level interests but 500KHz antenna’s, particularly mine are so inefficient that the radiated power will still be in the uW levels.

I decided to test the reproduce ability of the circuit, I think I was the first to try, there were a couple of head scratching moments before I discovered a couple of notation errors on the circuit diagram due to values left over from previous versions creeping through.

It was fun to be the Alpha tester of the design, particularly the last test, with my FT897 connected, my Netbook running WSPR and my power measurement 50 Ohm load on the output socket. Previous tests were with a short carrier to measure / set up levels. The 2 minutes of WSPR was a bit OTT for the load and some of the smoke inside was released, there is still some left in there though.:DD

I used the new pad cutter from GQRP Club and double sided PCB. I have used both copper planes as extra heat sink for the PA, coupling ‘studs’ for heat transfer being situated around the PA heat sink. Roger mentioned drift due to PA heat so I also used a metal box, that too coupled thermally to the PA.

Results using QRPp power levels previously were spectacular but many stations were not able to see my or Rogers’ signals. When Roger ran his test using the 472/500KHz Transverter many previously difficult or impossible stations appeared as if from the mist.

My first on air test in daylight brought reports from PA3FNY at 409Km a very good start. OPERA and QRSS tests will follow.