Electrical Supply Problem(s) 02 May

[Belperpete]

This hasn't been reported on national news I've seen. Was all this simply the dip in grid frequency and does anyone know what caused it? Why is such a modest (tiny?) drop in frequency such a deal breaker for modern trains?

It is often not the frequency variation itself, but the harmonics that it generates. A variation of just half a hertz in the main frequency, will generate significantly greater frequency spreads in the harmonic frequencies. Even if the equipment itself is immune to such frequencies, other equipment may not be, and so it may be required to shut down to avoid generating such harmonics.

For example, signalling, telecom and other control equipment may use Frequency Division Multiplex and similar equipment. It is important that traction doesn't generate frequencies that might interfere with the frequencies such systems use.

This is going to be more of a problem as the huge inertia of the big synchronous turbine generator sets of fossil fuel stations and the current nuclear stations go off line. I think this lack of inertia in the system caused the Iberian black out of a few years ago and makes restarting a grid much harder.

Agreed. It is a recognised problem that much of the "green" forms of energy don't provide the same frequency stability as traditional generator-based supplies. Even worse, they may automatically disconnect from the grid in such situations, to protect themselves, compounding the problem for the grid.

 

[dk1]

It seemed the Stadler units had far more tolerance to the low voltage issues than Alstom fleet on the GEML.

 

[edwin_m]

I can understand that it is impractical to provide equivalent protection to traction motor supplies, but surely the control equipment on rolling stock should be properly protected?

With the issue with the 700s, too much protection was actually the problem. The traction shut down on detecting a frequency deviation that was outside its limits, and couldn't be reset except by a techican attending hte train with a laptop.

It's probably reasonable for traction to shut down on these relatively rare occasions, but there needs to be a way of the driver resetting it, and that may be the case here. I haven't seen any problems reported here with the 700s, and a frequency deviation will affect everything across the grid, so it seems the settings on Aventras may just be more sensitive or harder to reset.

This is going to be more of a problem as the huge inertia of the big synchronous turbine generator sets of fossil fuel stations and the current nuclear stations go off line. I think this lack of inertia in the system caused the Iberian black out of a few years ago and makes restarting a grid much harder.

It is a recognised problem that much of the "green" forms of energy don't provide the same frequency stability as traditional generator-based supplies. Even worse, they may automatically disconnect from the grid in such situations, to protect themselves, compounding the problem for the grid.

More modern "green" systems can simulate the effects of inertia by tweaking the way their inverters feed power into the grid.

 

[Belperpete]

Overall though it wasn't too major of an incident for how large-scale it could've been. Imagine if it was the TL core instead of XR... we would likely still be feeling the delays and cancellations now.

It may not have been the TL core, but reports of significant disruption to TL and SE services due to a failure of a substation in the Deptford area.

== Doublepost prevention - post automatically merged: 3 May 2026 ==

This hasn't been reported on national news I've seen.

Considering the disruption caused, the lack of news reporting, even on local news, is surprising. Perhaps because it was a bank holiday Saturday, late afternoon, there were few journalists to pick up on it. And the news sources preoccupied with other things.

 

[saismee]

It's probably reasonable for traction to shut down on these relatively rare occasions, but there needs to be a way of the driver resetting it, and that may be the case here. I haven't seen any problems reported here with the 700s, and a frequency deviation will affect everything across the grid, so it seems the settings on Aventras may just be more sensitive or harder to reset.

I would say more sensitive, if they were harder to reset then it probably would've lasted much longer.

 

[GraphitProjekt]

and couldn't be reset except by a techican attending hte train with a laptop.

Luckily I don't think this was required for GA or WMT Aventras that were affected. Seemingly the driver could resolve the issue just by resetting the train. As a result, it looks (at least on paper) that GA was able to recover the service pretty effectively compared to EL, something else must have happened to EL.

It seemed the Stadler units had far more tolerance to the low voltage issues than Alstom fleet on the GEML.

Was there actually a low voltage issue in the London area or specifically just a low frequency issue, because my earlier assumption was that Pudding Mill Lane feeder actually went offline but I'm not sure if this is actually true.

== Doublepost prevention - post automatically merged: 3 May 2026 ==

It may not have been the TL core, but reports of significant disruption to TL and SE services due to a failure of a substation in the Deptford area.

