Phase balancing and OHLE

[mikeg]

Excuse me if I've used the wrong technical term. The National Grid is three phase. OHLE is single phase.

I believe there's no easy way to convert three phases into one, or at least this always used to be the case. How then does OHLE not create an imbalance between phases on the national grid. Obviously power can be taken from any two phases, but even if one set of tracks took from one couple of phases, and another from the other, surely there's still going to be on phase that's excessively loaded?

Is it balanced out on a national level with a significant imbalance on a local level? Just one of those random thoughts I've had...

Or doesn't it matter? I've always been taught that balancing load across the three phases is essential for the efficient functioning of any three phase electrical system.

 

[plugwash]

You can then use different phases on different lines to get some semblence of balance.

There was/is a technique invovling a "T transformer", which lets you turn three phases 120 degrees out of phase into two phases 90 degress out of phase while conserving balance so you only have to split your system into two peices rather than three. Not sure if this technique is used in the UK..

Either way though "some semblance of balance", is not really good enough for lower levels of the grid heirachy given how much power even a single train can draw. Historically OHLE systems had to be connected at a high level of the grid heirachy so the imbalances they created would not be too much for the grid to bear.

More recently, "static frequency converters" have been introduced, which convert via DC. These can draw a balanced load from the three phase grid while supplying single phase to loads, but they are more expensive and less efficient than simple transformers. Still they can be a much cheaper option as they can be connected at a lower level of the grid heirachy.

 

[contrex]

The AC railway does present an unbalanced load if fed one phase at a time, and historically the mitigation (such as it was) consisted of taking the feed at 240 or 400kV (more resilient bulk feed) and feeding the sectioned OHL from different phases. Nowadays, more and more, solid state electronics can be used to take a balance three-phase load and convert it to a single-phase output in substations called 'static frequency converters' (SFCs). These work by taking a three-phase balanced load from a high-voltage transmission line, converting it to DC and then to a single phase 25kV OLE supply. They do not affect the balance between the transmission line’s three phases. Hence, they can be fed from a local distribution network’s 33kV supply. Another advantage is the that the power output from adjacent substations can be made identical and thus the neutral section in between can be dispensed with if the sections either side are fed from SFCs .

 

[edwin_m]

Another advantage is the that the power output from adjacent substations can be made identical and thus the neutral section in between can be dispensed with if the sections either side are fed from SFCs .

Not the power output, the critical thing is to get the SFCs feeding the line in phase with each other. I assume is is done by synchronising the SFCs' outputs to one specific phase of the Grid.

 

[contrex]

Not the power output, the critical thing is to get the SFCs feeding the line in phase with each other. I assume is is done by synchronising the SFCs' outputs to one specific phase of the Grid.

I used the phrase 'power output' after reading an article about SFCs in the Rail Engineer - https://www.railengineer.co.uk/static-frequency-converters-improving-the-25kv/

its main role is to control the precise voltage, phase angle, and frequency of the output.

If this can be achieved, then the power output from adjacent substations can be made identical and thus the neutral section can be dispensed with if the sections either side are fed from SFCs . In the process, the control of the active power flow is better controlled and the reactive power reduced. That, in short, is what an SFC attempts to achieve.

I don't know that the single-phase SFC outputs need to be synchronised with any particular phase of the Grid, or with anything other than each other. I imagine the DC link in each SFC provides isolation from the Grid.

 

[edwin_m]

I used the phrase 'power output' after reading an article about SFCs in the Rail Engineer - https://www.railengineer.co.uk/static-frequency-converters-improving-the-25kv/

"Power output" is probably not the best wording in that article, and becomes incorrect when quoted out of context. To an engineer "power" means watts, but what the SFC controls is voltage and how it varies over time.

I don't know that the single-phase SFC outputs need to be synchronised with any particular phase of the Grid, or with anything other than each other. I imagine the DC link in each SFC provides isolation from the Grid.

Yes, the key thing is that it's essential they synchronise with each other to avoid the neutral section. The obvious way for them to do that is to follow the same phase of their own Grid supplies, as these will always be in phase right across the country. But they may do it in some other way.

 

[hwl]

"Power output" is probably not the best wording in that article, and becomes incorrect when quoted out of context. To an engineer "power" means watts, but what the SFC controls is voltage and how it varies over time.


Yes, the key thing is that it's essential they synchronise with each other to avoid the neutral section. The obvious way for them to do that is to follow the same phase of their own Grid supplies, as these will always be in phase right across the country. But they may do it in some other way.

They aren't synchronising yet that is the next stage and hasn't been done in the UK to eliminate neutral sections yet.
The existing primary transformer connections from 2 phases means the output used for traction isn't in sync with either grid phase so syncing SFC output to a grid phase may not be the best idea...

