Longest electrified stretch fed from one end?

[zwk500]

Isn't the phrase 'volts jolts, mills [milliamps] kills', because it's the current that does the damage, and even 25KV isn't always fatal.

 

[AM9]

Isn't the phrase 'volts jolts, mills [milliamps] kills', because it's the current that does the damage, and even 25KV isn't always fatal.

It's current flowing through the body, and particularly certain organs. So to get a certain current (>10mA sustained can be lethal, >100mA almost certain to be lethal) requires a voltage sufficient to overcome contact and body resistance. So in simple terms, at 750V a total resistance would need to be lower than 75kOhms for a severe shock and 7.5kOhms for a lethal shock. For 25kv, those resistances would be 2.5MOhms and 250kOhms respectively. Despite the 25kv needing a much higher resistance for survival, by design, coming into direct contact inadvertently with exposed conductors on those systems is much less likely, which is why it is inherently safer.

 

[Bald Rick]

Not necessarily. Ian Allan, in his autobiography, wrote of more than one Southern employee in WW2 who in the blackout stumbled over the live rail, but was saved by the voltage being generally reduced due to fuel shortage. Not a sure thing, but it can help.

I know people who have had a belt off the 750 and lived. My point, badly put, is that the voltage doesn’t really make a difference; it’s the current that’s the issue.

 

[LondonExile]

My point, badly put, is that the voltage doesn’t really make a difference; it’s the current that’s the issue.

The voltage does make a difference.

It's the current that kills, but Ohm's law (V=IR) also comes into play. At a basic level, rearrange the equation to get I=V/R, i.e. current is directly proportional to voltage. Now - if you're working with a current limited system where you simply cannot draw more than say 1mA, it's true that the voltage is going to be irrelevant, but if there is capacity to deliver a potentially lethal current, the voltage is still critical to the relevant safety of the system.

No electric system capable of powering trains is going to be current limited to the point that it is safe for humans to touch!

A car battery can potentially deliver 500A or so of current, but the fact it's only 12V means it has no chance of doing so through the human body.

 

[Taunton]

My point wasn't anything really to do with safety, but that it was consistently stated from HQ that the voltage at Lymington, far end of the section, was 750v, when in fact at the distance it is from the substation (original point of thread) you invariably only got a proportion of this, being a well-known issue to the electrical engineers.

The same problem used to afflict North Woolwich when it had the North London Line dc electrification. It seems to be a particular issue with dc single track branches, where the designers feel assured there will never be more than one train.

 

[Dai Corner]

My point wasn't anything really to do with safety, but that it was consistently stated from HQ that the voltage at Lymington, far end of the section, was 750v, when in fact at the distance it is from the substation (original point of thread) you invariably only got a proportion of this, being a well-known issue to the electrical engineers.

The same problem used to afflict North Woolwich when it had the North London Line dc electrification. It seems to be a particular issue with dc single track branches, where the designers feel assured there will never be more than one train.

As has been stated before, only under load. No current flowing means no voltage drop (ohm's law). Place a voltmeter across the relevant rails at Lymington or North Woolwich when there's no train taking power and it would read as near as dammit the same as at the feeder station.

If 450v was the minimum acceptable voltage that's what the designers would ensure was provided when the maximum required current was demanded.

 

[AM9]

My point wasn't anything really to do with safety, but that it was consistently stated from HQ that the voltage at Lymington, far end of the section, was 750v, when in fact at the distance it is from the substation (original point of thread) you invariably only got a proportion of this, being a well-known issue to the electrical engineers.

The same problem used to afflict North Woolwich when it had the North London Line dc electrification. It seems to be a particular issue with dc single track branches, where the designers feel assured there will never be more than one train.

It's a calcuated balance of operational reliability vs cost. f iIt is truly a single track with no parallel cables to reduce resistive losses, then the resistance of the steel, aluminium or composite conductor rail plus the steel running rail return will be in series with the supply. If there is a train at the remote end of the track drawing a high current (say 1000A for short periods), as suggested, the voltage at the end of the line will drop to c.450V. As soon as the driver backs off the traction power, the line voltage will rise to near 750V, - or even more if regen is allowed there (which I doubt). So in human terms, the risk is 750V for most of the time with occasional dips to as low as 450V.

