Friday, November 20, 2009

The Whys & Hows - Dixie Pixie


The "pixie" is a QRPp transceiver, it is a minimalist rig using the minimum components. It is fun to build and has a variety of mods to enhance its performance. Output power is in the range of 250mw to 500mw depending on the transistors. As you would expect from a rig like this it does have its problems and limitations, but there is lots of info on the web regarding mods and fixes for this rig. It can be made in an evening and can be an ongoing project for anyone interested in building and I am sure it will give anyone interested in home-brew QRP projects hours of tinkering time.

In this blog I will capture the steps needed to build the DIXIE PIXIE 2 via Manhattan Style and we will take the time to understand the various parts of the radio.

The PIXIE 2 It is available on the web as an inexpensive kit, my goal here is different, my goal is walk a newbie through through the process of creating the radio from scratch, and not just stuff components into a pre-fabricated board.

In preparing this course I researched the origins of the pixie dating back the former Soviet Union's Oleg Borodin RV3GM  and his "original" design The Micro-80 and the subsequent UK version FOXX Transceiver by George Burt GM30XX featured in SPRAT Magazine, and many others.

Follow this LINK to read about the origins and over 30 year history of this radio, as chronicled by UK's premier QRP magazine SPRAT.

The radio is an example in brevity, every parts in the design works double duty, truly a QRP classic and worth our study. In my design (click the image above) I selected the most FUNDAMENTAL parts, many designs you can find on the web are ... well let's just say "embellished heavily by what whatever components were handy at the time." -- Since we are starting from scratch and you don't any components, let's use the best we can find from some of the QRP's and Electronics suppliers, no need to start deviating too soon, although experimentation is allowed and encouraged, I will try and follow and lead you down the middle path.

In some ways the history of the PIXIE reads like a dime-store spy novel, in addition to the BEHIND IRON CURTAIN and UK versions there are also INDIAN, SWEDISH, GERMAN and dozens of AMERICAN versions, each the design is redone to suite the needs of new groups of builders adjusting for part availability, etc. For example many of the mid 90's US versions featured RADIO SHACK part numbers, well guess what fast forward a brief dozen years later (now - late 2009) and those parts are no longer available at you your local mall outlet.

So if you are new to electronics, new to HAM radio, or an old hand but always wanted dig deeper into how radio really work, here is your chance, book-mark this spot and come on back about once a week or so. I will add more and more layers until we have a  complete working radio.

72 ki4SGU

Tuesday, November 17, 2009

Design Specification - Dixie Pixie

We are up and running, by now you've created a handful of PADs and your brain is hurting from those IIT Electronic courses (don't worry we wont go through all of them, just the basics of the basics to get you conversant with all of the concepts)


We will be building a 80m PIXIE using a color burst crystal - Why??

-- The crystals are 10x cheaper than other HAM frequency crystals, are available world-wide either new or as scrap in old television sets (see TV Set Salvage as you can see one can practically build the whole PIXIE from one old TV) Also, through some extra ordinary good luck Color Burst Crystals are resonant at 3.579545Mc, which just happens to be in the 80m CW band, and well within TECH license portion of the band, that is right I said TECH.

-- These PIXIEs are not really super-duper DX gear, so making a few local & regional NVIS QSOs with your club mates will be as much we can reasonably expect. -- Later on, you'll have the skills to build another on 30m or 20m QRP rigs for real DX power, but by then you'll understand all of the underlying matters. -- This one, remember is to better understand the HOWs and WHYs of QRP construction and radio design, we will keep simple and cheap.

Afterwards, we will look at changing out the OUTPUT tuned circuits and crystals to explorer to other bands, or even be able to bend the crystal to get more tuning range out it. These radios are notorious for also being wide, a so an active cw and notch filters are already in the drawing board.

A simple 2 CMOS chip keyer, will make the kits more cw paddle friendly, and a PC interface will show us how to integrate our little radios to our computers for QRSS beacon operation. There your little kit with the patience of a saint (at least the patience of a machine) will happily chirp out for months making 250mw contacts with far away lands and you can watch via remote QRSS grabbers.

Go ahead and order a few Color Burst Crystals and hand full of 2n2222 transistors, you'll also need some 10nF & 100nF Caps, an assortment of resistors (330, 470, 1K2, 10K, and 22K and 4M7), and small QTY of 1n34A & 1n751A diodes, you'll need these next week, after the TG holidays.

NVIS and Hope! 72 ki4SGU

Sunday, November 15, 2009

BASIC Electronics - LECTURE2 - Resistors

Again, I say ... If you are new to electronics,  and are going to attempt the DIXIE PIXIE experience in the proper context you will need a good level of understanding of electrical principles.


In this lesson you should learn

  • What is a resistor
  • How to identify them by values and colors
  • how to measure there values with a meter
see WIKI on Resistors
Here is your second assignment.


LECTURE2 - Resistors

NOTE: If you are starting to follow along after 13NOV2009, don't panic. - Simply catch up at your own pace, although I would not recommended more than one lecture/lab session per day, and never skip the Labs. -- To the entire course see "Additional Support & Study Links" in the BLOG right hand side bar, as "Core - Basic Electronics - Lectures by Prof T.S.Natarajan".


