40m Pixie now on 7.030MHz

Thanks to Andy Cutland, my 40m Pixie is now on 7.030MHz, the QRP frequency. Although I have not seriously tried for other than local QSOs, I shall do so. When fitter, my intention is to switch between two crystal frequencies to increase my chances of QSOs. As the TX and RX netting is perfect and BC breakthrough quite reasonable, sensitivity not bad,  QSOs at decent ranges are entirely possible.

40m Pixie

I must have another go with this rig. As I mentioned before, a 7.030MHz crystal would be more useful than the 7.023MHz crystal it came with. The rig definitely works and the RX nets perfectly to the TX frequency. Sensitivity is better than I expected and BC breakthrough far less of an issue than I had expected. It is certainly a good little kit for the price. I need to check the Reverse Beacon Network in case I get out OK but don't hear any replies to my CQs.

40m Pixie

This afternoon, I'll be giving my 40m Pixie another outing on 7.023MHz. I'll take a look on the Reverse Beacon Network to see if I was noticed if I fail to make any 2-way QSOs.

I really need to check my junk box for 7.030MHz crystals as this would be a more useful QRP frequency. When fitter, I need to put it in a box!

UPDATE 1956z:  In the end I did not try the 40m Pixie today.

40m Pixie measurements

Pixie harmonic measurements

This afternoon I visited G6ALB to do some more accurate measurements of my completed 40m Pixie kit. Power out with the 12V battery was 400mW and with a fresh PP3 9V battery 200mW. On the 12V battery TX current is around 129mA and at 9V around 83mA. 2nd harmonic rejection was about 20dB. Even the 9V output is a useful level. Maybe the cased unit should include an internal 9V battery pack which is switched off when an external 12V pack is available?

Pixie on the bench

To be more useful, I should add a 7.030MHz crystal and have a switch to select this or the original 7.023MHz one. 7.030MHz is the QRP frequency. My experience is this can be quite busy.

I should also case the unit!  Drilling the case could be a challenge in my current state of health! Just doing the building and testing was exhausting.

First 40m Pixie QSO

At 1000z this morning I exchanged RST 579 reports with G6ALB who is 3km from me on my 40m Pixie on 7.023MHz CW. This was my first on-air QSO. Netting was perfect and I used the rig directly into my low Par triband antenna. An ATU might have helped. Andrew G6ALB said the channel was pretty busy but that I was a good RST579 with no hint of chirp. This afternoon we'll put the rig on his spectrum analyser. I gave Andrew 579 probably reflecting the poorer MDS of the Pixie. I have no idea of his power or antenna.

My initial goals have been met: I built the kit and have managed a QSO on the air with it. With the fatigue associated with my brain bleed still very apparent (I am well and truly shattered currently), I was well pleased. A few more QSOs with the 40m Pixie would be good, HI.

$10 well spent. Excellent little kit. FB little transceiver. Works surprisingly well.

40m Pixie - bench test went well

Well, it took a super-human effort on my part as a result of my extreme fatigue (brain bleed issue) but today I bench tested the little 40m Pixie kit I managed to build earlier this week. It is hard to explain just how exhausting this simple task was for me currently. I now need to recover my energy!

On receive a 1uV signal was perfectly readable and I could detect carrier on/off down to at least below 0.5uV. On TX it was putting out about 250mW into a 50 ohm load. On my antenna it was suffering some AM breakthrough, but less than I was expecting. I was using standard Walkman stereo headphones and a rechargeable 12V sealed lead acid battery. On a small 9V battery I'd expect about 100mW RF out, although this was not tried.

Overall, these results exceeded my expectations: it is more sensitive than I was expecting and the breakthrough of AM broadcasters is certainly OK in the day time. This is of course with no casing at all and no ATU. Using the fitted pot and fitted 1N4001 as a varicap it is easy to get the TX frequency and RX frequency to correspond, often an issue with very simple QRP rigs.

Next stage will be some on-air tests, but it most definitely works. Working the locals should certainly be possible. Sked with G6ALB (3km) arranged for 1000z Sunday on 7.023MHz CW.

UPDATE 2120z:  There is a contest in full swing and several Russians and Asiatic Russians copied on the little 40m Pixie. I did not stay on long but there was no apparent BC interference noticeable. So, on a quiet night I should able to use the rig in the evenings. The main issue on RX seems to be the lack of AF filtering, so one has to use "the filter that is between the ears" i.e.the brain.

40m Pixie transceiver

Yesterday I found my rechargeable 12V battery and charged it up. I also checked the Pixie I built as a kit for dry joints and shorts. As all looks good, I hope to test this on the air over the weekend or early next week.

There are a couple of locals I'd like to test with initially, although I am expecting to suffer from AM breakthrough mainly from broadcast stations very close to our 40m band. Probably there would be less breakthrough in the daytime? My antenna is short and low on 40m (a Par triband end-fed covering 10m/20m/40m) so any contacts will be good!

I can test it first on a signal generator and then a power meter. If it works, I'd expect a sensitivity of around -100dBm and a power out of around 200mW on 40m. We'll see. Even -80dBm and 100mW would be enough to work locals as long as AM breakthrough is at a manageable level.

Pixie (40m) build completed this evening

Completed 40m Pixie assembly

This evening, I completed the build of the 40m Pixie PCB assembly. It will get checked over for dry joints or shorts in the morning and air-tested shortly.

A very half-hearted attempt at the 6m UKAC this evening was aborted because of my poor voice but I only worked 3 locals in the short time I was on with my V2000 vertical.