 

[poffle]

"Power output" is probably not the best wording in that article, and becomes incorrect when quoted out of context. To an engineer "power" means watts, but what the SFC controls is voltage and how it varies over time.


Yes, the key thing is that it's essential they synchronise with each other to avoid the neutral section. The obvious way for them to do that is to follow the same phase of their own Grid supplies, as these will always be in phase right across the country. But they may do it in some other way.

I remember talking to a lecturer in college about how connecting two isolated grids together they would have a sort of oscilloscope where they could display the phases of both grids overpaid on each other and then trigger the circuit breaker to oin the segments when the phases matched.

This would have been a few decades ago.

I presume also they can operate the railway AC at a standard 50 Hz to avoid grid disturbances. Also avoids the issue with the Thameslink trains tripping out a few years when there was a major grid disturbances causing a significant frequency drop.

 

[swt_passenger]

Another advantage is the that the power output from adjacent substations can be made identical and thus the neutral section in between can be dispensed with if the sections either side are fed from SFCs .

It’s somewhat theoretical only at this stage, because there are so few completed SFC supplies to the railway, and one of them at least has been rather secretly postponed. (Marshall Meadows, near Berwick).

How many are planned in total, and are there any that are neighbours? I think it’s been said in previous discussions they aren’t seen as the future standards, they’ll still connect to 400 kV grid when it’s available, as seen at Braybrooke on the MML and in future at Kegworth.

 

[snowball]

As far as I know there are only one or two SFCs feeding the railway to date. The recently opened one at Hambleton, where the electrified ECML crosses the (perhaps one day electrified) Leeds-Hull line, is among the first.

In a previous thread I asked about synchronising SFCs along a route so that their outputs were all in phase, thus avoiding the need for a neutral section, but experts replied that it was possible but might not be worth the effort.

I think there are still a number of feeder stations where the input is only at 132kV, especially in Scotland, including the recently opened one at Ferguslie (which is not an SFC) and the proposed one at Tweedbank.

 

[edwin_m]

They aren't synchronising yet that is the next stage and hasn't been done in the UK to eliminate neutral sections yet.
The existing primary transformer connections from 2 phases means the output used for traction isn't in sync with either grid phase so syncing SFC output to a grid phase may not be the best idea...

The transformer will introduce a phase change too. Also, I imagine the phase angle of any particular section of the network might change during alternative feed arrangements. So it may not be a good idea to try to connect a SFC-fed to a conventional-fed section without a neutral section in between.

I remember talking to a lecturer in college about how connecting two isolated grids together they would have a sort of oscilloscope where they could display the phases of both grids overpaid on each other and then trigger the circuit breaker to oin the segments when the phases matched.

This would have been a few decades ago.

I presume also they can operate the railway AC at a standard 50 Hz to avoid grid disturbances. Also avoids the issue with the Thameslink trains tripping out a few years when there was a major grid disturbances causing a significant frequency drop.

Any train regenerating into an AC supply has to be synchronised with the exact frequency and phase of that supply, and all facilities supplying power to the Grid will have to do the same. The former is certainly done automatically every time regenerative braking is demanded, and I should imagine the phase matching is automatic for electricity suppliers too. This may also mean that SFCs can do something similar, and track the phase of the supply they are feeding so several can run in parallel. I believe it's tricky to synchronise two grids which aren't already connected and in phase, which is one reason why interconnectors between different countries tend to be DC (using SFCs to feed the destination grid).

Over the majority of the AC rail network where SFCs are not used, the overhead line is just fed through a transformer from the Grid connection. So its frequency is the same as the Grid frequency and its phase relationship to the Grid is fixed unless a particular section is switched to being fed from a transformer connected to a different Grid phase. SFCs could probably stay much closer to 50Hz, provided there was no reason to maintain a particular phase relationship with the Grid. The Thameslink problem was only as serious as it was because of software or setup errors in the trains' traction systems, which shut them down when the frequency went out of range but could only be restarted by a technician plugging in a laptop. I believe this has now been changed so the driver "re-booting" the train when the frequency has gone back within limits will restore normal working.

 

[contrex]

I remember talking to a lecturer in college about how connecting two isolated grids together they would have a sort of oscilloscope where they could display the phases of both grids overpaid on each other and then trigger the circuit breaker to oin the segments when the phases matched.

In the 1980s I worked at a company that made, among other things, a three-lamp contraption for syncing an alternator when bringing it on line, called a 'synchroscope'. The lamps (bulbs, rated for the full phase voltage) were mounted in a triangle on the front panel. Each lamp was connected between one phase of the alternator output and the corresponding phase of the system busbars. Out of sync conditions were seen as 'rotating' flashes of the lamps, and when they were all dark you closed the breaker.

