Is the leccy supply to the railway 'just enough'?

[Efini92]

A good example of not enough juice is around Preston. A few 331,397s and some 90s in the area at the same time quite often causes issues.

Then the sleeper rolls through and the CAF units have a melt down.

 

[HSTEd]

One of the advantages of railways operating to a timetable is that, within reason, supplies can be tailored to requirements.

People often rag on the BR electrification schemes for being done with extension leads or what not - but they delivered electrification on time and on budget, on budgets far smaller than the likes of Network Rail today.

An acceptable electrification scheme that delivers what was promised is preferable to an amazing one that fails miserably and kills the electrification programme - see the Great Western debacle.

 

[edwin_m]

On the third rail network, the Electrostars and Desiro fleets ran without regenerative braking when first introduced, so the power supply upgrades at the time must have been able to cope with that amount of demand. A few years later it was enabled, which must have given a bit of leeway in the power supply.

 

[paul1609]

On the third rail network, the Electrostars and Desiro fleets ran without regenerative braking when first introduced, so the power supply upgrades at the time must have been able to cope with that amount of demand. A few years later it was enabled, which must have given a bit of leeway in the power supply.

As I understand it at least on the Southern Region DC regenerated power can only be used by another train on the same substation or dissipated through network losses as the Substations do not have Invertors fitted to return power to the AC grid thus regeneration is good for scenarios such as the Brighton mainline but would be next to useless on something like the Uckfield Line where you have an hourly service on a single line. I think that on the current off peak timetable there is on ever 1 train south of Ashurst (Kent).

 

[swt_passenger]

As I understand it at least on the Southern Region DC regenerated power can only be used by another train on the same substation or dissipated through network losses as the Substations do not have Invertors fitted to return power to the AC grid thus regeneration is good for scenarios such as the Brighton mainline but would be next to useless on something like the Uckfield Line where you have an hourly service on a single line. I think that on the current off peak timetable there is on ever 1 train south of Ashurst (Kent).

On the DC Network the substations operate in parallel so the load and any regen is effectively shared over wider areas than individual substation positions would suggest.

 

[contrex]

On the DC Network the substations operate in parallel so the load and any regen is effectively shared over wider areas than individual substation positions would suggest.

Over how many substations?

 

[Bald Rick]

On the third rail network, the Electrostars and Desiro fleets ran without regenerative braking when first introduced, so the power supply upgrades at the time must have been able to cope with that amount of demand. A few years later it was enabled, which must have given a bit of leeway in the power supply.

Correct. Quite a lot more units running around since then though.

On the DC Network the substations operate in parallel so the load and any regen is effectively shared over wider areas than individual substation positions would suggest.

the DC substations also rectify from the AC distribution network, and typically the regenerated DC is not inverted back to AC.

Over how many substations?

it depends…


One thing though - even where there is no major load (ie other trains) in section to take a regenerated load, there is always the load of system losses, which as we know for big DC sections can sometimes be substantial.

 

[Ken H]

Correct. Quite a lot more units running around since then though.



the DC substations also rectify from the AC distribution network, and typically the regenerated DC is not inverted back to AC.



it depends…


One thing though - even where there is no major load (ie other trains) in section to take a regenerated load, there is always the load of system losses, which as we know for big DC sections can sometimes be substantial.

surely if there is no traffic load on the 3rd rail for a substation area, there can be no resistance in the rail because it wont be carrying current? But current does leak to earth, especially when its wet, so there will be some no-load losses.

 

[Bald Rick]

surely if there is no traffic load on the 3rd rail for a substation area, there can be no resistance in the rail because it wont be carrying current? But current does leak to earth, especially when its wet, so there will be some no-load losses.

err, yes!

 

[swt_passenger]

the DC substations also rectify from the AC distribution network, and typically the regenerated DC is not inverted back to AC.

I didn’t want to repeat many previous discussions, hence not mentioning that, but it was mentioned in post #35 anyway

 

[Bald Rick]

I didn’t want to repeat many previous discussions, hence not mentioning that, but it was mentioned in post #35 anyway

Sorry, I missed the original quote.

 

[swt_passenger]

Sorry, I missed the original quote.

No worries…

 

[Nicholas Lewis]

One thing though - even where there is no major load (ie other trains) in section to take a regenerated load, there is always the load of system losses, which as we know for big DC sections can sometimes be substantial.

System losses at the DC level are a direct function of how much current is being transmitted along the conductor rail if there are no trains the losses are limited to current leakage which is pretty insignificant in comparison especially with the advent of polymeric insulators and use of rail pads on concrete sleepers with well insulated rail clips. Biggest challenge on managing leakage current though is the keeping ballast away from the conductor rail especially when P Way do maintenance ballast top ups but don't the have the manpower or access to clear the ballast from around the conductor rail as ballast regulators have to keep the plough clear of con rail side.

the DC substations also rectify from the AC distribution network, and typically the regenerated DC is not inverted back to AC.

