Follow @jack_ganssle

The logo for The Embedded Muse

For novel ideas about building embedded systems (both hardware and firmware), join the 27,000+ engineers who subscribe to The Embedded Muse, a free biweekly newsletter. The Muse has no hype and no vendor PR. It takes just a few seconds (just enter your email, which is shared with absolutely no one) to subscribe.

This month we're giving away the Zeroplus Logic Cube logic analyzer that I review later in this issue. This is a top-of-the line model that goes for $2149.



By Jack Ganssle


Summary: The world runs on GPS, but only in recent years.

Every four years the Royal Western Yacht Club of Plymouth, England runs the OSTAR, a single-handed sailboat race from the UK to Newport, Rhode Island. In 1992 I participated with my, at the time, 30 year old wooden 35-footer. And two decades ago this week I abandoned that vessel, climbing onto a German container ship 350 miles from Newport after 29 days alone at sea.

But where, exactly, was Amber II on July 8 of that year? It's hard to say.

The previous year a friend and I sailed her to Newfoundland and thence to Plymouth. At the time GPS was hideously expensive. No one really knew where anything was. Bermuda was reported to be three miles SW of its charted position. Newfoundland's sailing directions warned that the island might be as much as 10 miles away from the location plotted on the chart. We had borrowed a GPS, one with military precision (pre-Clinton commercial GPSes had downgraded accuracy) from special friends in the government, but it took 20 minutes to compute a position. We didn't have enough AAs on board to run the thing very often, so navigated by sextant with only the occasional GPS fix.

1992 saw the collapse of the price of GPS - for $1000 ($1600 today) one could buy a unit which would report positions accurate to a couple of hundred meters. In the week prior to the race all but one of the 67 skippers bought a GPS in Plymouth via a special deal offered to the racers. My finances were at the breaking point, and a GPS was out of the question.

Celestial navigation has been used for centuries to determine position. The idea is simple: measure the angle between the sun (or star, planet or moon) and the horizon, and, if you know the time it's possible to calculate where you are. Think of it this way: At any given instant the heavenly object is directly above one location on Earth. That means the angle you measure puts you on a circle some distance from the object's ground position. Observe two bodies and your location is at the circles' intersection. (The circles generally intersect at two points thousands of miles apart; one presumably at least knows what ocean you're in).

The reality is a bit grittier. An observation error of one minute of arc (one sixtieth of a degree) induces an error of a mile. Yet at 50 degrees north the ocean is always rough, so the navigator is being tossed about violently while getting soaked in cold spray. It's impossible to sight two objects at the same time, so corrections are needed. Get the time wrong by four seconds and another mile of error creeps in. The sextant has errors, as does the clock, and the bodies in the sky don't move in quite the way predicted by the ephemeris tables, so correction after correction has to be applied, and the results carefully plotted carefully despite heavy seas.

Or, today, one merely looks at the screen and records GPS coordinates accurate to a handful of feet. In a single generation the entire history of navigation has been tossed overboard, replaced by cheap electronics with unprecedented accuracy.

In 1992 the Canadian P3 Orion located Amber II by homing in on my distress beacon. They found a ship but could not supply them an accurate vector to Amber II as their navigational tools weren't much better than mine. Today, one wouldn't bother to read off all of the digits of latitude and longitude from the electronic box as the precision far outstrips that required.

(In fact, most GPS users are clueless about the units' precision. An app on my iPhone displays coordinates to one-thousandth of a second of arc, which is about an inch, far exceeding the GPS's capabilities.)

The microprocessor has displaced all of traditional navigation techniques. Many sailors today barely know how to read a chart. One acquaintance navigated from below as his boat approached Virgin Gorda, reading data off the instruments. The crew on deck warned about the rocks clearly visible ahead, but they weren't shown on his display.

They hit the rocks.

Voyager, my current boat, is a 32 foot ketch, which is quite small for ocean sailing. But in that tiny envelope we carry three GPSes. And RADAR. Then there's the AIS which has a RADAR-like screen that displays the position of all ships within 30 miles, using data packets they all transmitted several times a minute on VHF frequencies. The RADAR detector alarms when it picks up a signal. A network connects the GPS to the AIS, the RADAR and the marine radio. A panic button will cause the radio to broadcast the boat's unique identification code and position to any nearby vessel. It's possible to add the autopilot to the network so it will automatically change course as needed, but that seems a silly feature to me.

Voyager is not a high-tech boat and does not even have a chart plotter, but she fairly bristles with electronics that sense ocean temperatures, ship positions and much more. None cost much and all are astonishingly reliable. And all are made possible by microprocessors.

But we also still carry my sextant, which I clutched while scrambling up the ship's cargo net. Today that Tamaya would cost around $2000, ten times what I paid for it 40 years ago, and the price of about twenty GPSes. It's not particularly accurate, by today's standards, but tradition still has value. So every year I take some sights to stay in practice. There's a certain satisfaction in repairing a diesel engine, fixing the rigging, and figuring position without the aid of millions of transistors.

But yesterday, at a West Marine store, my wife lovingly eyed a 12" full-color touch screen chart plotter. I told her we just don't have room for the thing.

She suggested making room by getting rid of the sextant.

(For the complete story of the 1992 OSTAR see

Published May 22, 2012