 

[yorkie]

Reopened for @GraphitProjekt to post an update

 

[GraphitProjekt]

People were talking about King's Lynn, but it only normally feeds to Littleport which is ~19 miles. Norwich feeds to Hales Street which is ~25miles. Both Norwich and Kings Lynn are single supply feeder stations with 10MVA transformers and Norwich has a slightly higher demand (2tph instead of 1tph and some hotel power for 755s).

But I also wanted to point out South Beach FS on the Ayrshire Coast line that feeds 24 miles to Lochwinnoch, 15 miles south to Ayr and 11 miles to Largs and the short Ardrossan branch. A total of about 42 route miles although South Beach is in the centre of those lines so I suppose the 24 miles is the longest reach it has - throws into question the correct way to count how much mileage a feeder is supplying. Also, a single 10 MVA transformer once again (although South Beach has a second one on standby should the first fail). I'm not familiar with the service pattern in this area, but I assume it's light.

 

[Nicholas Lewis]

People were talking about King's Lynn, but it only normally feeds to Littleport which is ~19 miles. Norwich feeds to Hales Street which is ~25miles. Both Norwich and Kings Lynn are single supply feeder stations with 10MVA transformers and Norwich has a slightly higher demand (2tph instead of 1tph and some hotel power for 755s).

But I also wanted to point out South Beach FS on the Ayrshire Coast line that feeds 24 miles to Lochwinnoch, 15 miles south to Ayr and 11 miles to Largs and the short Ardrossan branch. A total of about 42 route miles although South Beach is in the centre of those lines so I suppose the 24 miles is the longest reach it has - throws into question the correct way to count how much mileage a feeder is supplying. Also, a single 10 MVA transformer once again (although South Beach has a second one on standby should the first fail). I'm not familiar with the service pattern in this area, but I assume it's light.
View attachment 195703View attachment 195704

Its not that light for a single 10MVA transformer with quite a few trains being with the feeding area at any given time. Also the 380's work down here not sure if its the 3 or the 4cars but they certainly have a higher power draw than the 318's. Given there are two transformers at the feeder station they could install a split bar feeder station to provide some relief but most of the load is on Kilwinning to Ayr section.

 

[ABB125]

I'm not familiar with the service pattern in this area, but I assume it's light.

The standard off peak service pattern is, I believe, 2tph Glasgow to Ayr, and 1tph to each of Largs and Ardrossan. At peak times service can double in frequency (but seemingly only for an hour or so). Not sure how many services are double sets though

 

[GraphitProjekt]

Just had a quick look at trains through Kilwinning. Looks like 1tph to Largs and 1tph to Ardrossan Harbour, as well as up to 4tph to Ayr in the peak (2tph normally). So normally 4tph in each direction to Lochwinnoch and theoretically up to 6tph. But with 4tph that's 8 trains in a given hour. It'd be quite fiddly to figure out when all of those exit Lochwinnoch to find out how many trains would be moving simultaneously.

I got my numbers a bit messed up on the above. South Beach to Lochwinnoch should be about 14m 46ch. Kilwinning to Ayr is then 14m 64ch, South Beach to Largs is 11m 53ch, the Ardrossan branch is 1m 04ch, and the Byrehill connection is 0m 60ch. Inevitably some of those numbers will be a bit off. If you count all of those lines together it's 42m 07ch

As far as I could tell it's normally 380s on all those lines

== Doublepost prevention - post automatically merged: 11 Jan 2026 ==

Its not that light for a single 10MVA transformer with quite a few trains being with the feeding area at any given time.

Interestingly, all of the old Glasgow feeder stations for suburban or regional routes are quite small. The North Clyde line has Dalreoch and Parkhead which are both 10MVA supplies. Port Glasgow on Gourock/Wemyss electrification (which I think was mid 1960s like North Clyde) is a 2x 5MVA supply both of which are on the same phase running parallel (same situation for Dalreoch actually). The WCML ones are a lot more substantial: Eglinton Street (basically at Glasgow Central) is 2x 18MVA and Gowkthrapple (nearish Motherwell) is 2x 26MVA.

In recent years Scotrail have been adding feeder stations all of which I suspect are on the larger side, at least 20MVA although I don't have all the data regarding those. On the Ayreshire line, they recently (about 2022?) put in Fergulsie FS which takes over Lochwinnoch/Bishopston to around Shields Depot, ie: removing what used to be fed from Eglington Street


== Doublepost prevention - post automatically merged: 11 Jan 2026 ==

Is there a map which shows feeder station sites etc available publicly?

bit of a late reply, but yes only for 2009 though, so now very out of date but pretty much shows the situation as BR left it on privatisation since very little electrification happened between then and 2009 apart from HS1 which isn't on there anyway. The 2009 Electrification Route Utilisation Strategy (see page 28)

 

[ChilliSauce]

Whilst you can go for determining the branch line with the longest end feed ..
The key item is, if some or all of the branch is single or double track as for double track the OLE system on both lines will be parallelled together thus in effect doubling the cross sectional area of the feeding system ...so reducing the volt drop factor at the far end.

