Power Outage 30th Jan 2016: Severe weather batters Sydney

The words global warming and climate change are often said too quickly when it comes to the short term phenomena known as weather. The past fortnight, however, has been one of extremes in Sydney.

20160127_071136

Many days start off nice and sunny, only to see the sky darken in waves, as severe thunderstorms roll through the city, leaving paths of destruction in their wake. Whole weeks are condemned to an unpredictable schedule of hot summer sun and raging thunderstorms. In my short memory, I do not recall severe weather quite like this.

It was 14th January when severe weather hit the west, leaving suburbs near Quakers Hill needing much assistance from the SES to patch broken roofs where tiles were torn right off and trees were downed across roads and property, with clean-up taking several days. Later, on 25th January, more severe weather hit Campbelltown and 29th hit Blue Mountains, Emu Plains and Blacktown area leaving many without power. This month has truly been a spectacular month when it comes to severe weather.

messageImage_1454131945850

Image from Bureau of Meteorology IDN65156 Detailed Severe Thunderstorm Warning

That is, until yesterday, 30th January when three severe thunderstorms were battering Sydney from the north and west, sweeping through a majority of the suburbs. Rarely have I seen such a blanket coverage of the metropolitan area with storms, with one marked very dangerous. I was well prepared, having been warned by the Bureau of Meteorology and watching the rain radars quite carefully. Just half an hour earlier, I ensured the washing was all collected from the line outside.

When it came, it was truly impressive. Daylight turned to near night, as the darkness crept in, punctuated by very bright flashes of lightning. The crackling of thunder was truly startling, with the flashing caught in the peripheral vision no matter where you were looking. Around our area, the strikes were fairly close – and after three short dips in the power supply, the power went out entirely. Sigh.

20160130_170203

At first, in awe of the thunderstorm, I didn’t think so much about it. Of course, all of my servers were down and my PhD data recording experiment was in trouble, but I put that in the back of my mind. In fact, I was captivated by the rain that came right after, which, because of the summer sun and thinning sky, turned the air into a bright white fog.

20160130-1711-2314

The rain was so surprisingly strong and sideways that water did end up leaking down a wall. A close inspection in the roof space showed that the water was coming through a few nails put into the gyprock sheeting. Aside from rain, small hail was also experienced.

Eventually, sunlight returned and even some blue sky. But not the power. The back-up Telstra 3G Wi-Fi modem came out from hiding, along with power banks, LED torches and my Keysight U1461A meter. This proved to be another chance to log the voltage on the line just out of curiosity.

But that wasn’t all. A shift of winds bought a second severe storm cell our way merely an hour after the first one, ensuring that any help was not going to be on its way anytime soon.

A quick check of the Endeavour Energy Outage Management System didn’t give us good news:

Screenshot_2016-01-30-17-44-16

At first, the time to restoration was a little earlier, but then it quickly revised to 1am the following morning. Given that it was about 5:44pm at the screen-shot above, it would have been a long time without power. It wasn’t that long ago we lost power – and just our segment too. Just a block either side of our road were all enjoying the sweet goodness of electricity.

Such long disruption is understandable – unfavourable work conditions with limited crews and unusually high demands at odd hours with possibly complicated logistical needs to get parts to where they are needed. But regardless, it is still inconvenient. That night, we went out to dinner and when I returned, the power was still out and the fridge was starting to get a little borderline …

20160130_233021

… life by LED torch was starting to look like a scary movie. There was an eerie silence in the house – not a single hard drive or fan was spinning, and my ears started to tingle at the quietness. In the end, the power did return earlier than predicted, at just past 11:30pm at night. I spent the next few hours bringing everything back online, recovering from data corruption, etc.

storm-outage-voltage-profile

In all, the interruption lasted over six and a half hours, likely close to seven hours as I didn’t start data logging immediately after the power was lost.

Unlike the previous outage, this one seems to be a hard outage with virtually no voltage across the lines. The initial voltage profile as the storm was slowly heading to sea produced very interesting spikes of low magnitude which disappeared as the storm moved away. The spikes themselves generally had short duration, and because of the sampling rate of the data logging software, many short spikes may have been missed. Of the ones that were captured, it seems that it was able to somehow correlate with storm activity at least to some degree.

An hour later, the second storm also caused some spikes to appear, and the baseline voltage seemed to elevate ever-so-slightly. I wonder if that’s a result of lightning strike induction into the mains line acting like an antenna. The storms did pass, and the voltage seemed to steady below 5mV.

Towards the later night, it seemed that the decabit/K12 off-peak ripple signalling was leaking into the line at very low amplitude by some method of inductive coupling possibly. We can tell these are the decabit signalling as it happens at hourly or half-hourly intervals, at xx:25 and xx:50. According to my previous work, 21:55 was the first transmission slot after unscheduled street-lighting commands which were normally around 19:xx. That might explain one of the spikes near 19:45 which would not be storm related.

There was some elevation in the baseline voltage towards the end even though there was no storms at that time, which might have suggested crew were working with equipment at the time, or something else was happening.

In the end, I set up the experiment not knowing whether the data would show anything of interest. Ultimately, it seems to show that lightning strikes inductively coupled into the lines can be “received” but at very low amplitude, and mains off-peak ripple signalling can also be received as well.

As much as I am hoping to not have any further surprises when it comes to power outages, the bad weather isn’t over – tomorrow is another good candidate for more severe thunderstorm development. Lets hope it passes without any further impact …

About lui_gough

I’m a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!

This entry was posted in Electronics, Opinion and tagged , . Bookmark the permalink.

Error: Comment is Missing!