== Doublepost prevention - post automatically merged: 3 May 2026 ==

I can see that Thameslink were posting about this and had in fact ceased all services down the Greenwich/North Kent Line, but oddly I can't find much evidence of the issue on Southeastern's twitter.

 

[dk1]

Luckily I don't think this was required for GA or WMT Aventras that were affected. Seemingly the driver could resolve the issue just by resetting the train. As a result, it looks (at least on paper) that GA was able to recover the service pretty effectively compared to EL, something else must have happened to EL.


Was there actually a low voltage issue in the London area or specifically just a low frequency issue, because my earlier assumption was that Pudding Mill Lane feeder actually went offline but I'm not sure if this is actually true.

Not sure mate. Just something I read from one of our controllers. The 745s can tolerate the low voltage but it gives a very jumpy ride. We suffered between Norwich & Diss a couple of years back due to feeder issues. Climbing the bank from Trowse was lively to say the least.

 

[Belperpete]

More modern "green" systems can simulate the effects of inertia by tweaking the way their inverters feed power into the grid.

While the simulated inertia now being built into.modern green sources is a mitigation for what is being lost with the reduction in generator-produced supplies, my understanding is that it is nowhere near equivalent.

.In essence, a drop in frequency on the grid is caused by not enough energy being supplied to meet demand. And a rise in frequency by there being too much supply. Rotating generators can absorb this, as it takes significant power to speed up the mass of a generator to a higher frequency, so the generator absorbs some of the excess power. While the inertia of a generator will itself produce extra power if the grid attempts to slow down, in the same way as it will briefly continue to generate if you remove its power source.

While as you say some green energy systems can now simulate inertia, they are limited by how much energy they are capable of instantaneously absorbing or creating, before their self-protection kicks in and disconnects them. A solar farm, for example, would need to have additional capacity available to instantaneously produce extra power.

There is of course a limit to what a generator can do, brake or accelerate it too hard or for too long and it will fail, perhaps catastrophically. However, my understanding is that they are typically capable of withstanding a lot more before their protection systems kick in, compared to many green supply sources.

== Doublepost prevention - post automatically merged: 3 May 2026 ==

Not sure mate. Just something I read from one of our controllers. The 745s can tolerate the low voltage but it gives a very jumpy ride. We suffered between Norwich & Diss a couple of years back due to feeder issues. Climbing the bank from Trowse was lively to say the least.

Low voltage and low frequency are very different things.

 

[dk1]

Low voltage and low frequency are very different things.

Fair enough.

 

[Paul AC]

Someone better qualified can give specifics (I've stepped foot in the cab just about twice), but I believe ADD (automatic drop device) activations cause the emergency brake to apply. I imagine signal, if fed from the same supply, would appear blank.

I can't think of any good reason why an ADD activation should cause the emergency brake to apply. The driver ought to be allowed to coast to a suitable location then stop - similar to being able to override a Passenger Communication alarm, to avoid stopping in a tunnel or on a viaduct if possible; the 'best' place to stop being in a station platform.

Having said that, even if a train's traction system decides it doesn't like the supply frequency it should 'just' open the Vacuum Circuit Breaker (VCB) and keep it open until the driver performs some form of 'reset'. I see no point in dropping the pantograph: the roof-mounted voltage sensing transformer on all AC trains built in recent years should be able to tolerate a reasonable frequency variation, and as it is on the supply side of the VCB it can detect when the frequency (and voltage, if necessary) is back to normal.

If the Bombardier-designed trains (345s on Elizabeth line and 730s on the WCML) reacted badly to this supply problem, it means they didn't learn the lessons from the 2019 Siemens class 700 incident.

 

[GraphitProjekt]

Not sure mate. Just something I read from one of our controllers. The 745s can tolerate the low voltage but it gives a very jumpy ride. We suffered between Norwich & Diss a couple of years back due to feeder issues. Climbing the bank from Trowse was lively to say the least.

Hmm interesting, there is a Norwich feeder station directly on the Crown Point side of Trowse bridge, so I'd guess it was offline and the whole section fed from Stowmarket. Similarly I believe Kings Lynn feeder has been long term out of use (although GA don't go that way anymore)

 

[800301]

I can't think of any good reason why an ADD activation should cause the emergency brake to apply. The driver ought to be allowed to coast to a suitable location then stop - similar to being able to override a Passenger Communication alarm, to avoid stopping in a tunnel or on a viaduct if possible; the 'best' place to stop being in a station platform.

