My proposal to get rid of 3rd rail and convert routes to overhead lines

[Sir Felix Pole]

It would be politically 'brave' to spend billions on re-electrifying the Southern when other major cities and routes in the country have no electrified services at all - hardly a 'levelling-up' agenda. I am hopeful that a relaxation in the regulations will permit a modest infill of the annoying gaps in the Southern third-rail network. In the longer term I think that the 'country-end' of third-rail land will eventually be converted to 25kV AC overhead, but that the inner-suburbs and termini will never be done - the disruption and cost would be immense. The clearances on the former SE&CR routes are also notably tight.

 

[paul1609]

Will there be a demand to replace the rest of the country's third rail for anything other than "safety"? Especially since it is grandfathered in and accepted.

With the end of red zone working and with modern power electrics making it easier to do power isolations in the event of incidents has anybody risk assessed the remaining dc hazards? Id have thought in rural areas like the Uckfield Line it would be quite low.

 

[HSTEd]

With the end of red zone working and with modern power electrics making it easier to do power isolations in the event of incidents has anybody risk assessed the remaining dc hazards? Id have thought in rural areas like the Uckfield Line it would be quite low.

I tried to get a PhD doing this sort of thing, unfortunately I missed my 2.1 so lost the place and had to do another masters degree!

Pity, that looked like a fun project.

 

[hwl]

With the end of red zone working and with modern power electrics making it easier to do power isolations in the event of incidents has anybody risk assessed the remaining dc hazards? Id have thought in rural areas like the Uckfield Line it would be quite low.

The RSSB (with Arup) have been doing that since last year.

 

[A0wen]

Will there be a demand to replace the rest of the country's third rail for anything other than "safety"? Especially since it is grandfathered in and accepted.

Yes, efficiency comes into play, don't forget a number of the current 3rd rail EMUs (Electrostars and Desiros) have their performance limited on DC but not on AC.

So performance on 3rd rail is poorer when compared to OHLE.

If it were DC OHLE vs AC OHLE, I suspect it might not be a consideration - several countries in Europe run DC OHLE at 1500v or 3000v with 25kv AC on their network. 3rd rail is practically limited to 750v DC (yes, I know it's *possible* to go higher, but it's not common and has its own problems).

 

[Iliketrains123]

Yes, efficiency comes into play, don't forget a number of the current 3rd rail EMUs (Electrostars and Desiros) have their performance limited on DC but not on AC.

So performance on 3rd rail is poorer when compared to OHLE.

If it were DC OHLE vs AC OHLE, I suspect it might not be a consideration - several countries in Europe run DC OHLE at 1500v or 3000v with 25kv AC on their network. 3rd rail is practically limited to 750v DC (yes, I know it's *possible* to go higher, but it's not common and has its own problems).

Yes, I know OHLE is the superior standard. But where if the value for money in converting 36% of Britain's electrified network?
Sure the SWML could be converted but the government will never ever pay the astronomical costs of a full conversion. Nor will they outlaw existing 3rd rail as it is not prohibitively unsafe in it's current incarnation.

 

[A0wen]

Yes, I know OHLE is the superior standard. But where if the value for money in converting 36% of Britain's electrified network?
Sure the SWML could be converted but the government will never ever pay the astronomical costs of a full conversion. Nor will they outlaw existing 3rd rail as it is not prohibitively unsafe in it's current incarnation.

Given part of the push for carbon neutrality will also be reducing electricity consumption, then converting to AC OHLE as 3rd rail equipment reaches life expiry may be the way forward.

Most if not all new EMUs are dual standard capable, so there isn't even a risk of having to replace rolling stock unnecessarily. The only units where conversion would be difficult or expensive are those dating from BR days which will be nearing their end of life in any case.

 

[Mag_seven]

Moderator note: Split from https://www.railforums.co.uk/threads/lewisham-egress-2019.209846

My big take is that 3rd Rail is rubbish, we really shouldn’t install any more, and should have a presumption of moving to OLE (though of course places like Lewisham would be on the very late to never stages of that process....)

Thread reopened.

 

[zwk500]

Given part of the push for carbon neutrality will also be reducing electricity consumption, then converting to AC OHLE as 3rd rail equipment reaches life expiry may be the way forward.

This is only a viable option if either A. The 3rd rail network comes up for renewal in a convenient pattern to permit extentions of the continuous OLE network or B. Some solution is found to negate the massive inefficiencies (cost & Power) of OLE islands in the middle of the 3rd rail network.