-- Blog Entry 0.3 --

Friday, November 13, 2009

Manhattan Pads and the The Origin of the Term "Ugly Construction"



If you have been following along you have now completed reading most of the links on my site, you've watched a few of the Indian Institute of Technology videos and are now ready to try and actually build something. Albeit, not an actual radio yet, but you are inching to get some super glue stuck between your fingers.

Manhattan Pads - Are not New York City Apartments -- According to the QRP & Amateur and Short Wave Radio Electronics Experimenter's Web Site  

"A reference for Ugly Constructing is The 'Ugly Weekender' by Roger Hayward, KA7EXM and Wes Hayward, W7ZOI published in the August 1981 issue of QST. In fact, it was Wes and Roger who coined the term "Ugly Construction" when preparing this QST article.   
 [The article is available online to ARRL members - LINK]

A big part of the learning of the QST article was Ugly Construction! The term and the bread boarding technique emphasized the fact that there is no correlation between the "prettiness" of a construction project and the way it works. According to Wes, the goal had a couple of corollaries. 'First, people like myself who do NOT have the knack for doing pretty projects can still build successful radios. Second, is that we all need to look at our projects after the fact to discover what it is that really makes them work well. In the case of the Ugly Weekender, the thing that makes it fly is that there is a wonderful ground plane with that PC board material'. Indeed, this transmitter functions very well, especially after you temperature compensate the VFO. Wes also built versions for the 30 and 80 meter bands. This transmitter is a true QRP classic; both as a HAM radio transmitter and because it really promoted "ugly" ground-plane or dead bug bread boarding techniques to the scratch home-brew community.

OK, Hey but that is UGLY and not Manhattan, Where does that come from? Again quoting from the same QRP site
"Ugly Construction has its variations and the Manhattan Style Construction technique has become very popular. Manhattan or "paddy board" style uses small square or round pads cut or stamped from PC board which are glued copper side up on a large copper clad board which is also placed copper side up. The small pads or "islands"  are used for anchoring non-grounded components. Components soldered to the pads such as transistors or ICs are not positioned upside down like in standard Ugly Construction. Many Manhattan style builders use IC sockets as well. There is a tendency to make these projects look very nice and this is a wonderful ground-plane bread boarding technique for builders who have the time and ability to do this.

UGLY and Manhattan construction techniques benefits (also Manhattan can be beautiful )

  • A huge time and money saver for penny conscious QRPer
  • Experimenter can have total control over the design
  • Rapid and flexible bread boarding technique
  • A superior ground plane
  • No custom boards to order from QST or FAR circuits
  • No messy or dangerous chemicals
  • allows for easy re-mods and experimentation
You can see a lot of this all goes back quite a long ways and has many benefits. So now you've heard people taking about it, and you are ready to learn the details about it. First be sure and read the fantastic PRIMER "Manhattan Building Techniques by Chuck Adams, k7QO"

other sources for similar PRIMERS (also excellent, but related to only one radio project and therefore less general in scope) see;
  • Jason /nt7S Intro Paper for the Willamette Project -- PDF (nice overview of bench-tools & techniques)
  • Jim Kortge /k8IQY Notes for the IOWA-10  -- PDF  (nice organization & example of more complex project)


All done reading, Good! Now we are ready to try it ourselves.

Items required: 
  • HF Metal Punch (see text)
  • Super glue
  • some copper clad board stock (or scraps)
  • 25w-35w soldering Iron (See OSE in my Vendor List)
  • Optional - Digital Multi-Meter

Click on thumbnail pictures to make them larger


In order to make the copper pads the recommended way is to have a Deep Throat Metal Hand Punch available at Harbor Freight (ITEM 91510-2VGA). If you don't have one, I'm sure you can borrow one from one of the other build members -- As for non US builders, please email me concerning a suitable substitution in your country --

 

The various punch sets look like this, one can clearly see it just to the left of the calipers. The punch set is secured to the punch vise via a small set screw. The bigger of the fits under the punch, and the smaller on top (it makes the hole).  My Punch Kit cames with about 8 punch sets of varying sises, the second to smallest works best to make circuits pads.




Here one can see me busy punching away at the copper stock and making the actual pads. I have the larger punch, so the leverage is quite easy on the wrist, the smaller one available at harbor Freight requires a tad more muscle. For all my efforts, I will be rewarded with an easy to build and modify component to build QRP radios with. In my case, as you can see the punch is mounted on a small folding table. Works well enough and was surprisingly stable.



 And here is a close up photo the pad about to be made from the copper clad stock. The punched pads can be seen collecting at the bottom.









I would recommend this simple do-nothing circuit experiment to everyone. Take a few of your freshly prepared pads and play and experiment with the best ways to super-glue (cyanoacrylates)onto a copper board.


This will take the sting out of doing the same later on your PIXIE board, no gluing you fingers to your radio on the first try.