There are no inverter substations on NRs DC infrastructure but some other countries have trialled thyristor based systems. These would have created a headache in the UK with track circuit interreference although with high power IGBTs now available they could be deployed but would make trackside substations a lot more complicated compared to simple diode rectifiers.

The DC network perhaps needs to emulate F1 with the ability of the cars to recover a proportion of its braking energy on board and redeploy as required to drive up energy recover rates especially on the more rural lines with low train frequency.

 

[Bald Rick]

The DC network perhaps needs to emulate F1 with the ability of the cars to recover a proportion of its braking energy on board and redeploy as required to drive up energy recover rates especially on the more rural lines with low train frequency.

With batteries? Who’s have thought! And perhaps if they are big enough, you don’t need the third rail for some of the way...

 

[WAO]

Flywheels have been examined by LUL. They must operate at very high rotational speeds and safety is a question if one started to disintegrate or even become out of balance.
WAO

 

[Ken H]

With batteries? Who’s have thought! And perhaps if they are big enough, you don’t need the third rail for some of the way...

Or capacitors. No idea how that would work tho.
Or use the spare leccy to boost train heat or aircon.

 

[edwin_m]

As I understand it at least on the Southern Region DC regenerated power can only be used by another train on the same substation or dissipated through network losses as the Substations do not have Invertors fitted to return power to the AC grid thus regeneration is good for scenarios such as the Brighton mainline but would be next to useless on something like the Uckfield Line where you have an hourly service on a single line. I think that on the current off peak timetable there is on ever 1 train south of Ashurst (Kent).

True, but if a train is being supplied by regeneration from another train then it's not being supplied from the substation. So Network Rail is saving power, and the substation isn't being loaded so heavily and ultimately allowing more trains to be operated without extra substations.

 

[Nicholas Lewis]

With batteries? Who’s have thought! And perhaps if they are big enough, you don’t need the third rail for some of the way...

I should clarify that this would be energy recovery from braking which would get used next time your taking traction power again not enough to get it down the Uckfield line not even to Edenbridge Town!! DC regeneration (im ignoring AC which doesn't have the same issues) has improved DC credentials on energy usage but to improve it further needs something onboard the trains rather than dissipating 70-80% of regeneration as heat in the braking resistors. It would also reduce losses in the DC system as its the high current demand accelerating that drives the losses.

 

[Efini92]

True, but if a train is being supplied by regeneration from another train then it's not being supplied from the substation. So Network Rail is saving power, and the substation isn't being loaded so heavily and ultimately allowing more trains to be operated without extra substations.

How much could be generated from regen? It worked well on the woodhead line because of the gradients. Would that be the case in the southern region?

 

[Jozhua]

What happens if power requirements are exceeded? Do the trains just run slower, or do things start going pop?

Asides from the lead times, would I be mistaken for thinking that it might not be a massive deal? Getting the wires strung with the right clearances and gauging seems like the trickiest part, obviously grid interconnects aren't easy, but they seem like they would be a fairly standard, well known quantity. Probably doesn't require quite as much of a rail specific skillset and can use electricians/engineers who work on other large power projects.

I don't blame BR for underspeccing in the 80's - a lot of what they did got the railways out of a deep hole with 158's, HSTs, etc, which compared to their limited budgets, absolutely transformed travel in the UK. It's been 30/40 years on and the fact that, despite the railways being busier than ever, we are still choosing to not spend money upgrading the infrastructure properly, is really damming of our present day priorities.

I wonder whether power supply equipment could be moved to smaller electrification schemes with lower power requirements once it becomes overwhelmed on the ECML and the like? Might be a good way of getting some extra life out of it!

 

[edwin_m]

How much could be generated from regen? It worked well on the woodhead line because of the gradients. Would that be the case in the southern region?

Figures quoted are usually in the region of 15-20% saving. It works best on intensively-used lines with lots of station stops, where there's more likely to be another train nearby that can use the power generated. On less busy lines it works much better on the 25kV system, because on the third rail more of the power will be lost in line resistance before it gets to a train that can use it. So at a guess I'd say there was quite a saving in the inner London suburbs but much less somewhere like Bournemouth.

 

[AM9]

Figures quoted are usually in the region of 15-20% saving. It works best on intensively-used lines with lots of station stops, where there's more likely to be another train nearby that can use the power generated. On less busy lines it works much better on the 25kV system, because on the third rail more of the power will be lost in line resistance before it gets to a train that can use it. So at a guess I'd say there was quite a saving in the inner London suburbs but much less somewhere like Bournemouth.