The same will also apply in DC areas

I.e. Blackpool North might be end fed but it is double track throughout.

Walton on the Naze is double track as far as Thorpe Le Shoken then single.

Where as branches like Southminster are single line (apart from the passing loop)

 

[Kite159]

Whilst you can go for determining the branch line with the longest end feed ..
The key item is, if some or all of the branch is single or double track as for double track the OLE system on both lines will be parallelled together thus in effect doubling the cross sectional area of the feeding system ...so reducing the volt drop factor at the far end.

The same will also apply in DC areas

I.e. Blackpool North might be end fed but it is double track throughout.

Walton on the Naze is double track as far as Thorpe Le Shoken then single.

Where as branches like Southminster are single line (apart from the passing loop)

South Beach to Largs is single track as far as the OHLE goes (there is a 2nd track running alongside for a while, but that hasn't got OHLE. Also South Beach to Harbour is a single track branch line, it's double track towards Kilwinning after South Beach

 

[Taunton]

Interestingly, all of the old Glasgow feeder stations for suburban or regional routes are quite small. The North Clyde line has Dalreoch and Parkhead which are both 10MVA supplies. Port Glasgow on Gourock/Wemyss electrification (which I think was mid 1960s like North Clyde) is a 2x 5MVA supply both of which are on the same phase running parallel (same situation for Dalreoch actually).

The Glasgow suburban transformers were presumably replaced when the inner areas were converted from initial 6.25kV to 25kV (date?). On the north side this stretched from Dalmuir Park to Parkhead. There's a map of these initial 6.25kV areas on the last page here of the official accident report into the transformer explosions which happened shortly after Blue Train introduction

https://www.railwaysarchive.co.uk/documents/MoT_EMUFailures1962.pdf

 

[GraphitProjekt]

Also check out this website and the maps they have of the original system: http://www.scot-rail.co.uk/page/Glasgow+Suburban+Electrification+1960

My understanding was that all grid connections were still originally 132/25kV but then there were extra transformers to step down to 6.25kV within the railway side of the substation. So Parkhead has 2x 25kV supplies from Carntyne grid and then one 25/6.25 transformer within the FS equipment for the section Parkhead-Finnieston. On the other side, a substation that no longer exists at Westerton junction steps down Dalreoch's 25kV supply to feed towards Finnieston and also at Dalmuir Park stepping down for the Yoker line. There was also a 25kV cable from the WCML feeder station at "Motherwell" (maybe later replaced by Gowkthrapple?) to the Northern Electric Mainline at Coatbridge.

The fact that Finnieston is mentioned as a "sub feeder station" is a bit strange.

That Motherwell FS also supplied from there to Cathcart and the Neilston branch both at 25kV, but the Cathcart circle to Glasgow Central was 6.25kV I think with supply stepped down by the Cathcart substation but there was also an Eglinton Street FS at this time that may have been 6.25kV only until later upgraded? But that's a pure guess.

On the Anglia lines I think this generally wasn't the case and there were quite a few feeder stations built that were 6.25kV only and went away when the lines were upgraded - Silver Street and Wood Street on the lines to Southbury/Enfield and Chingford respectively. And on the LTS system: Gas Factory Jnc and Shoeburyness. There were also older 1.5kV DC substations that were converted to operate on 6.25kV off of the railway's 33kV supply (Bethnal Green, Stratford, Chadwell Heath, Gidea Park) and DC substations that did not run off the 33kV supply (Ramsden, Rayleigh (old), Prittlewell). Shenfield was totally rebuilt in 1960 off of a new supply and was 25kV from the start (and is still going today as is it's WAML counterpart Rye House from a similar era)

 

[MadMac]

Also check out this website and the maps they have of the original system: http://www.scot-rail.co.uk/page/Glasgow+Suburban+Electrification+1960

There was also a 25kV cable from the WCML feeder station at "Motherwell" (maybe later replaced by Gowkthrapple?) to the Northern Electric Mainline at Coatbridge.