I can see Basingstoke--Winchester-Weymouth being converted as part of Reading-Basingstoke-Salisbury-Romsey-Eastleigh/Southampton being wired for freight access to Southampton. Possibly some of the HS1 connections in east kent as well. But not wholesale conversion of the network given they are actively looking at putting in additional 3rd rail to meet decarbonisation goals. You'll save more carbon going Diesel > 3rd Rail than you will 3rd Rail > OLE.

Most if not all new EMUs are dual standard capable, so there isn't even a risk of having to replace rolling stock unnecessarily. The only units where conversion would be difficult or expensive are those dating from BR days which will be nearing their end of life in any case.

Yes almost all units running atm have provision for Dual voltage, but it's still a MASSIVE cost to fit the equipment in the Southern Region fleet.

 

[Bald Rick]

This is only a viable option if either A. The 3rd rail network comes up for renewal in a convenient pattern to permit extentions of the continuous OLE network or B. Some solution is found to negate the massive inefficiencies (cost & Power) of OLE islands in the middle of the 3rd rail network.

And C - someone comes up with a solution that cracks the nut of a switch from one system to the other at the end of the construction period that does not involve any lengthy time of both systems being in use.

 

[A0wen]

And C - someone comes up with a solution that cracks the nut of a switch from one system to the other at the end of the construction period that does not involve any lengthy time of both systems being in use.

Out of interest, is the problem that the DC supply is 3rd rail rather than OHLE?

Asking because on the continent and going back some time in Manchester there were places where 25kv AC and 1500v DC OHLE systems co-existed.

How does it work where 3rd rail and OHLE currently co exist e.g. Euston, Farringdon - City Thameslink?

 

[Bald Rick]

Out of interest, is the problem that the DC supply is 3rd rail rather than OHLE?

Asking because on the continent and going back some time in Manchester there were places where 25kv AC and 1500v DC OHLE systems co-existed.

How does it work where 3rd rail and OHLE currently co exist e.g. Euston, Farringdon - City Thameslink?

It’s a problem with AC vs D.C. and arrangements for the return current, both of which are via the running rails.

AC return current goes via ‘earth’ and is nominally at 0 volts.
D.C. return is a bonded system, which is slightly above 0 volts.

Therefore if you have the two together the D.C. return will try to find its way back to the AC substation. This can cause all many of problems with stray currents to the signalling, bonding, and even to metallic items near the railway. I have recounted before how stray DC currents caused the corrosion of the reinforcement in the concrete foundations of the OLE at Acton Central, to the extent that the masts could be pulled out quite easily.

The problem is exacerbated if there is a lot of D.C. traffic compared to the AC. For most of the locations where AC and D.C. share tracks, this isn’t an issue, and the stretches concerned relatively short.

The one place it is a major issue is the Thameslink core. There are some very expensive heavy duty electrical switching arrangements that, in effect, automatically change the return current path for each train as it moves along each section of track between Blackfriars and Farringdon (and vice versa). Even then there is still a lot of return current flowing around, and racing across the track circuit block joints is common, which means they are replaced much more frequently than would otherwise be the case.

Given the scale of traffic on the SWML though, it certainly would be an issue, if not quite as bad as on
Thameslink.

 

[Nicholas Lewis]

It’s a problem with AC vs D.C. and arrangements for the return current, both of which are via the running rails.

AC return current goes via ‘earth’ and is nominally at 0 volts.
D.C. return is a bonded system, which is slightly above 0 volts.

Therefore if you have the two together the D.C. return will try to find its way back to the AC substation. This can cause all many of problems with stray currents to the signalling, bonding, and even to metallic items near the railway. I have recounted before how stray DC currents caused the corrosion of the reinforcement in the concrete foundations of the OLE at Acton Central, to the extent that the masts could be pulled out quite easily.

The problem is exacerbated if there is a lot of D.C. traffic compared to the AC. For most of the locations where AC and D.C. share tracks, this isn’t an issue, and the stretches concerned relatively short.

The one place it is a major issue is the Thameslink core. There are some very expensive heavy duty electrical switching arrangements that, in effect, automatically change the return current path for each train as it moves along each section of track between Blackfriars and Farringdon (and vice versa). Even then there is still a lot of return current flowing around, and racing across the track circuit block joints is common, which means they are replaced much more frequently than would otherwise be the case.

Given the scale of traffic on the SWML though, it certainly would be an issue, if not quite as bad as on
Thameslink.

