M5FRA/G8FRA

Desoldering tweezers for SMD removal

M5FRA — Mon, 09 Jan 2012 22:19:40 +0000

After a recent post on the GQRP Yahoo group I have been asked by several people about the desoldering tweezers I use for removing SMDs. They plug into a Xytronic LF 1000 solder station. I have several different bit sizes as follows: 2mm, 3mm 5mm, 10mm, 20mm and 30mm. There are more available but these suited my needs.

I have also used the tweezers to remove blown chips, cleaned up the pads and replced ICs. Would not try that without them!

The best way to describe their use is via the photo gallery below.

(Click to enlarge image,navigate using << >>click on any image to close.)

Some tips

Use the tip that is just big enough for the job.
Work as quick as you can because the tweezers can put a lot of heat into the component.
A slight twist helps move the SMD off the pads.
Watch adjacent components as they can be damaged.
Do not squeeze the tweezers too hard or the removed SMD will ping off into oblivion!
Practice on duff boards before trying a repair.
Finally, do not be scared of SMDs, using them can be fun!

Band pass filter diode switching

M5FRA — Mon, 09 Jan 2012 08:36:22 +0000

The DC receiver project is progressing slowly. Yesterday I lashed up another band pass filter, BPF, using 10mm coils from Spectrum Communications. With a bit of fiddling it worked well. The PCB design is almost done and for the first time for me the BPFs will use all SMD components to keep the size and weight down.

Looking back to the last project convinced me that there had to be a better way of band switching than using relays. A trawl of the Internet brought up a relatively simple diode switched circuit which works well. The measured isolation between on/off states is 44db on 14 MHz which corresponds to predicted results, even with such a very crude lashup on the bench. That was with 0 dbm input so I am hoping it will be a few dbs better at normal signal levels and with better layout.

The switch uses 1N4148 diodes as they are cheap, usually 1p each when you buy 100, and readily available. What was slightly disappointing was the 33mA current draw, I had expected less and need to check the Iss (saturation) current for these diodes and see if it can be reduced. But 33ma is still better than the 2 relays used before.

Next is the RF stage which needs to be as linear as possible with about 15db gain.

(Click to enlarge image,navigate using << >>click on any image to close.)

Quadrantid meteor shower

M5FRA — Wed, 04 Jan 2012 08:40:09 +0000

For the first time I have been able to monitor the radio effects of a meteor shower. It was an interesting time with many pings from about 0400 onwards. The setup was far from ideal; an Icom 7200 as the RX using a long wire antenna resonated for 6m and Spectrum Lab on the PC. The RX was tuned to 46.95 MHz which is a meteor scatter data station on the Norfolk cost about 134 miles from here. Unfortunately that is still close enough to pick up some ground wave but it is the best I can find at present. The results were still amazing and this is something I want to do more of in the future.

Click to play >> MP3 recording of some Quadrantid meteor pings

(The warble on the tone is from the transmitter, it is transmitting digital signals and not a pure carrier.)

For a lot more information on the subject see Andy’s site (G7IZU) here.

(Click on the image to enlarge.)

Easy filters

M5FRA — Thu, 29 Dec 2011 17:45:32 +0000

A few months back I bought a VNWA from SDR-Kits. Now this was a big purchase, equal to a new rig or well on the way to one. So what can it do? The answer to that is that I am still finding out and learning how to use it.

Today is a cold, wet and windy Thursday afternoon just after Christmas, exactly the time to make some BPF’s, band pass filters, for yet another home brew project. The new rig is using the R2Pro direct conversion RX modules with a Tayloe mixer. Originally it was designed to use a ring diode mixer with discrete LNAs, low noise amplifiers, for each band. I say each band but I think this was originally intended to be a single band RX.

The original LNA design was well filtered to stop local oscillator radiation as well as giving some selectivity. Using a Tayloe mixer means that the LO is running at four times the signal frequency i.e. at around 40-41 MHz for the 30m band. This means that a simple low pass filter should stop any of the LO sig from going up the antenna so the need for extra filtering in the RX front end is eliminated.