The fact that Finnieston is mentioned as a "sub feeder station" is a bit strange.

That Motherwell FS also supplied from there to Cathcart and the Neilston branch both at 25kV, but the Cathcart circle to Glasgow Central was 6.25kV I think with supply stepped down by the Cathcart substation but there was also an Eglinton Street FS at this time that may have been 6.25kV only until later upgraded? But that's a pure guess.

The Feeder Station at Motherwell was on Platform 3 (5 at the time of opening). The building was still there until relatively recently. The cable to/from Coatbridge was in need of replacement at the time electrification to Gartsherrie was being looked at and was a factor in justifying the electrification.

 

[GraphitProjekt]

According to this old map:



It should be this building. I may be wrong though, it doesn't really look like a 1960s 25kV substation

The cable to/from Coatbridge was in need of replacement at the time electrification to Gartsherrie was being looked at and was a factor in justifying the electrification.

I've heard this as well. It seems clear to me that the point of the cable was to allow for a double outage at Parkhead, Dalreoch would feed from the west and Motherwell would from the east via the cable and a substation at Coatbridge. But even though Motherwell-Gartsherrie was wired, it doesn't seem like the capability to feed onto the North Clyde line is actually directly possible anymore and there is, as far as I can tell, no TSC at Coatbridge. Instead, if both Parkhead circuits go out, the line east of Parkhead is fed from Bathgate. (And for completeness, the section west of is would be fed by Eglinton Street via the Argyle line and Rutherglen + Finnieston TSCs.)

 

[MadMac]

According to this old map:

View attachment 195723

It should be this building. I may be wrong though, it doesn't really look like a 1960s 25kV substation
View attachment 195724


I've heard this as well. It seems clear to me that the point of the cable was to allow for a double outage at Parkhead, Dalreoch would feed from the west and Motherwell would from the east via the cable and a substation at Coatbridge. But even though Motherwell-Gartsherrie was wired, it doesn't seem like the capability to feed onto the North Clyde line is actually directly possible anymore and there is, as far as I can tell, no TSC at Coatbridge. Instead, if both Parkhead circuits go out, the line east of Parkhead is fed from Bathgate. (And for completeness, the section west of is would be fed by Eglinton Street via the Argyle line and Rutherglen + Finnieston TSCs.)

That’s definitely the building. I suspect it may have been rendered redundant when Gowkthrapple came online. There was at least one picture showing equipment being removed from it while the construction in the background was going on a few years ago.

On the question of the North and South being connected: bear in mind that the Whifflet-Sunnyside line was electrified in about 1990.

 

[RailwayRookie]

There’s a feeder at Kings Lynn. However as there are long stretches of single line up there, there is a second 25kV line to keep the other end live in the event of a fault on the OLE over the single track concurrent with an outage at Milton (Cambridge).

Late to the party as I wasnt on this forum for the original conversation, but to add some context.

The feeder station at Kings Lynn went bang at the end of 2024 so the entirety of that line is now fed from Milton Fen until a suitable replacement is put in.

 

[GraphitProjekt]

The feeder station at Kings Lynn went bang at the end of 2024 so the entirety of that line is now fed from Milton Fen until a suitable replacement is put in.

That's quite interesting! So we've been living in N-1 for years on end. If one of Milton Fen's supplies were to be unavailable, can the remaining one supply Litlington to Kings Lynn without requiring service curtailment?

I also heard that Dunton FS on the LTS lines (also sometimes known as Basildon FS) completely blew up a few years ago and a new building had to provided which was recovered off (I think) the former building at Biggleswade TSC which was going spare as all the 1980s equipment was being replaced by ECML PSU

 

[Nicholas Lewis]

On the Anglia lines I think this generally wasn't the case and there were quite a few feeder stations built that were 6.25kV only and went away when the lines were upgraded - Silver Street and Wood Street on the lines to Southbury/Enfield and Chingford respectively. And on the LTS system: Gas Factory Jnc and Shoeburyness. There were also older 1.5kV DC substations that were converted to operate on 6.25kV off of the railway's 33kV supply (Bethnal Green, Stratford, Chadwell Heath, Gidea Park) and DC substations that did not run off the 33kV supply (Ramsden, Rayleigh (old), Prittlewell). Shenfield was totally rebuilt in 1960 off of a new supply and was 25kV from the start (and is still going today as is it's WAML counterpart Rye House from a similar era)

Correct the LTS and parts of the GE/WA were originally run at 6.25kV off the 33kV distribution system formerly used to provide the 1500 V DC system using Scott connected transformers. These transformers allowed a three phase supply to be converted to two phase (2 x 6.25kv) but that rarely presented a fully balanced load as unlikely both supplies had the same load on it.

