Hi Brian

Yes, the initial development was motivated by the RSGB-hosted YOTA 2017
summer camp buildathon. It's a funny story. The RSGB sent me an email back
in February, asking for proposals for the buildathon. I saw that they were
planning to have 78 attendees from around IARU region 1, and then I saw
that their entire budget for the kits was 1000 pounds. Which came to around
$16/person at that time. I just laughed and clicked "delete". What can you
do for $16?

But then I remember well, it was a Friday, and my mind just kept coming
back to that deleted email from RSGB. I realised that the email must have
gone out to several kit producers. I speculated that most (if not all) of
those would be answering with a proposal for some shack accessory or some
simple toy, a CW practice oscillator perhaps. I thought that this is about
"Young persons ON THE AIR" so a proper transceiver would be best. I know
there are some very cheap radios around, particularly things like Chinese
Pixie clones on eBay etc. But I am not a fan of these radios, particularly
for newcomers. Even let's not start to discuss the build quality, component
quality and total lack of support. Even if a Pixie is perfect and working
as designed, it is still not great for beginners. The low output power
makes contacts hard and the very poor receiver performance makes them even
harder. A pixie might be a nice challenge for an experienced operator with
a tri-band beam but not a beginner. I know it well, from my own experience
- the Pixie was one of the first radios I ever built, and the performance
was so bad I never had a contact with it, it was just totally discouraging
for a newcomer. I thought - well, what if that $16 could be used to produce
a really clever radio, with top performance, and simple to build - that I
could supply at cost (or small subsidy) to the RSGB YOTA 2017 summercamp
buildathon - then sell the rest of the batch of 500 at a profit (I am a
business, after all). I wrote an email reply to RSGB with my idea.

So that Friday night I cold hardly sleep, my mind kept turning this problem
over and over. By the morning, I had most of the design figured out in my
mind. I would include built-in test and alignment equipment, so that it
could be built by ANYONE even with no test equipment at all: DVM, Frequency
counter, RF power meter, Signal generator - as well as being able to inject
the signal generator into the front end and have the microprocessor assist
with the alignment adjustments. I would use a receiver architecture with
top performance and all my experience with QRP Labs kits to date, to source
good quality components at good prices. The Si5351A synthesiser at the
heart of it, would be used with all three oscillator outputs to achieve
lots of functionality, like the signal generator etc. The microcontroller
in charge of everything, providing many convenience features like dual VFO,
split operation, stored messages and frequencies, and I could even add a CW
decoder using some simple DSP (Goertzel algorithm). I'd use a Class-E PA
design and aim for a full 3-5W output, a decent power with which making
contacts is easy, even without a fancy antenna.

At that stage I did not include the WSPR beacon, that was a later idea! It
is "free" since it requires no additional hardware, only firmware - and I
already had experience of WSPR encoding from my Ultimate3S and its
predecessors. The WSPR beacon was this later bonus feature, that provided
some link to the QRP Labs heritage and userbase. Also a nice introduction
to WPSR and digital modes, for CW operators who have never tried WSPR. And
frankly, an additional business opportunity for me - since a reasonable
proportion of QCX customers also purchase a QLG1 GPS kit
http://qrp-labs.com/qlg1 to go along with it - so adding the WSPR feature
provides an additional sales opportunity.

It was hard to wait until Monday (after the Friday's email), so I could
begin work on the first development version (photo attached). By the end of
the week, I was on air on 40m. Then there were months of work to
productionise this - to make a PCB layout, procure samples of all the
parts, build prototypes, manufacture the kits - and not least of all, write
the 138-page assembly manual! Later I changed the radio to 20m, and then
17m (the RSGB wanted their transceivers to be 17m (since it IS in fact,
2017 hi hi).

The YOTA kits were due for delivery at the end of July and I made it just
in time. Earlier in July I attended Friedrichsafen hamfest (largest in
Europe) where I had a small booth. I had a crowd of people round my table
solidly for all 3 days. On the first day I did not even take a 30 second
toilet break from before 9am to after 6pm. I demonstrated my pre-production
QCX, and took pre-orders. People brought me coffee and coke and chocolate
to sustain me. It was hectic! On the second day, a gentleman from the huge
shiny Kenwood booth opposite came over, before the show opened. He wanted
to know what was going on over here. So I said take a seat (there was only
one seat, normally nobody could use it, but this was pre-open so the crowds
weren't in). I showed him the QCX, and he ordered one! Selling a pre-order
to one of the Kenwood staff was a show highlight :-)

The YOTA buildathon went very well, by all accounts. When the kits went on
general sale on 21st August, the remainder of the first batch of 500 sold
in just over 24 hours - I was NOT expecting that! I immediately started
producing a new batch and taking pre-orders and that situation has
continued ever since, we are still fighting to cut down the waiting list,
get the QCX kits shipped out!

Some order statistics:

QCX is sold for 80, 60, 40, 30, 20 and 17m bands. 40m is most popular,
followed by 20m.
2,553 orders have been taken to date, and I am currently manufacturing the
6th batch of 500 kits.
QCX kits have been shipped to 54 countries
The US accounts for 46% of QCX purchases

There are quite a number of YouTube reviews etc now, see
http://qrp-labs.com/qcx/qcxvideos.html - and constructors are finding many
innovative ways to produce enclosures for their QCX kits:
http://qrp-labs.com/qcx/qcxgallery.html

Peter DL6DSA wrote me a nice email about his experiences of QCX during the
CQ WW DX contest:


*"With some pride I can report that my QCX was on 40m with 5 Watts last
weekend in the CQ WW DX Contest. At the end I have reached 175 QSO with 53
DXCC in 11 zones, 13x DX over 3000km. ODX was PJ2T with 8451km. Everything
only with my W3DZZ in the amount of residential buildings, 18m high.Once
again I would like to say that the RX is very well suited to fight in such
a worldwide "battle", even without AGC, which I no longer miss as much as
at first. It works best if you leave both hands on the buttons for tuning
and volume while searching the band. I have not experienced any overload,
great. At very, very strong stations you can already hear the other
sideband, but the difference is already very high."*

A great many people have now told me that the sensitivity is as good as
their top-end Kenwood, Icom, Elecraft, Yaesu etc; and weak signal handling,
dynamic range etc are also excellent. So, it seems I achieved my goal, I
took on the challenge to produce a radio with great performance, easy to
build, and low cost, all three at the same time.. and it seems I managed
it! The kit has been tremendously popular in just 3 months and I think that
it will probably become legendary!

73 Hans G0UPL
http://qrp-labs.com