Hey, but wait a darn sprekenzie die Minuten here, we've just glue a copper board to a copper board??! Wont that make a short circuit?? -- Well actually no!  Notice the resistance of my fingers is 27 Mega-Ohms, and not infinity (your results may vary, please don't email me about how to calibrate yourselves to 27MO).





But when I measure from the top of the pad to the ground plain, the resistance is essentially infinity (at least far as my meter is concerned)


My meter showing the measurement, and proving the soundness of the technique to easily, cheaply and electrically safely combine boards, pads and components harmoniously.




Here one can see the soldered component on the two Manhattan pads, once you have mastered this simple skill, one will be able to build anything one can dream of. Therefore, building more and more complex circuits requires one only to repeat these same steps over and over.


If you have completed this step, please drop me an email ki4SGU@arrl.net.

Next time we will build an Audio Amplifier.

Thursday, November 12, 2009

I hear and I forget; I see and I remember; I do and I understand.

About this proverb...  I hear and I forget; I see and I remember; I do and I understand. — The moral of this proverb, or a meaning could be You can only understand something by trying it yourself. Also said as… Telling me and I will forget. Showing me and I will remember. Involving me and I will understand. Also said as… You can’t understand until you walk a mile in someone else’s shoes. (This is a rough translation from a Chinese proverb)

If you are new to electronics,  and are going to attempt the DIXIE PIXIE experience in the proper context you will need a good level of understanding of electrical principles.


To achieve this end, I will share with this YOUTUBE channel which provides technical lectures from the Indian Institutes of Technology (IIT), Bangalore. -- We will use and review // Basic Electronics - Prof T. S. Natarajan. I will follow along and I be prepared to answer any questions which may arise, via email or the blog.

Each lecture consist of a hour long session where the first half is a mild lecture (with some math) and the second half is a lab. Where one is expected to follow along and build the apparatus and perform the measurements using real test equipment. (This will show what to buy, albeit you can do with less)

You will find that you can fake it, and skip this valuable step, but sooner or later when we are actually building, using and modifying your PIXIE, ie "the fun stuff", we will leave you behind in the dust because you will missing the essential underlying concepts. So my advise is settle down and plow through these free internet courses, you will be a stronger HAM for it.

Although the course is produced in India; The instructor's English is quite understandable to the American ear, and besides if you were to pay for the course at local junior college here in Atlanta, you would almost surly be assigned a foreign junior professor with a worst accent.


LECTURE1 - Introduction

NOTE: If you are starting to follow along after 12NOV2009, don't panic. - Simply catch up at your own pace, although I would not recommended more than one lecture/lab session per day, and never skip the Labs.

WARNING: The music is only for the first few seconds, it will pass - sorry that part is not so good.


-- Blog Entry 0.2 --

Wednesday, November 11, 2009

The role of the BLOG, and the pace of the project.

I got few panicked emails overnight concerning the DIXIE/PIXIE project, and a few kudos. Thanks for the later, and please allow me to try and explain the former. In the emails many folks pointed out that I had left out a lot of crucial details, this is true and it is by design.

Perhaps I did not make myself clear, the plan for the blog will be the fountain for my FUTURE updates, detailing the construction of the radio. So by definition a work in progress, a living document, the website you see today is but the start. As well as taking our time to understand the circuit as we build it. Therefore, there will be THEORY OF OPERATION sections -- IMHO many ham websites are rich in style and construction tips, but assume the builder understands the underlying electrical principals. -- I will also use the this forum to introduce main stream QRP ideas like QRSS, and more obscure issues like proper care and feeding of the LM386 OPAMP - which likes to squeal unless properly padded.

Of course not knowing how this will all end, I do not have a complete outline. However, I have already been working on the some of the first initial steps like how to make Manhattan Pads, a small power supply for experiments, and possibly a stand alone audio amp or oscillator circuit to get us warmed up.

The BLOG will perform double duty as a WIKI, as any one not reading this in the fall of 2009 should read the BLOG backwards, and each entry will numbered x.y, where x is major and y is minor.

As for the pace, everyone is welcomed to go at their own pace. As for me, I am planning on building three (3) PIXIEs between now and the first expected public & group build  (GB) at the STS in March 2010. The first 2 for experimentation, and the 3rd for pictures, so it will be prettier. - The blog assumes the follower will not be able to attend a structured group build in person and labors alone, therefore I will strive to be thorough. I will create a separate Power Point presentation for those few times when I present this concept in person. After the initial STS/GB I will begin to focus on mods like a 30m version for real worldwide QRSS reach, and a 2 chip CMOS keyer, and side-tone. This might seem slow, but I want to enjoy the journey, for every one step on the blog there may be hours of setup, ie construction, test, picture, measure, document re-peat process, etc. -- For those in a hurry I would suggest one of the other fine PIXIE sources list on the blog as "Other People's Pixie".  I will also make mistakes, to which you will all have a front row seat to, therefore there may be many re-dos and start overs, but real life is messy.

So sit back, grab a cup of hot green tea and join me every few days throughout the winter.

-- Blog Entry 0.1 --