 

[GRALISTAIR]

The feeder station at Kings Lynn went bang at the end of 2024 so the entirety of that line is now fed from Milton Fen until a suitable replacement is put in.

Without wishing to deviate off topic too much, surely Ely would be a strategically perfect location for a new gridfeeder? It would allow for easier future Electrification of the F2N corridor

 

[GraphitProjekt]

It absolutely would be, and for the Breckland Line. But there aren't really any high voltage power lines around that area other than building a whole new 400kV substation on the Burwell–Walpole line which would be quite an undertaking and not necessarily viable on the grid side.

For context of the scale, I circled Milton FS. I also think a theoretical railway substation on that line may not be independent enough from existing Milton or Kings Lynn.

 

[waverley47]

The feeder station at Kings Lynn went bang at the end of 2024 so the entirety of that line is now fed from Milton Fen until a suitable replacement is put in.

That particular feeder station was always a bit of a black sheep. I remember hearing from engineers back in the late 90's when discussing their experience with certain bits of electrification kit, who were very open about the fact that it was basically held together with duct tape, and someone was semi-permanently 'on call' should it go bang. I'm guessing that it finally blew up?

 

[GraphitProjekt]

Correct the LTS and parts of the GE/WA were originally run at 6.25kV off the 33kV distribution system formerly used to provide the 1500 V DC system using Scott connected transformers. These transformers allowed a three phase supply to be converted to two phase (2 x 6.25kv) but that rarely presented a fully balanced load as unlikely both supplies had the same load on it.

So I presume that when the scott transformers made 2 phases of 6.25kV they were distributed in either direction from the substation because they all had neutral sections added in at those locations and sometimes between them suggesting some level of double-end feeding which would be possible since they would be running off the same 33kV supply. I also think some of the 33kV circuits were handed over to the local electricity board when the lines went to 25kV as there was some mention in 2014 by UKPN about 'finally getting rid of those distribution circuits we have running in railway land'. Also, in the inner areas some of the brackets that originally held up the 33kV cables are now being used to hold the -25kV ATF trunks, which I think is an interesting moment of history repeating itself.

With Scott transformers, would it be possible to use the other phase as a -25kV ATF line on the same circuits so that both phases would be being used at once? Or are they not properly 180degrees apart. I've heard places like Russia use Scott transformers on 25kV lines to aid with phase imbalance but I've never seen an explanation of how that would work

 

[plugwash]

A scott configuration gives you two supplies 90 degrees apart.

 

[59CosG95]

That's quite interesting! So we've been living in N-1 for years on end. If one of Milton Fen's supplies were to be unavailable, can the remaining one supply Litlington to Kings Lynn without requiring service curtailment?

I also heard that Dunton FS on the LTS lines (also sometimes known as Basildon FS) completely blew up a few years ago and a new building had to provided which was recovered off (I think) the former building at Biggleswade TSC which was going spare as all the 1980s equipment was being replaced by ECML PSU

Another fact about Biggleswade TSC - it was actually replaced in the late 90s/early 00s as the neutral section was repositioned for whatever reason.

When I used to live in Peterborough, I lived very close to the Delivery Unit depot and saw the old Dunton FS building in there (this was circa 2018). So the Biggleswade replacement must have happened before then...

 

[GRALISTAIR]

A scott configuration gives you two supplies 90 degrees apart.

I am trying to get my head around that mathematically and engineering wise. I understand (a little - enough to be dangerous!) 3 phase supply so each phase is 2/3 pi radians -120 degrees different. I partially understand ATF feeding- pi radians - 180 degrees different, but 90 degrees? I shall have to google and study!

 

[GraphitProjekt]

Another fact about Biggleswade TSC - it was actually replaced in the late 90s/early 00s as the neutral section was repositioned for whatever reason.

When I used to live in Peterborough, I lived very close to the Delivery Unit depot and saw the old Dunton FS building in there (this was circa 2018). So the Biggleswade replacement must have happened before then...

I wonder how long the LTS lived without 2 of it's supplies, I mean, surely that would have needed service curtailment for at least multiple months - or was there that much headroom on LTS system that they could get away without it.