3rd rail north London

[HSTEd]

According to Wikipedia the skin effect at 50Hz gives an effective depth of about 10mm for copper and a bit more than 1mm for steel. However high voltage AC systems only use the rail to carry return currents for a maximum of a mile or so, before the current is diverted (using transformers again) into a return conductor.

Maybe using your decadent over-engineered Continental 25kV systems, but in glorious BR-land there are still large sections of track using simple rail return without even booster transformers.

After all, who is going to pay for them?

 

[Searchlight]

The Soviet Union, and their Russian successors, are another to have made the change from DC to AC. A substantial amount of the network was electrified at 3,000v DC up to the 1960s, when the policy change was made to 25Kv AC. All around Moscow and St Petersburg, and the busiest trunk routes, were already done at DC, but newer schemes moved to AC. From Moscow to Vladivostok the Trans Siberian western half is DC but its eastern, more recent electrification is AC. It can be difficult to identify in casual photographs which is which, as the loco and emu building industry have tended to use the same bodies and mechanical parts for both systems. The Russian Railways typically change locos at the substantial number of interface points between the two systems rather than use dual voltage locos. The recent high speed Sapsan emus built by Siemens are 3,000v DC on their initial Moscow to St Petersburg route, but now dual voltage units are being delivered for other lines.

Thanks for the info Taunton......I was under the impression that they used dual-system locos......Of course they COULD do if they wanted to.

 

[apk55]

The maximum available power for trains is normally quite limited for DC systems. 750V is really only good for about 2MW loads although parts of SE have stretched this to 4MW by having substations virtually every mile.
With 1.5KV it is possible to have 4MW loads with 3 mile substation spacing. Even so it would be almost impossible to run the current West coast service with a 1.5KV system, and the Dutch have to resort to assisting electric freight trains by diesel locos.

 

[Haydn1971]

Semi related I'd say... Will third rail systems start being replaced with overhead anytime soon ? In particular on heavy use routes like through London Bridge.

 

[mr_jrt]

Semi related I'd say... Will third rail systems start being replaced with overhead anytime soon ? In particular on heavy use routes like through London Bridge.

I doubt it.

The primary conversions will be simpler, isolated sections, not the complex 3rd-rail hellholes that are Clapham Junction, Waterloo and London Bridge.

I expect that if we see anything, it'll be the planned Basingstoke to Southampton line, followed by the Kent lines beyond Faversham and Ashford, to take advantage of the HS1 links.

There was talk of converting the BML south of Gatwick - I'd wager south of East Croydon would make more sense given Uckfield and East Grinstead were proposed for OHLE as well, though I would have though that once OHLE is at Hastings and the BML then converting the whole east coastway would make sense, so who knows what order that would happen.

 

[Class377/5]

Semi related I'd say... Will third rail systems start being replaced with overhead anytime soon ? In particular on heavy use routes like through London Bridge.

The idea would be for DC to be replaced from country towards London so places like Bridge and Clapham Junction would be done last.

There is a plan for wiring the BM from Croydon to Brighton taking advantage of the dual voltage rolling stock already on the line or 'easily' convertable stock. From May 2016, 6* of the 8 tph from Brighton to will be dual voltage stock. Problem is there is no funding for this.

* there will be a few DC only stock runs to London Bridge a day.

 

[Taunton]

Third rail is particularly straightforward to install; I recall seeing it go in on the single track far end of the Stratford to North Woolwich line in the 1980s where it was simplistically put down over a couple of miles in one pass in a weekend. Compared to what is happening on the GWML you need a huge advantage subsequently to overcome the investment. This is what was behind rolling it out across much of the Southern many years ago.

Low voltage DC can also be put straight into the motors; a 12-car 25Kv emu is carrying round three substations under its floor all the time, and it's only in recent years that these equipments have become more straightforward. Gerry Fiennes, in "I tried to Run a Railway", said that whoever suggested changing the quite new 1,500v DC out of Liverpool Street to AC around 1960 should have been "stoned as a false prophet". As he was the GM for the Great Eastern lines at the time, I think we can give him a bit of credit for experience at the sharp end of a changeover.

 

[edwin_m]

Gerry Fiennes, in "I tried to Run a Railway", said that whoever suggested changing the quite new 1,500v DC out of Liverpool Street to AC around 1960 should have been "stoned as a false prophet". As he was the GM for the Great Eastern lines at the time, I think we can give him a bit of credit for experience at the sharp end of a changeover.