Having said that, even if a train's traction system decides it doesn't like the supply frequency it should 'just' open the Vacuum Circuit Breaker (VCB) and keep it open until the driver performs some form of 'reset'. I see no point in dropping the pantograph: the roof-mounted voltage sensing transformer on all AC trains built in recent years should be able to tolerate a reasonable frequency variation, and as it is on the supply side of the VCB it can detect when the frequency (and voltage, if necessary) is back to normal.

If the Bombardier-designed trains (345s on Elizabeth line and 730s on the WCML) reacted badly to this supply problem, it means they didn't learn the lessons from the 2019 Siemens class 700 incident.

On a 387/80x if you receive an ADD, the affected pantograph should drop but you must press pan down regardless, there is no emergency brake application associated with an ADD but depending on what or how many pantographs you have up will depend on if you make an emergency brake application or not as per rule book

On an 80x if you lose line voltage completely while at a stand the pan will automatically drop and after a while the unit will run out of battery and completely shut down, depending on how long it’s left like this, it will require a generator to be plugged in to be able to restart the unit, this has happened quite a few times at Paddington when isolations have been taken or switching taking place without the units being aux’d off

 

[skyhigh]

I would hope that anyone competent to drive a freight train would very quickly notice a blank signal and activate a REC (which may have already happened with multiple ADD activations around the area).

You wouldn't make an emergency call for a blank signal, and you wouldn't do an emergency call for an ADD activation unless you believed there was a danger as a result of it (not to mention that OHLE supply being outside of operating parameters wouldn't activate the ADD anyway).

 

[davews]

The dip in frequency occurred around 4pm. Here at least the day had been sunny throughout and became suddenly cloudy around that time. This is similar to what happened in Spain where their issues occurred similarly when there was an unexpected big drop in solar. In this case the dip remained within specification but it would not have had to go much lower and there really would be a problem maintaining supply. The media don't seem to have noticed this time, next time they will...

 

[GraphitProjekt]

People talk a lot about the potential for grid batteries to fill in when the sun stops shining or wind stops blowing, I don't currently believe that to be plausible, but grid batteries will surely be able to buffer out that kind of situation for perhaps a few minutes to allow other sources to kick into action, maybe with enough batteries you could conceivably dispatch nuclear plants??

 

[saismee]

You wouldn't make an emergency call for a blank signal, and you wouldn't do an emergency call for an ADD activation unless you believed there was a danger as a result of it (not to mention that OHLE supply being outside of operating parameters wouldn't activate the ADD anyway).

Interesting, thank you. I would've expected that a blank signal would've been an emergency call but I suppose a normal call to the signaller makes sense.

Internal source said that there were ADD activations on the GEML, though it also had a feeder going down rather than just a fluctuation.

 

[GraphitProjekt]

Interesting, thank you. I would've expected that a blank signal would've been an emergency call but I suppose a normal call to the signaller makes sense.

Internal source said that there were ADD activations on the GEML, though it also had a feeder going down rather than just a fluctuation.

did you hear which feeder? Or even the general area?

 

[Elecman]

I doubt it would've been affected unless the substation (?) powering the equipment dropped out. I can't see anything internal to suggest that signals were affected, though they're not very high-priority if the trains themselves can't move either.

All the signalling supplies on the WCML are fed from the 25kV Traction supply through a UPS and have a 400v DNO secondary supply through the same UPS so should not have been affected unless the UPS was in bypass. Telecoms supplies also have UPS battery backed up supplies

 

[dk1]

Hmm interesting, there is a Norwich feeder station directly on the Crown Point side of Trowse bridge, so I'd guess it was offline and the whole section fed from Stowmarket. Similarly I believe Kings Lynn feeder has been long term out of use (although GA don't go that way anymore)

Yeah can’t remember the issue exactly. I was never that technically minded. Just drove them and hoped for the best lol.

 

[najaB]

A solar farm, for example, would need to have additional capacity available to instantaneously produce extra power.

This is where storage (battery, pumped hydro, etc.) come in to their own.

 

[Class 317]

It is often not the frequency variation itself, but the harmonics that it generates. A variation of just half a hertz in the main frequency, will generate significantly greater frequency spreads in the harmonic frequencies. Even if the equipment itself is immune to such frequencies, other equipment may not be, and so it may be required to shut down to avoid generating such harmonics.

For example, signalling, telecom and other control equipment may use Frequency Division Multiplex and similar equipment. It is important that traction doesn't generate frequencies that might interfere with the frequencies such systems use.


