Right up front, nothing in this post exposes sources or methods or any other secret-squirrel shit that might get me in trouble with the Canadian Navy.

I hope.

I spent a work term at the Canadian Naval Engineering Unit in Dartmouth ,Nova Scotia. Memory is a little bit fuzzy, but I think it might have been the summer of 1980 or 1981. It was an opportunity that allowed me some small level of autonomy when it came to delivering what I was asked to deliver.

One of the many projects (and one of the most rewarding) was to deliver a level of automation to their sonar self-noise trials. There’s an interesting instructional manual on self noise, on how to identify it and mitigate for it here. Essentially, the ship’s sonar is used to identify many things external to the ship, but in order to be effective, it can’t contribute to what it’s listening to. That ideal rarely exists, so a second best is to know what noise the ship makes when nothing else is present, and then remove those signals when actively using sonar for real.

The self-noise trial, prior to my project, involved two or more days off the coast, after steaming to deep water, where various parts of the ship were activated, one at a time, then recorded on a sonar sweep. This was used as a baseline and a reference when sweeping later. It’s also used to identify noise on the ship so it can be corrected. The old way was to create a radial chart off each beam of the ship that “identified” noise sources. It was very time consuming; the collection and analysis was manual.

I wrote a program on a sort of HP computer. Remember, this was in 1980-ish, so state of the art was pretty archaic. It interfaced with an HP spectrum analyser connected to the sonar system. It was a really basic collection program (written in Basic) that logged date, time and sweep scenario (which elements of the ship were off or on) and digitally stored the sweep.

All of this sounds really boring, and up to the test phase, it was.

Then one morning I was told to bring a bag the next day — I was going to sea to prove the program.

We left the dock in the late afternoon and I was shown an upper buck to sleep in for the overnight steam. Naive me thought the trip was solely for the self-noise test. Just for me.


Also on the schedule was some Bofors 57mm gun tests. If you don’t know (click the link) it’s a big fucking gun parked on the deck of the frigate. Now I realise there was absolutely no need to include me in the discussions of when and why these tests would be happening, but it would have been nice. I was asleep. Into deep REM sleep, I recall, when the first shot went off. I mentioned I was on the top buck, right? Smashed my head into the “ceiling” above the bunk. Naive me didn’t realise that more than a single shot would be required, either. When my heart rate finally dipped below three digits and I was trying to relax into some sort of hellish sleep, it went off again. Every minute for ten minutes.

Then it missed a minute. I’d adapted to the frequency by then, so the missed minute told my 20 year old brain they were finished and I could attempt sleep again.

They weren’t finished. They were on a break.

Another head smash and a dozen or more firings later and I slowly, tentatively slipped back to sleep. I got about four hours that night.

Breakfast was good, I guess. I don’t remember it, but naval food is generally awesome. Then I went with the sonar techs to the sonar room to set up. I remember being a little nervous, since this was the proof of concept, in the wild, of something I’d basically just been playing with on my desk, but there was also excitement.

Because the ship’s captain wanted to run the trial as quickly as possible, we were setting up in the sonar room while we steamed to our destination. The sonar room is at the very bow of the ship. It’s also below the waterline. We were crashing through pretty heavy seas and I was  in the part of the ship that took the brunt of the wave action. There was no horizon to use as a reference (below decks, below the water line) and the deck dropped out from under my feet with nauseating regularity.

I got seasick, the one and only time in my life. The CPO (I wish I could remember his name) got me to the head in time and I left the breakfast and probably some of the prior night’s meal in the toilet. “No wonder you didn’t feel good,” he said. “You had a stomach full of puke.”

The test was a success. I was told it reduced the trial from three days (plus steaming to and from the test location) to an afternoon. Smarter brains than mine took the digital traces and performed more detailed analysis and helped make the Canadian navy a very quiet navy.