That was probably true with the equipment only being a few years old (the trains continued to run, with transformers and rectifiers fitted to feed the original DC equipment). Might be a different story if the equipment is life-expired as we are told much of it is on the Southern.

 

[Class 170101]

Third rail is particularly straightforward to install; I recall seeing it go in on the single track far end of the Stratford to North Woolwich line in the 1980s where it was simplistically put down over a couple of miles in one pass in a weekend. Compared to what is happening on the GWML you need a huge advantage subsequently to overcome the investment. This is what was behind rolling it out across much of the Southern many years ago.

The power draw in DC mode was also quite limited over this section. Dual Voltage was available between Dalston and Channelsea but AC was the preferred method of operation.

 

[Taunton]

The power draw in DC mode was also quite limited over this section. Dual Voltage was available between Dalston and Channelsea but AC was the preferred method of operation.

The 25Kv on Dalston to Stratford was only progressively installed some years after the 3rd rail was put down east of Dalston. The section on from Stratford was always treated as a poor relation, services used to be cut short if there was late running so it was notably unreliable. I don't know where the substation was, it wouldn't surprise me if it was up at Stratford and the whole section was fed through the rail. The single line beyond Custom House would ensure there was only one train at a time at the voltage-drop extremity. Initially it was just former Southern 2-EPB units, 500 hp only and one power coach, later Class 313 with two motor coaches must have stretched things.

On introduction of the DC service in 1983 there was a major electrical problem with stray currents, to the extent that they afflicted the Victoria Line signalling 80 feet below at Highbury. It was a significant part of why the 2-EPBs were replaced after a few years by the dual voltage 313s, which coincided with the 25Kv wires being installed, originally just for freight and subsequently for the North of London Eurostars which never happened. All sorts of additional insulation and earth bonding was tried in the meantime. In the 1990s the North Woolwich to Richmond service was characterised by a considerable number of power changes between third rail and overhead as the wires were progressively extended in separate sections. Someone who is more of an electrical engineer than me can maybe explain why the stray currents were a problem with the DC but not the overhead AC on the same tracks.

 

[AM9]

Third rail is particularly straightforward to install; I recall seeing it go in on the single track far end of the Stratford to North Woolwich line in the 1980s where it was simplistically put down over a couple of miles in one pass in a weekend. Compared to what is happening on the GWML you need a huge advantage subsequently to overcome the investment. This is what was behind rolling it out across much of the Southern many years ago.

Your comparison is hardly relevant. The North Woolwich line is a short urban route and I doubt that any of the EMUs running on it ever got much above 60mph, plus as far as I can recall, they only ran single 3-car units anyway so that is less than 1000A peak draw. The GWML is a 125mph mainline with an intensive express service of 400 tonne trains. Each of these will be about 5300HP which means they could draw over 5000A from a 750VDC supply. So the 'huge investment', only part of which is necessary for OLE, is the only way to modernise the route in both capacity and services.

Low voltage DC can also be put straight into the motors; a 12-car 25Kv emu is carrying round three substations under its floor all the time, and it's only in recent years that these equipments have become more straightforward. Gerry Fiennes, in "I tried to Run a Railway", said that whoever suggested changing the quite new 1,500v DC out of Liverpool Street to AC around 1960 should have been "stoned as a false prophet". As he was the GM for the Great Eastern lines at the time, I think we can give him a bit of credit for experience at the sharp end of a changeover.

The problem with the GEML electrification was confined to one type of train only, i.e. the outer suburban units that had GEC electrical switchgear and transformers. The changeovers from 6.25kV to 25kV did not always work as intended and the 6.25kV transformer windings were hiot with the higher voltage. The results were sometimes spectacular, (I clearly remember passing unit No. 518 (the MBS well on fire) one day. These failure resulted in a major rework programme which must have frustrated attempts to run a service. The conversion of the wiring from DC to AC was comparatively painless as the clearances (and insulators!) for 1500VDC were adequate for 6.25kVac. It wasn't until the whole line was subsequently converted to 25kV that clearances were increased and insulators were replaced. Leaving the inner GE route at 1500VDC wasn't an issue once the decision to electrify out to Colchester and the Essex coast was taken. The fastest Clacton EMUs would draw over 2000A which together with the intensive service of local and outer-suburban EMUs would have required so many more feeds at a massive capital and running cost.
--- old post above --- --- new post below ---

.... Someone who is more of an electrical engineer than me can maybe explain why the stray currents were a problem with the DC but not the overhead AC on the same tracks.

