Six tornadoes hit the Ottawa-Gatineau region on Friday, the most powerful being an EF-3. As a result power was lost for 180,000 homes in the Ottawa region, and even as I write this Tuesday evening Hydro Ottawa reports there are still homes out of power. One of the major problems was the impact of damage to the Merivale transformer station, one of two major stations taking power from the provincial grid, and connecting it to the city grid.

Anyhow. That was a long intro to a short story about our kitchen clocks.

So we got our power back early Saturday (an outage of only about 14 hours). Andrea set the clocks in the kitchen to her cellphone at that point. This afternoon, I was making some lunch, and I noticed that weirdly my kitchen clocks (microwave and stove) were both running ten minutes slow.

I think I have a hypothesis as to why. I have no evidence for this! I just think it’s a cool hypothesis.

It’s based on some stuff I learned about a power grid dispute in Europe between Serbia and Kosovo. From what I understand of that situation (and my limited understanding of power grids), demand applied to an AC power grid acts sort of like resistance on a bicycle wheel. So long as you have enough power being applied to the wheel, you can keep it spinning at the same pace, but if you don’t have enough power the wheel slows down.

In the case of a power grid, the ‘speed’ of the wheel is the frequency of the grid (50hz in Europe, and 60hz in North America).

When Serbian power plants were not balancing the demand from Kosovo, this caused the frequency of the entire grid in Europe to drop, enough to cause noticeable skew to clocks that use the grid frequency to keep time.

My thinking is that what was happening in Ottawa over the weekend is that we got power back, but there wasn’t enough power in the grid to maintain a full 60hz AC signal. Here’s some back-of-the-envelope math: If we lost 10 minutes over three days lets say (Saturday noon to Tuesday noon), then that would be be a loss of only 10 x 60 x 60 (36,000) cycles out of the expected 3 x 24 x 60 x 60 x 60 (15,552,000) cycles, a skew of only 0.02%, which I think would be within the allowable range (there are cutoff frequencies where things will be disconnected from the grid to prevent damage).

And that was my random speculation about clocks for the day.

Edit: I think maybe if we could find historical frequency data for Hydro Ottawa, similar to what Swiss Grid provides here, but over the period of time from the storm, we could prove my hypothesis.