Sunday, February 19, 2012

Words of Wisdom from Farhan

I read the mail of the BITX20 group. Here is some good advice from Farhan:

I have often seen builders finishing an entire build, then powering it
up to face the frustration of a dead circuit. I suspect that the
trouble is with our kit building mind set. As a kit builder, we assume
that if it has worked well for a few hundred others, there is no
reason for it to not work for us. But the truth is more sobering ...
Of the hundred odd components, any of them could get swapped by
another, or a bad solder, wrong polarity, etc. can all conspire to
thwart your attempts. The bitx manuals are really some of the best
produced in the recent years and yet, even with leonard's videos,
troubleshooting kits is a challenge.
I am proposing a more elaborate, slower but surer approach to building the bitx.

It is as follows: build it one stage at a time, use one stage to test
the next. For instance, one could start with the bfo first. Just a
single transistor with the crystal. Then use an RF probe to check the
rf output. If there is no output, then sort that out before proceeding
to the next stage. With the addition of the buffer amp, the output
should go up. Then one could proceed to the audio amp. Injecting audio
from your mp3 player or computer could check that it works. Next,
replace the audio source with the mic amp, this tests the mic amp.
Now, if you add the two diode modulator, you should be able to receive
the dsb at 10 MHz on your HF transceiver.

This approach tests each stage individually and in isolation before
proceeding to the next. It also provides wholesome education to the
builder. In software industry, it is called a 'test driven
development' method of developing software.

In the end, this approach is no slower than the current approach,
except that surprises are not kept for the last.

I am sure that some of us can come out with a sequence of stages to
build where each stage is tested using the previous stage.

As much as bitx is about building it cheap, it is also about learning
your radio from inside. Bitx is also education on the cheap, don't
give up that opportunity.

- farhan VU2ESE

Our book: "SolderSmoke -- Global Adventures in Wireless Electronics"http://soldersmoke.com/book.htmOur coffee mugs, T-Shirts, bumper stickers: http://www.cafepress.com/SolderSmokeOur Book Store: http://astore.amazon.com/contracross-20

5 comments:

  1. Chuck Adams, K7QO, has written build-test-build-test instructions for the Small Wonder Labs SW30+ and SW40+. I followed his instructions to build the SW30+, and it was great to confirm each stage was OK before proceeding. BTW, only $60 for a widely respected 2.5W transceiver.

    Chuck has lots of interesting material at his website, k7qo.net.

    72, Al - va3iaw

    ReplyDelete
  2. So does Genesis :-)

    http://www.genesisradio.com.au/

    ReplyDelete
  3. I tend to build like this.

    An HF receiver that works usually tells you it's working - the hardest things to troubleshoot for an "amateur" like me are VHF projects.

    ReplyDelete
  4. Hello Bill, I built a transceiver from Solid State Design. I first built and adjusted the oscillators one at the time. After I completed the receiver, I built the transmitter. I used this 7 MHz TRX for 10 years and had great fun with it. 73, Bert

    ReplyDelete
  5. I agree wholeheartedly with Farhan; build and test stage-by-stage. Too many novice builders have become discouraged by spending hours on end building their first complex radio only to apply power and have it perform like a rock. Or worse yet, go up in smoke (and not the sweet rosin smoke either, HA!)

    This is the voice of experience speaking here.

    73.......Steve Smith WB6TNL

    ReplyDelete