How would you use electric trains on Harrow-On-The-Hill to Amersham?

[Energy]

Split from this thread, how would you use electric trains on this section which both Chiltern and London Underground run on?

You can have 4th rail and OHLE on the same track which won't require dual voltage trains

Indeed. Hence why I said dual-voltage stock was "easier", rather than the only way.

Dual-electrification of the line is difficult, and NR's West Midlands & Chilterns Route Study's Technical Appendices has the following on the matter (my bold):

To maximise the benefits of the Chiltern capacity and connectivity package, analysis has considered potential electrification options for the line between Aylesbury and Marylebone (via Amersham) which does not form part of the main line scope. The line includes infrastructure between Amersham and Harrow on the Hill that is owned by London Underground Limited (LUL) but shared operationally with Chiltern Railways (highlighted in red in Figure 4.7). Electrification options have been identified for this line, taking into account the DC 4th rail electrification on the section between Amersham and Harrow-on-the-Hill. As it would be challenging to operate and maintain an OLE (AC electrification) system alongside the 4th rail section, the introduction of AC electrification on part of the route (excluding the section between Amersham and Harrow on-the-Hill) has been identified as a potential option. It is proposed that this should be considered as part of any future development work assessing main line electrification. Key to this development work will be understanding the impact of different options on the future Chiltern rolling stock fleet. Progressing the option to part electrify (AC) the Aylesbury – Amersham – Marylebone route would require assessment of dual voltage or bi-mode rolling stock options which could operate as a combined fleet on both this route and the main line. The potential to use an independently powered EMU has also been considered, with initial modelling work supporting further development of this option.

Difficult but not impossible, it will depend on how they view the use of dual voltage or bimodes. I believe this section is 750v (it got upped from 630v as part of the 4 lines modernisation) so they should be able to use fairly standard dual voltage EMUs.

 

[edwin_m]

I'm not sure how "fairly standard" they would be for fourth rail plus AC. When in DC mode the return current would have to be directed to centre rail shoes rather than simply returned through the wheels. The shoes themselves might not be too difficult but would this require extra cables through the train to connect them to the traction package? I'm not sure also whether this might not cause some earthing issues when in AC mode, perhaps requiring some switching of the return current between the two, where the "safe" setting might be different depending on system so it couldn't be a traditional failsafe. I think we need some traction engineers to advise on all this!

Also Underground trains don't have bus cables, each power car feeds only from its own shoes, but the shoes on a third rail (or dual voltage) unit are electrically connected within the unit. Perhaps the section in question could be engineered to work with either type as it doesn't have tunnels where it might be considered a fire hazard, or transitions to NR third rail electrification where a gap is needed longer than the longest distance between electrically connected shoes.

The other option, ORR permitting, would be to forget 25kV and convert the whole lot to Network Rail standard third rail and signaling. The centre rail could bonded to the running rails so S Stock could still operate as it does on other shared sections. That would allow standard dual-voltage stock but would involve binning a lot of fairly new signaling.

 

[Bletchleyite]

Could you convert it to 25kV after Harrow-on-the-Hill (with the 25kV also going into Marylebone) and retrofit a pantograph and transformer to the S8s?

 

[HSTEd]

Also Underground trains don't have bus cables, each power car feeds only from its own shoes, but the shoes on a third rail (or dual voltage) unit are electrically connected within the unit. Perhaps the section in question could be engineered to work with either type as it doesn't have tunnels where it might be considered a fire hazard, or transitions to NR third rail electrification where a gap is needed longer than the longest distance between electrically connected shoes.

My understanding is that on the Class 313 at least this is not the case.
My understanding is that in third rail mode they do not use the bus line, to comply with LU practice on the Moorgate and other routes.

Could you convert it to 25kV after Harrow-on-the-Hill (with the 25kV also going into Marylebone) and retrofit a pantograph and transformer to the S8s?

Where would the pantograph and transformer go?
The trains weren't designed for it - London Underground trains don't even have a power bus along them for traction current.

 

[Bletchleyite]

My understanding is that on the Class 313 at least this is not the case.
My understanding is that in third rail mode they do not use the bus line, to comply with LU practice on the Moorgate and other routes.

I believe that's correct, though other PEP stock does have a DC bus line (e.g. Merseyrail units; there is a small sign on the shoegear pointing this out).

 

[Philip Phlopp]

Could you convert it to 25kV after Harrow-on-the-Hill (with the 25kV also going into Marylebone) and retrofit a pantograph and transformer to the S8s?

Yes and No. The electronics fit on the S8 supports 25kV AC OLE, the bodyshell doesn't.