I want to make a three band transceiver covering 40m, 30m and 20m. It will be mostly for /P use so needs to be small and light. There will be one RF amp with separate BPFs for each band. The last project used some very good toroid based. They were far from small and light so this time I wanted a different approach and decided to use canned coils from Spectrum Communications.

The photographs belows show the first tests of a simple BPF taken from a PDF information sheet from the GQRP club. The sheet shows the values of capacitor to use for TOKO coils which provided a close enough starting point to get instant success. In most cases all that was needed was to go the next value down for the resonating capacitors (C1 & C3.)

Getting back to the VNWA. It was really very easy to tune the coils. For the first time I could actually see what was happening and get figures for both forward and return loss and VSWR. This makes life much easier and produces a better filter response than is possible by tuning for maximum signal. The next step is to cascade 3 or 4 coils to try and improve the bandwidth although only operating at the CW end of the band means that I do not want full band coverage on 40 and 30. More fun on a wet Thursday afternoon!

(Click to enlarge image,navigate using << >>click on any image to close.)

Contesting update

M5FRA — Tue, 20 Dec 2011 14:32:53 +0000

Over the last few years there have been many debates about contests, the behaviour of contesters and what the ever increasing number of contests is doing to the hobby. Some say it is the only way forward, the TINA principle, there is no alternative (sounds familiar!) Others argue that far from being the saviour of ham radio contests are slowly killing it.

What has become apparent is that many national societies are actively pushing contesting and using it as way to attract newcomers. You could argue that shows desperation is setting in, or that contesters have quietly taken over positions of power, but the biggest danger is that amateur radio comes to be seen as nothing more that a series of contests.

Do not get me wrong, there is nothing wrong with having some contests where contesters behave like civilised human beings and respect that other, non-contest ops, have an equal right to their QSOs. Unfortunately that seems to be forgotten in the frantic rush for points.

Recently I was pleased to hear that our own national society, the RSGB, was using electronic verification of logs to ensure that contesters did not operate outside of band plans. Excellent I thought until it was pointed out that the RSGB has put contest preferred segments into band plans that contain ‘centres of activity’. On 80m the COA does not include the QRP frequency, great, but it does include the QRS COA where it is likely there will be newcomers trying to improve their Morse. The official response is that COAs are fair game, i.e. that contesters do not have to recognise them.

It seems that we have taken two steps forward and one backwards. What is the point of having COAs in the band plan is they do not have to be recognised?

The end result will be yet more ordinary working ops being forced off the band at weekends and being told to either go to the WARC bands or to go and do something else. That is just what many have done and they have not come back.

It really is about time that this problem was sorted out once and for all. The answer is very simple; contest ops adhere to the band plan at all times or are disqualified. No ifs, no buts no excuses. When this happens amateur radio will have a future.

High performance direct conversion receivers

M5FRA — Sun, 11 Dec 2011 09:45:11 +0000

I have always liked using DC, direct conversion receivers. Their sheer simplicity makes them attractive but the drawbacks can be irritating. They often suffer from breakthrough from broadcast, BC, stations and rarely have very good selectivity. I have heard of designs for high quality receivers but have been sceptical of the claims, until now!

The July/August edition of QEX included an article by AAoZZ which described a programmable phase locked loop, PPLL, VFO using the Si570. The VFO was used to drive a Tayloe mixer with several variations of receiver from a simple I Q output for use with PC programs to a full blown RX using an analogue audio board.

The article grabbed my attention because I had used a similar PPLL VFO for the last project. The other interesting part was using an analogue AF section to make a high quality DC receiver – the R2Pro project of a few years back; kits are still available from Kanga US

Yesterday the mixer was built on RF matrix board. I had already ordered the AF board, AF amp and 30m LNA and had a PA0KLT PPLL VFO built from a kit supplied by SDR_Kits. In the article Craig Johnson, AA0ZZ described building his version of the mixer ugly style. It was impossible to obtain DIL chips and all I could get were SOIC versions. Luckily the GQRP Club have some conversion boards from QRPme.