Agreed. It is a recognised problem that much of the "green" forms of energy don't provide the same frequency stability as traditional generator-based supplies. Even worse, they may automatically disconnect from the grid in such situations, to protect themselves, compounding the problem for the grid.

It's not a recognised problem but a recognised change in the solution to this problem. To keep frequency within parameters their is a large investment in fast response batteries and flywheels both already installed and being installed as renewables take a greater share of grid capacity.

 

[a_c_skinner]

I realise this isn't gridforums.co.uk but have I missed the precipitating event? This is a relatively low demand time of year and I think it wasn't a flat calm, interconnectors working and it was a bit sunny so generating capacity should have been OK.

 

[Gerald Fiennes]

So, is there any confirmation whether or not the Frequency excursion yesterday was the cause of the incident.

Pardon me for being blunt but, having attempted to use Crossrail from Reading yesterday is there no contingency plan if the tunnel section goes out rather than 'stop the job'?!!! For the £19bn it cost, one might have expected at least some sort of back-up. GWR EMUs were running down the relief lines, I think, but obvs don't call at all stations?

 

[mrmartin]

So, is there any confirmation whether or not the Frequency excursion yesterday was the cause of the incident.

Pardon me for being blunt but, having attempted to use Crossrail from Reading yesterday is there no contingency plan if the tunnel section goes out rather than 'stop the job'?!!! For the £19bn it cost, one might have expected at least some sort of back-up. GWR was running down the relief lines, I thinks?

Yes it will be. NESO (should) produce a report on what happened.

 

[Horizon22]

So, is there any confirmation whether or not the Frequency excursion yesterday was the cause of the incident.

Pardon me for being blunt but, having attempted to use Crossrail from Reading yesterday is there no contingency plan if the tunnel section goes out rather than 'stop the job'?!!! For the £19bn it cost, one might have expected at least some sort of back-up. GWR was running down the relief lines, I thinks?

It wasn't just the tunnel section. It was the fleet that was affected more than the actual OLE.

 

[Chonner]

Interesting - other than Langley and Iver they would be covered by ticket acceptance on London buses. No covering service west of Slough though.

I was on the 18:32 Class 387 Swindon crowd buster which had a stop order at Slough, Maidenhead & Twyford, although not shown on RTT. Those combined with the Didcots gave an effective 15 min frequency west of Slough except for Burnham and Taplow, although things weren't exactly running right time with all the extra stops.

 

[Gerald Fiennes]

It wasn't just the tunnel section. It was the fleet that was affected more than the actual OLE.

Thanks, but did the C345s actually freeze (a la the C700s during that notorious fault in 2019)? Could not a subfleet of them been organised to maintain the Reading all stations service?

The special stop order info is helpful. However, there was no mention of these at Reading Station at c.1800 when trying to reach the smaller stations. The indicators just showed the GW Didcots (at 2x per hour...). Perhaps a good time to have someone do it the old-fashioned way, through actual announcements by people and not rely solely on the computerised voices? Helps the railway show its human face, after all?!!!

 

[Belperpete]

This is where storage (battery, pumped hydro, etc.) come in to their own.

It takes time to bring pumped storage power on line, although if already running their generators will have the same inertial effect as other conventional generator power stations. Preventing this kind of frequency fluctuation needs something that can respond instantly, like the flywheels mentioned by cl 317.

My understanding is that Saturday's incident was caused by the failure of an underground grid feeder. ("underground" as in below ground, as opposed to overhead pylons, not "London underground").

 

[Horizon22]

Thanks, but did the C345s actually freeze (a la the C700s during that notorious fault in 2019)? Could not a subfleet of them been organised to maintain the Reading all stations service?

Many of them did come to a stop and needed remote fleet technician assistance. Not all, but by that point you’ve got trains 60+ minutes late, lines blocked, massive driver displacement etc.

 

[DelW]

In the current issue of Private Eye, their energy correspondent "Old Sparky" has a piece about grid stability and solar power, particularly the issues in dealing with sudden fluctuations from solar farms. Obviously no link available as PE is printed only (it's in issue 1674, p13).

One problem is allegedly that a lot of solar farms connect to local networks not at Grid level, so NESO has limited access to or control over their output intentions.

As well as the infamous Spanish grid collapse, it says there was a German problem earlier this year when solar generation was 30% different from the prediction.