DC is an unearthed system so as to encourage majority of current to return via running rails. The running rails on AC are earthed at every structure as well as at metallic overbridges, signal gantries and anything else that metallic within a few meters of the line. Thus the running rails are in effect earthed which you don't want in a DC system. The primary reason being if DC gets into a metallic structure and then escapes to natural earth corrosion will occur at that point and potentially weaken the structure over time. I was involved with original Snow Hill reopening and electrification and in mid 80's we undertook extensive testing in conjunction with LUL between Kentish Town and Moorgate to establish how much DC traction current would evade running rail return. Here we discovered that the superstructure of Smithfield market was made of cast iron and as cantilever arms were secured to it they had to be bonded to running rails so it was acting as giant earth farm that the DC current wanted to go through as to where it went we couldn't determine but it was 100's of Amps so it couldn't be permitted. New structures were installed to remove direct connections but ultimately it proved insufficient for the long term and as Bald Rick refers to above we created an island of 3rd rail separated from the main Southern DC system with a switching system driven by the trains position. This had a few faux pars over the years but has stood the test of time.

This isn't the only area of DC / AC interface in the UK. At Dollands Moor we isolated the OLE system from ET by the use of Isolation transformers basically 25kv:25kv. Then by having an earth wire system on all the structures and careful management of design the paths for DC current were minimised.

So with space in the Basingstoke area it would be perfectly feasible to provide a c/o over section East of Basingstoke. Whilst i wouldn't advocate wholesale re-electrification of the Southern area it would have made sense to have done west of Basingstoke given freight flows and XC traffic. Also as the Bournemouth Electrification traction system was low power and needed reinforcing it could have been converted to 25kv in lieu of doing that. This hasn't happened and it has been reinforced so given DofT propensity for BiMode now I doubt it will ever been changed now.

 

[21C101]

Then there is the issue if immunising the signalling to both a.c and d.c so it still works during the transition period and the civil engineering clearance works (bridge replacements etc).

Personally I can't see a.c. going in on the SWML other than the Up end of Basingstoke to Worting Junction and in the Redbridge area as part of a Basingstoke/Southampton - Salisbury - Bath/Exeter electrification (as well as, obviously, Basingstoke - Reading and Didcot - Coventry).

Wiring Basingstoke to Worting Junction would allow a.c. electric hauling of container freight to Southampton Docks via Andover and Romsey.

Redoubling of the Laverstock Loop and reinstatement of the through roads at Andover, along with southbound loop at Romsey and a northbound one just south of Dean, would be useful for providing looping for freight to allow passenger trains to pass, and in the case of Romsey might be useful for passenger services as well, reinstating the down loop platform.

If needed, mile long freight loops would also be relatively straightforward in engineering terms west of Grateley, as the formation was widened to three track for two miles west of Grateley for the Amesbury branch and is a straight formation in cutting where retaining walls could be used to make space for a fourth track where the formation is not already wide enough. If the loops started 1/4 mile west of Grateley the only bridge to be dealt with is a single, long disused, occupation crossing over the track

 

[Mikey C]

It would be politically 'brave' to spend billions on re-electrifying the Southern when other major cities and routes in the country have no electrified services at all - hardly a 'levelling-up' agenda. I am hopeful that a relaxation in the regulations will permit a modest infill of the annoying gaps in the Southern third-rail network. In the longer term I think that the 'country-end' of third-rail land will eventually be converted to 25kV AC overhead, but that the inner-suburbs and termini will never be done - the disruption and cost would be immense. The clearances on the former SE&CR routes are also notably tight.

Agreed it would be unbelievably expensive and disruptive, can you imagine wiring Clapham Junction?

The line to Weymouth west of Basingstoke would make sense to convert at some stage, especially as past Bournemouth it was done on the cheap and can only cope with shorter trains

 

[Nicholas Lewis]

Agreed it would be unbelievably expensive and disruptive, can you imagine wiring Clapham Junction?

The LBSCR wired it at 6.6kV and had not the LSWR not become dominant in the new Southern Railway leadership it would have been extended further but LSWR was already invested in 3rd Rail so it was removed. Mind you they managed the transition from AC to DC without any loss of service during 1928 and 29.

 

[Joshua483]

I would just string up the 3rd rail onto the OLE masts in Low space areas between bridges or tunnels and the OLE.

 

[HSTEd]

Agreed it would be unbelievably expensive and disruptive, can you imagine wiring Clapham Junction?