I really wanted to crack on with the Pixie build, although I was not expecting to finish the built today.  The small PCB based kit does represent very good value.

Pixie (40m) progress

40m Pixie kit

Pixie kit - being built

After carefully unpacking this kit today I checked all the parts were there and measured all values for correctness. The kit was very well packed and everything was there. Even at the price I paid ($10 with shipping by air included) it represents very good value.

The kit comes with a nice silk screened and well labelled PCB, crystal, all parts, schematic and a parts list. It would be hard to buy the bits for the price I paid. G1KQH paid even less!

I now realise that for close-up work I need better glasses!  Fortunately I have a small assembly aid that includes a magnifying glass. I am taking the build slowly and it is quite exhausting for me.  However, the actual assembly is going well. Last year, for me, this would have been impossible. I can still solder - big steps, HI.

Although the Pixie will be tested on the air, the important personal test is the building. It proves that, with some effort on my part, the world of experimental amateur radio is still there for me. You have no idea what this small step means to me.

Before too long I want to get out in the countryside and try again with my VLF and optical tests. It has been a long long long time.

Testing time - 40m Pixie

Tomorrow I am going to try to build my little 40m Pixie kit that I bought for just over £7 from China. This build is an attempt to see if I am "up to the job". I know the limited performance the Pixie is capable of and on 40m BC breakthrough may be a major issue here in the UK.

I really really miss doing real building and truly experimental amateur radio. This is a step on a long road. My health has been so poor in the last 16 months or so that there was no way I could have attempted this before. People have been very kind and offered me building help and help with antenna erection. These offers have come from all over the world - the world is filled with good and very kind people. There are far more good and kind people than evil ones.

By normal standards a little Pixie build would be trivial. For me it is a major step and test of my current abilities. I do so hope I can manage this! Please don't laugh - to me it is a major step.

Pixie kit from China

My low cost 40m Pixie kit - although G1KQH only paid about 60% of what I paid - is on its way from China but has not yet arrived here. I assume, with free shipping, this comes by sea. I am in no great hurry. As I mentioned before, I very much hope to build this unit as I have not built much in ages because of my brain bleed.

UPDATE 1600z:   My kit arrived in the post today.  Still to open the package, but judging by the speed this must have come by air. It looks well packed.

40m Pixie from AliExpress

From my bank statement I see my 40m Pixie kit from China, that has shipped, cost a little over £7. This includes all parts including crystal and PCB. At this price it would be hard to buy the individual parts. I very much want to build this as it will be the first amateur radio construction here in a very long time.   It is sufficiently simple and low cost that I should manage it, even if I am a little clumsy still with the soldering iron.

 No doubt Andrew G6ALB (3km) will help with a first test.

Another Pixie

Yet another variation on the Pixie arrived by email today from Sverre Holm in Norway who has LM386 mods to reduce BCI, provide mute and add sidetone.

See http://la3za.blogspot.no/2003/04/using-pin-7-of-lm386-to-reduce-bci-and.html .

This looks a very useful mod to this simplest of circuits. Sverre also added a much wider tuning range that helps to get contacts. Low power does not seem to be the main handicap.

Polyakov Pixie

Peter DL3PB has sent details of his version of the Pixie transceiver. He claims no broadcast breakthrough by using this configuration and lots of QSOs.

Peter DL2PB's Polyakov Pixie

Pixie and Micro 80

These little QRP transceivers first made an appearance some 20 odd years ago. They are extremely simple transceivers that use the TX PA as the RX mixer to save parts.

I built a Micro 80 some years ago and my best QSO was around 300km.  Biggest issue (for me) was broadcast breakthrough from strong broadcasters just above the 80m band. Variations included the Pixie 2 which added some refinements at the expense of more parts. Kits are available, but the circuits are so simple it is not work paying over the odds for these. The Micro 80 uses all discrete components whereas the Pixie and Pixie 2 use an LM386 for the RX audio.

If you build one, be prepared to fight for contacts. It is not the TX power that is the problem: it is the receiver that is the limitation.Given good conditions and little broadcast breakthrough, these rigs work.

See https://sites.google.com/site/g3xbmqrp3/hf/pixie .

The same basic schematic will work on any HF band with changes to the output filter. Watch out for chirp on the higher HF bands: don't be tempted to try to pull the crystal too much, especially on the higher HF bands.

The Pixie Files

The Pixie and Micro80 are  well known simple transceivers for any HF bands. I've built both in the past and they do work well considering their lack of complexity. The main limitation is the receiver which uses the TX PA as a simple mixer. For those wanting more information on the Pixie and its derivatives check out The Pixie Files which is a collection of articles from SPRAT.

Simplest possible SDR transceiver?

A challenge - what is the simplest possible SDR transceiver that can be built that would allow credible results on HF QRP? Does anyone know of any links to such ideas? Is there an "SDR equivalent" to the Pixie or similar ultra-simple QRP rigs, possibly supporting all modes but using just a handful of parts?

Micro80 4-transistor transceiver QSO

After some time without use, I resurrected the little Micro80 4 transistor transceiver for 80m last weekend. To increase the power a bit I've now got provision to run it from an external gell cell battery at 12v, so giving me about 2dB more RF out compared with the 9V internal pack. Using it with the bench PSU was impossible because of 50Hz hum pick-up. I also managed to improve the AM breakthrough by around 10dB at least by putting a 4n7 cap across the base-emitter junction of the first stage of the darlington in the RX AF. This makes it a usable radio on 80m at night now as it is not obliterated by the BC stations around 4MHz. Success soon followed with a solid QSO with G3XIZ in Biggleswade around 30 miles away late Sunday afternoon.