 

[172007]

Solar panels can be slightly over or under the 50hz when exporting, would this not make them out of phase with the street supply. I understand this is a problem when the local grid has enough property's with panels that it can go to +- a few hertz which can cause issues.

 

[GRALISTAIR]

Solar panels produce DC. The output has to go through an inverter.

 

[edwin_m]

Solar panels can be slightly over or under the 50hz when exporting, would this not make them out of phase with the street supply. I understand this is a problem when the local grid has enough property's with panels that it can go to +- a few hertz which can cause issues.

Do you have a source for this behaviour of solar panels? If a source is connected to a network with a significant phase difference, then large currents will flow between them, most likely causing some sort of overload. So I'd have thought that inverters on solar (and future vehicle to grid) units would have to track the mains frequency and phase very closely.

 

[GRALISTAIR]

Do you have a source for this behaviour of solar panels? If a source is connected to a network with a significant phase difference, then large currents will flow between them, most likely causing some sort of overload. So I'd have thought that inverters on solar (and future vehicle to grid) units would have to track the mains frequency and phase very closely.

Exactly my thoughts and understanding too.

 

[stuving]

I think there is a close parallel between directly connected SFCs and the need for grid-forming inverters for renewable energy sources feeding the national grid. In both cases inverters have to be given control software that makes them play nicely when connected in parallel. A power grid fed in parallel mainly by "normal" (grid-following) inverters is not stable; no doubt the same is true of a long stretch of OLE.

I don't think it's possible to take the grid as a common phase reference. The relevant question is "what are the phase differences between one point on the grid and another, and how do they vary with time and distance". Bear in mind that "grid" here will include the DNO (now ENO) networks that SFC feed stations are going to be connected to. I don't know the answer (I'm not a power engineer) but I guess it's "not so small that you can ignore them".

 

[edwin_m]

I think there is a close parallel between directly connected SFCs and the need for grid-forming inverters for renewable energy sources feeding the national grid. In both cases inverters have to be given control software that makes them play nicely when connected in parallel. A power grid fed in parallel mainly by "normal" (grid-following) inverters is not stable; no doubt the same is true of a long stretch of OLE.

I don't think it's possible to take the grid as a common phase reference. The relevant question is "what are the phase differences between one point on the grid and another, and how do they vary with time and distance". Bear in mind that "grid" here will include the DNO (now ENO) networks that SFC feed stations are going to be connected to. I don't know the answer (I'm not a power engineer) but I guess it's "not so small that you can ignore them".

I'm assuming the SFCs are, or could be, grid forming to get round this problem. Grid-following inverters need something else, such as a significant amount of supply from rotating alternators, to maintain frequency stability.

Electricity propagates like a wave at close to the speed of light, even though the electrons themselves move much less. So the difference between currents passing over diverse paths is likely to be tiny, even on a large grid. Provided all the SFCs on a route are fed from the same grid, I'm not clear why that couldn't be used as a phase reference (with appropriate measures to ensure they are all syncing to the same phase) as an alternative to using grid-forming inverters.

 

[poffle]

The bandwidth of light at 50Hz in a vacuum would be 6000 km. The phase differences caused by multipath transmission in GB wouldn't be significant.

Mainland Europe has a vastly larger synchronised stretching from Ukraine to the Atlantic.

 

[AM9]

I'm assuming the SFCs are, or could be, grid forming to get round this problem. Grid-following inverters need something else, such as a significant amount of supply from rotating alternators, to maintain frequency stability.

Electricity propagates like a wave at close to the speed of light, even though the electrons themselves move much less. So the difference between currents passing over diverse paths is likely to be tiny, even on a large grid. Provided all the SFCs on a route are fed from the same grid, I'm not clear why that couldn't be used as a phase reference (with appropriate measures to ensure they are all syncing to the same phase) as an alternative to using grid-forming inverters.

The problem in syncing the ac output of a group of SFCs to their grid feeds is that the phase of those feed points may not be constant. When routing of the national grid is changed for operational reasons (e.g. failures/maintenance, or back up to increase available power etc.), there may be a small but significant changs at one point wrt another. This would create an out of phase current which would manifest itself as a poor power factor.

 

[Ediswan]

The bandwidth of light at 50Hz in a vacuum would be 6000 km. The phase differences caused by multipath transmission in GB wouldn't be significant.

Mainland Europe has a vastly larger synchronised stretching from Ukraine to the Atlantic.

A 17 km path difference would make a 1 degree phase difference. The issue cannot be completely disregarded in GB.