Mainly because for the same power at the wheels, the 750VDC return current was over 33 times greater than the 25kVac one. Not only did this create ground currents and signal immunisation issues, but there was the ever-present threat of electrlytic corrosion which on the NLL would corrode the many plate girder bridges and even penetrate down to subterranean steel tunnels.

 

[ironstone11]

:
Originally Posted by Taunton View Post
Gerry Fiennes, in "I tried to Run a Railway", said that whoever suggested changing the quite new 1,500v DC out of Liverpool Street to AC around 1960 should have been "stoned as a false prophet". As he was the GM for the Great Eastern lines at the time, I think we can give him a bit of credit for experience at the sharp end of a changeover.
That was probably true with the equipment only being a few years old (the trains continued to run, with transformers and rectifiers fitted to feed the original DC equipment). Might be a different story if the equipment is life-expired as we are told much of it is on the Southern.
__________________

(the trains continued to run, with transformers and rectifiers fitted to feed the original DC equipment).

Only just! The problems were enormous leading to insufficient serviceable stock to run a service. The transformers, rectifiers and motors gave trouble. The 25kV/6.25kV tap changing transfomers being particularly troublesome.

As part of the bigger plan, the changeover needed to happen. Colchester to Clacton and Walton had been wired as a test bed at 25kV and the gap between Chelmsford and Colchester was ready to be filled.

In spite of the higher voltage and therefore lower currents, I have always felt the the acceleration of the 1.5kV units was far better than the 25kV units. Perhaps the 25kV units were more 'fragile' and the drivers had to be careful not to break them!

I think the fact remains that for a robust, reliable system, 1.5kV took a lot of beating. Which explains why so much 1.5/3.0kV is still in operation all over the world.

 

[SpacePhoenix]

Is the standards for the catenary the same between 1.5kV, 3kV and 25kV?

 

[Bald Rick]

Is the standards for the catenary the same between 1.5kV, 3kV and 25kV?

Not easy to answer now, as no one has installed 1500v or 3000v DC catenary in this country for quite a while. However, the contact wire for 1500v DC is noticeably thicker than for 25kV AC - there is still plenty of it knocking around on the GEML and presumably in East Manchester. This resulted in more supporting catenary to hold it up - usually in the form of a compound arrangement with an auxiliary (third) wire between the contact and catenary wire. This all added weight, which meant the supporting structures were bigger and / or more closely spaced.

So in construction terms, DC OLE needs to be heftier. Evidence, if you need it, is that in terms of wire in the air, the GE OLE renewal project takes out roughly twice as much copper (by weight) of the old DC kit as it puts back up for a given wire run.

 

[AM9]

Not easy to answer now, as no one has installed 1500v or 3000v DC catenary in this country for quite a while. However, the contact wire for 1500v DC is noticeably thicker than for 25kV AC - there is still plenty of it knocking around on the GEML and presumably in East Manchester. This resulted in more supporting catenary to hold it up - usually in the form of a compound arrangement with an auxiliary (third) wire between the contact and catenary wire. This all added weight, which meant the supporting structures were bigger and / or more closely spaced.

So in construction terms, DC OLE needs to be heftier. Evidence, if you need it, is that in terms of wire in the air, the GE OLE renewal project takes out roughly twice as much copper (by weight) of the old DC kit as it puts back up for a given wire run.

The othere differences involve the impact of higher voltages. Clearly the 25kV OLE needs clearances suitable for the voltage which I believe is now down to about 150mm on cabled sections. Much of that clearance is required because of oscillations and deflection resulting from the passage of pantographs whereas the absolute clearance required for a rigid conductor bar is probably little more than required to prevent flashover. This mechanical cleaqrance is also required for lower voltages so the actual minimum clearance is similar for both systems.
To avoid creepage (surface conduction through airborne contamination and water) it is necessary to increase the distance between ground and high voltage surfaces. This is done with the familiar ribbed insulators which can give over double the surface length for their physical end-to-end length. The ribs also cause rain to fall from them leaving dry sections in between. This is why they are also call 'sheds'. On lower voltage OLE, the insulators are simpler with maybe just one rib, but more substantial construction because of the additional weight of a heavier curerent catenary, as described above.
--- old post above --- --- new post below ---

Only just! The problems were enormous leading to insufficient serviceable stock to run a service. The transformers, rectifiers and motors gave trouble. The 25kV/6.25kV tap changing transfomers being particularly troublesome.

