4 ms·

The article doesn't mention a date. Presumably the "Wednesday" referred to is 18 Dec 2019. On a previous occasion when rail equipment failed because of a fluct

by bloak 7y ago

The article doesn't mention a date. Presumably the "Wednesday" referred to is 18 Dec 2019. On a previous occasion when rail equipment failed because of a fluctuation in supply voltage the electricity company was not at fault: the voltage had remained within specified limits. I find it a bit suspicious that the article doesn't mention what the change in voltage was this time, and what variation the specification allows. The penultimate paragraph is perhaps an implicit admission that the electricity supply was within its specification. I wonder if any of this is indirectly related to the official change in supply voltage, from 240 V to 230 V, in 1995. Nah, probably not; they can't have equipment that old, can they?

Nextgrid 7y ago
Did you get the voltages backwards? If the supply voltage was lowered wouldn't it mean that older equipment designed for 240V would have an even bigger safety margin when it comes to over voltage?
makomk 7y ago
It could be related to the official change in supply voltage, though not for the reason you're thinking. Although Britain's mains supply officially changed to 230V, in reality it's the same 240V as before and there's a wider tolerance band for voltages above the nominal 230V than below to make this work. So if Network Rail installed equipment that expected actual continental-style 230V, that could cause problems. It wouldn't be the first time that rail companies not paying attention to actual grid specs caused transport chaos this year. Back in August, a power failure caused load shedding that dropped a million houses off the grid but preserved power to the rail network as intended. However, one train company had bought Siemens trains that couldn't cope and immediately shut down when they saw a dip in frequency of the level required to cause automatic load shedding, and required an engineer visit to restart. That one set of trains effectively rendered all the measures taken to ensure the rail network retained power almost useless, as they turned into immobile blocks all over the network.
- avianlyric 7y ago
  > rail companies not paying attention to actual grid specs I would give the rail companies a little slack. They did specify trains that should have handled the frequency change. But Siemens didn’t build the trains to spec. The shutdown was unexpected and entirely down to Siemens not building their power supplies the specs set out by the rail company.
  - growse 7y ago
    This isn't accurate. Siemens delivered trains that could tolerate the power fluctuations at the time they were specified. The problem was that the national grid then changed (loosened) the specification of what an acceptable fluctuation was. The trains could not tolerate this new specification. What we can blame Siemens for is not providing an easier and quicker way to get the electrical systems back on their feet after a problem, instead relying on a specialist to physically visit a stranded train. I think they said they'd fix that one...
    - makomk 7y ago
      That's not true - the frequency spec hasn't changed in any way relevant to what happened. (I think there was some changes to rate-of-change-of-frequency disconnect specs for generators and possibly medium-term frequency accuracy, but neither of those things mattered here.) This kind of near-simultaneous failure of two large generators, tripping the low frequency demand disconnect, was always specified to cause a dip in frequency of a level those Siemens trains couldn't cope with. There was actually a very similar failure back in 2008. The only reason that didn't cause such chaos is because those trains hadn't been ordered yet.

by bloak 7y ago

Nextgrid 7y ago

Did you get the voltages backwards? If the supply voltage was lowered wouldn't it mean that older equipment designed for 240V would have an even bigger safety margin when it comes to over voltage?

makomk 7y ago

It could be related to the official change in supply voltage, though not for the reason you're thinking. Although Britain's mains supply officially changed to 230V, in reality it's the same 240V as before and there's a wider tolerance band for voltages above the nominal 230V than below to make this work. So if Network Rail installed equipment that expected actual continental-style 230V, that could cause problems. It wouldn't be the first time that rail companies not paying attention to actual grid specs caused transport chaos this year. Back in August, a power failure caused load shedding that dropped a million houses off the grid but preserved power to the rail network as intended. However, one train company had bought Siemens trains that couldn't cope and immediately shut down when they saw a dip in frequency of the level required to cause automatic load shedding, and required an engineer visit to restart. That one set of trains effectively rendered all the measures taken to ensure the rail network retained power almost useless, as they turned into immobile blocks all over the network.

avianlyric 7y ago
> rail companies not paying attention to actual grid specs I would give the rail companies a little slack. They did specify trains that should have handled the frequency change. But Siemens didn’t build the trains to spec. The shutdown was unexpected and entirely down to Siemens not building their power supplies the specs set out by the rail company.
- growse 7y ago
  This isn't accurate. Siemens delivered trains that could tolerate the power fluctuations at the time they were specified. The problem was that the national grid then changed (loosened) the specification of what an acceptable fluctuation was. The trains could not tolerate this new specification. What we can blame Siemens for is not providing an easier and quicker way to get the electrical systems back on their feet after a problem, instead relying on a specialist to physically visit a stranded train. I think they said they'd fix that one...
  - makomk 7y ago
    That's not true - the frequency spec hasn't changed in any way relevant to what happened. (I think there was some changes to rate-of-change-of-frequency disconnect specs for generators and possibly medium-term frequency accuracy, but neither of those things mattered here.) This kind of near-simultaneous failure of two large generators, tripping the low frequency demand disconnect, was always specified to cause a dip in frequency of a level those Siemens trains couldn't cope with. There was actually a very similar failure back in 2008. The only reason that didn't cause such chaos is because those trains hadn't been ordered yet.