5 ms·

Another interesting recent reminder of the interconnectedness of politics and electricity grids happened a couple of years ago, when all of Europe's frequency (

by nippoo 6y ago

Another interesting recent reminder of the interconnectedness of politics and electricity grids happened a couple of years ago, when all of Europe's frequency (and lots of grid-synced clocks) went out of spec because of a Kosovo/Serbian spat: https://www.theverge.com/2018/3/8/17095440/europe-clocks-running-slow-electricity-frequency-kosovo-serbia https://www.theverge.com/2018/3/8/17095440/europe-clocks-run...

rosstex 6y ago
This is so cool! Incoming rabbit hole of electric grid research...
_nalply 6y ago
Interesting. I don't understand everything so I have a question to the experts here: It seems that a small part of Kosovo stopped paying for electricity, so the electricity producers stopped producing for the parts. However this didn't work out, electricity still flowed to the nonpayants, because it seems that nobody cut the wires. This created an inbalance distorting the grid frequency, such that clocks in Europe ran a bit late (6 minutes). Sort of an Europe-wide soft brown-out. It's not told in the article, but I think the voltage was also a bit lower than it should be. Did I understand this correctly?
- probablypower 6y ago
  It is pretty much right. Neither country took responsibility for the demand (X), and so the market wasn't aware of the demand, and therefore the generation provided by the market was lacking X. When there is more energy being taken out of the grid than being provided, generators start to slow down. That is, the power is extracted from their rotational energy (inertia), meaning their rotational speed slows down (frequency decreases). There are mechanisms in place that fix any frequency deviations, but the assumption is that deviations in frequency are symmetrically distributed around 50 Hz (or 60 Hz for you americans). When you have consistently skewed errors (i.e. frequency always being pulled down by Kosovo), you end up with a 'time error' [0]. The time error was noticed fairly quickly (frequency and analogs to 'time error' are monitored in control rooms), but the issue was that no-one took responsibility for it and because it became a political issue it was difficult for other grid operators to step in and fix it. I'm not sure about voltage being lower in this case. Voltage is a very local phenomena, and it would be very unusual for control rooms to allow voltages to be 'lower than they should be', especially over a considerable time period. To do so they would be in violation of their grid codes. [0] https://www.nerc.com/docs/oc/rs/NERC%20Balancing%20and%20Frequency%20Control%20040520111.pdf https://www.nerc.com/docs/oc/rs/NERC%20Balancing%20and%20Fre...

by nippoo 6y ago

rosstex 6y ago

This is so cool! Incoming rabbit hole of electric grid research...

_nalply 6y ago

Interesting. I don't understand everything so I have a question to the experts here: It seems that a small part of Kosovo stopped paying for electricity, so the electricity producers stopped producing for the parts. However this didn't work out, electricity still flowed to the nonpayants, because it seems that nobody cut the wires. This created an inbalance distorting the grid frequency, such that clocks in Europe ran a bit late (6 minutes). Sort of an Europe-wide soft brown-out. It's not told in the article, but I think the voltage was also a bit lower than it should be. Did I understand this correctly?

probablypower 6y ago
It is pretty much right. Neither country took responsibility for the demand (X), and so the market wasn't aware of the demand, and therefore the generation provided by the market was lacking X. When there is more energy being taken out of the grid than being provided, generators start to slow down. That is, the power is extracted from their rotational energy (inertia), meaning their rotational speed slows down (frequency decreases). There are mechanisms in place that fix any frequency deviations, but the assumption is that deviations in frequency are symmetrically distributed around 50 Hz (or 60 Hz for you americans). When you have consistently skewed errors (i.e. frequency always being pulled down by Kosovo), you end up with a 'time error' [0]. The time error was noticed fairly quickly (frequency and analogs to 'time error' are monitored in control rooms), but the issue was that no-one took responsibility for it and because it became a political issue it was difficult for other grid operators to step in and fix it. I'm not sure about voltage being lower in this case. Voltage is a very local phenomena, and it would be very unusual for control rooms to allow voltages to be 'lower than they should be', especially over a considerable time period. To do so they would be in violation of their grid codes. [0] https://www.nerc.com/docs/oc/rs/NERC%20Balancing%20and%20Frequency%20Control%20040520111.pdf https://www.nerc.com/docs/oc/rs/NERC%20Balancing%20and%20Fre...