Ironically, the 3-car class 306 ('Shenfield) and 4-car class 307 ('Southend Victoria) trains didn't seem to suffer from the same faults. Their series-parallel connected traction motors meant that the DC bus was 1500V so the old S-P changeover switchgear was left in place and the pantograph/transformer/6.25kV-25kV selection switchgear was all mounted in an adjacent vehicle to provide the bus voltage.

 

[HSTEd]

Is the standards for the catenary the same between 1.5kV, 3kV and 25kV?

Not really - 1.5kV and 3kV systems have been engineered with essentially no insulators at all, simply using parafil ropes and fibreglass in place of the metal components and cables used on AC systems.
This has not been made practical at 25kV 50Hz yet.

 

[MarkyT]

On introduction of the DC service in 1983 there was a major electrical problem with stray currents, to the extent that they afflicted the Victoria Line signalling 80 feet below at Highbury. It was a significant part of why the 2-EPBs were replaced after a few years by the dual voltage 313s, which coincided with the 25Kv wires being installed, originally just for freight and subsequently for the North of London Eurostars which never happened. All sorts of additional insulation and earth bonding was tried in the meantime. In the 1990s the North Woolwich to Richmond service was characterised by a considerable number of power changes between third rail and overhead as the wires were progressively extended in separate sections. Someone who is more of an electrical engineer than me can maybe explain why the stray currents were a problem with the DC but not the overhead AC on the same tracks.

DC systems are designed with the running rail traction return path floating free of earth on insulated pads and with insulated retaining clips. The rails must not be earthed or connected to other surrounding metalwork at any point. This policy ensures the running rails themselves and any strengthening conductors bonded in parallel remain by far the most attractive path for the current to follow to return to the substation, and the method is very effective in reducing stray DC currents which can be so damaging.

For high voltage AC systems, the running rail return cannot float in the same way because accessible voltage can drift very much higher than with lower voltage DC. In such systems, everything metallic nearby must be bonded to the rails and Earth for safety. Whilst that does nothing to discourage stray current, with AC alone it doesn't cause the same corrosion problems as with DC and the current levels at high voltage are very much lower anyway for a given power load.

The two philosophies collide when trying to combine the two electrification systems on the same section of track, and where DC lines run closely parallel to each other. In order to safely manage accessible voltages, the comprehensive bonding of AC must be applied, but this promotes the high DC return current using alternative 'stray' paths. I think much of the NLL would have been affected by this problem. Ideally dual electrified areas should be avoided or kept as small as possible.

 

[HSTEd]

With SMPSs so cheap and available now it might be feasible to work out an analogue of the Booster transformer system for DC.
Force all the current to flow into a return busbar at points spaced every hundred metres or so along the track.

 

[AM9]

With SMPSs so cheap and available now it might be feasible to work out an analogue of the Booster transformer system for DC.
Force all the current to flow into a return busbar at points spaced every hundred metres or so along the track.

That sounds a whole lot more expensive and unreliable than electrifying properly with HVac ole.

 

[Taunton]

Not easy to answer now, as no one has installed 1500v or 3000v DC catenary in this country for quite a while. However, the contact wire for 1500v DC is noticeably thicker than for 25kV AC

I think the last 1500v put in was the Tyneside Metro extension beyond Sunderland, which opened in 2002, so not so long ago. I know they are lighter weight cars but the system is designed to handle 3 in multiple at frequent intervals. The overhead looks decidedly light compared to traditional rail standards.

DC systems are designed with the running rail traction return path floating free of earth on insulated pads and with insulated retaining clips. The rails must not be earthed

I can't see how that can be the case when street tramways, always DC, just run on plain rails laid in the street without these insulating elements.

Ironically, the 3-car class 306 ('Shenfield) and 4-car class 307 ('Southend Victoria) trains didn't seem to suffer from the same faults.

Different designer. The bad faults which led to explosions of the transformer/rectifier set were confined to those by another manufacturer.