The line to Weymouth west of Basingstoke would make sense to convert at some stage, especially as past Bournemouth it was done on the cheap and can only cope with shorter trains

You would probably have to use some exceptionally long gantries.
Trying to put posts in would be a nightmare.

 

[Bald Rick]

So with space in the Basingstoke area it would be perfectly feasible to provide a c/o over section East of Basingstoke.

Yep I agree with that. The issue is more about the rolling stock.

If (and it’s a big if) west of Basingstoke was switched to 25kV AC, there would be a period before the new stuff is fully brought into service where both systems were live. During this time, you would want as much as possible on the AC system to limit problems with the D.C. return. That means having a large fleet of (reliable) dual voltage trains in service (or ready to) immediately the AC goes live.

The only similar example I can think of is when the GEML switched from D.C. to AC over a long weekend in 1950. Whilst the infrastructure switch over was done in a few days, much of the rolling stock was converted from one system to the other and that took months. So there was a combination of new rolling stock and reduced services for a couple of months whilst it was done.

 

[A0wen]

Yep I agree with that. The issue is more about the rolling stock.

If (and it’s a big if) west of Basingstoke was switched to 25kV AC, there would be a period before the new stuff is fully brought into service where both systems were live. During this time, you would want as much as possible on the AC system to limit problems with the D.C. return. That means having a large fleet of (reliable) dual voltage trains in service (or ready to) immediately the AC goes live.

The only similar example I can think of is when the GEML switched from D.C. to AC over a long weekend in 1950. Whilst the infrastructure switch over was done in a few days, much of the rolling stock was converted from one system to the other and that took months. So there was a combination of new rolling stock and reduced services for a couple of months whilst it was done.

BIB - there's a few class 319s knocking around, nice proven tech on both AC and DC - job done ! (removes tongue from cheek).

 

[30907]

Yep I agree with that. The issue is more about the rolling stock.

If (and it’s a big if) west of Basingstoke was switched to 25kV AC, there would be a period before the new stuff is fully brought into service where both systems were live. During this time, you would want as much as possible on the AC system to limit problems with the D.C. return. That means having a large fleet of (reliable) dual voltage trains in service (or ready to) immediately the AC goes live.

The only similar example I can think of is when the GEML switched from D.C. to AC over a long weekend in 1950. Whilst the infrastructure switch over was done in a few days, much of the rolling stock was converted from one system to the other and that took months. So there was a combination of new rolling stock and reduced services for a couple of months whilst it was done.

1960? It helped that there was a whole load of new electrification in hand, so you could commission the new units early.
In the case of the LSW, you would ideally have a load of units you wanted to scrap/replace with new at the time for switching on the OLE. I'll leave someone else to predict possible dates (or decades?)...

 

[AM9]

Yep I agree with that. The issue is more about the rolling stock.

If (and it’s a big if) west of Basingstoke was switched to 25kV AC, there would be a period before the new stuff is fully brought into service where both systems were live. During this time, you would want as much as possible on the AC system to limit problems with the D.C. return. That means having a large fleet of (reliable) dual voltage trains in service (or ready to) immediately the AC goes live.

The only similar example I can think of is when the GEML switched from D.C. to AC over a long weekend in 1950. Whilst the infrastructure switch over was done in a few days, much of the rolling stock was converted from one system to the other and that took months. So there was a combination of new rolling stock and reduced services for a couple of months whilst it was done.

The only 1500VDC that needed converting to ac was the GEML lines to Shenfield, Chelmsford and Southend Victoria, mainly served by 92 3-car EMUs (what became class AM6) and 32 4-car outer suburban EMUs (what became class AM7). By the time that the power was switched over to ac in 1960, there were plenty of new ac only EMUs waiting for their bits of the Great Eastern lines to be electrified from scratch at 6.25/25kV or 25kV ac. For example, there were 112 class AM2 4-car units procured for the LT&S lines out of Fenchurch St, some of the 71 sets of class AM5/1 & AM5/2 3 & 4-car units destined for the Chingford, Enfield and Bishops Stortford lines, and 33 of the class AM8 4-car units that were waiting for the mainline east of Chelmsford to be wired.

 

[JohnElliott]

Yep I agree with that. The issue is more about the rolling stock.

If (and it’s a big if) west of Basingstoke was switched to 25kV AC, there would be a period before the new stuff is fully brought into service where both systems were live. During this time, you would want as much as possible on the AC system to limit problems with the D.C. return. That means having a large fleet of (reliable) dual voltage trains in service (or ready to) immediately the AC goes live.