It's also worth noting that the S8 stock is fully 750V DC compatible.

 

[Bletchleyite]

Yes and No. The electronics fit on the S8 supports 25kV AC OLE, the bodyshell doesn't.

Is there a reason the bodyshell could not be modified, or at worst a vehicle could be replaced with a new one with a pantograph well perhaps?

 

[popeter45]

isnt overground 4th rail just 3ed rail with the return rail bonded to the main tracks so 3ed rail units can also run on them (e.g. watford DC line?)
i would just full dual electrification to harrow on the hill then OHEL to marlybone

 

[MarkyT]

I'm not sure how "fairly standard" they would be for fourth rail plus AC. When in DC mode the return current would have to be directed to centre rail shoes rather than simply returned through the wheels. The shoes themselves might not be too difficult but would this require extra cables through the train to connect them to the traction package? I'm not sure also whether this might not cause some earthing issues when in AC mode, perhaps requiring some switching of the return current between the two, where the "safe" setting might be different depending on system so it couldn't be a traditional failsafe. I think we need some traction engineers to advise on all this!

Also Underground trains don't have bus cables, each power car feeds only from its own shoes, but the shoes on a third rail (or dual voltage) unit are electrically connected within the unit. Perhaps the section in question could be engineered to work with either type as it doesn't have tunnels where it might be considered a fire hazard, or transitions to NR third rail electrification where a gap is needed longer than the longest distance between electrically connected shoes.

The other option, ORR permitting, would be to forget 25kV and convert the whole lot to Network Rail standard third rail and signaling. The centre rail could bonded to the running rails so S Stock could still operate as it does on other shared sections. That would allow standard dual-voltage stock but would involve binning a lot of fairly new signaling.

An interesting option. 4-rail rolling stock, including Bakerloo tube trains and S7s, seems to work perfectly well using this arrangement on other shared sections, so I see no fundamental reason why it couldn't also apply on the shared outer reaches of the Metropolitan line. The new LUL signalling system is likely to use axle counters alongside beacon and odometry based positioning, so there shouldn't be any compatibility issues or complexity with traction return current flowing through the running rails as well as the centre return rail, and paralleling them all up periodically should reduce overall loop resistance and increase efficiency for all trains. The new Chiltern Aylesbury line fleet could also be battery assisted to limit the extent of new ground-level conductor. I would think a very good case could be made for new conductor rail where the Chiltern pair effectively shares the Metropolitan corridor, which is the entirety of the route from Harrow as far as Finchley Road, and using DC for this section would avoid any AC compatibility issues with the nearby LUL equipment and procedures. The remainder of the route into the terminus is also highly segregated with little chance of public ingress, and best modern practice might be employed for fine-grained local and control centre isolation switching of any proposed new 3rd rail for staff safety and emergency purposes, and having batteries on board the trains could allow isolations for maintenance work to take place while trains could continue to move under battery power. On the open-air sections north of Amersham, battery power could allow any new conductor rail provision to be confined to limited plain line areas well away from station platforms and any remaining level crossings, except perhaps at Aylesbury stations where short switched sections could be provided for static layover charging in sidings and platforms, only energised on demand when a train was present. The 3/4 rail system could also be used with dual-voltage rolling stock for AC wiring north of Amersham and possibly south of Neasden if required for the High Wycombe line.

 

[edwin_m]

isnt overground 4th rail just 3ed rail with the return rail bonded to the main tracks so 3ed rail units can also run on them (e.g. watford DC line?)

It is on the Network Rail sections where third rail stock also operates: DC lines plus the Richmond and Wimbledon branches of the District. But everywhere else (on the surface and underground) the fourth rail is isolated from the running rails, so the running rails only carry track circuit currents not traction return.

An interesting option. 4-rail rolling stock, including Bakerloo tube trains and S7s, seems to work perfectly well using this arrangement on other shared sections, so I see no fundamental reason why it couldn't also apply on the shared outer reaches of the Metropolitan line. The new LUL signalling system is likely to use axle counters alongside beacon and odometry based positioning, so there shouldn't be any compatibility issues or complexity with traction return current flowing through the running rails as well as the centre return rail, and paralleling them all up periodically should reduce overall loop resistance and increase efficiency for all trains. The new Chiltern Aylesbury line fleet could also be battery assisted to limit the extent of new ground-level conductor. I would think a very good case could be made for new conductor rail where the Chiltern pair effectively shares the Metropolitan corridor, which is the entirety of the route from Harrow as far as Finchley Road, and using DC for this section would avoid any AC compatibility issues with the nearby LUL equipment and procedures. The remainder of the route into the terminus is also highly segregated with little chance of public ingress, and best modern practice might be employed for fine-grained local and control centre isolation switching of any proposed new 3rd rail for staff safety and emergency purposes, and having batteries on board the trains could allow isolations for maintenance work to take place while trains could continue to move under battery power. On the open-air sections north of Amersham, battery power could allow any new conductor rail provision to be confined to limited plain line areas well away from station platforms and any remaining level crossings, except perhaps at Aylesbury stations where short switched sections could be provided for static layover charging in sidings and platforms, only energised on demand when a train was present. The 3/4 rail system could also be used with dual-voltage rolling stock for AC wiring north of Amersham and possibly south of Neasden if required for the High Wycombe line.