Initial tests involved feeding the I & Q outputs into the PC soundcard. This worked well but tests with headphones were better. The circuit as shown in the photographs will drive a pair of hifi ‘phones well enough for me to be able to copy CW sigs from about 2m away.

The RX performance is amazing, the background is VERY quiet, without an antenna attached there is just no noise. Sensitivity is good and overall this is easily the best DC receiver I have heard. Listening on 30M last night there was no BC breakthrough.

The next thing to build will be the analogue AF section, filters and the AF amp. I want the RX to be dual band, 30m and 20m, so will add an LNA for 20M.

The article also included a very simple TX strip so with some TX/RX switching this could be a useful, small transceiver.

(Click to enlarge image, navigate using << >> click on any image to close.)

SCAF demo files

M5FRA — Sat, 26 Nov 2011 12:17:16 +0000

I have had a few discussions recently about SCA filters, switched capacitive audio filters, and how effective they are. The best way to demonstrate is by listening to recordings made with a variety of receivers.

The first is a Small Wonder Lab Rockmite on 30m. The clicks/clunks are the sound of switching the filter in/out. This is an Idiom Press version.

All files are MP3 format so just click on a link to play.

30M Rock-Mite

Obviously with such a simple RX the difference is very obvious but the recording also shows how easy it is to make using simple rigs easier and pleasant on the ears!

The next recording is of a superhet Rx with a 3Khz wide SSB filter. The first part shows the effect of using the SCAF, it is then switched out. Obvious close in loud signals will have an effect on the RX and the SCAF is no substitute for a good crystal filter in such circumstances. But in many cases the SCAF can do almost as well. I hope this recording shows that.

20M Homebrew superhet

The last recording was taken from a ‘black box’ TX/RX with a reasonable filter. The effect of the SCAF is far less pronounced but it does clean up the signal and cut the noise level making the wanted sig easier to hear. This was the NEScaf version and the change in background noise was caused by rotating the bandwidth control.

Black box demo

The other supplier is the New England QRP Club who sell a kit priced at just $34 post paid to the UK. I have built one of these into the 8 band QRP transceiver.

The usual disclaimer applies, I have no connection with any company mentioned and have not been bribed into making this post!

Homebrew test equipment

M5FRA — Mon, 07 Nov 2011 09:28:32 +0000

Just made an ESR meter from the G0NQE design, details are here. Maybe it will help to avoid more exploding electrolytics! See page here.

Also working on an RF power meter from an article in the June 2001 QST by W7ZOI and W7PUA. More details later.

The DX code of conduct

M5FRA — Tue, 18 Oct 2011 06:24:04 +0000

I will listen, and listen, and then listen again before calling.
I will only call if I can copy the DX station properly.
I will not trust the DX cluster and will be sure of the DX station’s call sign before calling.
I will not interfere with the DX station nor anyone calling and will never tune up on the DX frequency or in the QSX slot.
I will wait for the DX station to end a contact before I call.
I will always send my full call sign.
I will call and then listen for a reasonable interval. I will not call continuously.
I will not transmit when the DX operator calls another call sign, not mine.
I will not transmit when the DX operator queries a call sign not like mine.
I will not transmit when the DX station requests geographic areas other than mine.
When the DX operator calls me, I will not repeat my call sign unless I think he has copied it incorrectly.
I will be thankful if and when I do make a contact.
I will respect my fellow hams and conduct myself so as to earn their respect.

Eight band homebrew CW QRP transceiver

M5FRA — Tue, 20 Sep 2011 14:47:35 +0000

At last the rig is complete, in its box and working. Not saying it is completely finished, there will always be improvements to make but it is working well on all bands. The Rx seems very good, better than I expected. The TX is good with adequate power on all bands. The PA stage is running cool at 5W. It is a real change not have cooling fans running every time I transmit!

More info and pictures later. See ‘Projects’ > ‘ Eight band homebrew CW transceiver’