 

[AM9]

I think the last 1500v put in was the Tyneside Metro extension beyond Sunderland, which opened in 2002, so not so long ago. I know they are lighter weight cars but the system is designed to handle 3 in multiple at frequent intervals. The overhead looks decidedly light compared to traditional rail standards.

I can't see how that can be the case when street tramways, always DC, just run on plain rails laid in the street without these insulating elements.

I don't know what current the Tyneside Metro traction system drew but I suspect that it would be less than 500A so the demand from the contact wire and return via the track would be lass than 300A per 2-car unit.

Different designer. The bad faults which led to explosions of the transformer/rectifier set were confined to those by another manufacturer.

Yes I know, (my one-time employer GEC). My point was about Gerry Fiennes problems, blaming ac electrification in general, when it was just a poor implementation on some (too many for comfort) of the new ac trains. The fact that the original DC stock modifications were generally trouble-free just shows that even in the early UK days, ac was not really a problem. Interestingly, the 307s survived longer than most of the other MKI EMUs despite their complex traction/supply system spread over 2 cars.

 

[Searchlight]

Third rail is particularly straightforward to install; I recall seeing it go in on the single track far end of the Stratford to North Woolwich line in the 1980s where it was simplistically put down over a couple of miles in one pass in a weekend. Compared to what is happening on the GWML you need a huge advantage subsequently to overcome the investment. This is what was behind rolling it out across much of the Southern many years ago.

Low voltage DC can also be put straight into the motors; a 12-car 25Kv emu is carrying round three substations under its floor all the time, and it's only in recent years that these equipments have become more straightforward. Gerry Fiennes, in "I tried to Run a Railway", said that whoever suggested changing the quite new 1,500v DC out of Liverpool Street to AC around 1960 should have been "stoned as a false prophet". As he was the GM for the Great Eastern lines at the time, I think we can give him a bit of credit for experience at the sharp end of a changeover.

Yes, they should have gone for 3Kv DC and pushed on with it! Thus
reducing substations by half........3Kv is not comparable with 1.5Kv......

 

[HSTEd]

That sounds a whole lot more expensive and unreliable than electrifying properly with HVac ole.

Not particularily - the SMPS would onyl be dealing with voltages in the range of 30-40V above or below ground.
That is the kind of stuff you order from Alibaba by the truckload.

 

[Taunton]

Interestingly, the 307s survived longer than most of the other MKI EMUs despite their complex traction/supply system spread over 2 cars.

I think because they were early 1950s build and without the asbestos body insulation that came into vogue just when the other units were built.

Regarding recent experience with 1500v overhead, the Hong Kong Metro with their huge, heavy and frequent trains use this, with the system also being extended well into this century. There was much UK input (GEC) into the design and development of the electrical system, and they had a particularly close association with the Tyneside system. Several of the key management and technical staff went between the two, and the first Hong Kong cars were tested on the Newcastle test track. The vehicles for both were built by Metro Cammell.

 

[apk55]

With SMPSs so cheap and available now it might be feasible to work out an analogue of the Booster transformer system for DC.
Force all the current to flow into a return busbar at points spaced every hundred metres or so along the track.

You would be looking at power supplies of several hundred KW. They might only have to deal with 10 to 30V BUT the full traction current of up to almost 7KA.

 

[MarkyT]

I can't see how that can be the case when street tramways, always DC, just run on plain rails laid in the street without these insulating elements.

Systems have changed over history. First generation electric tramways in UK, like contemporary systems abroad, had electrical bonding to underground metal utility pipes and other metal work in the belief that that would stop current travelling through the earth to reach those assets. In fact that just equipped the rails with huge earth electrodes to further promote the phenomenon and could sometimes cause large currents to flow through those third party assets if there was a discontinuity in the rail bonding. Early modern UK tramways used collection meshes beneath the rails to attempt to 'mop up' any stray current. These also seem to have been counterproductive whether connected to the substation negative or just left floating. Guidance today recommends the simpler, cheaper, insulated floating rail system with frequent cross-bonding between parallel rails and tracks, and strengthening conductors where necessary. Rail fixing solutions that support this are available and are used for the latest new tramways, system refurbishments and extensions.

For an interesting bedtime read I suggest the following ORR publication:

http://orr.gov.uk/__data/assets/pdf_file/0012/5070/TTGN3.pdf

DESIGN STANDARDS STRAY CURRENT MANAGEMENT
Tramway Technical Guidance Note 3