That's kind of what I was thinking, but I'd put a changeover to 25kV north of Amersham as DC systems get increasingly inefficient on faster and more rural routes, and it would be harder to justify a less safe system to ORR. If at all possible I'd also use 25kV out of Marylebone towards High Wycombe and as far as possible towards Harrow. The fourth rail system is probably less problematic than third rail for earthing and stray current in proximity to 25kV, because the return is in a proper insulated rail not the running rails. These issues were overcome on the Tilbury line many years ago without apparent problems.

 

[Thomas31]

Instead of faffing about fitting pantographs to the S8 trains etc, why not just re-extend the Met to Aylesbury and reroute all Chiltern services via High Wycombe?

 

[bramling]

Split from this thread, how would you use electric trains on this section which both Chiltern and London Underground run on?


Difficult but not impossible, it will depend on how they view the use of dual voltage or bimodes. I believe this section is 750v (it got upped from 630v as part of the 4 lines modernisation) so they should be able to use fairly standard dual voltage EMUs.

From an operational perspective, I’d be inclined to go with:

* A hybrid 750V DC 4th rail system from Harrow to Amersham similar to that used on the DC Lines and Wimbledon/Richmond branches. This would avoid Chiltern trains having to have negative shoegear. For simplicity it could perhaps be prudent to convert Watford and Chesham as well.

* Extend 3rd rail to Great Missenden and have AC/DC interfaces there and Harrow. This may require segregating Harrow platforms 1 and 2 from LU and using them for Chiltern only, which would require some track layout changes to maintain a reversing facility for LU trains. I suspect there may be some bridges in the Harrow area which would require costly rebuilds.

Others would have to comment on any technical impediments to this, but this setup would appear to be the most simple. Having Chiltern trains carry around a set of negative shoe gear is to my mind undesirable and should be avoided, and having S stocks convey AC equipment even more undesirable for a number of reasons.

 

[Philip Phlopp]

Is there a reason the bodyshell could not be modified, or at worst a vehicle could be replaced with a new one with a pantograph well perhaps?

Ask Bombardier. I've no idea if it's possible and what further modifications may be required, such as the provision of an electrical bus to connect the output from the transformer to the existing traction equipment. The cost of conversion for the rolling stock plus the cost of converting the line itself is not going to be cheap. Project Thor (the Voyager bi-mode plan) was sunk mainly on the cost of converting one vehicle to carry a pantograph/transformer, so making the S8 stock 25kV AC OLE compatible is not the correct direction to be looking.

I'm not sure how "fairly standard" they would be for fourth rail plus AC. When in DC mode the return current would have to be directed to centre rail shoes rather than simply returned through the wheels. The shoes themselves might not be too difficult but would this require extra cables through the train to connect them to the traction package? I'm not sure also whether this might not cause some earthing issues when in AC mode, perhaps requiring some switching of the return current between the two, where the "safe" setting might be different depending on system so it couldn't be a traditional failsafe. I think we need some traction engineers to advise on all this!

Also Underground trains don't have bus cables, each power car feeds only from its own shoes, but the shoes on a third rail (or dual voltage) unit are electrically connected within the unit. Perhaps the section in question could be engineered to work with either type as it doesn't have tunnels where it might be considered a fire hazard, or transitions to NR third rail electrification where a gap is needed longer than the longest distance between electrically connected shoes.

The other option, ORR permitting, would be to forget 25kV and convert the whole lot to Network Rail standard third rail and signaling. The centre rail could bonded to the running rails so S Stock could still operate as it does on other shared sections. That would allow standard dual-voltage stock but would involve binning a lot of fairly new signaling.

There's no need to bin the Thales SelTrac CBTC signalling system - modern rolling stock is capable of handling two or more signalling systems. The Class 345 stock manages three - TPWS, ETCS and Siemens CBTC (just about). The Thales SelTrac system is a modern immunised system which is technically capable of operating in exactly the same manner regardless of whether the electrification system is 750V DC third-rail, +500/-250V DC fourth-rail (and 25kV AC, which I mention purely for the purposes of the changeover areas). The rolling stock itself is capable of operating on both third-rail and fourth-rail electrification with no significant issues.