The only similar example I can think of is when the GEML switched from D.C. to AC over a long weekend in 1950. Whilst the infrastructure switch over was done in a few days, much of the rolling stock was converted from one system to the other and that took months. So there was a combination of new rolling stock and reduced services for a couple of months whilst it was done.

I may be oversimplifying, but wouldn't the approach be to fit the 444s and 450s with AC pickups and transformers before or during the installation of OHLE?

 

[mr_jrt]

That seemed to me to be the rather obvious solution as well. Having a unit converted in advance to a dual voltage unit running on DC isn't going to do any harm. Start nice and early so you can do so at a comfortable pace and you can be doing it whilst all the other preparatory works (like re-siting the signalling, piling for the masts, etc) happens.

Actually, would it make more sense to switch to ETCS in-cab signalling and do away with the lineside equipment rather than go though the process of moving it around? That could even be done in an earlier project to reduce the scope of the conversion work.

 

[21C101]

You would probably have to use some exceptionally long gantries.
Trying to put posts in would be a nightmare.

The LBSC installed huge cantilever gantries there that spanned all four of their tracks, with a support on one side only.

 

[zwk500]

I may be oversimplifying, but wouldn't the approach be to fit the 444s and 450s with AC pickups and transformers before or during the installation of OHLE?

Yes it would, but you would then need to send each converted unit to OLE land for a period of fault-free running to accept the modification into service.

That seemed to me to be the rather obvious solution as well. Having a unit converted in advance to a dual voltage unit running on DC isn't going to do any harm. Start nice and early so you can do so at a comfortable pace and you can be doing it whilst all the other preparatory works (like re-siting the signalling, piling for the masts, etc) happens.

Actually, would it make more sense to switch to ETCS in-cab signalling and do away with the lineside equipment rather than go though the process of moving it around? That could even be done in an earlier project to reduce the scope of the conversion work.

It would make more sense to switch to in-cab ETCS so you can place OLE support without worrying about signal sighting and the like. It would however mean designing ETCS to be DC and AC compatible, which is an additional cost (ETCS uses loops and beacons in the 4ft, not sure if they'd be affected by DC currents). It's all a giant set of compromises.

 

[paul1609]

Yep I agree with that. The issue is more about the rolling stock.

If (and it’s a big if) west of Basingstoke was switched to 25kV AC, there would be a period before the new stuff is fully brought into service where both systems were live. During this time, you would want as much as possible on the AC system to limit problems with the D.C. return. That means having a large fleet of (reliable) dual voltage trains in service (or ready to) immediately the AC goes live.

The only similar example I can think of is when the GEML switched from D.C. to AC over a long weekend in 1950. Whilst the infrastructure switch over was done in a few days, much of the rolling stock was converted from one system to the other and that took months. So there was a combination of new rolling stock and reduced services for a couple of months whilst it was done.

How far does the West of Basingstoke conversion go? Are you going to dual electrify Eastleigh and St Denys so Portsmouth and Southern services can change over?

 

[Bald Rick]

I may be oversimplifying, but wouldn't the approach be to fit the 444s and 450s with AC pickups and transformers before or during the installation of OHLE?

Possibly, but then you have to figure out how long it takes to do each unit, how many units you are willing to have being modified at any one time, and how long it will take to do the fleet.

How far does the West of Basingstoke conversion go? Are you going to dual electrify Eastleigh and St Denys so Portsmouth and Southern services can change over?

No idea!

 

[mr_jrt]

Rather than switching back and forth, it might make more sense to start at the Weymouth end and work your way east - that way you only have one "temporary" changeover point to worry about at any given time until you reach St Denys, where you could have something a bit more longer term in place on 3/4 until you were ready to start switching over towards Portsmouth. As areas switch over the recovered assets could then be used elsewhere (i.e. Uckfield, Marshlink, the North Downs, or even just renewals).

 

[JamesT]

I may be oversimplifying, but wouldn't the approach be to fit the 444s and 450s with AC pickups and transformers before or during the installation of OHLE?

Rather than trying to convert DC units into dual voltage, wouldn’t it be easier to cascade in stock that can already do it, e.g. Electrostars? Given the length of time a conversion programme would take, newer DC stock would be used to replace older end of life stock until the programme finished or they hit end of life in turn. The areas that were donors for the dual voltage would get new AC stock instead. (I assume there must be some cost involved in having a train capable of both, so if you’re only going to have one form of electrification in the future you just buy trains capable of that)