 

[stuving]

The phase differences I'm talking about are related to the impedance of power lines connecting NGET substations. These are small, in ohms, but still absorb significant power (losses) and VARs. There are figures available for all NGET lines, but quoted in a per-unit system I'm not familiar with. What is clear, however, is that the reactance is roughly ten times the resistance. So if a current flows through the line - which is what it's there for - there has to be a voltage across its ends.

That's about the limit of my understanding of this stuff; how the routing of power through the grid is controlled is beyond it. One thing I have confirmed recently is that the 400 kV grid (aka supergrid) is directly connected throughout - no 400:400 kV transformers are involved. However, there are compensators - essentially series capacitors - in the two big cross-border links with the Scottish subgrid.

 

[Elecman]

The grid bulk supply points are connected across different 2 phases at adjacent feeders so in old money Feeder One may be Red/Yellow - Feeder2 Yellow/Blue phases and Feeder3 Blue/Red phases thus balancing the intakes over the grid itself

 

[contrex]

Solar panels can be slightly over or under the 50hz when exporting, would this not make them out of phase with the street supply. I understand this is a problem when the local grid has enough property's with panels that it can go to +- a few hertz which can cause issues.

A grid-tied inverter MUST be locked for frequency and phase when exporting, just as a rotating alternator must. Otherwise bad things will happen. It is a trivial matter to make the inverter's 50 Hz oscillator match the external utility supply in frequency and phase.

 

[Taunton]

The bandwidth of light at 50Hz in a vacuum would be 6000 km. The phase differences caused by multipath transmission in GB wouldn't be significant.

Mainland Europe has a vastly larger synchronised stretching from Ukraine to the Atlantic.

It did used to be significant within Britain in television transmissions I was involved with in the 1970s, a live transition from London studio to Scotland studio (550km) caused a significant frame roll on everyone's television sets until it settled down. You could see it happen sometimes in the evening news. There was a technical means to overcome this by making the remote studio camera use the London synchronisation pulses, but it seemingly wasn't always switched on.

It was long a significant problem (in fact the key problem) for Eurovision. The live juries in the various countries were long audio only because of this.

 

[contrex]

It did used to be significant within Britain in television transmissions I was involved with in the 1970s, a live transition from London studio to Scotland studio (550km) caused a significant frame roll on everyone's television sets until it settled down. You could see it happen sometimes in the evening news. There was a technical means to overcome this by making the remote studio camera use the London synchronisation pulses, but it seemingly wasn't always switched on.

It was long a significant problem (in fact the key problem) for Eurovision. The live juries in the various countries were long audio only because of this.

In London, from 1960 on, we used to get frame rolling on our Stella 405 line set during periods of heavy load in the winter, accompanied by reduced scan height and width.

 

[najaB]

A grid-tied inverter MUST be locked for frequency and phase when exporting, just as a rotating alternator must. Otherwise bad things will happen. It is a trivial matter to make the inverter's 50 Hz oscillator match the external utility supply in frequency and phase.

This is also one way that they can detect grid failure. They will try to set their frequency a little higher/lower than the grid input. If the grid is up they'll sense the current that leads to those 'bad things', if the grid is down then there won't be any opposing current. That's how pure grid-tie inverters know to shut down if the mains supply fails, and inverters with a battery backup will disconnect their input side to avoid back-feeding into the grid and potentially harming line workers.

 

[Nottingham59]

Is it balanced out on a national level with a significant imbalance on a local level? Just one of those random thoughts I've had...

To answer your question, yes it can lead to imbalances. To minimise the impact of these the railway has to take its supply from 275kV or 400kV circuits, as these lines will carry 1 or 2GW each, so the imbalance from taking just 40 or 80kVA (or less) from only two of the three phases is manageable.

But 400kVA transformers are incredibly expensive and take many years to acquire from the National Grid, which is one reason why electrification projects are so costly and slow.

Static Frequency Converters effectively convert three phases to DC and then convert that back into single phase AC. So they can take a balanced load from the grid, and tap into much cheaper grid connections. But we don't use them much in the UK.

They're much more common in Germany where the railway supply is 16.7Hz and you have to convert the frequency. In the past this was done with motor-generator sets, where a 50Hz three-phase motor with three windings around the stator drove a shaft at 16.7rpm, which in turn drove a single-phase generator at 16.7Hz. It's all done with solid-state power electronics now, of course.

HTH

 

[Irascible]

As I'm not a power engineer, is there anything impractical about syncing SFC output using a clock rather than their input?

 

[najaB]

As I'm not a power engineer, is there anything impractical about syncing SFC output using a clock rather than their input?

If you mean in terms of feeding power back into the grid, then the main issue is that grid frequency isn't fixed and instead varies slightly based on load.