The correct answer is to modify the shared section to NR compatible 750V DC, convert fully to NR specification 750V DC third-rail (and perhaps a few miles more to suit changeover locations for both options, as bramling suggests) whilst retaining the LUL specification Thales SelTrac CBTC signalling, and fit the new dual-voltage Chiltern EMU fleet with SelTrac CBTC signalling.

 

[bramling]

Ask Bombardier. I've no idea if it's possible and what further modifications may be required, such as the provision of an electrical bus to connect the output from the transformer to the existing traction equipment. The cost of conversion for the rolling stock plus the cost of converting the line itself is not going to be cheap. Project Thor (the Voyager bi-mode plan) was sunk mainly on the cost of converting one vehicle to carry a pantograph/transformer, so making the S8 stock 25kV AC OLE compatible is not the correct direction to be looking.



There's no need to bin the Thales SelTrac CBTC signalling system - modern rolling stock is capable of handling two or more signalling systems. The Class 345 stock manages three - TPWS, ETCS and Siemens CBTC (just about). The Thales SelTrac system is a modern immunised system which is technically capable of operating in exactly the same manner regardless of whether the electrification system is 750V DC third-rail, +500/-250V DC fourth-rail (and 25kV AC, which I mention purely for the purposes of the changeover areas). The rolling stock itself is capable of operating on both third-rail and fourth-rail electrification with no significant issues.

The correct answer is to modify the shared section to NR compatible 750V DC, convert fully to NR specification 750V DC third-rail (and perhaps a few miles more to suit changeover locations for both options, as bramling suggests) whilst retaining the LUL specification Thales SelTrac CBTC signalling, and fit the new dual-voltage Chiltern EMU fleet with SelTrac CBTC signalling.

Out of interest, does the hybrid DC system carry any disbenefits compared to a pure 4-rail system in this setup? I know 4 rails is preferred in tunnels to avoid corrosion issues, however presumably this isn’t a major issue in the open?

 

[MarkyT]

It is on the Network Rail sections where third rail stock also operates: DC lines plus the Richmond and Wimbledon branches of the District. But everywhere else (on the surface and underground) the fourth rail is isolated from the running rails, so the running rails only carry track circuit currents not traction return.

That's kind of what I was thinking, but I'd put a changeover to 25kV north of Amersham as DC systems get increasingly inefficient on faster and more rural routes, and it would be harder to justify a less safe system to ORR. If at all possible I'd also use 25kV out of Marylebone towards High Wycombe and as far as possible towards Harrow. The fourth rail system is probably less problematic than third rail for earthing and stray current in proximity to 25kV, because the return is in a proper insulated rail not the running rails. These issues were overcome on the Tilbury line many years ago without apparent problems.

If Marylebone - High Wycombe and beyond at 25kV was possible, then yes a dual voltage Aylesbury line fleet with wires north of Amersham would be best. I'd take wires about a mile north of the junction at Neasden to changeover to 3rd/4th rail on the move just south of Wembley Park. This would avoid any further potential Bridgework between there and Harrow, so only around 6 single track miles of additional ground-level conductor provision required, and it should be possible to feed the power from existing LUL DC substations through this area, with some strengthening work carried as necessary of course.

 

[Philip Phlopp]

Out of interest, does the hybrid DC system carry any disbenefits compared to a pure 4-rail system in this setup? I know 4 rails is preferred in tunnels to avoid corrosion issues, however presumably this isn’t a major issue in the open?

Hybrid DC and indeed third-rail DC in general needs a little more attention paid to management of return and stray electrical currents, but you should be paying attention to those issues anyway during design and managing it during routine inspections.

 

[MarkyT]

Extend 3rd rail to Great Missenden and have AC/DC interfaces there and Harrow. This may require segregating Harrow platforms 1 and 2 from LU and using them for Chiltern only, which would require some track layout changes to maintain a reversing facility for LU trains. I suspect there may be some bridges in the Harrow area which would require costly rebuilds.

As I suggested in my other post if changeover on the move is acceptable, most economical would be to changeover just south of Wembley Park with new DC north thereof. Similarly, no need to take the conductor rail as far as Great Missenden. Changeover could take place just north of Amersham.

Others would have to comment on any technical impediments to this, but this setup would appear to be the most simple. Having Chiltern trains carry around a set of negative shoe gear is to my mind undesirable and should be avoided, and having S stocks convey AC equipment even more undesirable for a number of reasons.

