Article

Where a Smith chart program travels in thirty-three years

I wrote QuickSmith in 1993 for my own bench. I was not trying to publish anything. Over the years it turned up in an ARRL journal, on a CD glued into the back of a Japanese antenna book, and in a steady trickle of email from people I have never met. This is the part of a program's life that never shows up in its source code.

1993: a program for one bench

The Smith chart is older than almost everything else on an RF engineer's desk, and in 1993 most of us still worked it the way Phillip Smith intended: on paper, with a compass and a straight edge, rotating a transparent overlay to add a length of line. It works. It is also slow, and it punishes you for wanting to try one more value.

So I wrote a program to do the arithmetic. Windows 3.1, Visual Basic, a ladder of components down the middle and a chart beside it. Type a value, watch the dot move. That is the whole idea, and it has not changed since.

I gave it away, because that is what you did. There was no store to put it in, no licence to enforce, and no particular reason to think anyone outside my building would want it.

2010: QEX, and Harold Kinley

In July 2010 Harold Kinley, WA4GIB, wrote to say that his article about QuickSmith had just appeared in QEX, the ARRL's experimenter's journal, and did I want a copy of the magazine, because he could not find my address. ARRL's own summary of that issue puts it plainly: he "describes how to solve several impedance matching problems by using Nathan Iyer's free Smith Chart computer program in Using QuickSmith, Part 1." Part 2 ran in the September/October issue.

What strikes me, reading that exchange back, is how ordinary it was. He posted me the magazine. I wrote back to say the article was clearly written, which it was. A reader had already written to me, unprompted, to say he liked it. Harold mentioned that the article had brought in a few orders for his books, and seemed pleased about that in the mild way of someone who had been doing this a long time.

He had. Harold wrote The Radioman's Manual of RF Devices, Principles and Practices and the Standard Radio Communications Manual, which was translated into Spanish and sold in a dozen countries. He was a land-mobile radio technician by trade, and it shows in the writing: every explanation ends up back at something you can measure.

Harold became a silent key in March 2017, at 72. I am glad the last thing I said to him was that his article was well written.

2012: a CD in the back of a Japanese book

In October 2012 an email arrived from Katsumi Ooi, JA5COY, in Japan. He was writing a book on antennas and matching for CQ Publishing, and some years earlier he had put QuickSmith on the CD that came with an earlier one of his, a print run of 5,000 copies. Now he wanted to include it again, with an introduction, in the new book. Could he have permission?

He could, of course, and I said so the same day. No money changed hands in either direction. It had not occurred to me that five thousand copies of a program I wrote on a Windows 3.1 machine were sitting on a shelf on the other side of the Pacific.

His most recent one is パソコンでスッキリ! 電波とアンテナとマッチング, roughly "radio waves, antennas and matching, made clear on a PC". It still ships with a CD of RF software, which tells you something about how durable the idea is: a chart, a ladder, and something to do the arithmetic for you.

The email that would not stop

There is one more thread running through all of this, and it is the reason this site exists.

In that same 2010 message to Harold I mentioned, almost in passing, that I had been getting requests to update QuickSmith for 64-bit Windows 7, and that the hard part was finding the time. That was sixteen years ago. The requests never stopped. Every few months, someone would write: it will not install, it will not run, is there a version that works on my machine.

A Windows program written in 1993 has a slow-motion expiry date built into it. Not because the mathematics goes stale, since the Smith chart of 1939 is the Smith chart of today, but because the ground underneath it keeps moving. 16-bit to 32-bit to 64-bit. Visual Basic runtimes that stop shipping. An installer that a modern Windows will not trust.

So in 2026 I rebuilt the whole thing to run in a browser, which is the one platform that has not thrown anything away in thirty years. There is nothing to install and nothing to go out of date. If you want the engineering detail, that is a separate article: the solver, the chart and the tests, rebuilt without changing how the program is used.

What has not changed

The .sch file format. A design saved in 1993 still opens today, which I admit is partly stubbornness and partly a promise. The worked examples are the same examples with the same numbers. The program is still free, and it is now open source under the MIT licence, so this cannot happen to it again: if I stop, someone else can carry it.

And the thing itself is unchanged. You put a component in the ladder and you watch the dot move toward the middle. Everything since 1993, the QEX articles, the CD in the Japanese book, the rebuild, has been in service of that one moment when the dot lands in the centre and the mismatch is gone.

If you have used QuickSmith in a class, a book or an article, I would like to hear about it: feedback@quicksmith.online. I am still finding out where it went.