Wasn't the Woodhead line a special case benefitting from a fairly uniform train type, i.e. class 76 locos with long slow drags up each side of to Pennines matched by long falls in the opposite direction. This meant that there was an almost continuous demand for the regenerated energy for much of the day. On a metro or commuter line, the sheer density of traffic will likely aggregate to give a fairly continuous regen supply matching the demand. With ac electrification, the situation is much less critical because the regen. AC is available direct from the trains and usually is directly available to a much larger population of trains. An additional gain is the practicality of feeding back power into the grid where there is always a much more constant demand in which to sink the energy.

 

[Bald Rick]

How much could be generated from regen?

On intensive metro lines with lots of stops 40% is not unheard of.

What happens if power requirements are exceeded? Do the trains just run slower, or do things start going pop?

Firstly the voltage drops, which does mean there’s less power at the train (so it accelerates a little more slowly), but then the breakers start tripping in the substations.

Power is a curious thing though - most power equipment is given a ‘continuous’ rating, ie the limit it can operate at indefinitely; and then timebound ratings which is a higher limit it can operate at for this time, (e.g. 1 hour, or 1 minute).

 

[hwl]

On intensive metro lines with lots of stops 40% is not unheard of.




Firstly the voltage drops, which does mean there’s less power at the train (so it accelerates a little more slowly), but then the breakers start tripping in the substations.

Power is a curious thing though - most power equipment is given a ‘continuous’ rating, ie the limit it can operate at indefinitely; and then timebound ratings which is a higher limit it can operate at for this time, (e.g. 1 hour, or 1 minute).

And as the voltage starts to drop the distribution losses (up until the shoes) go up, 3rd rail regen has some useful indirect effects at increasing the average 3rd rail voltage and reducing losses which improves the overall power supply situation more than might be expected looking at it simplistically.

 

[dm1]

Modern trains are designed to reduce the power they draw if the voltage drops too far and refuse to move at all if it drops below a certain threshold.

The inability for the driver to reset this system on the train was the cause of all the Class 700s on Thameslink getting stuck when the power went out a while back.

 

[Ken H]

And as the voltage starts to drop the distribution losses (up until the shoes) go up, 3rd rail regen has some useful indirect effects at increasing the average 3rd rail voltage and reducing losses which improves the overall power supply situation more than might be expected looking at it simplistically.

Can they? Are the Up and down supplied from a common supply and connected to each other away from the sub station? Or does the track paralleling hut(TPH) do that?
(No idea what a TPH does BTW)

 

[dm1]

Can they? Are the Up and down supplied from a common supply and connected to each other away from the sub station? Or does the track paralleling hut(TPH) do that?
(No idea what a TPH does BTW)

I'm no expert, but I suspect that's exactly what a track paralleling hut does. Connecting the two lines to one another at regular intervals reduces the overall impedance (resistance) of the system and therefore reduces losses and reduces the voltage drop.

In basic terms, it also reduces the distance current has to travel between a regenerating train on one line and am accelerating train on the other, reducing losses.

 

[AM9]

Modern trains are designed to reduce the power they draw if the voltage drops too far and refuse to move at all if it drops below a certain threshold.

The inability for the driver to reset this system on the train was the cause of all the Class 700s on Thameslink getting stuck when the power went out a while back.

Without going into a debate about the even, wasn't the TL issue more concerning the mains frequency momentarily dropping outside the acceptable range for the lightweight 'low iron' transformers fitted to the class 700s, exacerbated by the removal of driver's ability to reset the system. Low voltage events (like temporary power failures) seem to be driver resettable because of the number of possible causes, but a high performance transformer has limited safe working outside its specified frequency range so needs more intervention for such an unlikely event.

 

[XAM2175]

The inability for the driver to reset this system on the train was the cause of all the Class 700s on Thameslink getting stuck when the power went out a while back.

Without going into a debate about the even, wasn't the TL issue more concerning the mains frequency momentarily dropping outside the acceptable range for the lightweight 'low iron' transformers fitted to the class 700s, exacerbated by the removal of driver's ability to reset the system.

Yes, the "Big 700 and 717 Sit-Down Event" was caused by the supply frequency falling out of range, not voltage. The drivers should have been able to reset the train - and indeed several were - but the trains that couldn't be reset had a fault in their operating software. There was no deliberate removal of the driver's ability to perform a reset.

Obviously this only affected units on AC power at the time of the frequency deviation.

 

[hwl]

Without going into a debate about the even, wasn't the TL issue more concerning the mains frequency momentarily dropping outside the acceptable range for the lightweight 'low iron' transformers fitted to the class 700s, exacerbated by the removal of driver's ability to reset the system. Low voltage events (like temporary power failures) seem to be driver resettable because of the number of possible causes, but a high performance transformer has limited safe working outside its specified frequency range so needs more intervention for such an unlikely event.

Actually more embarrassing - the wrong frequency values being put in the software for the limits, the start shedding load value was use as the value to shed all load...
The frequency never went below the shed all load value in the standards so no resets should have been need let alone drivers not being able to do them!
Swiss cheese failure model at work there.