Agreed. This system could allow standard trains to be used.

Out of interest, does the hybrid DC system carry any disbenefits compared to a pure 4-rail system in this setup? I know 4 rails is preferred in tunnels to avoid corrosion issues, however presumably this isn’t a major issue in the open?

I don't think there are any significant disadvantages on surface sections, as long as the running rails are laid on insulating pads to reduce leakage and insulating inserts on clips and bolts are used, which must be the case already because track circuits have always been used for train detection historically. The new signalling uses digital beacons and odometry for LUL train positioning combined with axle counters, which are used specifically for the fixed blocks planned to be used for the Chiltern trains, so there should be no issues with future TC compatibility with the return current flowing in the rails as well as the centre rail. Thus insulated joints should theoretically be able to be removed widely and there would be no need to employ any complex impedance bond systems. Widespread parallel bonding between the running rails, including between the two running rails of the same track and to parallel tracks should thus be possible resulting in the total traction loop resistance being reduced compared to just using the centre conductor alone or the running rail alone, so performance for all trains should actually be improved. Widespread parallel bonding should also result in greater resilience with many alternative legitimate low resistance paths reducing the risk of faults diverting current into damaging leakage paths.

 

[Bletchleyite]

The correct answer is to modify the shared section to NR compatible 750V DC, convert fully to NR specification 750V DC third-rail (and perhaps a few miles more to suit changeover locations for both options, as bramling suggests) whilst retaining the LUL specification Thales SelTrac CBTC signalling, and fit the new dual-voltage Chiltern EMU fleet with SelTrac CBTC signalling.

And then presumably 25kV from Amersham to Aylesbury? This all makes a lot of sense.

 

[MarkyT]

.The correct answer is to modify the shared section to NR compatible 750V DC, convert fully to NR specification 750V DC third-rail (and perhaps a few miles more to suit changeover locations for both options, as bramling suggests) whilst retaining the LUL specification Thales SelTrac CBTC signalling, and fit the new dual-voltage Chiltern EMU fleet with SelTrac CBTC signalling.

It is the new signalling that makes this all practical, as the existing installation uses track circuits widely, which, as on the Southern, can be made compatible with traction return current going through the running rails clearly, but need masses of heavy complex impedance bonds to do so. With the new system employing digital beacon and odemetry and axle counters, its train detection elements will share no electrical commonality with running rail return current, so all that complexity is avoided , and troublesome insulated block joints can be removed, hopefully making for a much more reliable railway. The current plan for the shared section is to provide conventional colour light signals which the Chiltern trains will use on a fixed block basis, whose occupancy will be determined by axle counters. Train protection will continue to be provided by traditional mechanical train stop arms. I've suggested somewhere else on these forums already that if the Thales system has outputs that can drive this old tech 'underlay' those exact same outputs could switch active ETCS balises, via LEU (lineside encoding unit) interface, in the same positions as the train stops to directly replace them and avoid a brand new 'native ETCS' fleet requiring a complex custom SelTrac emulation, or indeed 1909 vintage trip cocks!

 

[apk55]

Normal third rail equipment is designed so that the power circuit is isolated from ground and the return path only connects to ground via axle brushes. Normaly the link to the axle brushes is made easily disconectable for test purposes (eg flash testing to find an insulation fault) so it would be quite easy to convert to fourth rail operation. However there is the problem of shoe beams under power bogies and also an extra contactor would be required to isolate the center shoes from equipment.

 

[Philip Phlopp]

I've suggested somewhere else on these forums already that if the Thales system has outputs that can drive this old tech 'underlay' those exact same outputs could switch active ETCS balises, via LEU (lineside encoding unit) interface, in the same positions as the train stops to directly replace them and avoid a brand new 'native ETCS' fleet requiring a complex custom SelTrac emulation, or indeed 1909 vintage trip cocks!

Yes - if you can make SelTrac drive ETCS L2 and avoid having to fit the hypothetical Chiltern EMU stock with SelTrac, even better.

 

[MarkyT]

Yes - if you can make SelTrac drive ETCS L2 and avoid having to fit the hypothetical Chiltern EMU stock with SelTrac, even better.

It the active balises simply replaced the trainstops, it would actually be L1 limited supervision in the LUL area, but that can now be supported natively in the latest ETCS baseline, which I assume a notional new fleet would be delivered with. 'True' L2 would require a much more complex direct interface between the Seltrac interlocking level equivalent and ETCS L2 compatible radio block centres covering the area, which would probably be as difficult as a custom STM solution making the ETCS compatible with some limited SelTrac functionality on board!