[time-nuts] Blackout in Europe and power line frequency jump

Rusty Dekema rdekema at gmail.com
Thu Nov 9 05:52:41 UTC 2006


On 11/6/06, Marco IK1ODO <IK1ODO at spin-it.com> wrote:
>
> Saturday evening part of Europe experienced a blackout, caused, it
> seems, by a single failure in Northern Germany, then propagated up to
> southern Italy.
>
> A friend that routinely monitors the 0-120Hz ULF/ELF frequency
> spectrum captured the power line frequency jump following the fault.
> Hre are the links:
>
>
>
Thanks for sending this. Those graphs were very interesting. I've certainly
never seen a frequency deviation of that magnitude or duration before! (Too
bad I wasn't monitoring in 2003 during the US/East blackout! I was just on
the edge of the grid collapse region at the time; our line voltage went
absolutely insane for 1-2 minutes on one or two occasions, but we never lost
power completely. I bet _that_ would have made for some interesting
waveforms.)

On the other hand, I don't take my own measurements yet, either. What got me
started on power grid monitoring was the US Department of Energy's "Gridwise
Screensaver". This is a screensaver (and a standalone app) which downloads
10 samples per second of power grid frequency monitoring data from a server
measuring wall-plug frequency in some DoE-funded college laboratory in the
western United States.

This system appears to have been designed by an engineering undergraduate
working as an intern or co-op, and although it is not a bad effort at all as
such, it is not what one might call well-put-together. Also, the servers on
which it relies for the frequency data are often down for days or weeks at a
time.

As a silver lining, I did notice that you can get the DoE frequency servers
to spit raw data at you, consisting of 10 frequency values per second. (You
can do this by connecting with 'telnet' or 'netcat' to the monitoring port
on an operable server and simply saving the data to a file.) This did at
least allow me to do some rudimentary analysis in Matlab, although I did not
obtain any particular conclusions, nor did I observe any especially
out-of-the-ordinary events in the short time I was monitoring this.

Your report, though, inspires me to look into more reliable and precise ways
to measure the frequency of one's local grid. I had not actually considered
monitoring electric fields "off the air" before, but it seems like this
might be fairly practical.

>Renato told me that during the frequency jump a faint 50Hz line was
>still present. Possibly some part of the network that had detached
>from the rest, or is the Russian network not interconnected with the
>European Union?

Could be either as far as I know. I don't know the nature of the Russian/EU
interconnect (if one exists), but it is possible that the two networks are
connected but only over high-voltage DC links. This is how, for example, the
US Eastern and US Western interconnect zones are linked, and the
rectifiers/inverters on each end of the links allows for independent
frequencies in the two systems.

On the subject of faint lines, I wonder if that line just below 20Hz is from
16.6Hz railway electrical networks. I could be wrong but I think they use
these in Germany, although I don't know about Italy.

Finally, if I may ask, I am curious why your friend monitors the 0-120Hz
ULF/ELF frequency
spectrum. Is the power grid the primary target of this monitoring? I don't
know what else would be interesting in that range, but maybe there is
something I don't know about.


Rusty



More information about the Time-nuts_lists.febo.com mailing list